Redis/redis-For-Windows
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 3 Wiki Activity

Compare commits

base: Redis/redis-For-Windows:v2.0.0-rc1
Branches Tags
Redis/redis-For-Windows:develop Redis/redis-For-Windows:win-5.0 Redis/redis-For-Windows:win-6.2
Redis/redis-For-Windows:5.0.14.1 Redis/redis-For-Windows:v5.0.14.1 Redis/redis-For-Windows:v5.0.14 Redis/redis-For-Windows:v5.0.10 Redis/redis-For-Windows:v5.0.9 Redis/redis-For-Windows:v4.0.14.2 Redis/redis-For-Windows:v4.0.14.1 Redis/redis-For-Windows:v4.0.14 Redis/redis-For-Windows:v4.0.2.3-alpha Redis/redis-For-Windows:v4.0.2.2-alpha Redis/redis-For-Windows:v4.0.2.1-alpha Redis/redis-For-Windows:v4.0.2-alpha Redis/redis-For-Windows:win-3.2.100 Redis/redis-For-Windows:win-3.0.504 Redis/redis-For-Windows:win-3.0.503 Redis/redis-For-Windows:win-2.8.2402 Redis/redis-For-Windows:win-3.0.502 Redis/redis-For-Windows:win-2.8.2401 Redis/redis-For-Windows:3.2.1 Redis/redis-For-Windows:win-3.2.0000-preview Redis/redis-For-Windows:3.2.0 Redis/redis-For-Windows:3.2.0-rc3 Redis/redis-For-Windows:3.2.0-rc2 Redis/redis-For-Windows:win-2.8.2400 Redis/redis-For-Windows:win-3.0.501 Redis/redis-For-Windows:3.2-rc1 Redis/redis-For-Windows:3.0.6 Redis/redis-For-Windows:2.8.24 Redis/redis-For-Windows:win-3.0.500 Redis/redis-For-Windows:win-3.0.500-rc2 Redis/redis-For-Windows:win-3.0.500-rc1 Redis/redis-For-Windows:win-2.8.2104 Redis/redis-For-Windows:3.0.5 Redis/redis-For-Windows:2.8.23 Redis/redis-For-Windows:win-3.0.300-beta1 Redis/redis-For-Windows:win-2.8.2103 Redis/redis-For-Windows:2.8.22 Redis/redis-For-Windows:3.0.4 Redis/redis-For-Windows:win-3.0.300-alpha3 Redis/redis-For-Windows:win-2.8.2102 Redis/redis-For-Windows:win-3.0.300-alpha2 Redis/redis-For-Windows:win-3.0.100-alpha1 Redis/redis-For-Windows:3.0.3 Redis/redis-For-Windows:win-2.8.2101 Redis/redis-For-Windows:win-2.8.21 Redis/redis-For-Windows:3.0.2 Redis/redis-For-Windows:2.8.21 Redis/redis-For-Windows:2.8.20 Redis/redis-For-Windows:3.0.1 Redis/redis-For-Windows:win-2.8.19.1 Redis/redis-For-Windows:3.0.0 Redis/redis-For-Windows:3.0.0-rc6 Redis/redis-For-Windows:3.0.0-rc5 Redis/redis-For-Windows:win-2.8.19-rc1 Redis/redis-For-Windows:win-2.8.19 Redis/redis-For-Windows:3.0.0-rc4 Redis/redis-For-Windows:win-2.8.19-pre2 Redis/redis-For-Windows:win-2.8.17.4 Redis/redis-For-Windows:3.0.0-rc3 Redis/redis-For-Windows:3.0.0-rc2 Redis/redis-For-Windows:win-2.8.17.3 Redis/redis-For-Windows:win-2.8.19-pre1 Redis/redis-For-Windows:win-2.8.17.2 Redis/redis-For-Windows:win-2.8.17.1 Redis/redis-For-Windows:2.8.19 Redis/redis-For-Windows:2.8.18 Redis/redis-For-Windows:3.0.0-rc1 Redis/redis-For-Windows:2.8.17 Redis/redis-For-Windows:2.8.16 Redis/redis-For-Windows:2.8.15 Redis/redis-For-Windows:win2.8.12 Redis/redis-For-Windows:2.8.14 Redis/redis-For-Windows:3.0.0-beta8 Redis/redis-For-Windows:2.8.13 Redis/redis-For-Windows:3.0.0-beta7 Redis/redis-For-Windows:win_2.8.9 Redis/redis-For-Windows:2.8.12 Redis/redis-For-Windows:2.8.11 Redis/redis-For-Windows:3.0.0-beta6 Redis/redis-For-Windows:2.8.10 Redis/redis-For-Windows:3.0.0-beta5 Redis/redis-For-Windows:3.0.0-beta4 Redis/redis-For-Windows:2.6.14_msopentech Redis/redis-For-Windows:2.8.4_msopentech Redis/redis-For-Windows:msopentech_2.8.4 Redis/redis-For-Windows:3.0.0-beta3 Redis/redis-For-Windows:2.8.9 Redis/redis-For-Windows:2.8.8 Redis/redis-For-Windows:3.0.0-beta2 Redis/redis-For-Windows:2.8.7 Redis/redis-For-Windows:2.8.6 Redis/redis-For-Windows:3.0.0-beta1 Redis/redis-For-Windows:2.8.5 Redis/redis-For-Windows:2.8.4 Redis/redis-For-Windows:2.8.3 Redis/redis-For-Windows:2.8.2 Redis/redis-For-Windows:2.8.1 Redis/redis-For-Windows:2.8.0 Redis/redis-For-Windows:2.8.0-rc6 Redis/redis-For-Windows:2.8.0-rc5 Redis/redis-For-Windows:2.8.0-rc4 Redis/redis-For-Windows:2.8.0-rc3 Redis/redis-For-Windows:2.8.0-rc2 Redis/redis-For-Windows:2.8.0-rc1 Redis/redis-For-Windows:2.6.14 Redis/redis-For-Windows:Win2.6.8 Redis/redis-For-Windows:2.6.13 Redis/redis-For-Windows:2.6.12 Redis/redis-For-Windows:2.6.11 Redis/redis-For-Windows:2.6.10 Redis/redis-For-Windows:2.6.9 Redis/redis-For-Windows:2.6.8 Redis/redis-For-Windows:2.6.7 Redis/redis-For-Windows:2.6.6 Redis/redis-For-Windows:2.6.5 Redis/redis-For-Windows:2.6.4 Redis/redis-For-Windows:2.6.3 Redis/redis-For-Windows:2.6.2 Redis/redis-For-Windows:2.6.1 Redis/redis-For-Windows:2.6.0 Redis/redis-For-Windows:2.6.0-rc8 Redis/redis-For-Windows:2.6.0-rc7 Redis/redis-For-Windows:2.6.0-rc6 Redis/redis-For-Windows:2.6.0-rc5 Redis/redis-For-Windows:2.6.0-rc4 Redis/redis-For-Windows:2.6.0-rc3 Redis/redis-For-Windows:2.6.0-rc2 Redis/redis-For-Windows:2.6.0-rc1 Redis/redis-For-Windows:2.4.11 Redis/redis-For-Windows:2.4.10 Redis/redis-For-Windows:2.4.9 Redis/redis-For-Windows:2.4.8 Redis/redis-For-Windows:2.4.6_win32 Redis/redis-For-Windows:2.4.7 Redis/redis-For-Windows:2.4.6 Redis/redis-For-Windows:2.4.5 Redis/redis-For-Windows:2.4.4 Redis/redis-For-Windows:2.4.3 Redis/redis-For-Windows:2.4.2 Redis/redis-For-Windows:2.4.1 Redis/redis-For-Windows:2.4.0 Redis/redis-For-Windows:2.4.0-rc8 Redis/redis-For-Windows:2.4.0-rc7 Redis/redis-For-Windows:2.4.0-rc6 Redis/redis-For-Windows:2.4.0-rc5 Redis/redis-For-Windows:2.4.0-rc4 Redis/redis-For-Windows:2.4.0-rc3 Redis/redis-For-Windows:with-deprecated-diskstore Redis/redis-For-Windows:2.4.0-rc2 Redis/redis-For-Windows:2.4.0-rc1 Redis/redis-For-Windows:2.2.4 Redis/redis-For-Windows:2.2.3 Redis/redis-For-Windows:3.0-alpha0 Redis/redis-For-Windows:2.2.2 Redis/redis-For-Windows:2.2.1 Redis/redis-For-Windows:2.2.0 Redis/redis-For-Windows:2.2.0-rc4 Redis/redis-For-Windows:2.3-alpha0 Redis/redis-For-Windows:2.2.0-rc3 Redis/redis-For-Windows:2.2.0-rc2 Redis/redis-For-Windows:2.2.0-rc1 Redis/redis-For-Windows:2.2-alpha6 Redis/redis-For-Windows:2.2-alpha5 Redis/redis-For-Windows:2.2-alpha4 Redis/redis-For-Windows:2.2-alpha3 Redis/redis-For-Windows:2.2-alpha2 Redis/redis-For-Windows:2.2-alpha1 Redis/redis-For-Windows:2.2-alpha0 Redis/redis-For-Windows:v2.1.1-watch Redis/redis-For-Windows:v2.0.0-rc1 Redis/redis-For-Windows:v1.3.12 Redis/redis-For-Windows:v1.3.11 Redis/redis-For-Windows:v1.3.10 Redis/redis-For-Windows:v1.3.9 Redis/redis-For-Windows:v1.3.8 Redis/redis-For-Windows:v1.3.7 Redis/redis-For-Windows:1.3.6 Redis/redis-For-Windows:vm-playpen
..
compare: Redis/redis-For-Windows:2.6.6
Branches Tags
Redis/redis-For-Windows:win-5.0 Redis/redis-For-Windows:develop Redis/redis-For-Windows:win-6.2
Redis/redis-For-Windows:5.0.14.1 Redis/redis-For-Windows:v5.0.14.1 Redis/redis-For-Windows:v5.0.14 Redis/redis-For-Windows:v5.0.10 Redis/redis-For-Windows:v5.0.9 Redis/redis-For-Windows:v4.0.14.2 Redis/redis-For-Windows:v4.0.14.1 Redis/redis-For-Windows:v4.0.14 Redis/redis-For-Windows:v4.0.2.3-alpha Redis/redis-For-Windows:v4.0.2.2-alpha Redis/redis-For-Windows:v4.0.2.1-alpha Redis/redis-For-Windows:v4.0.2-alpha Redis/redis-For-Windows:win-3.2.100 Redis/redis-For-Windows:win-3.0.504 Redis/redis-For-Windows:win-3.0.503 Redis/redis-For-Windows:win-2.8.2402 Redis/redis-For-Windows:win-3.0.502 Redis/redis-For-Windows:win-2.8.2401 Redis/redis-For-Windows:3.2.1 Redis/redis-For-Windows:win-3.2.0000-preview Redis/redis-For-Windows:3.2.0 Redis/redis-For-Windows:3.2.0-rc3 Redis/redis-For-Windows:3.2.0-rc2 Redis/redis-For-Windows:win-2.8.2400 Redis/redis-For-Windows:win-3.0.501 Redis/redis-For-Windows:3.2-rc1 Redis/redis-For-Windows:3.0.6 Redis/redis-For-Windows:2.8.24 Redis/redis-For-Windows:win-3.0.500 Redis/redis-For-Windows:win-3.0.500-rc2 Redis/redis-For-Windows:win-3.0.500-rc1 Redis/redis-For-Windows:win-2.8.2104 Redis/redis-For-Windows:3.0.5 Redis/redis-For-Windows:2.8.23 Redis/redis-For-Windows:win-3.0.300-beta1 Redis/redis-For-Windows:win-2.8.2103 Redis/redis-For-Windows:2.8.22 Redis/redis-For-Windows:3.0.4 Redis/redis-For-Windows:win-3.0.300-alpha3 Redis/redis-For-Windows:win-2.8.2102 Redis/redis-For-Windows:win-3.0.300-alpha2 Redis/redis-For-Windows:win-3.0.100-alpha1 Redis/redis-For-Windows:3.0.3 Redis/redis-For-Windows:win-2.8.2101 Redis/redis-For-Windows:win-2.8.21 Redis/redis-For-Windows:3.0.2 Redis/redis-For-Windows:2.8.21 Redis/redis-For-Windows:2.8.20 Redis/redis-For-Windows:3.0.1 Redis/redis-For-Windows:win-2.8.19.1 Redis/redis-For-Windows:3.0.0 Redis/redis-For-Windows:3.0.0-rc6 Redis/redis-For-Windows:3.0.0-rc5 Redis/redis-For-Windows:win-2.8.19-rc1 Redis/redis-For-Windows:win-2.8.19 Redis/redis-For-Windows:3.0.0-rc4 Redis/redis-For-Windows:win-2.8.19-pre2 Redis/redis-For-Windows:win-2.8.17.4 Redis/redis-For-Windows:3.0.0-rc3 Redis/redis-For-Windows:3.0.0-rc2 Redis/redis-For-Windows:win-2.8.17.3 Redis/redis-For-Windows:win-2.8.19-pre1 Redis/redis-For-Windows:win-2.8.17.2 Redis/redis-For-Windows:win-2.8.17.1 Redis/redis-For-Windows:2.8.19 Redis/redis-For-Windows:2.8.18 Redis/redis-For-Windows:3.0.0-rc1 Redis/redis-For-Windows:2.8.17 Redis/redis-For-Windows:2.8.16 Redis/redis-For-Windows:2.8.15 Redis/redis-For-Windows:win2.8.12 Redis/redis-For-Windows:2.8.14 Redis/redis-For-Windows:3.0.0-beta8 Redis/redis-For-Windows:2.8.13 Redis/redis-For-Windows:3.0.0-beta7 Redis/redis-For-Windows:win_2.8.9 Redis/redis-For-Windows:2.8.12 Redis/redis-For-Windows:2.8.11 Redis/redis-For-Windows:3.0.0-beta6 Redis/redis-For-Windows:2.8.10 Redis/redis-For-Windows:3.0.0-beta5 Redis/redis-For-Windows:3.0.0-beta4 Redis/redis-For-Windows:2.6.14_msopentech Redis/redis-For-Windows:2.8.4_msopentech Redis/redis-For-Windows:msopentech_2.8.4 Redis/redis-For-Windows:3.0.0-beta3 Redis/redis-For-Windows:2.8.9 Redis/redis-For-Windows:2.8.8 Redis/redis-For-Windows:3.0.0-beta2 Redis/redis-For-Windows:2.8.7 Redis/redis-For-Windows:2.8.6 Redis/redis-For-Windows:3.0.0-beta1 Redis/redis-For-Windows:2.8.5 Redis/redis-For-Windows:2.8.4 Redis/redis-For-Windows:2.8.3 Redis/redis-For-Windows:2.8.2 Redis/redis-For-Windows:2.8.1 Redis/redis-For-Windows:2.8.0 Redis/redis-For-Windows:2.8.0-rc6 Redis/redis-For-Windows:2.8.0-rc5 Redis/redis-For-Windows:2.8.0-rc4 Redis/redis-For-Windows:2.8.0-rc3 Redis/redis-For-Windows:2.8.0-rc2 Redis/redis-For-Windows:2.8.0-rc1 Redis/redis-For-Windows:2.6.14 Redis/redis-For-Windows:Win2.6.8 Redis/redis-For-Windows:2.6.13 Redis/redis-For-Windows:2.6.12 Redis/redis-For-Windows:2.6.11 Redis/redis-For-Windows:2.6.10 Redis/redis-For-Windows:2.6.9 Redis/redis-For-Windows:2.6.8 Redis/redis-For-Windows:2.6.7 Redis/redis-For-Windows:2.6.6 Redis/redis-For-Windows:2.6.5 Redis/redis-For-Windows:2.6.4 Redis/redis-For-Windows:2.6.3 Redis/redis-For-Windows:2.6.2 Redis/redis-For-Windows:2.6.1 Redis/redis-For-Windows:2.6.0 Redis/redis-For-Windows:2.6.0-rc8 Redis/redis-For-Windows:2.6.0-rc7 Redis/redis-For-Windows:2.6.0-rc6 Redis/redis-For-Windows:2.6.0-rc5 Redis/redis-For-Windows:2.6.0-rc4 Redis/redis-For-Windows:2.6.0-rc3 Redis/redis-For-Windows:2.6.0-rc2 Redis/redis-For-Windows:2.6.0-rc1 Redis/redis-For-Windows:2.4.11 Redis/redis-For-Windows:2.4.10 Redis/redis-For-Windows:2.4.9 Redis/redis-For-Windows:2.4.8 Redis/redis-For-Windows:2.4.6_win32 Redis/redis-For-Windows:2.4.7 Redis/redis-For-Windows:2.4.6 Redis/redis-For-Windows:2.4.5 Redis/redis-For-Windows:2.4.4 Redis/redis-For-Windows:2.4.3 Redis/redis-For-Windows:2.4.2 Redis/redis-For-Windows:2.4.1 Redis/redis-For-Windows:2.4.0 Redis/redis-For-Windows:2.4.0-rc8 Redis/redis-For-Windows:2.4.0-rc7 Redis/redis-For-Windows:2.4.0-rc6 Redis/redis-For-Windows:2.4.0-rc5 Redis/redis-For-Windows:2.4.0-rc4 Redis/redis-For-Windows:2.4.0-rc3 Redis/redis-For-Windows:with-deprecated-diskstore Redis/redis-For-Windows:2.4.0-rc2 Redis/redis-For-Windows:2.4.0-rc1 Redis/redis-For-Windows:2.2.4 Redis/redis-For-Windows:2.2.3 Redis/redis-For-Windows:3.0-alpha0 Redis/redis-For-Windows:2.2.2 Redis/redis-For-Windows:2.2.1 Redis/redis-For-Windows:2.2.0 Redis/redis-For-Windows:2.2.0-rc4 Redis/redis-For-Windows:2.3-alpha0 Redis/redis-For-Windows:2.2.0-rc3 Redis/redis-For-Windows:2.2.0-rc2 Redis/redis-For-Windows:2.2.0-rc1 Redis/redis-For-Windows:2.2-alpha6 Redis/redis-For-Windows:2.2-alpha5 Redis/redis-For-Windows:2.2-alpha4 Redis/redis-For-Windows:2.2-alpha3 Redis/redis-For-Windows:2.2-alpha2 Redis/redis-For-Windows:2.2-alpha1 Redis/redis-For-Windows:2.2-alpha0 Redis/redis-For-Windows:v2.1.1-watch Redis/redis-For-Windows:v2.0.0-rc1 Redis/redis-For-Windows:v1.3.12 Redis/redis-For-Windows:v1.3.11 Redis/redis-For-Windows:v1.3.10 Redis/redis-For-Windows:v1.3.9 Redis/redis-For-Windows:v1.3.8 Redis/redis-For-Windows:v1.3.7 Redis/redis-For-Windows:1.3.6 Redis/redis-For-Windows:vm-playpen
2007 Commits
v2.0.0-rc1 .. 2.6.6
Author SHA1 Message Date
antirez b41f4e5a48 Redis 2.6.6 2012-11-28 18:53:09 +01:00
antirez 21b26915ce Jemalloc updated to version 3.2.0. 2012-11-28 18:42:41 +01:00
antirez 3ccfb5a4d6 Redis 2.6.5 2012-11-22 17:03:27 +01:00
antirez 9120275dc9 EVALSHA is now case insensitive. 
EVALSHA used to crash if the SHA1 was not lowercase (Issue #783).
Fixed using a case insensitive dictionary type for the sha -> script
map used for replication of scripts.
 2012-11-22 15:51:01 +01:00
antirez de00a5a092 Fix integer overflow in zunionInterGenericCommand(). 
This fixes issue #761.
 2012-11-22 15:28:45 +01:00
antirez 2b45ef020c Test: MULTI state is cleared after EXECABORT error. 2012-11-22 10:36:36 +01:00
antirez 52bd3d8a39 Test: make sure EXEC fails after previous transaction errors. 2012-11-22 10:36:32 +01:00
antirez 37d978d9b4 Test: MULTI/EXEC tests moved into multi.tcl. 2012-11-22 10:36:28 +01:00
antirez 41f0f927c9 Safer handling of MULTI/EXEC on errors. 
After the transcation starts with a MULIT, the previous behavior was to
return an error on problems such as maxmemory limit reached. But still
to execute the transaction with the subset of queued commands on EXEC.

While it is true that the client was able to check for errors
distinguish QUEUED by an error reply, MULTI/EXEC in most client
implementations uses pipelining for speed, so all the commands and EXEC
are sent without caring about replies.

With this change:

1) EXEC fails if at least one command was not queued because of an
error. The EXECABORT error is used.
2) A generic error is always reported on EXEC.
3) The client DISCARDs the MULTI state after a failed EXEC, otherwise
pipelining multiple transactions would be basically impossible:
After a failed EXEC the next transaction would be simply queued as
the tail of the previous transaction.
 2012-11-22 10:36:20 +01:00
antirez 5ab4151d7f Children creating AOF or RDB files now report memory used by COW. 
Finally Redis is able to report the amount of memory used by
copy-on-write while saving an RDB or writing an AOF file in background.

Note that this information is currently only logged (at NOTICE level)
and not shown in INFO because this is less trivial (but surely doable
with some minor form of interprocess communication).

The reason we can't capture this information on the parent before we
call wait3() is that the Linux kernel will release the child memory
ASAP, and only retain the minimal state for the process that is useful
to report the child termination to the parent.

The COW size is obtained by summing all the Private_Dirty fields found
in the "smap" file inside the proc filesystem for the process.

All this is Linux specific and is not available on other systems.
 2012-11-20 12:32:19 +01:00
antirez e95ca66320 zmalloc_get_private_dirty() function added (Linux only). 
For non Linux systmes it just returns 0.

This function is useful to estimate copy-on-write because of childs
saving stuff on disk.
 2012-11-20 12:32:14 +01:00
antirez 681c191006 zmalloc: kill unused __size parameter in update_zmalloc_stat_alloc() macro. 2012-11-14 13:07:48 +01:00
antirez d85a09dfc6 MIGRATE: fix default timeout to 1000 milliseconds. 
When a timeout <= 0 is provided we set a default timeout of 1 second.
It was set to 1 millisecond for an error resulting from a recent change.
 2012-11-12 18:58:23 +01:00
antirez 26a48504cb MIGRATE timeout should be in milliseconds. 
While it is documented that the MIGRATE timeout is in milliseconds, it
was in seconds instead. This commit fixes the problem.
 2012-11-12 18:57:55 +01:00
antirez ae3aeca828 MIGRATE: fix fd leak due to missing close on error. 2012-11-12 18:56:29 +01:00
antirez 5a8234d2c9 Copyright date fixed in COPYING file. 2012-11-08 19:14:29 +01:00
antirez 2861cd84f9 Redis 2.6.4 2012-11-08 18:47:58 +01:00
antirez 8e30092f06 Make clear that contributing code to the Redis project means to release it under the terms of the BSD license. 2012-11-08 18:43:37 +01:00
antirez d288ee655f BSD license added to every C source and header file. 2012-11-08 18:33:13 +01:00
antirez 89514f80e9 Redis 2.6.3 2012-11-06 22:07:51 +01:00
Runzhen Wang 3f006679b1 fix a typo in redis.h line 595 comment 2012-11-02 12:12:01 +01:00
antirez 36f026a3a0 More robust handling of AOF rewrite child. 
After the wait3() syscall we used to do something like that:

    if (pid == server.rdb_child_pid) {
        backgroundSaveDoneHandler(exitcode,bysignal);
    } else {
        ....
    }

So the AOF rewrite was handled in the else branch without actually
checking if the pid really matches. This commit makes the check explicit
and logs at WARNING level if the pid returned by wait3() does not match
neither the RDB or AOF rewrite child.
 2012-11-01 22:41:57 +01:00
Salvatore Sanfilippo cc017c5839 Merge pull request #743 from Cofyc/fixtypo 
fix typo in comments (redis.c, networking.c)
 2012-11-01 14:26:01 -07:00
antirez b6ffa85f11 Unix socket clients properly displayed in MONITOR and CLIENT LIST. 
This also fixes issue #745.
 2012-11-01 22:12:45 +01:00
antirez 4917a6a8b3 32 bit build fixed on Linux. 
It failed because of the way jemalloc was compiled (without passing the
right flags to make, but just to configure). Now the same set of flags
are also passed to the make command, fixing the issue.

This fixes issue #744
 2012-11-01 15:37:38 +01:00
Yecheng Fu ed44a74e72 fix typo in comments (redis.c, networking.c) 2012-11-01 18:14:55 +08:00
YAMAMOTO Takashi 47ff443b53 fix a typo in a comment 2012-10-31 09:29:56 +01:00
antirez 4d9bd53530 Invert two sides of if expression in SET to avoid a lookup. 
Because of the short circuit behavior of && inverting the two sides of
the if expression avoids an hash table lookup if the non-EX variant of
SET is called.

Thanks to Weibin Yao (@yaoweibin on github) for spotting this.
 2012-10-31 09:27:48 +01:00
antirez 5bf0997ff8 No longer used macro rdbIsOpcode() removed. 2012-10-30 19:10:56 +01:00
antirez ec0f483de2 help.h update (adds bitop, bitcount, evalsha...) 2012-10-30 18:57:28 +01:00
antirez 0dd6241314 Redis 2.6.2 2012-10-26 16:20:24 +02:00
antirez 3b1a41b172 Ctrl+w support in linenoise. 2012-10-26 16:06:36 +02:00
antirez 9b3b110004 Marginally more robust glibc version test for sync_file_range detection. 2012-10-26 15:55:10 +02:00
charsyam b25b0dc5f4 patch config.h for sync_file_range 2012-10-26 09:30:35 +02:00
antirez 2b5fc529a4 Redis 2.6.1 2012-10-25 22:07:27 +02:00
antirez be407c015c Fix compilation on Linux kernels or glibc versions lacking sync_file_range(). 
This fixes issue #667.

Many thanks to Didier Spezia for the fix.
 2012-10-25 22:01:07 +02:00
antirez 1eb9145089 Update memory peak stats while loading RDB / AOF. 2012-10-24 12:21:34 +02:00
antirez 5eec376c2f Redis 2.6.0 2012-10-22 23:27:18 +02:00
antirez 99d7dbe669 A filed called slave_read_only added in INFO output. 
This was an important information missing from the INFO output in the
replication section.

It obviously reflects if the slave is read only or not.
 2012-10-22 19:22:48 +02:00
Greg Hurrell a61705dd95 Fix (cosmetic) typos in dict.h 2012-10-22 11:56:06 +02:00
Schuster 16144589f2 redis-check-dump now understands dumps produced by Redis 2.6 
(Commit message from @antirez as it was missign in the original commits,
also the patch was modified a bit to still work with 2.4 dumps and to
avoid if expressions that are always true due to checked types range)

This commit changes redis-check-dump to account for new encodings and
for the new MSTIME expire format. It also refactors the test for valid
type into a function.

The code is still compatible with Redis 2.4 generated dumps.

This fixes issue #709.
 2012-10-22 11:53:43 +02:00
antirez a25b25f4ef Default memory limit for 32bit instanced moved from 3.5 GB to 3 GB. 
In some system, notably osx, the 3.5 GB limit was too far and not able
to prevent a crash for out of memory. The 3 GB limit works better and it
is still a lot of memory within a 4 GB theorical limit so it's not going
to bore anyone :-)

This fixes issue #711
 2012-10-22 10:45:55 +02:00
antirez ab55180883 Differentiate SCRIPT KILL error replies. 
When calling SCRIPT KILL currently you can get two errors:

* No script in timeout (busy) state.
* The script already performed a write.

It is useful to be able to distinguish the two errors, but right now both
start with "ERR" prefix, so string matching (that is fragile) must be used.

This commit introduces two different prefixes.

-NOTBUSY and -UNKILLABLE respectively to reply with an error when no
script is busy at the moment, and when the script already executed a
write operation and can not be killed.
 2012-10-22 10:31:46 +02:00
NanXiao a03c32702b Update src/redis-benchmark.c 
The code of current implementation:

if (c->pending == 0) clientDone(c);
In clientDone function, the c's memory has been freed, then the loop will continue: while(c->pending). The memory of c has been freed now, so c->pending is invalid (c is an invalid pointer now), and this will cause memory dump in some platforams(eg: Solaris).

So I think the code should be modified as:
if (c->pending == 0)
{
clientDone(c);
break;
}
and this will not lead to while(c->pending).
 2012-10-18 11:05:47 +02:00
antirez 2164523244 Fix MULTI / EXEC rendering in MONITOR output. 
Before of this commit it used to be like this:

MULTI
EXEC
... actual commands of the transaction ...

Because after all that is the natural order of things. Transaction
commands are queued and executed *only after* EXEC is called.

However this makes debugging with MONITOR a mess, so the code was
modified to provide a coherent output.

What happens is that MULTI is rendered in the MONITOR output as far as
possible, instead EXEC is propagated only after the transaction is
executed, or even in the case it fails because of WATCH, so in this case
you'll simply see:

MULTI
EXEC

An empty transaction.
 2012-10-16 17:41:39 +02:00
antirez c3ff470889 Merge remote-tracking branch 'origin/2.6' into 2.6 2012-10-11 18:36:18 +02:00
antirez 0e25c0ccf4 Allow AUTH when Redis is busy because of timedout Lua script. 
If the server is password protected we need to accept AUTH when there is
a server busy (-BUSY) condition, otherwise it will be impossible to send
SHUTDOWN NOSAVE or SCRIPT KILL.

This fixes issue #708.
 2012-10-11 18:35:52 +02:00
antirez f4eb4b3352 CONTRIBUTING file updated. 2012-10-06 12:04:38 +02:00
antirez 500dddc264 Redis 2.6.0 RC8 (2.5.14) 2012-10-05 19:08:55 +02:00
Salvatore Sanfilippo a6305f13d5 Merge pull request #544 from dvirsky/2.6 
fixed install script to rewrite the default config
 2012-10-05 03:30:58 -07:00
antirez 99c3338c23 Hash function switched to murmurhash2. 
The previously used hash function, djbhash, is not secure against
collision attacks even when the seed is randomized as there are simple
ways to find seed-independent collisions.

The new hash function appears to be safe (or much harder to exploit at
least) in this case, and has better distribution.

Better distribution does not always means that's better. For instance in
a fast benchmark with "DEBUG POPULATE 1000000" I obtained the following
results:

    1.6 seconds with djbhash
    2.0 seconds with murmurhash2

This is due to the fact that djbhash will hash objects that follow the
pattern `prefix:<id>` and where the id is numerically near, to near
buckets. This improves the locality.

However in other access patterns with keys that have no relation
murmurhash2 has some (apparently minimal) speed advantage.

On the other hand a better distribution should significantly
improve the quality of the distribution of elements returned with
dictGetRandomKey() that is used in SPOP, SRANDMEMBER, RANDOMKEY, and
other commands.

Everything considered, and under the suspect that this commit fixes a
security issue in Redis, we are switching to the new hash function.
If some serious speed regression will be found in the future we'll be able
to step back easiliy.

This commit fixes issue #663.
 2012-10-05 11:16:40 +02:00
antirez 05e06e1543 Warn when configured maxmemory value seems odd. 
This commit warns the user with a log at "warning" level if:

1) After the server startup the maxmemory limit was found to be < 1MB.
2) After a CONFIG SET command modifying the maxmemory setting the limit
is set to a value that is smaller than the currently used memory.

The behaviour of the Redis server is unmodified, and this wil not make
the CONFIG SET command or a wrong configuration in redis.conf less
likely to create problems, but at least this will make aware most users
about a possbile error they committed without resorting to external
help.

However no warning is issued if, as a result of loading the AOF or RDB
file, we are very near the maxmemory setting, or key eviction will be
needed in order to go under the specified maxmemory setting. The reason
is that in servers configured as a cache with an aggressive
maxmemory-policy most of the times restarting the server will cause this
condition to happen if persistence is not switched off.

This fixes issue #429.
 2012-10-05 10:56:35 +02:00
antirez e2f2dab3ef Include time.h in ae.c as we now use time(). 2012-10-05 10:10:34 +02:00
Jokea e480c279bc Force expire all timer events when system clock skew is detected. 
When system time changes back, the timer will not worker properly
hence some core functionality of redis will stop working(e.g. replication,
bgsave, etc). See issue #633 for details.

The patch saves the previous time and when a system clock skew is detected,
it will force expire all timers.

Modiifed by @antirez: the previous time was moved into the eventLoop
structure to make sure the library is still thread safe as long as you
use different event loops into different threads (otherwise you need
some synchronization). More comments added about the reasoning at the
base of the patch, that's worth reporting here:

/* If the system clock is moved to the future, and then set back to the
 * right value, time events may be delayed in a random way. Often this
 * means that scheduled operations will not be performed soon enough.
 *
 * Here we try to detect system clock skews, and force all the time
 * events to be processed ASAP when this happens: the idea is that
 * processing events earlier is less dangerous than delaying them
 * indefinitely, and practice suggests it is. */
 2012-10-04 19:30:48 +02:00
antirez 0c19880ce6 "Timeout receiving bulk data" error message modified. 
The new message now contains an hint about modifying the repl-timeout
configuration directive if the problem persists.

This should normally not be needed, because while the master generates
the RDB file it makes sure to send newlines to the replication channel
to prevent timeouts. However there are times when masters running on
very slow systems can completely stop for seconds during the RDB saving
process. In such a case enlarging the timeout value can fix the problem.

See issue #695 for an example of this problem in an EC2 deployment.
 2012-10-04 11:49:17 +02:00
antirez 2ba962714a "SORT by nosort" (skip sorting) respect sorted set ordering. 
When SORT is called with the option BY set to a string constant not
inclduing the wildcard character "*", there is no way to sort the output
so any ordering is valid. This allows the SORT internals to optimize its
work and don't really sort the output at all.

However it was odd that this option was not able to retain the natural
order of a sorted set. This feature was requested by users multiple
times as sometimes to call SORT with GET against sorted sets as a way to
mass-fetch objects can be handy.

This commit introduces two things:

1) The ability of SORT to return sorted sets elements in their natural
ordering when `BY nosort` is specified, accordingly to `DESC / ASC` options.
2) The ability of SORT to optimize this case further if LIMIT is passed
as well, avoiding to really fetch the whole sorted set, but directly
obtaining the specified range.

Because in this case the sorting is always deterministic, no
post-sorting activity is performed when SORT is called from a Lua
script.

This commit fixes issue #98.
 2012-10-03 18:46:48 +02:00
antirez 151b606c11 Revert "Scripting: redis.NIL to return nil bulk replies." 
This reverts commit e061d797d7.

Conflicts:

	src/scripting.c
 2012-10-01 10:10:03 +02:00
antirez f1466e11cd Scripting: add helper functions redis.error_reply() and redis.status_reply(). 
A previous commit introduced Redis.NIL. This commit adds similar helper
functions to return tables with a single field set to the specified
string so that instead of using 'return {err="My Error"}' it is possible
to use a more idiomatic form:

    return redis.error_reply("My Error")
    return redis.status_reply("OK")
 2012-09-28 16:58:38 +02:00
antirez e061d797d7 Scripting: redis.NIL to return nil bulk replies. 
Lua arrays can't contain nil elements (see
http://www.lua.org/pil/19.1.html for more information), so Lua scripts
were not able to return a multi-bulk reply containing nil bulk
elements inside.

This commit introduces a special conversion: a table with just
a "nilbulk" field set to a boolean value is converted by Redis as a nil
bulk reply, but at the same time for Lua this type is not a "nil" so can
be used inside Lua arrays.

This type is also assigned to redis.NIL, so the following two forms
are equivalent and will be able to return a nil bulk reply as second
element of a three elements array:

    EVAL "return {1,redis.NIL,3}" 0
    EVAL "return {1,{nilbulk=true},3}" 0

The result in redis-cli will be:

    1) (integer) 1
    2) (nil)
    3) (integer) 3
 2012-09-28 14:19:15 +02:00
Erik Dubbelboer 04779bdfe1 Fixed some spelling errors in the comments 2012-09-27 13:22:42 +02:00
Erik Dubbelboer e04be06e89 Added consts keyword where possible 2012-09-27 13:12:05 +02:00
antirez c4cbffa3a1 Final merge of Sentinel into 2.6. 
After cherry-picking Sentinel commits a few spurious issues remained
about references to Redis Cluster that is not present in the 2.6 branch.
 2012-09-27 13:10:36 +02:00
antirez dfb7194cba Sentinel: Support for AUTH. 2012-09-27 13:06:17 +02:00
antirez b8ce9a84c5 Sentinel: reply -IDONTKNOW to get-master-addr-by-name on lack of info. 
If we don't have any clue about a master since it never replied to INFO
so far, reply with an -IDONTKNOW error to SENTINEL
get-master-addr-by-name requests.
 2012-09-27 13:06:12 +02:00
antirez 1f8bd82332 Sentinel: more easy master redirection if master is a slave. 
Before this commit Sentienl used to redirect master ip/addr if the
current instance reported to be a slave only if this was the first INFO
output received, and the role was found to be slave.

Now instead also if we find that the runid is different, and the
reported role is slave, we also redirect to the reported master ip/addr.

This unifies the behavior of Sentinel in the case of a reboot (where it
will see the first INFO output with the wrong role and will perform the
redirection), with the behavior of Sentinel in the case of a change in
what it sees in the INFO output of the master.
 2012-09-27 13:06:05 +02:00
antirez ef792fc950 Sentinel: do not crash against slaves not publishing the runid. 
Older versions of Redis (before 2.4.17) don't publish the runid field in
INFO. This commit makes Sentinel able to handle that without crashing.
 2012-09-27 13:06:01 +02:00
antirez de499f7f7e Sentinel: INFO command implementation. 2012-09-27 13:05:58 +02:00
antirez b65f3c2176 Sentinel: add Redis execution mode to INFO output. 
The new "redis_mode" field in the INFO output will show if Redis is
running in standalone mode, cluster, or sentinel mode.
 2012-09-27 13:05:53 +02:00
antirez 161e137c55 Sentinel: Sentinel-side support for slave priority. 
The slave priority that is now published by Redis in INFO output is
now used by Sentinel in order to select the slave with minimum priority
for promotion, and in order to consider slaves with priority set to 0 as
not able to play the role of master (they will never be promoted by
Sentinel).

The "slave-priority" field is now one of the fileds that Sentinel
publishes when describing an instance via the SENTINEL commands such as
"SENTINEL slaves mastername".
 2012-09-27 13:05:49 +02:00
antirez d480b9ce7f Sentinel: suppress harmless warning by initializing 'table' to NULL. 
Note that the assertion guarantees that one of the if branches setting
table is always entered.
 2012-09-27 13:05:45 +02:00
antirez fa23fc3363 Sentinel: send SCRIPT KILL on -BUSY reply and SDOWN instance. 
From the point of view of Redis an instance replying -BUSY is down,
since it is effectively not able to reply to user requests. However
a looping script is a recoverable condition in Redis if the script still
did not performed any write to the dataset. In that case performing a
fail over is not optimal, so Sentinel now tries to restore the normal server
condition killing the script with a SCRIPT KILL command.

If the script already performed some write before entering an infinite
(or long enough to timeout) loop, SCRIPT KILL will not work and the
fail over will be triggered anyway.
 2012-09-27 13:05:41 +02:00
antirez fc0a0d4aa7 Sentinel: fixed a crash on script execution. 
The call to sentinelScheduleScriptExecution() lacked the final NULL
argument to signal the end of arguments. This resulted into a crash.
 2012-09-27 13:05:38 +02:00
antirez ea9bec50c6 Sentinel: SENTINEL FAILOVER command implemented. 
This command can be used in order to force a Sentinel instance to start
a failover for the specified master, as leader, forcing the failover
even if the master is up.

The commit also adds some minor refactoring and other improvements to
functions already implemented that make them able to work when the
master is not in SDOWN condition. For instance slave selection
assumed that we ask INFO every second to every slave, this is true
only when the master is in SDOWN condition, so slave selection did not
worked when the master was not in SDOWN condition.
 2012-09-27 13:05:33 +02:00
antirez 26a340095d Sentinel: client reconfiguration script execution. 
This commit adds support to optionally execute a script when one of the
following events happen:

* The failover starts (with a slave already promoted).
* The failover ends.
* The failover is aborted.

The script is called with enough parameters (documented in the example
sentinel.conf file) to provide information about the old and new ip:port
pair of the master, the role of the sentinel (leader or observer) and
the name of the master.

The goal of the script is to inform clients of the configuration change
in a way specific to the environment Sentinel is running, that can't be
implemented in a genereal way inside Sentinel itself.
 2012-09-27 13:05:30 +02:00
antirez 524b79d231 Sentinel: when leader in wait-start, sense another leader as race. 
When we are in wait start, if another leader (or any other external
entity) turns a slave into a master, abort the failover, and detect it
as an observer.

Note that the wait-start state is mainly there for this reason but the
abort was yet not implemented.

This adds a new sentinel event -failover-abort-race.
 2012-09-27 13:05:26 +02:00
antirez 201ed6d42e Sentinel: sentinelRefreshInstanceInfo() comments improved a bit. 2012-09-27 13:05:19 +02:00
antirez 7c9bfe1036 Sentinel: sentinel.conf self-documenation improved. 2012-09-27 13:05:15 +02:00
antirez 3da75e2ca4 Sentinel: abort failover when in wait-start if master is back. 
When we are a Leader Sentinel in wait-start state, starting with this
commit the failover is aborted if the master returns online.

This improves the way we handle a notable case of net split, that is the
split between Sentinels and Redis servers, that will be a very common
case of split becase Sentinels will often be installed in the client's
network and servers can be in a differnt arm of the network.

When Sentinels and Redis servers are isolated the master is in ODOWN
condition since the Sentinels can agree about this state, however the
failover does not start since there are no good slaves to promote (in
this specific case all the slaves are unreachable).

However when the split is resolved, Sentinels may sense the slave back
a moment before they sense the master is back, so the failover may start
without a good reason (since the master is actually working too).

Now this condition is reversible, so the failover will be aborted
immediately after if the master is detected to be working again, that
is, not in SDOWN nor in ODOWN condition.
 2012-09-27 13:05:11 +02:00
antirez e328e41a3a Sentinel: scripts execution engine improved. 
We no longer use a vanilla fork+execve but take a queue of jobs of
scripts to execute, with retry on error, timeouts, and so forth.

Currently this is used only for notifications but soon the ability to
also call clients reconfiguration scripts will be added.
 2012-09-27 13:05:07 +02:00
Jan-Erik Rediger 8a8e560b87 Include sys/wait.h to avoid compiler warning 
gcc warned about an implicit declaration of function 'wait3'.
Including this header fixes this.
 2012-09-27 13:04:51 +02:00
antirez 0d0975f2b6 Sentinel: don't start a failover as leader if there is no good slave. 2012-09-27 13:04:48 +02:00
Jeremy Zawodny af41f6cffb comment fix 
improve English a bit. :-)
 2012-09-27 13:04:44 +02:00
antirez 999fe0d352 Sentinel: ability to execute notification scripts. 2012-09-27 13:04:41 +02:00
mrb f1057534e7 Fix warning in redis.c for sentinel config load 2012-09-27 13:04:37 +02:00
mrb fcc8bf9974 Some cleanup in sentinel.conf 2012-09-27 13:04:32 +02:00
antirez 374eed7d2a Sentinel: abort failover if no good slave is available. 
The previous behavior of the state machine was to wait some time and
retry the slave selection, but this is not robust enough against drastic
changes in the conditions of the monitored instances.

What we do now when the slave selection fails is to abort the failover
and return back monitoring the master. If the ODOWN condition is still
present a new failover will be triggered and so forth.

This commit also refactors the code we use to abort a failover.
 2012-09-27 13:04:28 +02:00
antirez 2085fdb1f4 Sentinel: reset pending_commands in a more generic way. 2012-09-27 13:04:24 +02:00
antirez f8a19e32e2 Prevent a spurious +sdown event on switch. 
When we reset the master we should start with clean timestamps for ping
replies otherwise we'll detect a spurious +sdown event, because on
+master-switch event the previous master instance was probably in +sdown
condition. Since we updated the address we should count time from
scratch again.

Also this commit makes sure to explicitly reset the count of pending
commands, now we can do this because of the new way the hiredis link
is closed.
 2012-09-27 13:04:19 +02:00
antirez 7c39b55d42 Sentinel: debugging message removed. 2012-09-27 13:04:16 +02:00
antirez e47236d8d4 Sentinel: changes to connection handling and redirection. 
We disconnect the Redis instances hiredis link in a more robust way now.
Also we change the way we perform the redirection for the +switch-master
event, that is not just an instance reset with an address change.

Using the same system we now implement the +redirect-to-master event
that is triggered by an instance that is configured to be master but
found to be a slave at the first INFO reply. In that case we monitor the
master instead, logging the incident as an event.
 2012-09-27 13:04:12 +02:00
antirez 8ab7e998d1 Sentinel: check that instance still exists in reply callbacks. 
We can't be sure the instance object still exists when the reply
callback is called.
 2012-09-27 13:04:08 +02:00
antirez e01a415d37 Sentinel: more robust failover detection as observer. 
Sentinel observers detect failover checking if a slave attached to the
monitored master turns into its replication state from slave to master.
However while this change may in theory only happen after a SLAVEOF NO
ONE command, in practie it is very easy to reboot a slave instance with
a wrong configuration that turns it into a master, especially if it was
a past master before a successfull failover.

This commit changes the detection policy so that if an instance goes
from slave to master, but at the same time the runid has changed, we
sense a reboot, and in that case we don't detect a failover at all.

This commit also introduces the "reboot" sentinel event, that is logged
at "warning" level (so this will trigger an admin notification).

The commit also fixes a problem in the disconnect handler that assumed
that the instance object always existed, that is not the case. Now we
no longer assume that redisAsyncFree() will call the disconnection
handler before returning.
 2012-09-27 13:04:02 +02:00
antirez d26a8fb4db Fixed an error in the example sentinel.conf. 2012-09-27 13:03:55 +02:00
antirez 5b5eb192f5 Typo. 2012-09-27 13:03:49 +02:00
antirez 120ba3922d First implementation of Redis Sentinel. 
This commit implements the first, beta quality implementation of Redis
Sentinel, a distributed monitoring system for Redis with notification
and automatic failover capabilities.

More info at http://redis.io/topics/sentinel
 2012-09-27 13:03:41 +02:00
antirez 2812b945f0 Test for SRANDMEMBER with <count>. 2012-09-21 11:56:04 +02:00
antirez 31fe053a62 SRANDMEMBER <count> leak fixed. 
For "CASE 4" (see code) we need to free the element if it's already in
the result dictionary and adding it failed.
 2012-09-21 11:56:00 +02:00
antirez dd94771578 Added the SRANDMEMBER key <count> variant. 
SRANDMEMBER called with just the key argument can just return a single
random element from a Redis Set. However many users need to return
multiple unique elements from a Set, this is not a trivial problem to
handle in the client side, and for truly good performance a C
implementation was required.

After many requests for this feature it was finally implemented.

The problem implementing this command is the strategy to follow when
the number of elements the user asks for is near to the number of
elements that are already inside the set. In this case asking random
elements to the dictionary API, and trying to add it to a temporary set,
may result into an extremely poor performance, as most add operations
will be wasted on duplicated elements.

For this reason this implementation uses a different strategy in this
case: the Set is copied, and random elements are returned to reach the
specified count.

The code actually uses 4 different algorithms optimized for the
different cases.

If the count is negative, the command changes behavior and allows for
duplicated elements in the returned subset.
 2012-09-21 11:55:56 +02:00
antirez 8b6b1b27cc Fix compilation on FreeBSD. Thanks to @koobs on twitter. 2012-09-17 12:45:57 +02:00
antirez 4403862623 Redis 2.5.13 (2.6.0 RC7). 2012-09-17 11:02:49 +02:00
antirez 174518ffb7 .gitignore modified to be more general with less entries. 2012-09-17 10:50:07 +02:00
antirez f444e2afdc A reimplementation of blocking operation internals. 
Redis provides support for blocking operations such as BLPOP or BRPOP.
This operations are identical to normal LPOP and RPOP operations as long
as there are elements in the target list, but if the list is empty they
block waiting for new data to arrive to the list.

All the clients blocked waiting for th same list are served in a FIFO
way, so the first that blocked is the first to be served when there is
more data pushed by another client into the list.

The previous implementation of blocking operations was conceived to
serve clients in the context of push operations. For for instance:

1) There is a client "A" blocked on list "foo".
2) The client "B" performs `LPUSH foo somevalue`.
3) The client "A" is served in the context of the "B" LPUSH,
synchronously.

Processing things in a synchronous way was useful as if "A" pushes a
value that is served by "B", from the point of view of the database is a
NOP (no operation) thing, that is, nothing is replicated, nothing is
written in the AOF file, and so forth.

However later we implemented two things:

1) Variadic LPUSH that could add multiple values to a list in the
context of a single call.
2) BRPOPLPUSH that was a version of BRPOP that also provided a "PUSH"
side effect when receiving data.

This forced us to make the synchronous implementation more complex. If
client "B" is waiting for data, and "A" pushes three elemnents in a
single call, we needed to propagate an LPUSH with a missing argument
in the AOF and replication link. We also needed to make sure to
replicate the LPUSH side of BRPOPLPUSH, but only if in turn did not
happened to serve another blocking client into another list ;)

This were complex but with a few of mutually recursive functions
everything worked as expected... until one day we introduced scripting
in Redis.

Scripting + synchronous blocking operations = Issue #614.

Basically you can't "rewrite" a script to have just a partial effect on
the replicas and AOF file if the script happened to serve a few blocked
clients.

The solution to all this problems, implemented by this commit, is to
change the way we serve blocked clients. Instead of serving the blocked
clients synchronously, in the context of the command performing the PUSH
operation, it is now an asynchronous and iterative process:

1) If a key that has clients blocked waiting for data is the subject of
a list push operation, We simply mark keys as "ready" and put it into a
queue.
2) Every command pushing stuff on lists, as a variadic LPUSH, a script,
or whatever it is, is replicated verbatim without any rewriting.
3) Every time a Redis command, a MULTI/EXEC block, or a script,
completed its execution, we run the list of keys ready to serve blocked
clients (as more data arrived), and process this list serving the
blocked clients.
4) As a result of "3" maybe more keys are ready again for other clients
(as a result of BRPOPLPUSH we may have push operations), so we iterate
back to step "3" if it's needed.

The new code has a much simpler semantics, and a simpler to understand
implementation, with the disadvantage of not being able to "optmize out"
a PUSH+BPOP as a No OP.

This commit will be tested with care before the final merge, more tests
will be added likely.
 2012-09-17 10:26:50 +02:00
antirez b58f03a0e8 Make sure that SELECT argument is an integer or return an error. 
Unfortunately we had still the lame atoi() without any error checking in
place, so "SELECT foo" would work as "SELECT 0". This was not an huge
problem per se but some people expected that DB can be strings and not
just numbers, and without errors you get the feeling that they can be
numbers, but not the behavior.

Now getLongFromObjectOrReply() is used as almost everybody else across
the code, generating an error if the number is not an integer or
overflows the long type.

Thanks to @mipearson for reporting that on Twitter.
 2012-09-11 10:34:48 +02:00
antirez efb54f0593 Match printf format with actual type in genRedisInfoString(). 2012-09-10 12:43:20 +02:00
antirez 58889867bb BITCOUNT regression test for #582 fixed for 32 bit target. 
Bug #582 was not present in 32 bit builds of Redis as
getObjectFromLong() will return an error for overflow.

This commit makes sure that the test does not fail because of the error
returned when running against 32 bit builds.
 2012-09-05 17:50:24 +02:00
Haruto Otake 4c3d419013 BITCOUNT: fix segmentation fault. 
remove unsafe and unnecessary cast.
until now, this cast may lead segmentation fault when end > UINT_MAX

setbit foo 0 1
bitcount  0 4294967295
=> ok
bitcount  0 4294967296
=> cause segmentation fault.

Note by @antirez: the commit was modified a bit to also change the
string length type to long, since it's guaranteed to be at max 512 MB in
size, so we can work with the same type across all the code path.

A regression test was also added.
 2012-09-05 16:20:21 +02:00
Saj Goonatilleke 0671d88cab Bug fix: slaves being pinged every second 
REDIS_REPL_PING_SLAVE_PERIOD controls how often the master should
transmit a heartbeat (PING) to its slaves.  This period, which defaults
to 10, is measured in seconds.

Redis 2.4 masters used to ping their slaves every ten seconds, just like
it says on the tin.

The Redis 2.6 masters I have been experimenting with, on the other hand,
ping their slaves *every second*.  (master_last_io_seconds_ago never
approaches 10.)  I think the ping period was inadvertently slashed to
one-tenth of its nominal value around the time REDIS_HZ was introduced.
This commit reintroduces correct ping schedule behaviour.
 2012-09-05 16:01:01 +02:00
antirez 5ddee9b7d5 Scripting: Force SORT BY constant determinism inside SORT itself. 
SORT is able to return (faster than when ordering) unordered output if
the "BY" clause is used with a constant value. However we try to play
well with scripting requirements of determinism providing always sorted
outputs when SORT (and other similar commands) are called by Lua
scripts.

However we used the general mechanism in place in scripting in order to
reorder SORT output, that is, if the command has the "S" flag set, the
Lua scripting engine will take an additional step when converting a
multi bulk reply to Lua value, calling a Lua sorting function.

This is suboptimal as we can do it faster inside SORT itself.
This is also broken as issue #545 shows us: basically when SORT is used
with a constant BY, and additionally also GET is used, the Lua scripting
engine was trying to order the output as a flat array, while it was
actually a list of key-value pairs.

What we do know is to recognized if the caller of SORT is the Lua client
(since we can check this using the REDIS_LUA_CLIENT flag). If so, and if
a "don't sort" condition is triggered by the BY option with a constant
string, we force the lexicographical sorting.

This commit fixes this bug and improves the performance, and at the same
time simplifies the implementation. This does not mean I'm smart today,
it means I was stupid when I committed the original implementation ;)
 2012-09-05 01:19:47 +02:00
antirez fd2a8951bf Send an async PING before starting replication with master. 
During the first synchronization step of the replication process, a Redis
slave connects with the master in a non blocking way. However once the
connection is established the replication continues sending the REPLCONF
command, and sometimes the AUTH command if needed. Those commands are
send in a partially blocking way (blocking with timeout in the order of
seconds).

Because it is common for a blocked master to accept connections even if
it is actually not able to reply to the slave requests, it was easy for
a slave to block if the master had serious issues, but was still able to
accept connections in the listening socket.

For this reason we now send an asynchronous PING request just after the
non blocking connection ended in a successful way, and wait for the
reply before to continue with the replication process. It is very
unlikely that a master replying to PING can't reply to the other
commands.

This solution was proposed by Didier Spezia (Thanks!) so that we don't
need to turn all the replication process into a non blocking affair, but
still the probability of a slave blocked is minimal even in the event of
a failing master.

Also we now use getsockopt(SO_ERROR) in order to check errors ASAP
in the event handler, instead of waiting for actual I/O to return an
error.

This commit fixes issue #632.
 2012-09-03 11:48:27 +02:00
antirez 42a239b888 Scripting: Reset Lua fake client reply_bytes after command execution. 
Lua scripting uses a fake client in order to run commands in the context
of a client, accumulate the reply, and convert it into a Lua object
to return to the caller. This client is reused again and again, and is
referenced by the server.lua_client globally accessible pointer.

However after every call to redis.call() or redis.pcall(), that is
handled by the luaRedisGenericCommand() function, the reply_bytes field
of the client was not set back to zero. This filed is used to estimate
the amount of memory currently used in the reply. Because of the lack of
reset, script after script executed, this value used to get bigger and
bigger, and in the end on 32 bit systems it triggered the following
assert:

    redisAssert(c->reply_bytes < ULONG_MAX-(1024*64));

On 64 bit systems this does not happen because it takes too much time to
reach values near to 2^64 for users to see the practical effect of the
bug.

Now in the cleanup stage of luaRedisGenericCommand() we reset the
reply_bytes counter to zero, avoiding the issue. It is not practical to
add a test for this bug, but the fix was manually tested using a
debugger.

This commit fixes issue #656.
 2012-08-31 11:08:53 +02:00
antirez 851ac9d072 Sentinel: added documentation about slave-priority in redis.conf 2012-08-31 10:30:51 +02:00
antirez 48d26a483d Sentinel: Redis-side support for slave priority. 
A Redis slave can now be configured with a priority, that is an integer
number that is shown in INFO output and can be get and set using the
redis.conf file or the CONFIG GET/SET command.

This field is used by Sentinel during slave election. A slave with lower
priority is preferred. A slave with priority zero is never elected (and
is considered to be impossible to elect even if it is the only slave
available).

A next commit will add support in the Sentinel side as well.
 2012-08-31 10:30:29 +02:00
antirez edfaa64f49 Scripting: require at least one argument for redis.call(). 
Redis used to crash with a call like the following:

    EVAL "redis.call()" 0

Now the explicit check for at least one argument prevents the problem.

This commit fixes issue #655.
 2012-08-31 10:28:36 +02:00
antirez 13732168a5 Incrementally flush RDB on disk while loading it from a master. 
This fixes issue #539.

Basically if there is enough free memory the OS may buffer the RDB file
that the slave transfers on disk from the master. The file may
actually be flused on disk at once by the operating system when it gets
closed by Redis, causing the close system call to block for a long time.

This patch is a modified version of one provided by yoav-steinberg of
@garantiadata (the original version was posted in the issue #539
comments), and tries to flush the OS buffers incrementally (every 8 MB
of loaded data).
 2012-08-28 12:47:35 +02:00
antirez 06bd3b9acd Fix a forget zmalloc_oom() -> zmalloc_oom_handler() replacement. 2012-08-24 15:41:49 +02:00
antirez 5de75120ba Better Out of Memory handling. 
The previous implementation of zmalloc.c was not able to handle out of
memory in an application-specific way. It just logged an error on
standard error, and aborted.

The result was that in the case of an actual out of memory in Redis
where malloc returned NULL (In Linux this actually happens under
specific overcommit policy settings and/or with no or little swap
configured) the error was not properly logged in the Redis log.

This commit fixes this problem, fixing issue #509.
Now the out of memory is properly reported in the Redis log and a stack
trace is generated.

The approach used is to provide a configurable out of memory handler
to zmalloc (otherwise the default one logging the event on the
standard output is used).
 2012-08-24 13:03:40 +02:00
antirez 32095c4057 redis-benchmark: disable big buffer cleanup in hiredis context. 
This new hiredis features allows us to reuse a previous context reader
buffer even if already very big in order to maximize performances with
big payloads (Usually hiredis re-creates buffers when they are too big
and unused in order to save memory).
 2012-08-22 11:34:03 +02:00
antirez 7fcba9fd9a hiredis library updated. 
This version of hiredis merges modifications of the Redis fork with
latest changes in the hiredis repository.

The same version was pushed on the hiredis repository and will probably
merged into the master branch in short time.
 2012-08-22 11:33:57 +02:00
Pieter Noordhuis 2f44452612 Set p to its new offset before modifying it 2012-08-22 11:33:52 +02:00
Pieter Noordhuis 89bf6f58fd Add ziplist test for deleting next to last entries 2012-08-22 11:33:47 +02:00
Tobias Schwab 013189e7db Fix version numbers 2012-08-02 14:38:42 +02:00
antirez 73d3e8751b Redis 2.5.12 (2.6 RC6). 2012-08-01 12:06:03 +02:00
Michael Parker 628890e43e Use correct variable name for value to convert. 
Note by @antirez: this code was never compiled because utils.c lacked the
float.h include, so we never noticed this variable was mispelled in the
past.

This should provide a noticeable speed boost when saving certain types
of databases with many sorted sets inside.
 2012-07-31 11:50:51 +02:00
Saj Goonatilleke 4c0c1fff5a Truncate short write from the AOF 
If Redis only manages to write out a partial buffer, the AOF file won't
load back into Redis the next time it starts up.  It is better to
discard the short write than waste time running redis-check-aof.
 2012-07-31 10:58:16 +02:00
Saj Goonatilleke f00b0844c9 New in INFO: aof_last_bgrewrite_status 
Behaves like rdb_last_bgsave_status -- even down to reporting 'ok' when
no rewrite has been done yet.  (You might want to check that
aof_last_rewrite_time_sec is not -1.)
 2012-07-31 10:58:12 +02:00
Steeve Lennmark 889e443ce5 Check that we have connection before enabling pipe mode 2012-07-22 17:19:05 +02:00
antirez 3bb3f12539 Allow Pub/Sub in contexts where other commands are blocked. 
Redis loading data from disk, and a Redis slave disconnected from its
master with serve-stale-data disabled, are two conditions where
commands are normally refused by Redis, returning an error.

However there is no reason to disable Pub/Sub commands as well, given
that this layer does not interact with the dataset. To allow Pub/Sub in
as many contexts as possible is especially interesting now that Redis
Sentinel uses Pub/Sub of a Redis master as a communication channel
between Sentinels.

This commit allows Pub/Sub to be used in the above two contexts where
it was previously denied.
 2012-07-22 17:18:12 +02:00
antirez 82675c86a6 Don't assume that "char" is signed. 
For the C standard char can be either signed or unsigned, it's up to the
compiler, but Redis assumed that it was signed in a few places.

The practical effect of this patch is that now Redis 2.6 will run
correctly in every system where char is unsigned, notably the RaspBerry
PI and other ARM systems with GCC.

Thanks to Georgi Marinov (@eesn on twitter) that reported the problem
and allowed me to use his RaspBerry via SSH to trace and fix the issue!
 2012-07-18 12:01:43 +02:00
jokea 8a8e01f4a7 mark fd as writable when EPOLLERR or EPOLLHUP is returned by epoll_wait. 2012-07-09 12:15:07 +02:00
antirez d3d567428a Typo in comment. 2012-07-07 17:24:40 +02:00
antirez dbd8c753c4 REPLCONF internal command introduced. 
The REPLCONF command is an internal command (not designed to be directly
used by normal clients) that allows a slave to set some replication
related state in the master before issuing SYNC to start the
replication.

The initial motivation for this command, and the only reason currently
it is used by the implementation, is to let the slave instance
communicate its listening port to the slave, so that the master can
show all the slaves with their listening ports in the "replication"
section of the INFO output.

This allows clients to auto discover and query all the slaves attached
into a master.

Currently only a single option of the REPLCONF command is supported, and
it is called "listening-port", so the slave now starts the replication
process with something like the following chat:

    REPLCONF listening-prot 6380
    SYNC

Note that this works even if the master is an older version of Redis and
does not understand REPLCONF, because the slave ignores the REPLCONF
error.

In the future REPLCONF can be used for partial replication and other
replication related features where there is the need to exchange
information between master and slave.

NOTE: This commit also fixes a bug: the INFO outout already carried
information about slaves, but the port was broken, and was obtained
with getpeername(2), so it was actually just the ephemeral port used
by the slave to connect to the master as a client.
 2012-07-07 17:24:33 +02:00
antirez b3f28b90d2 Fixed comment typo into time_independent_strcmp(). 2012-06-21 14:26:00 +02:00
antirez 4b3865cbdb Fixed a timing attack on AUTH (Issue #560). 
The way we compared the authentication password using strcmp() allowed
an attacker to gain information about the password using a well known
class of attacks called "timing attacks".

The bug appears to be practically not exploitable in most modern systems
running Redis since even using multiple bytes of differences in the
input at a time instead of one the difference in running time in in the
order of 10 nanoseconds, making it hard to exploit even on LAN. However
attacks always get better so we are providing a fix ASAP.

The new implementation uses two fixed length buffers and a constant time
comparison function, with the goal of:

1) Completely avoid leaking information about the content of the
password, since the comparison is always performed between 512
characters and without conditionals.
2) Partially avoid leaking information about the length of the
password.

About "2" we still have a stage in the code where the real password and
the user provided password are copied in the static buffers, we also run
two strlen() operations against the two inputs, so the running time
of the comparison is a fixed amount plus a time proportional to
LENGTH(A)+LENGTH(B). This means that the absolute time of the operation
performed is still related to the length of the password in some way,
but there is no way to change the input in order to get a difference in
the execution time in the comparison that is not just proportional to
the string provided by the user (because the password length is fixed).

Thus in practical terms the user should try to discover LENGTH(PASSWORD)
looking at the whole execution time of the AUTH command and trying to
guess a proportionality between the whole execution time and the
password length: this appears to be mostly unfeasible in the real world.

Also protecting from this attack is not very useful in the case of Redis
as a brute force attack is anyway feasible if the password is too short,
while with a long password makes it not an issue that the attacker knows
the length.
 2012-06-21 12:01:07 +02:00
antirez 0c9cf45270 Redis 2.5.11 (2.6 RC5). 2012-06-15 13:44:17 +02:00
antirez 6fe9c402a2 Fix c->reply_bytes computation in setDeferredMultiBulkLength() 
In order to implement reply buffer limits introduced in 2.6 and useful
to close the connection under user-selected circumastances of big output
buffers (for instance slow consumers in pub/sub, a blocked slave, and so
forth) Redis takes a counter with the amount of used memory in objects
inside the output list stored into c->reply.

The computation was broken in the function setDeferredMultiBulkLength(),
in the case the object was glued with the next one. This caused the
c->reply_bytes field to go out of sync, be subtracted more than needed,
and wrap back near to ULONG_MAX values.

This commit fixes this bug and adds an assertion that is able to trap
this class of problems.

This problem was discovered looking at the INFO output of an unrelated
issue (issue #547).
 2012-06-15 10:11:27 +02:00
antirez 8361d6c406 ziplistFind(): don't assume that entries are comparable by encoding. 
Because Redis 2.6 introduced new integer encodings it is no longer true
that if two entries have a different encoding they are not equal.

An old ziplist can be loaded from an RDB file generated with Redis 2.4,
in this case for instance a small unsigned integers is encoded with a
16 bit encoding, while in Redis 2.6 a more specific 8 bit encoding
format is used.

Because of this bug hashes ended with duplicated values or fields lookup
failed, causing many bad behaviors.
This in turn caused a crash while converting the ziplist encoded hash into
a real hash table because an assertion was raised on duplicated elements.

This commit fixes issue #547.

Many thanks to Pinterest's Marty Weiner and colleagues for discovering
the problem and helping us in the debugging process.
 2012-06-15 10:11:23 +02:00
Ted Nyman e612508d38 Standardize punctuation in redis-cli help. 
Right there is a mix of help entries ending with periods or
without periods. This standardizes the end of command as without
periods, which seems to be the general custom in most unix tools,
at least.
 2012-06-13 10:27:23 +02:00
antirez 1364395f18 Added a new hash fuzzy tester. 
The new fuzzy tester also removes elements from the hash instead of just
adding random fields. This should increase the probability to find bugs
in the implementations of the hash type internal representations.
 2012-06-12 15:20:16 +02:00
antirez 0b8441c1b2 Dump ziplist hex value on failed assertion. 
The ziplist -> hashtable conversion code is triggered every time an hash
value must be promoted to a full hash table because the number or size of
elements reached the threshold.

If a problem in the ziplist causes the same field to be present
multiple times, the assertion of successful addition of the element
inside the hash table will fail, crashing server with a failed
assertion, but providing little information about the problem.

This code adds a new logging function to perform the hex dump of binary
data, and makes sure that the ziplist -> hashtable conversion code uses
this new logging facility to dump the content of the ziplist when the
assertion fails.

This change was originally made in order to investigate issue #547.
 2012-06-12 00:49:32 +02:00
antirez a85c89854f New test: hash ziplist -> hashtable encoding conversion. 
A new stress test was added to stress test the code converting a ziplist
into an hash table.

In this commit also randomValue helper function was modified to also
return negative values.
 2012-06-11 15:21:44 +02:00
dvir volk 0ea1a9c452 fixed server install script to rewrite the default configuration file and not a template, and removed the old config template 2012-06-08 16:03:18 +03:00
antirez 1a3e9d951b Version bumped to 2.5.10 (2.6 RC4) + Release Notes. 2012-06-07 11:41:38 +02:00
antirez 3f12656781 EVAL replication test: less false positives. 
wait_for_condition is now used instead of the usual "after 1000" (that
is the way to sleep in Tcl). This should avoid to find the replica in
a state where it is loading the RDB in memory, returning -LOADING error.

This test used to fail when running the test over valgrind, due to the
added latencies.
 2012-06-02 23:32:01 +02:00
Alex Mitrofanov 591c9e6543 Fixed RESTORE hash failure (Issue #532) 
(additional commit notes by antirez@gmail.com):

The rdbIsObjectType() macro was not updated when the new RDB object type
of ziplist encoded hashes was added.

As a result RESTORE, that uses rdbLoadObjectType(), failed when a
ziplist encoded hash was loaded.
This does not affected normal RDB loading because in that case we use
the lower-level function rdbLoadType().

The commit also adds a regression test.
 2012-06-02 10:27:31 +02:00
antirez ebdcd723a7 RDB type loading functions clarified in comments. 
Improved comments to make clear that rdbLoadType() just loads a
general TYPE in the context of RDB that can be an object type or an
expire type, end-of-file, and so forth.

While rdbLoadObjectType() enforces that the type is a valid Object Type
otherwise it returns -1.
 2012-06-02 10:27:27 +02:00
antirez 8f0658cdd0 BITOP bug when called against non existing keys fixed. 
In the issue #529 an user reported a bug that can be triggered with the
following code:

flushdb
set a
"\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00"
bitop or x a b

The bug was introduced with the speed optimization in commit 8bbc076
that specializes every BITOP operation loop up to the minimum length of
the input strings.

However the computation of the minimum length contained an error when a
non existing key was present in the input, after a key that was non zero
length.

This commit fixes the bug and adds a regression test for it.
 2012-05-31 21:51:52 +02:00
antirez 9a8d51add5 2.6 Release notes: added more new features to the list. 2012-05-27 17:17:22 +02:00
antirez 473f3090f2 Tests modified to account for INFO fields renaming. 
Commit 33e1db36fa modified the name of a
few INFO fields. This commit changes the Redis test to account for this
changes.
 2012-05-25 16:25:07 +02:00
antirez 69396249e5 Release notes: more info about 2.4 -> 2.6 migration. 
Migration information moved at the end of the file as now they start to
be too long to stay at the top, but a warning was added at the top of
the file to remember the user to check.

Added information about INFO fields that changed name between 2.4 and
2.6.
 2012-05-25 12:24:20 +02:00
antirez 6b4d92e21d Four new persistence fields in INFO. A few renamed. 
The 'persistence' section of INFO output now contains additional four
fields related to RDB and AOF persistence:

 rdb_last_bgsave_time_sec       Duration of latest BGSAVE in sec.
 rdb_current_bgsave_time_sec    Duration of current BGSAVE in sec.
 aof_last_rewrite_time_sec      Duration of latest AOF rewrite in sec.
 aof_current_rewrite_time_sec   Duration of current AOF rewrite in sec.

The 'current' fields are set to -1 if a BGSAVE / AOF rewrite is not in
progress. The 'last' fileds are set to -1 if no previous BGSAVE / AOF
rewrites were performed.

Additionally a few fields in the persistence section were renamed for
consistency:

 changes_since_last_save -> rdb_changes_since_last_save
 bgsave_in_progress -> rdb_bgsave_in_progress
 last_save_time -> rdb_last_save_time
 last_bgsave_status -> rdb_last_bgsave_status
 bgrewriteaof_in_progress -> aof_rewrite_in_progress
 bgrewriteaof_scheduled -> aof_rewrite_scheduled

After the renaming, fields in the persistence section start with rdb_ or
aof_ prefix depending on the persistence method they describe.
The field 'loading' and related fields are not prefixed because they are
unique for both the persistence methods.
 2012-05-25 12:20:35 +02:00
antirez 8bbc0768b8 BITOP command 10x speed improvement. 
This commit adds a fast-path to the BITOP that can be used for all the
bytes from 0 to the minimal length of the string, and if there are
at max 16 input keys.

Often the intersected bitmaps are roughly the same size, so this
optimization can provide a 10x speed boost to most real world usages
of the command.

Bytes are processed four full words at a time, in loops specialized
for the specific BITOP sub-command, without the need to check for
length issues with the inputs (since we run this algorithm only as far
as there is data from all the keys at the same time).

The remaining part of the string is intersected in the usual way using
the slow but generic algorith.

It is possible to do better than this with inputs that are not roughly
the same size, sorting the input keys by length, by initializing the
result string in a smarter way, and noticing that the final part of the
output string composed of only data from the longest string does not
need any proecessing since AND, OR and XOR against an empty string does
not alter the output (zero in the first case, and the original string in
the other two cases).

More implementations will be implemented later likely, but this should
be enough to release Redis 2.6-RC4 with bitops merged in.

Note: this commit also adds better testing for BITOP NOT command, that
is currently the faster and hard to optimize further since it just
flips the bits of a single input string.
 2012-05-24 15:25:13 +02:00
antirez f0d962c0ec BITOP: handle integer encoded objects correctly. 
A bug in the implementation caused BITOP to crash the server if at least
one one of the source objects was integer encoded.

The new implementation takes an additional array of Redis objects
pointers and calls getDecodedObject() to get a reference to a string
encoded object, and then uses decrRefCount() to release the object.

Tests modified to cover the regression and improve coverage.
 2012-05-24 15:25:10 +02:00
antirez c19bfe9fcf BITCOUNT performance improved. 
At Redis's default optimization level the command is now much faster,
always using a constant-time bit manipualtion technique to count bits
instead of GCC builtin popcount, and unrolling the loop.

The current implementation performance is 1.5GB/s in a MBA 11" (1.8 Ghz
i7) compiled with both GCC and clang.

The algorithm used is described here:

http://graphics.stanford.edu/~seander/bithacks.html
 2012-05-24 15:25:06 +02:00
antirez e53a1fdd60 bitop.c renamed bitops.c 
bitop.c contains the "Bit related string operations" so it seems more
logical to call it bitops instead of bitop.
This also makes it matching the name of the test (unit/bitops.tcl).
 2012-05-24 15:24:57 +02:00
antirez 951213e2d2 Bit operations tests improved. 
Fuzzing tests of BITCOUNT / BITOP are iterated multiple times.
The new BITCOUNT fuzzing test uses random strings in a wider interval of
lengths including zero-len strings.
 2012-05-24 15:24:35 +02:00
antirez d12a68d415 popcount() optimization for speed. 
We run the array by 32 bit words instead of processing it byte per byte.
If the code is compiled using GCC __builtin_popcount() builtin function
is used instead.
 2012-05-24 15:24:31 +02:00
antirez 7ffa248c7c BITCOUNT refactoring. 
The low level popualtion counting function is now separated from the
BITCOUNT command implementation, so that the low level function can be
further optimized and eventually used in other contexts if needed.
 2012-05-24 15:24:27 +02:00
antirez 84ec6706cf Bit-related string operations moved to bitop.c 
All the general string operations are implemented in t_string.c, however
the bit operations, while targeting the string type, are better served
in a specific file where we have the implementations of the following
four commands and helper functions:

    GETBIT
    SETBIT
    BITOP
    BITCOUNT

In the future this file will probably contain more code related to
making the BITOP and BITCOUNT operations faster.
 2012-05-24 15:24:14 +02:00
antirez 1f40cdd0e5 BITOP and BITCOUNT tests. 
The Redis implementation is tested against Tcl implementations of the
same operation. Both fuzzing and testing of specific aspects of the
commands behavior are performed.
 2012-05-24 15:22:45 +02:00
antirez 4a789decaa New commands: BITOP and BITCOUNT. 
The motivation for this new commands is to be search in the usage of
Redis for real time statistics. See the article "Fast real time metrics
using Redis".

http://blog.getspool.com/2011/11/29/fast-easy-realtime-metrics-using-redis-bitmaps/

In general Redis strings when used as bitmaps using the SETBIT/GETBIT
command provide a very space-efficient and fast way to store statistics.
For instance in a web application with users, every user can be
associated with a key that shows every day in which the user visited the
web service. This information can be really valuable to extract user
behaviour information.

With Redis bitmaps doing this is very simple just saying that a given
day is 0 (the data the service was put online) and all the next days are
1, 2, 3, and so forth. So with SETBIT it is possible to set the bit
corresponding to the current day every time the user visits the site.

It is possible to take the count of the bit sets on the run, this is
extremely easy using a Lua script. However a fast bit count native
operation can be useful, especially if it can operate on ranges, or when
the string is small like in the case of days (even if you consider many
years it is still extremely little data).

For this reason BITOP was introduced. The command counts the number of
bits set to 1 in a string, with optional range:

BITCOUNT key [start end]

The start/end parameters are similar to GETRANGE. If omitted the whole
string is tested.

Population counting is more useful when bit-level operations like AND,
OR and XOR are avaialble. For instance I can test multiple users to see
the number of days three users visited the site at the same time. To do
this we can take the AND of all the bitmaps, and then count the set bits.

For this reason the BITOP command was introduced:

BITOP [AND|OR|XOR|NOT] dest_key src_key1 src_key2 src_key3 ... src_keyN

In the special case of NOT (that inverts the bits) only one source key
can be passed.

The judicious use of BITCOUNT and BITOP combined can lead to interesting
use cases with very space efficient representation of data.

The implementation provided is still not tested and optimized for speed,
next commits will introduce unit tests. Later the implementation will be
profiled to see if it is possible to gain an important amount of speed
without making the code much more complex.
 2012-05-24 15:22:42 +02:00
antirez bec74b1b9c Add aof_rewrite_buffer_length INFO field. 
The INFO output, persistence section, already contained the field
describing the size of the current AOF buffer to flush on disk. However
the other AOF buffer, used to accumulate changes during an AOF rewrite,
was not mentioned in the INFO output.

This commit introduces a new field called aof_rewrite_buffer_length with
the length of the rewrite buffer.
 2012-05-24 15:22:39 +02:00
antirez edb952532d Allow an AOF rewrite buffer > 2GB (Fix for issue #504). 
During the AOF rewrite process, the parent process needs to accumulate
the new writes in an in-memory buffer: when the child will terminate the
AOF rewriting process this buffer (that ist the difference between the
dataset when the rewrite was started, and the current dataset) is
flushed to the new AOF file.

We used to implement this buffer using an sds.c string, but sds.c has a
2GB limit. Sometimes the dataset can be big enough, the amount of writes
so high, and the rewrite process slow enough that we overflow the 2GB
limit, causing a crash, documented on github by issue #504.

In order to prevent this from happening, this commit introduces a new
system to accumulate writes, implemented by a linked list of blocks of
10 MB each, so that we also avoid paying the reallocation cost.

Note that theoretically modern operating systems may implement realloc()
simply as a remaping of the old pages, thus with very good performances,
see for instance the mremap() syscall on Linux. However this is not
always true, and jemalloc by default avoids doing this because there are
issues with the current implementation of mremap().

For this reason we are using a linked list of blocks instead of a single
block that gets reallocated again and again.

The changes in this commit lacks testing, that will be performed before
merging into the unstable branch. This fix will not enter 2.4 because it
is too invasive. However 2.4 will log a warning when the AOF rewrite
buffer is near to the 2GB limit.
 2012-05-24 15:22:35 +02:00
antirez 8152d0c046 Dead code removed from replication.c. 
The user @jokea noticed that the following line of code into
replication.c made little sense:

    addReplySds(slave,sdsempty());

Investigating a bit I found that this was introduced by commit 6208b3a7
three years ago in the early stages of Redis. The code apparently is not
useful at all, so I'm removing it.

This change will not be backported into 2.4 so that in the rare case
this should introduce a bug, we'll have a chance to detect it into the
development branch. However following the code path it seems like the
code is not useful at all, so the risk is truly small.
 2012-05-24 15:22:31 +02:00
antirez 9b6035bc7f TODO file removed. 
The list of things to do is since long time in two places:

1) Github issues.
2) I've a private TOOD list of random ideas, what makes sense is later
moved to github issues. So github is anyway the true source of things to
do.
 2012-05-23 11:52:47 +02:00
antirez 27fc5bf582 Use comments to split aof.c into sections. 
This makes the code more readable, it is still not the case to split the
file itself into three different files, but the logical separation
improves the readability especially since new commits are going to
introduce an additional section.
 2012-05-23 11:52:40 +02:00
antirez dfa90b5969 Redis test: include bug report on crash. 
Due to a change in the format of the bug report in case of crash of
failed assertion the test suite was no longer able to properly log it.
Instead just a protocol error was logged by the Redis TCL client that
provided no clue about the actual problem.

This commit resolves the issue by logging everything from the first line
of the log including the string REDIS BUG REPORT, till the end of the
file.
 2012-05-23 11:52:35 +02:00
jokea a5a037bf81 Set fd to writable when poll(2) detects POLLERR or POLLHUP event. 2012-05-23 11:33:25 +02:00
antirez 4dada1b5bc Fixed issue #516 (ZINTERSTORE mixing sets and zsets). 
Weeks ago trying to fix an harmless GCC warning I introduced a bug in
the ziplist-encoded implementations of sorted sets.

The bug completely broke zuiNext() iterator, that is used in the
ZINTERSTORE and ZUNIONSTORE implementation, so those two commands are no
longer reliable starting from Redis version 2.4.12 and latest 2.6.0-RC
releases.

This commit fixes the problem and adds a regression test.
 2012-05-23 11:12:38 +02:00
antirez afc6bd1b48 Deleted jemalloc.orig from /deps. 
In the commit upgrading jemalloc to version 3.0.0 I added the old
version of Jemalloc in the 'jemalloc.orig' directory for an error.
This commit removes the not useful version of jemalloc.
 2012-05-16 12:22:29 +02:00
antirez 4934f93dfb Jemalloc updated to 3.0.0. 
Full changelog here:

http://www.canonware.com/cgi-bin/gitweb.cgi?p=jemalloc.git;a=blob_plain;f=ChangeLog;hb=master

Notable improvements from the point of view of Redis:

1) Bugfixing.
2) Support for Valgrind.
3) Support for OSX Lion, FreeBSD.
 2012-05-16 11:20:44 +02:00
Salvatore Sanfilippo 14beba7868 Merge pull request #496 from pietern/2.6-makeinstall 
Fix `install` target on OSX (see #495)
 2012-05-15 02:18:03 -07:00
Salvatore Sanfilippo 404223547c Merge pull request #510 from pietern/2.6-eventport 
Support for Illumos event ports
 2012-05-15 02:16:55 -07:00
Pieter Noordhuis d318803f94 Whitespace 2012-05-14 11:06:34 -07:00
Dave Pacheco 72f30bcd30 use port_getn instead of port_get 2012-05-14 11:01:35 -07:00
Dave Pacheco 2aa1efb8a5 first cut at event port support 2012-05-14 11:01:33 -07:00
antirez e67d014d9a Added time.h include in redis-cli. 
redis-cli.c uses the time() function to seed the PRNG, but time.h was
not included. This was not noticed since sys/time.h is included and was
enough in most systems (but not correct). With Ubuntu 12.04 GCC
generates a warning that made us aware of the issue.
 2012-05-14 17:43:31 +02:00
antirez eee680541b activeExpireCycle(): better precision in max time used. 
activeExpireCycle() can consume no more than a few milliseconds per
iteration. This commit improves the precision of the check for the time
elapsed in two ways:

1) We check every 16 iterations instead of the main loop instead of 256.
2) We reset iterations at the start of the function and not every time
   we switch to the next database, so the check is correctly performed
   every 16 iterations.
 2012-05-14 17:43:26 +02:00
antirez a8a981a834 Impovements for: Redis timer, hashes rehashing, keys collection. 
A previous commit introduced REDIS_HZ define that changes the frequency
of calls to the serverCron() Redis function. This commit improves
different related things:

1) Software watchdog: now the minimal period can be set according to
REDIS_HZ. The minimal period is two times the timer period, that is:

    (1000/REDIS_HZ)*2 milliseconds

2) The incremental rehashing is now performed in the expires dictionary
as well.

3) The activeExpireCycle() function was improved in different ways:

- Now it checks if it already used too much time using microseconds
  instead of milliseconds for better precision.
- The time limit is now calculated correctly, in the previous version
  the division was performed before of the multiplication resulting in
  a timelimit of 0 if HZ was big enough.
- Databases with less than 1% of buckets fill in the hash table are
  skipped, because getting random keys is too expensive in this
  condition.

4) tryResizeHashTables() is now called at every timer call, we need to
   match the number of calls we do to the expired keys colleciton cycle.

5) REDIS_HZ was raised to 100.
 2012-05-14 17:43:23 +02:00
antirez f7f2b2610e Redis timer interrupt frequency configurable as REDIS_HZ. 
Redis uses a function called serverCron() that is very similar to the
timer interrupt of an operating system. This function is used to handle
a number of asynchronous things, like active expired keys collection,
clients timeouts, update of statistics, things related to the cluster
and replication, triggering of BGSAVE and AOF rewrite process, and so
forth.

In the past the timer was called 1 time per second. At some point it was
raised to 10 times per second, but it still was fixed and could not be
changed even at compile time, because different functions called from
serverCron() assumed a given fixed frequency.

This commmit makes the frequency configurable, so that it is simpler to
pick a good tradeoff between overhead of this function (that is usually
very small) and the responsiveness of Redis during a few critical
circumstances where a lot of work is done inside the timer.

An example of such a critical condition is mass-expire of a lot of keys
in the same second. Up to a given percentage of CPU time is used to
perform expired keys collection per expire cylce. Now changing the
REDIS_HZ macro it is possible to do less work but more times per second
in order to block the server for less time.

If this patch will work well in our tests it will enter Redis 2.6-final.
 2012-05-14 17:43:07 +02:00
antirez f078d5628d Comment improved so that the code goal is more clear. Thx to @agladysh. 2012-05-12 09:33:29 +02:00
antirez 3a401464e9 More incremental active expired keys collection process. 
If a large amonut of keys are all expiring about at the same time, the
"active" expired keys collection cycle used to block as far as the
percentage of already expired keys was >= 25% of the total population of
keys with an expire set.

This could block the server even for many seconds in order to reclaim
memory ASAP. The new algorithm uses at max a small amount of
milliseconds per cycle, even if this means reclaiming the memory less
promptly it also means a more responsive server.
 2012-05-12 09:33:24 +02:00
antirez 064223107e If the computer running the Redis test is slow, we revert to --clients 1 to avoid false positives. 2012-05-12 09:33:19 +02:00
antirez 25496f4700 redis-cli pipe mode: handle EINTR properly as well so that SIGSTOP/SIGCONT are handled correctly. 2012-05-12 09:33:15 +02:00
antirez 346825c7ed redis-cli pipe mode: handle EAGAIN while writing to socket. 2012-05-12 09:33:11 +02:00
antirez dd4e8203b2 redis-cli --pipe for mass import. 2012-05-12 09:33:06 +02:00
antirez 91d18504c2 Fix PREFIX typo in Makefile. 2012-05-09 20:45:13 +02:00
antirez f580a3e3a0 Allow PREFIX to be overrided in Makefile. 2012-05-09 10:34:58 +02:00
antirez 8afa5b7072 Fixed typo in RC3 release notes. 2012-05-06 10:22:07 +02:00
antirez af2455be29 More complete release notes for 2.5.9 2012-05-06 10:15:40 +02:00
antirez 184b8e78c6 Redis 2.5.9 (2.6 RC3). 2012-05-06 10:11:54 +02:00
Salvatore Sanfilippo d23d73c746 Merge pull request #500 from pietern/2.6-zipfix 
Compare integers in ziplist regardless of encoding
 2012-05-06 01:03:15 -07:00
Pieter Noordhuis 0ef889274f Compare integers in ziplist regardless of encoding 
Because of the introduction of new integer encoding types for ziplists
in the 2.6 tree, the same integer value may have a different encoding in
different versions of the ziplist implementation. This means that the
encoding can NOT be used as a fast path in comparing integers.
 2012-05-04 17:26:24 -07:00
Pieter Noordhuis 1b1cf8e7bf Fix install target on OSX (see #495) 2012-05-04 11:54:25 -07:00
quiver 7f10703b33 fix several bugs of init.d scripts 
- PIDFILE environ variable was not properly retrieved
- chkconfig command failed
 2012-05-04 13:42:03 +02:00
antirez 0cf10e8e86 syncio.c read / write functions reworked for correctness and performance. 
The new implementation start reading / writing before blocking with
aeWait(), likely the descriptor can accept writes or has buffered data
inside and we can go faster, otherwise we get an error and wait.

This change has effects on speed but also on correctness: on socket
errors when we perform non blocking connect(2) write is performed ASAP
and the error is returned ASAP before waiting.

So the practical effect is that now a Redis slave is more available if it
can not connect to the master, previously the slave continued to block on
syncWrite() trying to send SYNC, and serving commands very slowly.
 2012-05-02 22:45:12 +02:00
antirez 9b43b1ef4d Remove useless trailing space in SYNC command sent to master. 2012-05-02 21:48:08 +02:00
antirez 0b08d64882 Use specific error if master is down and slave-serve-stale-data is set to no. 
We used to reply -ERR ... message ..., now the reply is
instead -MASTERDOWN ... message ... so that it can be distinguished
easily by the other error conditions.
 2012-05-02 17:14:45 +02:00
antirez 96d9c6cb8b Add a note abotu read-only slaves in 2.6 release notes. 2012-05-02 16:35:53 +02:00
antirez 0f07781538 Redis 2.5.8 (2.6.0 RC2). 2012-05-02 12:17:21 +02:00
antirez 1858da2faa Test "Turning off AOF kills the background writing child if any" is now more reliable. 2012-05-02 11:40:55 +02:00
Salvatore Sanfilippo 83b1092cf2 Merge pull request #488 from pietern/2.6-safekeys 
Use safe dictionary iterator from KEYS (fixes #487)
 2012-05-01 01:50:25 -07:00
Pieter Noordhuis 9311d2b527 Use safe dictionary iterator from KEYS 
Every matched key in a KEYS call is checked for expiration. When the key
is set to expire, the call to `getExpire` will assert that the key also
exists in the main dictionary. This in turn causes a rehashing step to
be executed. Rehashing a dictionary when there is an iterator active may
result in the iterator emitting duplicate entries, or not emitting some
entries at all. By using a safe iterator, the rehash step is omitted.
 2012-04-30 10:16:20 -07:00
Harmen 8520066d7b Show problem with 'keys' command with specific command sequence. 2012-04-30 09:51:23 -07:00
antirez 8f984bef29 Properly wait the slave to sync with master in BRPOPLPUSH test. 2012-04-30 11:32:02 +02:00
antirez dd418873db A more lightweight implementation of issue 141 regression test. 2012-04-29 17:16:47 +02:00
antirez b330de57ff Added "read-only slaves" in new features section of 2.6 release notes. 2012-04-27 23:06:02 +02:00
antirez 18759c927a yet another typo fixed in release notes. 2012-04-27 17:04:03 +02:00
antirez 8b97442c20 Fixed release notes typo 2012-04-27 17:01:44 +02:00
antirez 7c5d96d98e Redis 2.5.7 (2.6 RC1) 2012-04-27 16:40:07 +02:00
antirez 603adb2b29 memtest.c fixed to actually use v1 and v2 in memtest_fill_value(). 2012-04-27 16:28:31 +02:00
antirez 748f206e3d Release notes updated with the new 2.6 features. 2012-04-27 16:07:42 +02:00
antirez 2c0aae760f redis-cli commands description in help.h updated. 2012-04-27 15:57:17 +02:00
antirez b1aa7183d7 Update makefile dependencies. 2012-04-27 15:56:16 +02:00
antirez 77a75fdef5 Set LUA_MASKCOUNT hook more selectively. Fixes issue #480. 
An user reported a crash with Redis scripting (see issue #480 on
github), inspection of the kindly provided strack trace showed that
server.lua_caller was probably set to NULL. The stack trace also slowed
that the call to the hook was originating from a point where we just
used to set/get a few global variables in the Lua state.

What was happening is that we did not set the timeout hook selectively
only when the user script was called. Now we set it more selectively,
specifically only in the context of the lua_pcall() call, and make sure
to remove the hook when the call returns. Otherwise the hook can get
called in random contexts every time we do something with the Lua
state.
 2012-04-27 11:47:30 +02:00
antirez e69e76d442 Re-introduce -g -rdynamic -ggdb when linking, fixing strack traces. 
A previous commit removed -g -rdynamic -ggdb as LDFLAGS, not allowing
Redis to produce a stack trace wth symbol names on crash.
This commit fixes the issue.
 2012-04-27 11:47:27 +02:00
antirez 9db4cea235 Produce the stack trace in an async safe way. 2012-04-27 11:47:22 +02:00
antirez a28ab2a92b Don't use an alternative stack for SIGSEGV & co. 
This commit reverts most of c575766202, in
order to use back main stack for signal handling.

The main reason is that otherwise it is completely pointless that we do
a lot of efforts to print the stack trace on crash, and the content of
the stack and registers as well. Using an alternate stack broken this
feature completely.
 2012-04-27 11:47:18 +02:00
antirez b1ee7da75a Redis test: More reliable BRPOPLPUSH replication test. 
Now it uses the new wait_for_condition testing primitive.
Also wait_for_condition implementation was fixed in this commit to properly
escape the expr command and its argument.
 2012-04-27 11:47:15 +02:00
antirez d7bad544dc Redis test: scripting EVALSHA replication test more reliable. 
A new primitive wait_for_condition was introduced in the scripting
engine that makes waiting for events simpler, so that it is simpler to
write tests that are more resistant to timing issues.
 2012-04-27 11:47:12 +02:00
David Tran 8111a803cb Spelling: s/synchrnonization/synchronization 2012-04-27 11:47:08 +02:00
antirez c16bf717de Explicitly use bash for install_server.sh. Fixes issue #397 2012-04-24 19:35:40 +02:00
antirez 12a042f863 redis-check-dump now is RDB version 6 ready. 2012-04-24 19:35:36 +02:00
antirez 717145c1a3 Spurious debugging printf removed. 2012-04-24 19:35:33 +02:00
antirez dcd4efe9ef Added two new encodings to ziplist.c 
1) One integer "immediate" encoding that can encode from 0 to 12 in the
encoding byte itself.
2) One 8 bit signed integer encoding that can encode 8 bit signed small
integers in a single byte.

The idea is to exploit all the not used bits we have around in a
backward compatible way.
 2012-04-24 19:35:29 +02:00
antirez 62bfa66284 rdbLoad() should check REDIS_RDB_VERSION instead of hardcoded number. 2012-04-24 19:35:25 +02:00
antirez dd51571cc3 ziplist.c: added comments about the new 24 bit encoding. 2012-04-24 19:35:22 +02:00
Grisha Trubetskoy ad91404a32 Add a 24bit integer to ziplists to save one byte for ints that can 
fit in 24 bits (thanks to antirez for catching and solving the two's compliment
bug).

Increment REDIS_RDB_VERSION to 6
 2012-04-24 19:35:18 +02:00
antirez 9de5d4600a A few compiler warnings suppressed. 2012-04-24 19:35:03 +02:00
antirez 38b60dea54 Fix and refactoring of code used to get registers on crash. 
This fixes compilation on FreeBSD (and possibly other systems) by
not using ucontext_t at all if HAVE_BACKTRACE is not defined.
Also the ifdefs to get the registers are modified to explicitly test for the
operating system in the first level, and the arch in the second level
of nesting.
 2012-04-24 19:34:15 +02:00
antirez 69b30cfcb6 Ziplist encoding now tested with negative integers as well. 2012-04-24 19:34:08 +02:00
antirez 0a8a1e78dc New time limit for protocol desync test set to 30 seconds to reduce false positives. 2012-04-24 19:34:03 +02:00
antirez 537dafab84 Remove loadfile() access from the scripting engine. 2012-04-24 19:33:58 +02:00
Michael Schlenker 7d6bf7956e Replace unnecessary calls to echo and cat 
Tcl's exec can send data to stdout itself, no need to call cat/echo for
that usually.
 2012-04-24 19:33:54 +02:00
antirez e337b26050 Even inside #if 0 comments are comments. 2012-04-21 21:49:34 +02:00
antirez 590d55a206 Limit memory used by big SLOWLOG entries. 
Two limits are added:

1) Up to SLOWLOG_ENTRY_MAX_ARGV arguments are logged.
2) Up to SLOWLOG_ENTRY_MAX_STRING bytes per argument are logged.
3) slowlog-max-len is set to 128 by default (was 1024).

The number of remaining arguments / bytes is logged in the entry
so that the user can understand better the nature of the logged command.
 2012-04-21 21:00:33 +02:00
antirez c11a01a030 redis.conf AOF section comments improved. 2012-04-21 12:17:33 +02:00
antirez 30e89410d4 README now makes clear that our support for solaris derived systems is "best effort". 2012-04-21 12:17:29 +02:00
antirez abfd08f5ad New tests related to scripts max execution time. 2012-04-19 23:49:46 +02:00
antirez ca577d162a SHUTDOWN NOSAVE now can stop a non returning script. Issue #466. 2012-04-19 23:36:04 +02:00
antirez d54943b76d Currenly not used code in dict.c commented out. 2012-04-18 23:56:15 +02:00
antirez c3312760fe Tests for scripting PRNG. 2012-04-18 23:50:27 +02:00
antirez 1d82bbd432 cr16.c removed from 2.6 branch, was not used. 2012-04-18 23:41:00 +02:00
antirez 0122cc4f42 redis-cli --bigkeys output modified to be simpler to read.. 2012-04-18 21:23:42 +02:00
antirez e10768518c redis-cli --bigkeys 2012-04-18 21:23:39 +02:00
antirez 5c45ae1f7b Test SDIFF with first set empty. 2012-04-18 21:23:35 +02:00
antirez 7a2065ef33 Test SINTER against same integer elements, but different set encoding. 2012-04-18 21:23:31 +02:00
antirez eb624e3416 Test SINTER with non existing key. 2012-04-18 21:23:15 +02:00
antirez ff5e31f74b Added an SMOVE test where src and dest key are the same. 2012-04-18 21:23:07 +02:00
antirez 24982f2bbc New hash fuzzing test. 2012-04-18 21:23:04 +02:00
antirez 60ef787efa Document mostly dead code in RPOPLPUSH implementation. 2012-04-18 21:23:00 +02:00
antirez bec200ec39 Explicit RPOP/LPOP tests. 2012-04-18 21:22:56 +02:00
antirez a00fcaa671 Test LINSERT syntax error. 2012-04-18 21:22:52 +02:00
antirez 8d12645569 Test LINDEX out of range index. 2012-04-18 21:22:48 +02:00
Salvatore Sanfilippo d00f51405b Merge pull request #455 from pietern/2.6-mkopts 
Persist Makefile flags (2.6)
 2012-04-18 07:26:36 -07:00
antirez ae55245d1c Marginally cleaner lookupKeyByPattern() implementation. 
just fieldobj itself as sentinel of the fact a field object is used or
not, instead of using the filed length, that may be confusing both for
people and for the compiler emitting a warning.
 2012-04-18 11:41:06 +02:00
antirez 212bb9ca2e More robust maxclients test. 2012-04-18 11:41:03 +02:00
antirez 727d6dd52a Two small fixes to maxclients handling. 
1) Don't accept maxclients set to < 0
2) Allow maxclients < 1024, it is useful for testing.
 2012-04-18 11:40:59 +02:00
antirez a1090c1193 Added test for SORT corner case: pattern ending with just "->". 2012-04-17 18:26:56 +02:00
antirez 68ee18558a lookupKeyByPattern() used by SORT GET/BY rewritten. Fixes issue #460. 
lookupKeyByPattern() was implemented with a trick to speedup the lookup
process allocating two fake Redis obejcts on the stack. However now that
we propagate expires to the slave as DEL operations the lookup of the
key may result into a call to expireIfNeeded() having the stack
allocated object as argument, that may in turn use it to create the
protocol to send to the slave. But since this fake obejcts are
inherently read-only this is a problem.

As a side effect of this fix there are no longer size limits in the
pattern to be used with GET/BY option of SORT.

See https://github.com/antirez/redis/issues/460 for bug details.
 2012-04-17 18:26:52 +02:00
antirez 96aeca4b9d Less false positives in maxclients test, hopefully. 2012-04-17 10:04:59 +02:00
Pieter Noordhuis d0cd262fdf Persist make settings and trigger rebuild if anything changes 2012-04-13 18:05:51 -07:00
Pieter Noordhuis 864229585a Don't set flags recursively 2012-04-13 18:05:51 -07:00
Pieter Noordhuis 3126e08757 Remove unused LIBS variable 2012-04-13 18:05:51 -07:00
Pieter Noordhuis 61e8825e89 First set defaults, then do composition 2012-04-13 18:05:51 -07:00
Pieter Noordhuis 95bc195158 Question mark assignment is not needed 2012-04-13 18:05:51 -07:00
Pieter Noordhuis cb481f432a Ignore gcov/lcov artifacts 2012-04-13 18:05:51 -07:00
Pieter Noordhuis 0b27a55f30 The lcov target shouldn't clean 
This is not needed because every change in compiler/linker flags
triggers a cleanup.
 2012-04-13 18:05:51 -07:00
Pieter Noordhuis 6cedb4d489 Rename ADD_*FLAGS -> REDIS_*FLAGS, REDIS_*FLAGS -> FINAL_*FLAGS 
This reflects that REDIS_*FLAGS will only be used for compilation of
Redis and not for its dependencies. Similarly, that FINAL_*FLAGS are
composed of other variables and holds the options that are finally
passed to the compiler and linker.
 2012-04-13 18:05:51 -07:00
antirez 59333ffd37 New test for scripting engine: DECR_IF_GT. 2012-04-13 16:23:46 +02:00
antirez 9a2dd1eff9 EVAL errors are more clear now. 2012-04-13 16:23:42 +02:00
antirez c9853f537b Use Lua tostring() before concatenation. 2012-04-13 16:23:38 +02:00
antirez d63a1716eb Tests modified to match the new global protection implementation. 2012-04-13 16:23:34 +02:00
antirez d2906893e8 mt.declared is no longer needed. 
Lua global protection can now be simpified becuase we no longer have the
global() function. It's useless to occupy memory with this table, it is
also not faster because the metamethods we use are only called when a
global object does not exist or we are trying to create it from a
script.
 2012-04-13 16:23:29 +02:00
antirez 97cab30993 Stop access to global vars. Not configurable. 
After considering the interaction between ability to delcare globals in
scripts using the 'global' function, and the complexities related to
hanlding replication and AOF in a sane way with globals AND ability to
turn protection On and Off, we reconsidered the design. The new design
makes clear that there is only one good way to write Redis scripts, that
is not using globals. In the rare cases state must be retained across
calls a Redis key can be used.
 2012-04-13 16:23:21 +02:00
antirez 6255a5ae66 Globals protection global() function modified for speed and correctness. 2012-04-13 16:22:47 +02:00
antirez e387dc52a0 Tests for lua globals protection. 2012-04-13 16:22:43 +02:00
antirez 3e6a4463e0 Scripting: globals protection can now be switched on/off. 2012-04-13 16:22:35 +02:00
antirez 430602b26c Protect globals access in Lua scripting. 2012-04-13 16:21:46 +02:00
antirez 236adc2809 A few var names fixed in Makefile. 
I modified it for error in a previous commit doing search & replace.
 2012-04-13 16:15:39 +02:00
antirez d298825803 Print arch bits with redis-server -v 2012-04-13 16:12:32 +02:00
antirez 206568257a memtest.c: integer overflow fixed. 2012-04-13 16:12:28 +02:00
antirez 50fb330399 Make gcov fixed. 2012-04-13 16:12:24 +02:00
antirez 69ac4d063d Makefile now introduces Redis-specific CFLAGS / LDFLAGS. Gcov target fixed. Added comments to describe how it works. 2012-04-13 16:12:04 +02:00
Erik Dubbelboer b9cd703b5b added explanation for the magic 511 backlog number 2012-04-11 18:32:10 +02:00
antirez 79e3df9d72 Makefile.dep updated. 2012-04-11 18:30:23 +02:00
antirez d5ec389585 make dep: redirect output to Makefile.dep. 2012-04-11 18:29:31 +02:00
antirez 336ba6a152 Make inline functions rioRead/Write/Tell static. This fixes issue #447. 2012-04-11 12:03:25 +02:00
antirez ef278d110f Macros ULONG_ONEZERO / ULONG_ZEROONE were inverted in #ifdef to test 32/64 bit arch. 2012-04-11 12:03:21 +02:00
Pieter Noordhuis de07849e0d Clean up Makefiles 
Remove unused variables. Instead of overriding non-standard variables
such as ARCH and PROF, use standard variables CFLAGS and LDFLAGS to
override Makefile settings. Move dependencies generated by `make dep` to
a separate file.
 2012-04-11 12:01:29 +02:00
Pieter Noordhuis e74bec56fa Everything x86 is little endian 2012-04-11 11:59:53 +02:00
antirez 3ba5eab774 Minor MIGRATE implementation simplification about ttl handling. 2012-04-10 20:04:21 +02:00
antirez b9aa332843 Check write(2) return value to avoid warnings, because in this context failing write is not critical. 2012-04-10 20:02:33 +02:00
antirez 3f64694e71 Version 2.5.6. 2012-04-10 16:34:51 +02:00
antirez fdf8bd4025 Test for maxclients. 2012-04-10 16:28:21 +02:00
antirez e95740392b dump/restore fixed to use the new crc64 API. 2012-04-10 16:27:36 +02:00
antirez 39d1e350d9 It is now possible to enable/disable RDB checksum computation from redis.conf or via CONFIG SET/GET. Also CONFIG SET support added for rdbcompression as well. 2012-04-10 16:27:03 +02:00
antirez 7f4f86f427 RDB files now embed a crc64 checksum. Version of RDB bumped to 5. 2012-04-10 16:26:58 +02:00
antirez 1bcb45d118 Fixed compilation of new rio.c changes (typos and so forth.) 2012-04-10 16:26:52 +02:00
antirez b4b923b04b Add checksum computation to rio.c 2012-04-10 16:26:45 +02:00
antirez bb99f42596 crc64.c modified for incremental computation. 2012-04-10 16:26:41 +02:00
antirez 9ba4d5a3bb rio.c file somewhat documented so that the casual reader can understand what's going on without reading the code. 2012-04-10 16:26:37 +02:00
antirez 08211b25d3 Added new test to check that "CONFIG appendonly no" actually kills the background AOF operation in progress if any. 2012-04-08 10:43:33 +02:00
antirez 55951f9005 For coverage testing use exit() instead of _exit() when termiating saving children. 2012-04-08 10:43:29 +02:00
antirez 2cf3f071a5 Tests for MONITOR. 2012-04-07 11:27:04 +02:00
antirez b162e6f133 New client info field added to CLIENT LIST output: multi, containing the length of the current pipeline. Test modified accordingly. 2012-04-07 11:27:00 +02:00
antirez 4f0bd607d9 Never used function stringObjectEqualsMs() removed. 2012-04-07 11:26:56 +02:00
antirez 7dc1d2ba17 Removed dead code: function rdbSaveTime() is no longer used since RDB now saves expires in milliseconds. 2012-04-07 11:26:52 +02:00
antirez eb6bc2e047 Two new tests for BGREWRTIEAOF. 
Check for scheduled rewrite if a BGSAVAE is in progress.
Check for error if a rewrite is already in progress.
 2012-04-07 11:26:47 +02:00
antirez 3984108474 redis.tcl: no longer leave unread replies if an error happens during a MULTI/EXEC block. 2012-04-07 11:26:42 +02:00
antirez 8d3f5ce9db New INFO field in persistence section: bgrewriteaof_scheduled. 2012-04-06 21:13:33 +02:00
antirez 8382453ecf version bumped to 2.5.5 2012-04-06 12:27:31 +02:00
jokea a3ec16f0c2 implement aeWait using poll(2). Fixes issue #267. 2012-04-06 11:49:16 +02:00
antirez 3aad0de2e9 expireGenericCommand(): better variable names and a top-comment that describes the function's behavior. 2012-04-06 00:40:36 +02:00
Premysl Hruby 81a28fe131 for (p)expireat use absolute time, without double recomputation 2012-04-06 00:40:31 +02:00
Premysl Hruby b518a83525 fix mstime() ommited while comparing if key is already expired 2012-04-06 00:40:26 +02:00
Premysl Hruby 9f899440d2 add support for generation of lcov coverage reports 2012-04-06 00:40:22 +02:00
Premysl Hruby 937abebbe1 future-proof version comparison 2012-04-05 11:04:28 +02:00
antirez c537980d98 On slow computers, 10 seconds are not enough for this heavy replication test. 2012-04-05 11:04:23 +02:00
Premysl Hruby c9adc81e2e allocate alternate signal stack, change of sigaction flags for sigterm 2012-04-05 11:04:19 +02:00
Premysl Hruby b811b334ae in kill_server send the signal once, then wait for up to 5sec before sending lethal SIGKILL 2012-04-05 11:04:14 +02:00
Premysl Hruby 4d57e44839 new option for choosing number of test clients to run 2012-04-05 11:04:09 +02:00
Premysl Hruby 6c52d5ce27 remove mentions of VM in comments 2012-04-05 11:04:03 +02:00
antirez 09eb448746 Structure field controlling the INFO field master_link_down_since_seconds initialized correctly to avoid strange INFO output at startup when a slave has yet to connect to its master. 2012-04-04 18:33:13 +02:00
antirez fa2a27cfeb New "os" field in INFO output providing information about the operating system. 2012-04-04 18:33:09 +02:00
antirez a5b75e9a86 SLAVEOF is not a write command. 2012-04-04 18:33:04 +02:00
antirez df35d87336 Print milliseconds of the current second in log lines timestamps. Sometimes precise timing is very important for debugging. 2012-04-04 18:32:59 +02:00
antirez 3c413d59c8 redis-cli help.h updated. Script to generate it updated as well. 2012-04-03 15:29:47 +02:00
antirez e3fd3ccd72 More MIGRATE tests. 2012-04-03 15:14:03 +02:00
antirez fb8409a5c8 Another fix for MIGRATE. 2012-04-03 15:13:50 +02:00
antirez bde80cb28b Two fixed for MIGRATE: fix TTL propagation and fix transferring of data bigger than 64k. 2012-04-03 14:57:28 +02:00
antirez fbce475270 When the user-provided 'maxclients' value is too big for the max number of files we can open, at least try to search the max the OS is allowing (in steps of 256 filedes). 2012-04-03 14:55:18 +02:00
antirez 31e2156c4b redis-trib.rb removed from 2.6 branch. 2012-04-03 10:31:51 +02:00
antirez b739506936 MIGRATE test modified because the implementation changed. 2012-04-02 16:41:11 +02:00
antirez 37cc07dd41 MIGRATE now let the client distinguish I/O errors and timeouts from other erros. 2012-04-02 16:41:02 +02:00
antirez 285df7b0e0 Version bumped 2.4.5 2012-04-02 14:27:04 +02:00
antirez 8cf8974a03 DUMP, RESTORE, MIGRATE tests. 2012-04-02 14:22:07 +02:00
antirez bff31e129d New verions of DUMP, RESTORE, MIGRATE back ported from unstable to 2.6 2012-04-02 14:21:59 +02:00
antirez 179ee2db78 CRC64 implementation added to Redis code base. 2012-04-02 13:53:43 +02:00
antirez 11dae1711f Write RDB magic using a REDIS_RDB_VERSION define that is defined inside rdb.h 2012-04-02 13:52:42 +02:00
ThePicard f08aa2bf48 Fixed a typo in install_server.sh 2012-04-01 10:15:38 +02:00
antirez fa34ba3908 syncio.c calls in replication.c fixed for the new millisecond timeout API. 2012-03-31 11:26:46 +02:00
antirez 4ccf671cc6 Better syncio.c with millisecond resolution. 2012-03-31 11:26:42 +02:00
antirez 88bd32f1b7 Purely aesthetic code change. 2012-03-30 10:40:04 +02:00
Joseph Jang ae15f75089 Fixed a memory leak with replication 
occurs when two or more dbs are replicated and at least one of them is >db10
 2012-03-30 10:40:01 +02:00
antirez 0f51e3c564 Regression test for issue 417 (memory leak when replicating to DB with id >= 10) 2012-03-30 10:39:56 +02:00
antirez f1e38b352e Fixed typo in comment: "te" -> "the". 2012-03-29 09:34:07 +02:00
antirez ed4d4f1145 Fix for slaves chains. Force resync of slaves (simply disconnecting them) when SLAVEOF turns a master into a slave. 2012-03-29 09:34:03 +02:00
antirez ee704a0ff1 Test for redis.sha1hex(). 2012-03-29 09:33:57 +02:00
Nathan Fritz 2f2e6ad487 added redis.sha1hex(string) as lua scripting function. 
(The original implementation was modified by @antirez to conform Redis
coding standards.)
 2012-03-29 09:33:52 +02:00
antirez e51f7d2c17 Fixes for redisLogFromHandler(). 2012-03-28 14:57:09 +02:00
antirez be4f8cccaa Log from signal handlers is now safer. 2012-03-28 14:57:04 +02:00
antirez f3e159bc93 Redis test: regexp to check if valgrind reported errors modified. Now we no longer look at the total count because this includes "possibly lost" bytes that are not interesting for Redis (tons of false positives because of how sds.c works). 2012-03-28 13:18:32 +02:00
antirez dd896e1b8d define zlibc_free() in a way that is not shadowed by jemalloc. 2012-03-28 13:18:29 +02:00
antirez 59d884af8c Produce the watchlog warning log in a way that is safer from a signal handler. Fix a memory leak in the backtrace generation function. 2012-03-28 13:18:25 +02:00
antirez e4cd5838fc Mask SIGALRM everything but in the main thread. 
This is required to ensure that the signal will be delivered to the main
thread when the watchdog timer expires.
 2012-03-28 13:18:21 +02:00
antirez 1507be5b01 Correctly set the SIGARLM timer for the software watchdog. 2012-03-28 13:18:17 +02:00
antirez fc030ac7de Redis software watchdog. 2012-03-28 13:18:13 +02:00
huangz1990 085aaef325 fix typo 2012-03-27 23:04:40 +02:00
Premysl Hruby 92dd4e4301 fix time() instead of mstime() in expireIfNeeded 2012-03-27 23:04:34 +02:00
Premysl Hruby 56ff70f8e0 use server.unixtime instead of time(NULL) where possible (cluster.c not checked though) 2012-03-27 23:04:27 +02:00
Premysl Hruby 38d6976c03 declare hashDictType as external too 2012-03-27 23:04:21 +02:00
Premysl Hruby efd412f908 remove disk-store related comments 2012-03-27 23:04:16 +02:00
antirez 9ea95e6c64 SIGSEGV handler refactored so that we can reuse stack trace and current client logging functionalities in other contexts. 2012-03-27 10:33:45 +02:00
antirez d0407c2d85 CONFIG RESETSTAT resets two more fields. 2012-03-25 11:43:42 +02:00
antirez 81f32c7b65 New INFO field aof_delayed_fsync introduced. 
This new field counts all the times Redis is configured with AOF enabled and
fsync policy 'everysec', but the previous fsync performed by the
background thread was not able to complete within two seconds, forcing
Redis to perform a write against the AOF file while the fsync is still
in progress (likely a blocking operation).
 2012-03-25 11:43:38 +02:00
antirez 754643c1c7 Version is now 2.5.3. 2012-03-25 11:18:24 +02:00
antirez 0387e9c199 convert-zipmap-hash-on-load false positive fixed. 
Apparently because the sample RDB file was not copied before every test
Redis had a chance to replace it with a newly written one, so that the
next test could fail.
 2012-03-25 10:57:34 +02:00
antirez 29e1976855 When running the test in valgrind mode, pass the right flags to show memory leaks stack traces but only including the "definitely lost" items. 2012-03-24 12:06:56 +01:00
antirez 188a17ed56 Add used allocator in redis-server -v output. 2012-03-24 11:52:56 +01:00
antirez 9542d9d8d7 convert-zipmap-hash-on-load test enabled 2012-03-24 11:42:39 +01:00
antirez 2d04eef425 Contextualize comment. 2012-03-23 20:20:43 +01:00
antirez 925927fc8f Merge remote-tracking branch 'origin/2.6' into 2.6 2012-03-23 20:20:13 +01:00
antirez 27688de10e RDB load of different encodings test added. 2012-03-23 20:22:22 +01:00
antirez 9b962d10a4 Fixed memory leak in hash loading. 2012-03-23 20:21:19 +01:00
antirez ab0603812d RDB load of different encodings test added. 2012-03-23 15:23:01 +01:00
antirez b64281cc0e Big endian fix. The bug was introduced because of a typo. 2012-03-23 12:50:19 +01:00
antirez 52192552bd Replicate HINCRBYFLOAT as HSET. 2012-03-23 10:24:25 +01:00
antirez b54cdfb2b7 Code style hack. 2012-03-22 18:16:41 +01:00
antirez 7b22c44ccd Result of INCRBYFLOAT and HINCRBYFLOAT should never be in exponential form, and also should never contain trailing zeroes. This is not possible with vanilla printf() format specifiers, so we alter the output. 2012-03-22 18:14:32 +01:00
antirez 9aba884b34 Comments about security of slave-read-only in redis.coinf. 2012-03-22 10:09:58 +01:00
antirez 3f7ad83398 Correctly create shared.oomerr as an sds string. 2012-03-22 10:09:54 +01:00
antirez 38bb45223a DEBUG should not be flagged as w otherwise we can not call DEBUG DIGEST and other commands against read only slaves. 2012-03-22 10:09:49 +01:00
antirez 054061685a Support for read-only slaves. Semantical fixes. 
This commit introduces support for read only slaves via redis.conf and CONFIG GET/SET commands. Also various semantical fixes are implemented here:

1) MULTI/EXEC with only read commands now work where the server is into a state where writes (or commands increasing memory usage) are not allowed. Before this patch everything inside a transaction would fail in this conditions.

2) Scripts just calling read-only commands will work against read only
slaves, when the server is out of memory, or when persistence is into an
error condition. Before the patch EVAL always failed in this condition.
 2012-03-22 10:09:43 +01:00
antirez 7c1cec2525 Redis cluster stuff removed from 2.6 redis.conf file. 2012-03-20 13:07:26 +01:00
antirez 0380c13bbe redis_init_script template updated. 2012-03-20 13:06:50 +01:00
antirez 518e7202b2 Fixed typo in 2.6 release notes. 2012-03-19 22:10:18 +01:00
antirez f597910121 Merge remote-tracking branch 'origin/2.6' into 2.6 2012-03-19 19:30:30 +01:00
antirez 24b0942275 Suppress warnings compiling redis-cli with certain gcc versions. 2012-03-19 19:29:06 +01:00
antirez 6e6bbac7a5 Read-only flag removed from PUBLISH command. 2012-03-19 19:18:51 +01:00
antirez a7ef5ce1b0 Memory addressing test implemented. 2012-03-19 14:02:57 +01:00
antirez ea693f0282 More memory tests implemented. Default number of iterations lowered to a more acceptable value of 50. 2012-03-18 18:03:49 +01:00
antirez 74760d3ccd Fixed typo. 2012-03-18 17:28:03 +01:00
antirez 32f62ed6d0 Number of iteration of --test-memory is now 300 (several minutes per gigabyte). Memtest86 and Memtester links are also displayed while running the test. 2012-03-18 17:24:48 +01:00
antirez d4e6ce3e97 On crash suggest to give --test-memory a try. 2012-03-18 11:43:09 +01:00
antirez d605fdabfa Memory test function now less boring thanks to screen-wide progress bar. 2012-03-18 11:43:02 +01:00
antirez f4df22d1c5 Hem... actual memtest.c file added. 2012-03-18 11:42:56 +01:00
antirez 78d6a02b0c First implementation of --test-memory. Still a work in progress. 2012-03-18 11:42:47 +01:00
antirez f1eaf57200 Fix for issue #391. 
Use a simple protocol between clientsCron() and helper functions to
understand if the client is still valind and clientsCron() should
continue processing or if the client was freed and we should continue
with the next one.
 2012-03-15 20:51:39 +01:00
antirez 749817b7c3 Version bumped to 2.5.2 2012-03-14 15:37:47 +01:00
antirez 9fa9ccb04e Reclaim space from the client querybuf if needed. 2012-03-14 15:36:27 +01:00
antirez 6934832e53 sds.c: sdsAllocSize() function added. 2012-03-14 15:36:24 +01:00
antirez f9322fb8ed sds.c new function sdsRemoveFreeSpace(). 
The new function is used in order to resize the string allocation so
that only the minimal allocation possible is used, removing all the free
space at the end of the string normally used to improve efficiency of
concatenation operations.
 2012-03-14 15:36:20 +01:00
antirez 5e473cd8fd Call all the helper functions needed by clientsCron() as clientsCronSomething() for clarity. 2012-03-14 15:36:16 +01:00
antirez 3e8fcb6d03 CLIENT LIST test modified to reflect the new output. 2012-03-13 18:08:59 +01:00
antirez cfa4b57cb0 Process async client checks like client timeouts and BLPOP timeouts incrementally using a circular list. 2012-03-13 18:08:53 +01:00
antirez 57a5e54ddc Added a qbuf-free field to CLIENT LIST output. 2012-03-13 18:08:50 +01:00
antirez 41e8e5cb8e Client creation time in redisClient structure. New age field in CLIENT LIST output. 2012-03-13 18:08:47 +01:00
antirez 56de4964ee c->bufpos initialization moved for aesthetics. 2012-03-13 18:08:43 +01:00
antirez a74ab6478c RDB hashes loading, fixed another bug in the loading of HT-encoded hashes: when the hash entry is too big for ziplist, add the field, then convert. The code used to break before the new entry was inserted, resulting into missing fields in the loaded Hash object. 2012-03-13 11:01:08 +01:00
antirez ee61a4b99e RDB hashes loading fixed removing the assertion that failed every time an HT-encoded hash was loaded. 2012-03-13 09:49:21 +01:00
quiver dfc2545470 fix typo of redis.conf 2012-03-10 13:23:31 +01:00
antirez b014c1f211 RDB4 support in redis-check-dump. 2012-03-10 12:38:59 +01:00
antirez 37180ed9cd RDB version is no 4, because small hashes are now encoded as ziplists, so older versions of Redis will not understand this format. 2012-03-10 12:35:15 +01:00
antirez 7551f2a0b1 Version is now 2.5.1, first unstable release of Redis 2.6 2012-03-10 12:29:47 +01:00
antirez 00b3336334 Build dependencies updated. 2012-03-10 12:28:14 +01:00
antirez 571e257db1 Redis 2.6 branch obtained from unstable removing all the cluster related code. 2012-03-10 12:26:37 +01:00
antirez 87faf90696 hash-max-zipmap-... renamed hash-max-ziplist-... in defalt conf for tests. 2012-03-10 12:14:17 +01:00
antirez 753bb3dcbd More vertical space saved. 2012-03-10 11:19:17 +01:00
antirez c0caa1cf54 Minor code aesthetic change to use Redis code base style rule of saving vertical space when possible. 2012-03-10 11:09:43 +01:00
antirez d22248ae99 Removed handling of deprecated hash-max-zipmap-entries nad hash-map-zipmap-value. Pieter is too good with users ;). Better to have them switch to a saner configuration ASAP after the 2.6 upgrade. 2012-03-10 10:41:39 +01:00
antirez addc032756 Added a top-function comment to rioWriteHashIteratorCursor() to better specify what the function does. Not immediately clear from the name. 2012-03-10 10:36:51 +01:00
antirez 8562798308 Merge conflicts resolved. 2012-03-09 22:07:45 +01:00
antirez 250e7f6908 Instantaneous ops/sec figure in INFO output. 2012-03-08 16:15:37 +01:00
antirez 0823e48fb9 Support for all the redis.conf fields in CONFIG GET. config.c refactored a bit. 2012-03-08 12:14:23 +01:00
antirez 91d664d6ce run_id added to INFO output. 
The Run ID is a field that identifies a single execution of the Redis
server. It can be useful for many purposes as it makes easy to detect if
the instance we are talking about is the same, or if it is a different
one or was rebooted. An application of run_id will be in the partial
synchronization of replication, where a slave may request a partial sync
from a given offset only if it is talking with the same master. Another
application is in failover and monitoring scripts.
 2012-03-08 10:13:36 +01:00
antirez 44f508f1a8 clusterGetRandomName() generalized into getRandomHexChars() so that we can use it for the run_id field as well. 2012-03-08 10:08:44 +01:00
antirez 4d3bbf3590 By default Redis refuses writes with an error if the latest BGSAVE failed (and at least one save point is configured). However people having good monitoring systems may prefer a server that continues to work, since they are notified that there are problems by their monitoring systems. This commit implements the ability to turn the feature on or off via redis.conf and CONFIG SET. 2012-03-07 18:02:26 +01:00
antirez c25e7eafef Refuse writes if can't persist on disk. 
Redis now refuses accepting write queries if RDB persistence is
configured, but RDB snapshots can't be generated for some reason.
The status of the latest background save operation is now exposed
in the INFO output as well. This fixes issue #90.
 2012-03-07 13:05:53 +01:00
antirez e31b615e62 Better MONITOR output, now includes client ip:port or the lua string if the command was executed by the scripting engine. 2012-03-07 12:12:15 +01:00
antirez 7b845b6228 anetPeerToString() automatically populates ip/port with something that may be provided to the user as output in case of errors. 2012-03-07 11:30:30 +01:00
antirez 9494f1f15b TIME command. 2012-03-07 10:38:01 +01:00
antirez 60893c6cc6 redis-cli: CSV output added, used for the --slave mode. 2012-02-29 17:43:06 +01:00
antirez b8283ab218 Initial implementation of redis-cli --slave support. 2012-02-29 17:10:21 +01:00
antirez a950a84303 Ping the slave using the standard protocol instead of the inline one. 2012-02-29 16:33:54 +01:00
antirez c1db214eeb Better implementation for BRPOP/BLPOP in the non blocking case. 2012-02-29 14:41:57 +01:00
antirez cd8bdea31b lpush arguments vector rewrite modified for more speed and to memory leak removal. 2012-02-29 13:38:30 +01:00
antirez b67feecacd Version bumped to 2.9.5 2012-02-29 00:54:52 +01:00
antirez 78d6a22dc3 Better system for additional commands replication. 
The new code uses a more generic data structure to describe redis operations.
The new design allows for multiple alsoPropagate() calls within the scope of a
single command, that is useful in different contexts. For instance there
when there are multiple clients doing BRPOPLPUSH against the same list,
and a variadic LPUSH is performed against this list, the blocked clients
will both be served, and we should correctly replicate multiple LPUSH
commands after the replication of the current command.
 2012-02-29 00:46:50 +01:00
antirez eeb34eff52 Added a new API to replicate an additional command after the replication of the currently executed command, in order to propagte the LPUSH originating from RPOPLPUSH and indirectly by BRPOPLPUSH. 2012-02-28 18:03:08 +01:00
antirez d8b1228bf6 propagate() prototype added to redis.h 2012-02-28 16:20:41 +01:00
antirez edba65d090 Var renamed into pushGenericCommand() to better reflect what it means. 2012-02-28 16:17:55 +01:00
antirez ad08d059d0 Added command propagation API. 2012-02-28 16:17:00 +01:00
Salvatore Sanfilippo 64ef44d568 Merge pull request #362 from fawek/mac-build-fix 
Fix the build on Mac OS X, where features.h is non-existent
 2012-02-26 07:18:43 -08:00
Jakub Wieczorek 1d6628c088 Fix the build on Mac OS X, where features.h is non-existent 2012-02-26 15:13:05 +01:00
Premysl Hruby 80ff1fc6d0 use gcc sync builtins for memory counter when possible 2012-02-26 11:45:52 +01:00
Premysl Hruby 59132e4212 fix redis-benchmark memory leak 2012-02-26 10:01:27 +01:00
antirez 36e6637dfd lua_cmsgpack.c added 2012-02-24 15:46:25 +01:00
antirez 63505e0b59 Lua_cmsgpack added to Redis scripting. 2012-02-24 15:45:16 +01:00
antirez f6da155bde redis-benchmark: a few fixes to pipelining implementation. 2012-02-23 15:39:26 +01:00
antirez b9474282f8 redis-benchmark: Initial support for pipelining. 2012-02-23 15:02:43 +01:00
antirez 62c394e219 unstable version called 2.9.4 2012-02-22 17:44:18 +01:00
antirez 92e984dbcd The universe is 80 columns wide, accept it. 2012-02-22 17:22:40 +01:00
Salvatore Sanfilippo dbb27a0a90 Merge pull request #304 from bradvoth/unstable 
Simple install script changes
 2012-02-22 08:19:54 -08:00
antirez a5bd08487f Fix for issue #306, thanks to tchajed (on github) for the pull request. The original patch was reworked a bit. 2012-02-22 16:07:06 +01:00
antirez 4eb3b3e970 Redis-cli should accept help command even if there is no connection to the server. 2012-02-22 15:16:11 +01:00
antirez fcdeb98568 even less false positive on obuf-limits test. 2012-02-22 11:25:30 +01:00
antirez 467d703ead Merge remote-tracking branch 'origin/unstable' into unstable 2012-02-22 09:52:29 +01:00
antirez feb282883b zzlIsInRange() now is capable of handling empty sorted sets that may end inside the data set when loading very old RDB files produced by early-stage versions of Redis. 2012-02-22 09:52:10 +01:00
Salvatore Sanfilippo 30c06ae6b4 Merge pull request #347 from pietern/unstable-segv 
Force SIGSEGV without HAVE_BACKTRACE (unstable)
 2012-02-21 10:29:55 -08:00
Pieter Noordhuis ca2344f991 Solaris has backtrace(3) 2012-02-21 10:20:28 -08:00
Pieter Noordhuis 57be47758e Also force SIGSEGV without HAVE_BACKTRACE 2012-02-21 10:20:01 -08:00
antirez 7c96b467c1 Fixed undefined behavior in *INCR style functions overflow detection. Sorry clang! 2012-02-21 18:26:11 +01:00
antirez fe7be46025 bzero -> memset 2012-02-21 10:06:04 +01:00
antirez ad0df18fdc Fix 32bit build of hiredis 2012-02-19 20:49:20 +01:00
antirez 8e65b6a33d Applied a few modifications to hiredis to tune it for speed (redis-benchmark) and ability to read more deeply nested replies (redis-cli). 2012-02-19 20:33:59 +01:00
antirez b66e5add82 The hiredis lib shipped with Redis was updated to latest version. 2012-02-19 20:26:50 +01:00
Pieter Noordhuis cb598cdd59 Don't expire keys when loading an RDB after a SYNC 
The cron is responsible for expiring keys. When keys are expired at
load time, it is possible that the snapshot of a master node gets
modified. This can in turn lead to inconsistencies in the data set.

A more concrete example of this behavior follows. A user reported a
slave that would show an monotonically increase input buffer length,
shortly after completing a SYNC. Also, `INFO` output showed a single
blocked client, which could only be the master link. Investigation
showed that indeed the `BRPOP` command was fed by the master. This
command can only end up in the stream of write operations when it did
NOT block, and effectively executed `RPOP`. However, when the key
involved in the `BRPOP` is expired BEFORE the command is executed, the
client executing it will block. The client in this case, is the master
link.
 2012-02-16 17:21:13 +01:00
antirez d347348109 Issue #330 regression test. 2012-02-16 16:13:40 +01:00
antirez a400a9b2d7 Now HINCRBY can detect overflows too. Fix for issue #330. 2012-02-16 16:09:08 +01:00
antirez 6e09ad1c15 Return ASAP from checkClientOutputBufferLimits() if c->reply_bytes is zero. 2012-02-16 12:20:56 +01:00
antirez 48a32944e6 Merge branch 'issue327' into unstable 2012-02-16 09:40:27 +01:00
antirez 3b0600c777 Add stdint.h in endianconv.h to fix issue #336. 2012-02-15 12:21:04 +01:00
antirez 4a701b3801 redis-check-aof is now large files safe also on 32 bit systems. 2012-02-14 19:57:51 +01:00
antirez 120a36f22b add -f flag to cp when installing, otherwise stopping the server is 
needed when installing a new Redis version. Thanks to Scott Kevill.
Fixes issue #335.
 2012-02-14 16:15:24 +01:00
antirez 7a3e372025 endian.c/h -> endianconv.c/h to avoid issues with broken libraries search paths. 2012-02-14 16:11:46 +01:00
antirez 18aa2b87b6 Merge remote-tracking branch 'origin/unstable' into unstable 2012-02-14 16:02:04 +01:00
antirez 6136a16bd1 intset.c endianess fixes. 2012-02-14 15:35:50 +01:00
Salvatore Sanfilippo 5e985e795d Merge pull request #334 from lsbardel/quantredis 
added lua struct c extension
 2012-02-13 15:05:59 -08:00
lsbardel 2f75bbab02 added lua struct c extension 2012-02-13 21:05:21 +00:00
antirez 66d1b021ec ziplist.c endianess fixes, chapter 5. 2012-02-09 17:09:01 +01:00
antirez cab1105c6e ziplist.c endianess fixes, chapter 4. 2012-02-09 16:36:25 +01:00
antirez 3fa19b7dfc ziplist.c endianess fixes, chapter 3. 2012-02-09 16:28:35 +01:00
antirez 8e0ef249a2 more ziplist.c endianess fixes 2012-02-08 23:20:39 +01:00
antirez 5653847714 ziplist.c fixes for bigendian 2012-02-08 22:59:35 +01:00
antirez ac834d237a A few small BSD related fixes. 2012-02-08 22:24:59 +01:00
antirez f9ef912c66 more practical maxmemory+slaves hint in redis.conf 2012-02-08 00:20:46 +01:00
antirez cebb7b92ce redis.conf updated with new maxmemory semantics 2012-02-08 00:17:27 +01:00
antirez b129c6df45 debugging messages removed from freeMemoryIfNeeded() 2012-02-08 00:10:20 +01:00
antirez 609baba8a2 Fixes to c->reply_bytes computation, and debug messages to closely study the behavior of memory pressure + slaves + maxmemory + blocked slaves. 2012-02-07 17:41:31 +01:00
antirez 01e95705f8 Fixes to 2.6 release notes file 2012-02-07 15:08:38 +01:00
antirez 442246dde2 Precision of getClientOutputBufferMemoryUsage() greatily improved, see issue #327 for more information. 2012-02-07 13:05:36 +01:00
antirez 8b7c3455b9 freeMemoryIfNeeded() minor refactoring 2012-02-06 16:56:42 +01:00
antirez c1ef6ffe8a Also remove size of AOF buffers from used memory when doing the math for freeMemoryIfNeeded() 2012-02-06 16:35:43 +01:00
antirez c2513ecb98 A first (work in progress) release notes for 2.6 2012-02-05 11:08:01 +01:00
antirez 3508899944 INSTALL now redirects the user to README 2012-02-05 09:38:41 +01:00
antirez 7441fcdd56 Redis Manifesto moved from src to root dir 2012-02-05 09:37:08 +01:00
antirez f6b32c14f4 This fixes issue #327, is a very complex fix (unfortunately), details: 
1) sendReplyToClient() now no longer stops transferring data to a single
client in the case we are out of memory (maxmemory-wise).

2) in processCommand() the idea of we being out of memory is no longer
the naive zmalloc_used_memory() > server.maxmemory. To say if we can
accept or not write queries is up to the return value of
freeMemoryIfNeeded(), that has full control about that.

3) freeMemoryIfNeeded() now does its math without considering output
buffers size. But at the same time it can't let the output buffers to
put us too much outside the max memory limit, so at the same time it
makes sure there is enough effort into delivering the output buffers to
the slaves, calling the write handler directly.

This three changes are the result of many tests, I found (partially
empirically) that is the best way to address the problem, but maybe
we'll find better solutions in the future.
 2012-02-04 14:05:54 +01:00
antirez 355f859134 Use less memory when emitting the protocol, by using more shared objects for commonly emitted parts of the protocol. 2012-02-04 08:58:37 +01:00
antirez ce8b772be7 Now Lua scripts dispatch Redis commands properly calling the call() function. In order to make this possible call() was improved with a new flags argument that controls how the Redis command is executed. 2012-02-02 16:30:52 +01:00
antirez d876678b5e Set a 3.5 GB maxmemory limit with noeviction policy if a 32 bit instance without user-provided memory limits is detected. 2012-02-02 10:26:20 +01:00
antirez 75eaac5c74 Added a server.arch_bits field instead of computing it at runtime for INFO. 2012-02-02 10:23:31 +01:00
antirez b80b1c5914 Only incremnet stats for key miss/hit when the key is semantically accessed in read-only. 2012-02-01 21:51:20 +01:00
antirez c17947287a Added tests checking ability of the scripting engine to reorder the output of commands with a random output regarding signle elements position in the multi bulk reply. 2012-02-01 17:49:03 +01:00
antirez 44d77bb217 A few SORT tests made more resistant to false negatives resulitng from poor randomization of Redis hash function with one byte inputs. 2012-02-01 17:37:48 +01:00
antirez 01f75bf352 New SORT tests checking the new more deterministic behavior of SORT sorting algorithm. 2012-02-01 17:17:52 +01:00
antirez de79a2ee12 Make SORT BY <constant> STORE ... to always produce the same output by force sorting, so that we have deterministic replication of this command. 2012-02-01 17:05:45 +01:00
antirez 2c861050c1 SORT is now more deterministic: does not accept to compare by score items that have scores not representing a valid double. Also items with the same score are compared lexycographically. At the same time the scripting side introduced the ability to sort the output of SORT when sort uses the BY <constant> optimization, resulting in no specific ordering. Since in this case the user may use GET, and the result of GET can be null, converted into false as Lua data type, this commit also introduces the ability to sort Lua tables containining false, only if the first (faster) attempt at using just table.sort with a single argument fails. 2012-02-01 15:22:28 +01:00
antirez 548efd91e5 Order output of commands returning random arrays using table.sort when called from Lua, partially fixing issue #165. The issue is yet not completely fixed since we can't add the REDIS_CMD_SORT_FOR_SCRIPT flag in SORT currently, both because it may contain NULLs and because it is not cool to re-sort everything at every call when instead this should be sorted only if BY <constant> is used. 2012-01-31 16:09:21 +01:00
antirez 8dd19d818f Fixed redis-benchmark --help output typo 2012-01-31 11:43:32 +01:00
antirez 3c08fdae71 64 bit instances are no longer limited to have at max 2^32-1 elements in lists. 2012-01-31 10:35:52 +01:00
antirez fc4ed4299b minimal change to obuf-limits.tcl test to make sure there are no false positives with 32bit instances as well. 2012-01-30 21:08:10 +01:00
antirez d1c4c94ebf Merge remote-tracking branch 'origin/unstable' into unstable 2012-01-30 10:40:28 +01:00
Salvatore Sanfilippo eb7d19ed69 Merge pull request #319 from fawek/lua-error-location 
Lua reports line numbers off by one in error messages
 2012-01-30 01:40:17 -08:00
antirez 89f6f6abcf setKey(): call the higher level wrapper setModifiedKey() instead of touchWatchedKey() even if currently they are exactly the same. 2012-01-30 10:27:50 +01:00
Salvatore Sanfilippo 824ea0177f Merge pull request #321 from mkwiatkowski/ticket227 
SORT with STORE removes key if result is empty. This fixes issue #227.
 2012-01-30 01:25:34 -08:00
Michal Kwiatkowski a0bf8d0ad8 SORT with STORE removes key if result is empty. This fixes issue #227. 2012-01-30 07:36:49 +01:00
Jakub Wieczorek 4d776dba3c Lua reports line numbers off by one in error messages 2012-01-29 14:53:49 +01:00
antirez 6c29410117 false positive in expire tests mitigated with a sleep, but other solutions exist if needed later. 2012-01-26 16:45:08 +01:00
antirez 7f7a13b27b Less false positives for obuf-limits.tcl tests 2012-01-26 16:08:24 +01:00
Pieter Noordhuis d3ea4c86a8 Update default configuration 2012-01-25 13:37:43 -08:00
Pieter Noordhuis 80586cb894 Test that zipmap from RDB is correctly converted 2012-01-25 13:28:11 -08:00
antirez da9c4cc147 Added test for client output buffer limit (soft limit). 2012-01-25 18:34:56 +01:00
antirez efc8f6c1a2 Added test for client output buffer limit (hard limit). 2012-01-25 18:11:04 +01:00
antirez c715c9b8bf Fixed typo in getClientLimitClassByName() 2012-01-25 18:07:56 +01:00
antirez a30a8ae7ba Merge branch 'unstable' into limits 2012-01-25 16:59:58 +01:00
antirez 4e97c2c71a Fixed another possible bug in cluster.c found by clang --analyze. 2012-01-25 16:59:54 +01:00
antirez d329031fad Fixed another possible bug in cluster.c found by clang --analyze. 2012-01-25 16:59:36 +01:00
antirez 9b810f1cde Merge branch 'unstable' into limits 2012-01-25 16:46:53 +01:00
antirez 6710ff24d0 Fixed a non critical bug signaled by clang static analyzer thanks to Mukund Sivaraman for reporting it: there was a not initialized field populating the cluster message header, but it is always fixed at later time before sending the packet. 2012-01-25 16:46:35 +01:00
antirez c69ad1f24e Merge branch 'unstable' into limits 2012-01-25 10:41:25 +01:00
antirez ecc5702145 aeCreateEventLoop() cleanup on error unified in a single block (original 
patch by Mukund Sivaraman, modified by me to make it simpler and to use
my coding style).
 2012-01-25 10:37:32 +01:00
Mukund Sivaraman caa63a3821 If aeApiCreate() fails, there's probably not much one can do, but in the name of consistency... 2012-01-25 10:27:37 +01:00
antirez 2f0f0d95c0 lenght -> length 2012-01-24 15:33:15 +01:00
antirez 3cbce4f45e redis.conf typo fixed in the client-output-buffer-limit documentation 2012-01-24 14:46:23 +01:00
antirez 7957c67604 after all closing a client for output buffer limit overcoming is a WARNING level message. 2012-01-24 12:03:48 +01:00
antirez c8a607f2b6 client-output-buffer-limit documented in redis.conf 2012-01-24 11:23:23 +01:00
antirez 7fe8d49a70 Client output buffer limits: configuration of parameters for the different classes of clients implemented. 2012-01-24 10:43:30 +01:00
antirez 06b3dced99 asyncCloseClientOnOutputBufferLimitReached() now ignores clients with REDIS_CLOSE_ASAP flag already set. Return value of the function changed from int to void since it is not used. Fixed logging of the client scheduled to be closed. 2012-01-24 09:32:39 +01:00
antirez 51669c5ac1 client buffer handling refactoring and optimization 2012-01-23 17:15:49 +01:00
antirez 3c0602ff41 CLIENT LIST test fixed (there is a new omem field in the output) 2012-01-23 16:19:49 +01:00
antirez 8c43e66346 actually call the function to async free clients in serverCron(). 2012-01-23 16:17:22 +01:00
antirez 7eac2a75a4 Implementation of the internals that make possible to terminate clients overcoming configured output buffer (soft and hard) limits. 2012-01-23 16:12:37 +01:00
antirez 890da62eea Merge branch 'unstable' into limits 2012-01-23 10:36:07 +01:00
antirez b362c111da fixed typo in hahs function seed default value. It is no longer used but fixed to retain the old constant as default anyway. 2012-01-22 01:40:23 +01:00
antirez 06c5523a88 typo in comment fixed 2012-01-21 23:34:06 +01:00
antirez a48c8d873b Fix for hash table collision attack. We simply randomize hash table initialization value at startup time. 2012-01-21 23:30:13 +01:00
antirez 447ebf3bc7 Better looking registers/stack dump 2012-01-20 16:40:43 +01:00
antirez 632da60583 added support to dump registers on crash on Linux x32 2012-01-20 14:37:50 +01:00
antirez eea8c7a4f8 added support to dump registers on crash on Linux x64 2012-01-20 12:54:15 +01:00
antirez d4d208595c all the stack trace related functions are now in debug.c. Now Redis dumps registers and stack content on crash. Currently osx supported, adding Linux right now. 2012-01-20 12:20:45 +01:00
Brad Voth bad2b8e6ec updated utils/install_server.sh to support chkconfig boxes (redhat/ 
centos) as well as debian/ubuntu
 2012-01-18 10:09:42 -05:00
antirez 498dc5557c Introduced three client limit classes: normal, slave, pubsub 2012-01-17 12:43:01 +01:00
antirez 3853c16839 Track the length of the client pending output buffers (still to transfer) in a new field in the client structure. 2012-01-17 12:23:25 +01:00
antirez f7ccc4830b Merge remote-tracking branch 'origin/unstable' into unstable 2012-01-16 16:50:41 +01:00
antirez 4aac3ff247 It is now posible to flush all the previous saving points in redis.conf by using a save directive with a single empty string argument, like it happens for CONFIG SET save. 2012-01-16 16:50:24 +01:00
Salvatore Sanfilippo a11c874777 Merge pull request #279 from kashif/patch-1 
Fixed some typos in redis.conf.
 2012-01-16 03:44:00 -08:00
antirez ebdfad69dc Don't change the replication state if SLAVE OF is called with arguments specifying the same master we are already connected with. This fixes issues #290. 2012-01-16 11:29:47 +01:00
antirez 1596d6a61e Obsolete SDS_ABORT_ON_OOM removed from sds.c 2012-01-16 10:08:33 +01:00
antirez bd068b15c2 sds.c no longe pre-allocate more than 1MB of free space ahead. This fixes issue #252. 2012-01-16 10:03:38 +01:00
antirez 7faa1f075b added a comment on top of the zslRandomLevel() function 2012-01-16 09:39:04 +01:00
antirez 06312eed86 Possible fix for false positives in issue 141 regression test 2012-01-12 16:24:54 +01:00
antirez 00010fa96f On crash print information about the current client (if any), command vector, and object associated to first argument assuming it is a key. 2012-01-12 16:02:57 +01:00
antirez d8bd12f984 error in comment fixed 2012-01-11 20:25:51 +01:00
antirez 5db904bdc5 show GCC version in INFO output. 2012-01-10 18:37:16 +01:00
antirez 68fe1b9b99 version bumped to 2.9.3 2012-01-07 12:54:48 +01:00
antirez 414c3deac1 Regression test for the main problem causing issue #141. Minor changes/fixes/additions to the test suite itself needed to write the test. 2012-01-06 17:28:40 +01:00
antirez 06f8a473fc Regression tests for protocol desync bug related to Issue #141 2012-01-06 12:54:29 +01:00
Pieter Noordhuis fe45840201 Implements ziplistFind 
To improve the performance of the ziplist implementation, some
functions have been converted to macros to avoid unnecessary stack
movement and duplicate variable assignments.
 2012-01-03 16:13:42 -08:00
Pieter Noordhuis ebd85e9a45 Encode small hashes with a ziplist 2012-01-02 22:14:10 -08:00
Pieter Noordhuis 9ea54feef0 string2* functions take a const pointer 2012-01-02 15:24:50 -08:00
antirez 11e0c4c55b Protections against protocol desyncs, leading to infinite query buffer growing, due to nul-terms in specific bytes of the request or indefinitely long multi bulk or bulk count strings without newlines. This bug is related to Issue #141 as well. 2011-12-31 16:09:46 +01:00
antirez f42e2f1bd7 Protocol and I/O related defines moved into a separated section of redis.h 2011-12-31 15:37:33 +01:00
antirez 35c6032cfa A few no longer used defines removed from redis.h 2011-12-31 15:34:02 +01:00
antirez 1824e3a3a3 Fixed replication when multiple slaves are attaching at the same time. The output buffer was not copied correctly between slaves. This fixes issue #141. 2011-12-30 19:40:43 +01:00
Kashif Rasul 57c0cf8b85 Fixed some typos in redis.conf. 2011-12-28 22:14:23 +01:00
antirez 2ebd2720b3 Merge remote-tracking branch 'origin/unstable' into unstable 2011-12-23 09:34:28 +01:00
antirez 9678c37577 Added regression test for ZUNIONSTORE creating NaN (github issue #264) 2011-12-23 09:34:06 +01:00
antirez 256356ffd8 Prevent NaN scores in sorted sets resulting from calls to ZUNIONSTORE and ZINTERSTORE. 2011-12-23 09:27:31 +01:00
Salvatore Sanfilippo d9b952211c Merge pull request #268 from diegok/unstable 
Fixed a config docs typo.
 2011-12-22 07:00:42 -08:00
diegok ce6628dae7 Fixed a config docs typo. 2011-12-22 12:22:23 +01:00
antirez d334281c80 Merge remote-tracking branch 'origin/unstable' into unstable 2011-12-21 17:13:31 +01:00
antirez e51b79f315 sligthly better AOF related logs 2011-12-21 17:12:23 +01:00
antirez b941417c1b Log when the child rewriting the AOF is killed by stopAppendOnly() 2011-12-21 16:37:22 +01:00
antirez 1844f9900f server.replstate -> server.repl_state 2011-12-21 12:23:18 +01:00
antirez f48cd4b90c some RDB server struct fields renamed. 2011-12-21 12:22:13 +01:00
antirez ff2145adac more AOF server struct fields renamed. 2011-12-21 12:17:02 +01:00
antirez 2c915bcf6d AOF fileds in the global server state, and define names, renamed with more consistent names. More work to do. 2011-12-21 11:58:42 +01:00
antirez e394114d95 AOF refactoring, now with three states: ON, OFF, WAIT_REWRITE. 2011-12-21 10:31:34 +01:00
antirez c6ac7d0302 server.appendonly -> server.aof_state, and many comments added in the server global state structure in the process. 2011-12-21 10:05:32 +01:00
antirez b4fb720b10 A test modified to pass with the new AOF start/stop semantics. 2011-12-21 09:24:14 +01:00
antirez 202e3091d7 Redis test: when assertion fails print not just the expression but also expanded values in the error message. 2011-12-21 09:23:22 +01:00
antirez efb6022529 Do not propagate DEBUG LOADAOF 2011-12-20 17:52:57 +01:00
antirez af77acc310 startAppendOnly() fixed to set server.appendonly to 0 if the start failed. 2011-12-20 17:00:50 +01:00
antirez fc96409f91 Merge remote-tracking branch 'origin/unstable' into aof-fixes 2011-12-20 10:29:33 +01:00
Salvatore Sanfilippo 7d5209d7d3 Merge pull request #258 from miaout17/bugfix-247 
Fix issue #247 and add a test case
 2011-12-19 05:38:31 -08:00
BigCat f54c299091 I found that no test will fail even if I broke the SORT x LIMIT y z code. 
Added a naive unit test for SORT-LIMIT command.
 2011-12-19 19:49:34 +08:00
BigCat 706b32e0e0 Fix issue #247 : Accepting non-integer parameters when shouldn't 
Using `getLongFromObjectOrReply` instead of `atoi` if possible.
The following functions are modified.

* lrangeCommand
* ltrimCommand
* lremCommand
* lindexCommand
* lsetCommand
* zunionInterGenericCommand
* genericZrangebyscoreCommand
* sortCommand
 2011-12-19 19:48:35 +08:00
antirez 0e59a94728 version bumped to 2.9.2 2011-12-19 10:55:31 +01:00
antirez 3ae5a308db unit/introspection.tcl added 2011-12-19 10:21:50 +01:00
antirez f4bddefecd unit/introspection added among tests executed by default 2011-12-19 10:18:27 +01:00
antirez 0a466a7542 Fixed memleak in CLIENT INFO, added simple test that will work as regression test on mac os x and in the CI when running over valgrind. This fixes issue #256 2011-12-19 10:16:37 +01:00
antirez a244a13b4c added assertion in zslInsert() that ensures the inserted element score is not NaN 2011-12-18 11:12:58 +01:00
antirez 24accab2aa Merge remote-tracking branch 'origin/unstable' into unstable 2011-12-16 09:56:12 +01:00
antirez 18d0ef4bf0 ae.c: solved a memory leak with no practical effects (since the event 
loop is never destroyed in Redis). Thanks to @anydot for noticing it.
 2011-12-16 09:55:06 +01:00
antirez 087f414090 Fix: when aof_write_rewrite is true don't append on the AOF buffer but accumulate the differences for the rewrite. 2011-12-15 20:03:28 +01:00
antirez e7a2e7c1f7 AOF fixes in the context of replicaiton (when AOF is used by slave) and CONFIG SET appendonly yes/no. 2011-12-15 16:07:49 +01:00
antirez 36dda9554a ae_epoll.c typo introduced in the previous commit fixed. 2011-12-15 11:50:15 +01:00
antirez e074416be4 Max limit to 10k clients removed, this implements feature request on issue #194 2011-12-15 11:42:40 +01:00
antirez 503d87a818 List connected slaves with ip,port,state information in INFO, as requested by github issue #219 2011-12-14 15:11:11 +01:00
antirez 19951d965a version bumped to 2.9.1 2011-12-13 18:07:21 +01:00
antirez e2f313899b redis-cli --eval implemented 2011-12-13 16:22:28 +01:00
antirez da33edbc58 Merge remote-tracking branch 'origin/unstable' into unstable 2011-12-13 11:16:34 +01:00
antirez 54ecc0e777 Hash type AOF rewrite using HMSET. 2011-12-13 11:10:51 +01:00
antirez 7df9b1412b Sorted Set type AOF rewrite using variadic ZADD. 2011-12-13 11:10:46 +01:00
antirez 8d875ccb7a Set type AOF rewrite using variadic SADD. 2011-12-13 11:10:41 +01:00
antirez 570bbcf8e0 Added tests for AOF rewrite. 2011-12-13 11:10:36 +01:00
antirez 38c06fa037 List type AOF rewrite using variadic RPUSH for the linked list encoding. 2011-12-13 11:10:31 +01:00
antirez 5b25009656 Lists AOF rewrite using variadic RPUSH (work in progress) 2011-12-13 11:10:21 +01:00
antirez 954cc9d0f6 Redis test: vaoid two false positives while running under valgrind. 2011-12-10 13:28:32 +01:00
antirez 43093dff2d Redis test: two redundant tests removed as they tend to create issues when running the test with valgrind. 2011-12-07 18:31:39 +01:00
antirez b1d08d4540 Redis test: wait more time for the server to start if it is running using valgrind. 2011-12-07 17:51:21 +01:00
antirez dfcf5a0ac8 Redis test port allocation fixed. 2011-12-07 17:15:17 +01:00
antirez 85b69afede EXPIRE tests more tolerant to random latency while the test is running. 2011-12-07 12:44:27 +01:00
antirez 846bcd9abe Redis test: handle inability to start the server in a better way. 2011-12-07 11:47:38 +01:00
antirez fb293ccbdd Fixed a theoretical memory leak with no practical effects in ae_kqueue.c and ae_epoll.c, thanks to magicyang87 for reporting it. 2011-12-07 08:58:29 +01:00
antirez 237194b760 Fixed SORT bugs (issue #224) with regression tests. 2011-12-01 16:08:05 +01:00
antirez 1576520cc0 Document the redis-server cli options in the README 2011-12-01 15:40:49 +01:00
antirez 9268a5b5fd usage information updated to reflect the new way to pass options to Redis from the command line. 2011-12-01 13:50:36 +01:00
antirez 27ccb94a02 Even when loglevel is warning the server should log that it started. 2011-12-01 13:45:19 +01:00
antirez 67c6f0f630 Support for command line configuration options for redis-server. 2011-12-01 13:44:53 +01:00
antirez 4c8bd905a0 -h is now alias for --help when running redis-server. 2011-12-01 12:18:22 +01:00
antirez 39ca1713d7 I like when main() is the last function in the file. 2011-12-01 12:15:44 +01:00
antirez 27acd7aa89 Replication bug fixed: now non blocking connect is also forced to follow the configured replication timeout. 2011-11-30 15:35:16 +01:00
antirez c6f9ee88fa default log level is now notice. 2011-11-30 15:04:40 +01:00
antirez 38b957d8ba --quiet option implemented in the Redis test. Output improved a bit. 2011-11-29 16:29:12 +01:00
antirez 63fd13996d show initial querybuf bytes on querybuf overflow. 2011-11-28 11:12:55 +01:00
antirez e7ef418ccd two new AOF related INFO fields that can be interesting information for debugging. 2011-11-28 11:04:03 +01:00
antirez 53a2af3941 still problems with the AOF race regression test, disabling it again for now 2011-11-28 10:35:30 +01:00
antirez 86de66e32f jemalloc gitignore updated to jemalloc 2.2.5 2011-11-25 16:42:10 +01:00
Salvatore Sanfilippo 2753acf1d6 Merge pull request #208 from jbergstroem/jemalloc-2.2.5 
Update to jemalloc 2.2.5
 2011-11-25 07:29:55 -08:00
antirez 3e0a975e07 log client protocol errors for log level >= verbose 2011-11-25 16:09:34 +01:00
antirez 674df1eef5 Added regression tests for issue #209 2011-11-25 12:27:29 +01:00
Salvatore Sanfilippo 07ac2daaaa Merge pull request #209 from woowenjie/unstable 
clean REDIS_DIRTY_CAS when discard Command runs. otherwise the next MULT...
 2011-11-25 03:13:10 -08:00
woowenjie f371e721fa clean REDIS_DIRTY_CAS when discard Command runs. otherwise the next MULTI/EXEC may fail in the same RedisClient 2011-11-25 10:34:05 +08:00
antirez fa5af017d9 better bug report info on crash 2011-11-24 15:47:26 +01:00
antirez 45e7a1ce00 minor refactoring to networking.c adding a separated function to get a string representing the current state of all the connected clients. 2011-11-24 15:04:42 +01:00
antirez 2c74a9f948 last executed command in CLIENT LIST output. 2011-11-24 14:56:34 +01:00
jbergstroem 1d03c1c98a Update to jemalloc 2.2.5 2011-11-23 21:36:25 +01:00
antirez 3c95e7212e new counter in INFO output: rejected_connections with number of dropped connections because of maxclients limit reached. 2011-11-23 18:38:12 +01:00
antirez 1b2962c61f Merge remote-tracking branch 'origin/unstable' into unstable 2011-11-22 14:27:53 +01:00
antirez 1bcfa0f643 More fixes in the Redis command table, to make sure all the keys are detected by Redis Cluster. 2011-11-22 14:27:32 +01:00
Salvatore Sanfilippo 4e5f63a60c Merge pull request #199 from jerem/patch-1 
Fixed a typo in comments.
 2011-11-22 02:43:28 -08:00
antirez 39da5d1f21 Fixed command table for SETEX and PSETEX causing the expire time to be hashed by Redis cluster instead of the key. 2011-11-22 10:16:15 +01:00
antirez 7501c66f2e comment describing the Redis command table enhanced 2011-11-22 10:13:45 +01:00
antirez afd0f06b75 Fixed bug in getClientInfoString() that was not rendering the N (no flags) special flag correctly. 2011-11-21 16:19:30 +01:00
antirez becf5fdb0c Close client connection and log the event when the client input buffer reaches 1GB. 2011-11-21 16:17:51 +01:00
antirez 6621b8ffa1 show active events in client file descriptor in CLIENT LIST. 2011-11-21 16:06:03 +01:00
antirez f14479c796 new ae.c API to get current events by file descriptor. 2011-11-21 16:05:29 +01:00
antirez 491c1c4e04 added output list and buffer length, query buffer size, to CLIENT LIST output. 2011-11-21 15:54:49 +01:00
antirez 5e84f56ea6 file .prerequisites added to gitignore 2011-11-21 15:35:54 +01:00
antirez 17d25a33e2 code generating the CLIENT LIST output refactored to have a function that is able to render a single client into a client info string. 2011-11-21 15:34:32 +01:00
antirez 08a879af6e Added sdscatsds() to sds.c/h 2011-11-21 15:34:00 +01:00
antirez 7e14a20801 Fixed a comment typo in the command table. 2011-11-21 10:55:52 +01:00
Jérémy Bethmont 92a157eae5 Fixed a typo in comments. 2011-11-21 09:38:14 +08:00
antirez 40f3ff21b7 Merge remote-tracking branch 'origin/unstable' into unstable 2011-11-18 16:30:26 +01:00
antirez 0b14e44118 redis.conf updated to reflect the new scripting timeout semantics. 2011-11-18 16:29:22 +01:00
antirez 5c63922691 aof-race test reenabled now that port allocation of the test suite is smarter 2011-11-18 15:53:47 +01:00
Salvatore Sanfilippo 794e6ce14f Merge pull request #173 from jasondavies/typo 
Fix some minor typos.
 2011-11-18 05:34:56 -08:00
antirez e41677b4cf prevent (more) commands from Lua scripts 2011-11-18 14:23:38 +01:00
antirez 4ab8695d53 New script timeout semantics and SCRIPT KILL implemented. SHUTDOWN NOSAVE and SHUTDOWN SAVE implemented. 2011-11-18 14:10:48 +01:00
Salvatore Sanfilippo 5c85257b96 Merge pull request #193 from lamby/sparc-crash 
Fix crash on SPARC due to improper alignment of allocated memory
 2011-11-16 09:23:19 -08:00
Chris Lamb 5ae979bf57 Fix crash on SPARC due to improper alignment of allocated memory 
I believe that you should be able to drop 'defined(__sun)' completely
from this clause, as Solaris on x86 hardware probably does not have
strict alignment requirements, but I don't have a way to test that.

Thanks to Jurij Smakov <jurij@wooyd.org>.

Signed-off-by: Chris Lamb <lamby@debian.org>
 2011-11-16 17:21:16 +00:00
Pieter Noordhuis 42c6a5da3d Don't force rebuild when params didn't change 2011-11-16 17:49:06 +01:00
antirez bf758397a1 more valgrind (and other archs) friendly testing of floating number related features. 2011-11-16 14:40:50 +01:00
antirez cd4aba36cd valgrind handles floating point numbers differently for some reason, so using "simpler" numbers to make tests happy. 2011-11-16 13:35:22 +01:00
Salvatore Sanfilippo 32adb53156 Merge pull request #192 from pietern/makefile 
Add separate clean/distclean targets to Makefile
 2011-11-16 01:29:42 -08:00
Pieter Noordhuis a63a3d4ddb Update README on allocators 2011-11-15 13:28:57 -08:00
Pieter Noordhuis 5bb2c88e94 Rebuild source when allocator changes 
To do so, the Makefile stores the contents of the MALLOC environment
variable in a file named .make-malloc. When the contents of this file
and the MALLOC variable are not equal, it forces a rebuild of the Redis
source tree.

A side-effect of this change is that choosing an allocator can now be
done using the single MALLOC variable instead of USE_TCMALLOC,
USE_JEMALLOC and so forth. These variables continue to work for
backwards compatibility.
 2011-11-15 13:09:34 -08:00
Pieter Noordhuis 4b8a63941d Rebuild deps/ and src/ when ARCH changes 
This change moves the build instructions for dependencies to a separate
Makefile in deps/. The ARCH environment variable is stored in a
.make-arch file in the same directory as the Makefile. The contents of
this file is read and compared to the current ARCH, and, on a mismatch
triggers rebuilding the entire source tree.

When file .make-arch exists and matches with ARCH from the environment,
the dependencies are assumed to already be built.

The new "clean" target only cleans the Redis source tree, not its
dependencies. To clear the dependencies as well, the "distclean" target
can be used.
 2011-11-15 12:41:35 -08:00
Pieter Noordhuis 321a0440c0 Use CCLINK as last argument 2011-11-15 09:39:38 -08:00
Pieter Noordhuis 61c0e893aa CCLINK already includes ALLOC_LINK 2011-11-15 09:36:13 -08:00
Salvatore Sanfilippo 7b81e5a16c Merge pull request #181 from redsquirrel/unstable 
Fixing inject: it was failing when there was 1 or >2 sources
 2011-11-15 06:36:53 -08:00
antirez 3c5680ed9d HINCRBYFLOAT tests 2011-11-15 15:26:10 +01:00
antirez 68bfe993c8 HINCRBYFLOAT implemented 2011-11-15 15:09:39 +01:00
antirez d4a3cfed9c Merge branch 'unstable' into incrbyfloat 2011-11-14 15:59:56 +01:00
antirez 70023978fd test for the new more strict behavior about number parsing 2011-11-14 15:39:55 +01:00
antirez c9df799b2c INCRBYFLOAT tests 2011-11-14 15:37:13 +01:00
antirez d93f9a8644 string to number API is now more strict not accepting spaces before or after the number. A few tests converted to match the new error messages using the word float instead of double. 2011-11-14 15:34:44 +01:00
antirez 5244d6e54e rewrite INCRBYFLOAT as SETs for AOF/replication 2011-11-14 10:15:13 +01:00
antirez 5574b53eae INCRBYFLOAT implementation 2011-11-12 19:27:35 +01:00
antirez 762eea07ca Added test to make sure Redis evicts expired keys actively (and not just in a lazy fashion). 2011-11-12 11:27:38 +01:00
antirez 4be855e757 Fixed issues with expire introduced with latest millisecond resolution feature. Many time_t were not converted to long long, and one time() call was not replaced with mstime(). 2011-11-12 01:04:27 +01:00
antirez 3570629f90 set default client timeout to zero inside redis.h as well 2011-11-11 17:18:35 +01:00
antirez 64c7499eb8 new tests for EXPIRE family functions, new millisecond precision, and new millisecond argument variants. 2011-11-11 15:11:50 +01:00
antirez 12d293ca6e high resolution expires API modified to use separated commands. AOF transation to PEXPIREAT of all the expire-style commands fixed. 2011-11-10 17:52:02 +01:00
Dave Hoover 57f8021cb2 Fixing inject: it was failing when there was 1 or >2 sources 2011-11-09 19:06:12 -06:00
antirez dab5332f95 Fixed a few typos 2011-11-09 21:59:27 +01:00
antirez bdbdb02edb Fixed bug breaking rdbSaveMillisecondTime() in 32 bit systems. Thanks to @anydot (Přemysl Hrubý) 2011-11-09 18:47:48 +01:00
antirez 52d46855d9 TTL, EXPIRE and EXPIREAT now support the milliseconds input/output form 2011-11-09 18:05:35 +01:00
antirez b0b74486e1 use "EXPIREAT ... ms" form when rewriting the AOF 2011-11-09 17:20:14 +01:00
antirez 7dcc10b65e Initial support for key expire times with millisecond resolution. RDB version is now 3, new opcoded added for high resolution times. Redis is still able to correctly load RDB version 2. Tests passing but still a work in progress. API to specify milliseconds expires still missing, but the precision of normal expires is now already improved and working. 2011-11-09 16:51:19 +01:00
antirez 14ed10d957 dict set/get macros for integers fixed. 2011-11-09 13:39:59 +01:00
antirez 2c2b208537 added mstime() to get UNIX time in milliseconds. 2011-11-09 00:03:03 +01:00
antirez 6c578b764a dict.c: added macros to get signed/unsigned integer values from hash 
entry. Field name of hash entry union modified for clarity.
 2011-11-08 23:59:53 +01:00
antirez b28d0461b4 32bit build fixed, broken by a previous commit fixing build on Solaris 2011-11-08 21:00:52 +01:00
antirez aa9a61ccd7 dict.c: added macros in dict.h to set signed and unsigned 64 bit values directly inside the hash entry without using additional memory. 2011-11-08 19:41:29 +01:00
antirez c0ba9ebe13 dict.c API names modified to be more coincise and consistent. 2011-11-08 17:07:55 +01:00
antirez 71a50956b1 dict.c: added two lower level methods for directly manipulating hash entries. This is useful in order to set 64 bit integers as values directly inside the hash entry (in order to save memory), without casting, and even in 32 bit builds. 2011-11-08 16:57:20 +01:00
antirez 66d8818cb7 Merge branch 'unstable' into ttlres 2011-11-08 16:15:08 +01:00
antirez b90314588f useless double if removed. 2011-11-08 11:26:06 +01:00
antirez b0a2e34059 yet another #if REDIS_MBULK_BIG_ARG removed. 2011-11-08 11:24:12 +01:00
antirez 53272781d0 Multi bulk optimization for creating big objects without copying data is no longer optional, #ifdefs removed. Also debugging messages removed. 2011-11-08 11:22:40 +01:00
antirez 65330badb9 hiredis/redis changes for speed with big payloads: read buffer size set 
to 16k, request buffer size is no longer destroyed when emtpy and large
(better fix needed). Redis clients static output buffer set to 16k as
well.
 2011-11-08 10:59:59 +01:00
antirez d5a8018287 speed-regression.tcl: command line options to select tests, data size, and number of requests. 2011-11-07 17:18:50 +01:00
antirez 55758a5fab speed-regression.tcl: move tests, data size, requests in global vars that will be changed via command line options. 2011-11-07 16:52:55 +01:00
antirez 85bc9b06b7 speed-regression.tcl script: killing previously tested instance fixed. Don't run if there is already an instance running in the same port. 2011-11-07 16:00:15 +01:00
antirez 84c6bdfcd0 speed-regression.tcl script: obtain test names dynamically. 2011-11-07 15:35:01 +01:00
antirez d9747b496e redis-benchmark: ability to run selected tests. Better help with examples. 2011-11-07 11:29:37 +01:00
antirez 9f080a01fa first version of the speed regression test 2011-11-04 20:45:46 +01:00
antirez 7b86f5e6d5 csv output for redis-benchmark 2011-11-04 14:49:24 +01:00
antirez 410dfe90b2 PING / MSET benchmarks enabled again. 2011-11-04 11:20:19 +01:00
antirez ca908473e8 A comment moved a few lines for clarity. 2011-11-04 11:18:15 +01:00
antirez 94d490b9f6 Added a define to set the size threshold to enable the multi bulk parsing big objects optimization. 2011-11-04 11:16:11 +01:00
antirez 826b5beb9c further optimizations for the multi bulk protocol parsing code when big objects are transmitted to Redis. 2011-11-03 15:53:40 +01:00
antirez 9217095572 optimized object creation in multi-bulk protocol parsing 2011-11-02 17:30:19 +01:00
antirez b8d743e181 sdsIncrLen() / sdsMakeRoomFor() used to avoid copying to intermediate buffer while reading the client query. 2011-11-02 16:52:45 +01:00
antirez a54806ac6a testhelp.h now exits with retcode 1 on failed tests. 2011-11-02 16:52:10 +01:00
antirez dd5fbedf7b I/O buffer length enlarged 2011-11-02 16:51:33 +01:00
antirez d0b2a9b223 sdsMakeRoomFor() exposed as public API. sdsIncrLen() added. Both the changes make it possible to copy stuff from a system call to an sds buffer without the need of an additional buffer and copying overhead. 2011-11-02 16:50:59 +01:00
antirez 3526724594 added a comment to sdsMakeRoomFor() to make it clear what the function actually does. 2011-11-02 16:09:14 +01:00
antirez 6a7841eb09 added an union in the dict.h structure to store 64 bit integers directly into hash table entries. 2011-11-02 15:28:45 +01:00
Jason Davies b91cbf66bb Fix minor typos. 2011-11-01 19:58:09 +00:00
antirez ef23f3ac92 Script max execution time test disabled for now since it is no longer enforced. 2011-10-31 16:09:07 +01:00
antirez d569f39a77 removed the vmpointer structure that is no longer user in Redis >= 2.5.x 2011-10-31 15:47:12 +01:00
antirez aba4adb7dc use timeout 0 by default, as this is a common source of problems. 2011-10-31 15:43:09 +01:00
Salvatore Sanfilippo 404cd6450d Merge pull request #170 from bnoordhuis/fix-sunos-build 
build: fix sunos build, compile lua with __C99FEATURES__=1
 2011-10-31 03:21:09 -07:00
antirez 85ccd57632 added more clarifications in redis.conf about ther right value to set as replication timeout. 2011-10-31 11:17:21 +01:00
antirez aeecbdfae3 1fe4cd5 2011-10-31 11:14:24 +01:00
Herbert G. Fischer f15e33a884 fixed typo in comments 2011-10-31 11:13:37 +01:00
antirez 8996bf7720 7c6da73 2011-10-31 11:13:28 +01:00
antirez 58732c23d5 maxclients configuration is now implemented dealing with the actual process rlimits. Setting maxclients to 0 no longer makes sense and is now invalid, the new default is 10000. 
See issue #162 for more information.
 2011-10-31 10:49:27 +01:00
antirez 1eceb85ebc It is now possible to disable password authentication via CONFIG SET. Thanks to Jan Oberst. This closes issue #171. 2011-10-31 09:57:06 +01:00
Ben Noordhuis b83e95830f build: fix sunos build, compile lua with __C99FEATURES__=1 2011-10-30 03:20:00 +00:00
antirez db6a2e7f78 Fixed a memory leak in redis-cli.c 2011-10-28 17:43:04 +02:00
antirez e1cf460cf1 sds.c single quotes support 2011-10-28 17:31:40 +02:00
antirez d8ba159b6c use the more generic -BUSY error for idle scripts 2011-10-27 20:56:12 +02:00
antirez 115e3ff39e If a Lua script executes for more time than the max time specified in the configuration Redis will log a warning, and will start accepting queries (re-entering the event loop), returning -SLOWSCRIPT error for all the commands but SHUTDOWN that remains callable. 2011-10-27 14:49:10 +02:00
Nathan Florea bb48c5fa67 Set errno to 0 before calling strtol. 2011-10-26 13:01:24 +02:00
antirez b508aeb994 Clear the AOF rewrite scheduled flag once an AOF rewrite is triggered. Fix for issue #161, probably fixing 159 as well. 2011-10-26 12:53:30 +02:00
antirez e089e61019 Don't timeout scripts if the instance is a slave. 2011-10-25 19:54:10 +02:00
antirez e5abf6ef19 SCRIPT LOAD now returns the SHA1 instead of +OK 2011-10-25 14:46:15 +02:00
antirez 8cb8d417b1 Lau scripts default max execution time set to 5 seconds. 2011-10-25 12:10:15 +02:00
antirez e8c993f0fb Fixes for the scripting refactoring and new commands. Tests for the new features. 2011-10-25 11:19:15 +02:00
antirez a9b07ac459 SCRIPT LOAD implemented, scripting eval command refactoring. 2011-10-25 10:25:59 +02:00
antirez 070e39454d SCRIPT command for introspection and control of the scripting environment. 2011-10-24 22:47:00 +02:00
antirez ca1f766a55 Better installation info inside README file. 2011-10-23 12:29:26 +02:00
antirez a84e7e46ec removed a few entries for gitignore 2011-10-23 12:16:07 +02:00
antirez f013f40003 Fixed a few warnings compiling on Linux. 2011-10-23 10:57:01 +02:00
antirez 9e6a9f30ea A past commit removed the inclusion of redis.h from rdb.c, completely breaking 32 bit builds under Linux. 2011-10-23 10:42:16 +02:00
antirez 9ed32ba083 Redis.call is now split into two variants of the same function. Redis.call will raise an error by default. Redis.pcall will return the error object instead. 2011-10-20 16:02:23 +02:00
antirez 2f5abbfeb7 Merge branch 'unstable' of github.com:/antirez/redis into unstable 2011-10-20 11:18:10 +02:00
antirez 73fac227a0 use signalModifiedKey to set the key as dirty in the context of WATCH for both MIGRATE and RESTORE. 2011-10-20 11:17:30 +02:00
Salvatore Sanfilippo 3a02bd96e7 Merge pull request #147 from janoberst/unstable 
Cleaned up redis root directory a tiny bit and changed links to github issues
 2011-10-20 00:53:22 -07:00
antirez 34296ae890 Lua cjson include paths now use the local includes. 2011-10-19 16:57:54 +02:00
antirez 8f2af26c2c Merge remote-tracking branch 'origin/unstable' into unstable 2011-10-19 16:42:35 +02:00
antirez 15108778d5 JSON support for Lua scripting, based on work from @lp, thanks!. We are using the good and fast cjson by Mark Pulford. 2011-10-19 16:42:10 +02:00
Jan Oberst 61e02c4f6c Switched issues URL to Github in BUGS 2011-10-18 14:28:23 -04:00
Jan Oberst 3fb1c8a0bc Actually let's just cpy over the README from stable. Missed the updated version there. 2011-10-18 14:28:07 -04:00
Jan Oberst e491a1a1a8 Cleaned up redis root directory and changed links for redis.io and github issues and pull requests 2011-10-18 14:14:16 -04:00
Salvatore Sanfilippo d0dc3d8204 Merge pull request #146 from smly/fixed-redis-check-dump 
Memory leak in redis-check-dump
 2011-10-18 05:13:25 -07:00
smly 046f70f758 Fixed a small memory leak I found with valgrind 2011-10-18 20:18:26 +09:00
antirez 76e772f39a Return from syncWithMaster() ASAP if the event fired but the instance is no longer a slave. This should fix Issue #145. 2011-10-18 11:15:11 +02:00
antirez 6856c7b4d6 First implementation of the ASKING command. Semantics still to verify. 2011-10-17 17:35:23 +02:00
antirez e0aab1fc79 MIGRATE +NOKEY error was missing the final CRLF causing a protocol error. 2011-10-17 16:44:08 +02:00
antirez bfbc16ae83 Fixed a typo causing segfault on MIGRATE 2011-10-17 16:39:05 +02:00
antirez c0875a77a1 Regression test for issue #142 added 2011-10-17 10:41:46 +02:00
antirez 13cd1515f9 FLUSHALL now prevents rdbSave() from resetting the dirty counter, so that the command will get replicated and put inside the AOF. This fixes issue #142 2011-10-17 10:31:47 +02:00
antirez 42a6fcd6c5 FLUSHALL will only perform a blocking SAVE if RDB persistence is configured. 2011-10-17 10:31:34 +02:00
antirez 6d61e5bf5b Exit with Fatal error at startup if the RDB file signature or version is wrong. 
Ref: issue #103
 2011-10-14 16:59:38 +02:00
antirez 2e63cfe20d redis-check-dump: RDB version 2 now supported. 2011-10-14 16:52:52 +02:00
antirez 80ad7189e8 More informative error when DEBUG RELOAD fails. 2011-10-14 14:31:33 +02:00
antirez dd9ad57062 redis-trib: fix for a slot allocation bug. 2011-10-13 15:16:03 +02:00
antirez 2b9ce0192e small comment added 2011-10-13 14:51:29 +02:00
antirez ab52d1f4a8 Fix for issue #132. Now AUTH raises an error if no server password is configured. 2011-10-10 22:21:17 +02:00
Salvatore Sanfilippo 70cb03e172 Merge pull request #126 from florean/unstable 
Unlink Unix socket file on shutdown
 2011-10-10 13:05:16 -07:00
Nathan Florea 8523876503 Added a config directive for a Unix socket mask 
Added a configuration directive to allow a user to specify the
permissions to be granted to the Unix socket file.  I followed
the format Pieter and Salvatore discusses in issue #85 (
https://github.com/antirez/redis/issues/85).
 2011-10-10 11:21:15 -07:00
antirez d83eda48e9 Issue #131. stime/utime reported in INFO was inverted. Fixed thanks to Didier Spezia. 2011-10-10 15:31:55 +02:00
antirez 4ab18a3331 Fix for bug #128 about the RENAME command. 2011-10-10 15:21:19 +02:00
antirez 02a3e58258 Added two new tests for RENAME, currently both will fail because of bug #128. 2011-10-10 12:01:13 +02:00
Nathan Florea 56209f720a Unlink Unix socket file on shutdown 2011-10-09 00:46:03 -07:00
antirez ff15dba093 not used parameter for redisLog() removed. Error message grammar fixed. 
Thanks to @ardsrk.
 2011-10-08 10:59:04 +02:00
antirez d38ef52085 Redis Cluster: process node to node CLUSTERMSG_TYPE_PUBLISH messages and send it to the local clients. 2011-10-07 16:34:16 +02:00
antirez c563ce463b propagate PUBLISH messages using the redis cluster nodes bus. Still need to process the incoming packets of that type. Work in progress. 2011-10-07 15:37:34 +02:00
antirez 623131d408 Totally hackish and dirty, but working, support for Redis Cluster in redis-cli 2011-10-05 19:55:33 +02:00
antirez 9465d83efd A node can populate a slot if a message from a trusted slot is received claiming ownership of this slot, that is currently empty or served by a node in FAIL state. However this feature was broken since calling clusterAddSlot() was not enough as the slot bit is already set in the node that pong us. We need to directly alter the table. This commit fixes the issue. 2011-10-05 17:40:02 +02:00
antirez dbf6bca431 Merge remote-tracking branch 'origin/unstable' into unstable 2011-10-05 16:08:54 +02:00
antirez 2aa26d2a63 CONFIG SET/GET support for loglevel 2011-10-05 16:08:35 +02:00
antirez ad7a86fbe0 just minor aesthetic changes to cluster.c 2011-10-05 16:02:45 +02:00
Salvatore Sanfilippo 093433c16d Merge pull request #118 from dvirsky/unstable 
fixes to install scripts
 2011-10-05 03:49:52 -07:00
antirez eab0e26e03 replaced redisAssert() with redisAssertWithInfo() in a shitload of places. 2011-10-04 18:43:03 +02:00
antirez bab205f787 redisAssertWithClientInfo() is now redisAssertWithInfo() that is also able to report an optional object. The client is also optional. Specifying NULL will prevent dumping the not available information (either client or object). 2011-10-04 18:05:26 +02:00
antirez e3e6993510 Introduced a redisAssert() variant that is able to show information about the client in the context where the failed assertion was detected. 2011-10-04 17:22:29 +02:00
dvir volk 9a01957f68 fixes to install script 2011-10-04 16:58:55 +02:00
antirez 909aee2c6c redis-trib: create subcommand fixed calling node info loading method. Was needed after recent refactoring. 2011-10-04 15:53:07 +02:00
antirez 24bfb570ee Redis test ports selection made more robust. This prevents the test from hanging if an already bound port is selected but the TCP server listening to it does not cause a protocol error with a Redis client PING. Also base port moved away from the range near to the Redis Cluster gossip ports. 2011-10-04 10:05:21 +02:00
antirez 0bb5160cb0 Revert "Use the new install script as make install target. Message about install script requiring root changed a bit to make it more evident." 
After talking with Pieter he changed my mind about this, it is better to
have a simpler install script that works everywhere, and the complex one
can be always executed if needed by hand. We'll make possibly a new
target for the full featured installation script, and even suggest it
after a Make install.

This reverts commit f1e60d7530.
 2011-10-03 16:04:44 +02:00
antirez 37d650032a Rewrite MIGRATE as DEL for AOF/replication. Also increment the dirty counter to both force replication and persistence. 2011-10-03 15:51:10 +02:00
antirez 2a95c944ff increment server.dirty in restore command to make sure it is replicated and increments the dirty count for persistence concerns. 2011-10-03 15:45:14 +02:00
Salvatore Sanfilippo cd73df62af Merge pull request #114 from pietern/unstable-zcount 
Use rank information to compute ZCOUNT
 2011-10-03 06:20:23 -07:00
Pieter Noordhuis 0cfc894066 Remove ZCOUNT branches from generic RANGEBYSCORE code 2011-10-03 14:23:31 +02:00
Pieter Noordhuis 62d774e5ba Use element rank instead of iterating in ZCOUNT 2011-10-03 14:14:43 +02:00
antirez f1e60d7530 Use the new install script as make install target. Message about install script requiring root changed a bit to make it more evident. 2011-10-03 13:28:54 +02:00
dvir volk 78f56a5aee changed license to gpl BSD. LOL 2011-10-03 11:35:30 +02:00
dvir volk c01043baf1 changed license to gpl v3 2011-10-03 11:29:01 +02:00
dvir volk 001f8da256 fixes to install script and template 2011-10-03 10:58:43 +02:00
dvir volk 9210e70173 Added installer and config template renderer 2011-10-03 10:48:07 +02:00
antirez 13c7e5ef29 Exact variant of CRC16 specified into crc16.c 2011-10-02 14:05:29 +02:00
antirez 66c146592a when processing gossip packets make sure to also update the node bitmap when associating slots to nodes. Fixed simply using the appropriate helper function to add a slot. 2011-10-01 15:11:07 +02:00
antirez 7112580cb3 fixed a few problems with redis-trib resharding. 2011-10-01 01:31:46 +02:00
antirez a0189bbe75 redis-trib: initial working implementation of cluster rehashing, more work needed to make it more correct, more reliable. 2011-09-30 22:18:07 +02:00
antirez d38d2fdfdc Fixed two bugs in the CLUSTER SETSLOT ... NODE subcommand 2011-09-30 22:17:24 +02:00
antirez 9a38df8799 redis-trib: actual slot migration work in progress, more work needed. 2011-09-30 19:21:36 +02:00
antirez c5954c1966 Clear the importing status from the slot if a CLUSTER SETSLOT command permanently assigns the slot to the importing node. 2011-09-30 19:20:56 +02:00
antirez 6c0047c430 redis-trib: better slots allocation strategy for resharding 2011-09-30 18:41:25 +02:00
antirez 841cd7684c redis-trib: initial slots allocation strategy, still to improve for better balancing when there are just a few slots to move. 2011-09-30 18:28:51 +02:00
antirez 85b514d165 redis-trib: reshard command work in progress 2011-09-29 16:06:49 +02:00
antirez 9514aa3251 redis-trib: minor refactoring creating a method to load the cluster information starting from a single node 2011-09-29 15:45:10 +02:00
antirez 5d8f25da36 redis-trib: don't crash loading node info for nodes without assigned slots 2011-09-29 15:31:02 +02:00
antirez 3b5289a04c added comment about SETSLOT NODE form. 2011-09-29 15:26:01 +02:00
antirez 7afc3a9611 Fixed command table for RESTORE to make it accepting keys only for the right slots when cluster is enabled. 2011-09-29 15:14:06 +02:00
antirez d7021b081b redis-trib: added cluster state to node info string 2011-09-29 15:04:23 +02:00
antirez 92dd76c8ee redis-trib: all node information into a single hash 2011-09-29 14:44:27 +02:00
antirez b08c9dd280 redis-trib: when loading node info also fetch the output of CLUSTER INFO. 2011-09-29 13:18:09 +02:00
antirez f639f99180 redis-trib cluster check command: check that all the 4096 slots are covered 2011-09-29 12:50:46 +02:00
antirez 0cae060a26 Compilation fixed on OpenBSD making sure that _XOPEN_SOURCE is set to 700 in fmacros.h 2011-09-29 10:20:03 +02:00
antirez d3f7fbfc09 initial node loading info code for redis-trib 2011-09-28 19:08:35 +02:00
antirez 1ef8b0a96d show node with myself flag always as connected in CLUSTER NODES output. 2011-09-28 18:28:36 +02:00
antirez 07c152a7bd cluster config file entries added to the example configuration file 2011-09-28 15:55:52 +02:00
antirez 002d5626e0 Scripting engine now only loads selected libraries, using code originally contributed by @loopole. 2011-09-27 18:46:23 +02:00
antirez 882b65ceaf a few entries removed from the old TODO list (done) 2011-09-27 16:07:38 +02:00
antirez 7864ef8551 new tests for the scripting engine: not allowed commands and write commands after random commands. 2011-09-27 15:39:41 +02:00
antirez 9f772cc237 Return errors if a write command is called inside a Lua script after a random command was called. See https://github.com/antirez/redis/issues/95 for more information. 2011-09-27 15:30:31 +02:00
antirez 15ef605340 Deny commands flagged as REDIS_CMD_NOSCRIPT from Lua scripts 2011-09-27 13:57:10 +02:00
antirez b60ed6e812 added the NOSCRIPT and RANDOM command flags 2011-09-27 13:45:46 +02:00
antirez 5d02b00f56 command table refactoring to make it simpler adding new flags 2011-09-26 15:40:39 +02:00
antirez e108bab043 Lua math.random and math.randomseed replaced with our own version based on redisSrand48(). Seed initialized to the same value at every EVAL/EVALSHA call. 2011-09-23 15:40:58 +02:00
antirez 69786b262d Defined macro with bigger number that redisLrand48() can output. 2011-09-23 15:06:07 +02:00
antirez cbdac04a5e libc neutral random function derived from a drand48() implementation added. Will be used to replace Lua's math.random implementation. 2011-09-23 14:51:48 +02:00
antirez f96a8a8054 rioInitWithFile nad rioInitWithBuffer functions now take a rio structure pointer to avoid copying a structure to return value to the caller. 2011-09-22 16:00:40 +02:00
antirez 69cecb511f make sure to return just 1 for rio.c write when the target is a buffer, as we do when the target is a file. 2011-09-22 15:54:08 +02:00
antirez 4c0462972e comment on top of the _rio structure modified for correctness as actually fwrite/fread semantics is different in general, but was 0/1 in our old usage before rio.c as we always used 1 as number items, and the actual number of bytes to read as item length. 2011-09-22 15:47:48 +02:00
antirez f9c6f39b2b merge conflicts resolved 2011-09-22 15:15:26 +02:00
Salvatore Sanfilippo 7c748c061e Merge pull request #97 from jvain/redis-cli 
redis-cli segfaults with single numeric argument greater than zero
 2011-09-21 13:32:24 -07:00
Juri M. Vainonen aee7f99760 fixed a small bug that caused redis-cli to segfault when given single numeric parameter greater that zero. 2011-09-21 23:22:14 +03:00
antirez 5833cf1676 TODO updated 2011-09-21 19:13:43 +02:00
antirez 0e793fbe84 TODO updated after moving more things into the github issues system 2011-09-20 15:34:49 +02:00
antirez 29177b4d08 TODO updated 2011-09-20 14:50:01 +02:00
antirez 2e5e29815f no longer valid design documents removed 2011-09-20 11:15:55 +02:00
Salvatore Sanfilippo 0ace6ce463 Merge pull request #72 from datafueled/spelling 
Spelling fixes.
 2011-09-20 02:12:14 -07:00
Salvatore Sanfilippo ac1f6fd07a Merge pull request #78 from vambo/patch-1 
Corrected typo in documentation for R_Zero, R_PosInf, R_NegInf, R_Nan;
 2011-09-20 02:00:58 -07:00
Salvatore Sanfilippo 5bae401107 Merge pull request #79 from annulen/mac_ppc 
Fixed compilation on Mac/PPC
 2011-09-20 01:58:55 -07:00
Salvatore Sanfilippo 18be93e0a4 Merge pull request #74 from kmerenkov/issue_620 
[issue 620] don't segfault if number of arguments is less than 1
 2011-09-20 01:55:34 -07:00
antirez eef17490d4 Include port number in error when can't bind lisetning port on startup. 2011-09-20 10:48:21 +02:00
antirez e7aec180e8 Fixed comment typo 2011-09-19 17:50:07 +02:00
antirez 47e7f9acbd Comment out things in bio.c that are currently not useful but that may be useful in the future. 2011-09-19 17:06:27 +02:00
antirez 77ca5fcba6 Emit a log message when AOF fsync is still in progress but we are forced to write from the main thread since two seconds already elapsed. 2011-09-19 16:52:13 +02:00
antirez a60b397b6d use aof_fsync wrapper instead of fsync(2) in bio.c 2011-09-16 15:53:01 +02:00
antirez 37183f14e4 Merge remote-tracking branch 'origin/unstable' into bg-aof-2 2011-09-16 15:44:14 +02:00
antirez dcdfd005a0 commenting out aof-race test for now, waiting for a fix to issue #80 2011-09-16 15:43:03 +02:00
antirez af4707a80c Merge remote-tracking branch 'origin/unstable' into bg-aof-2 2011-09-16 12:36:33 +02:00
antirez a3fcd6bc21 fixed flushAppendOnlyFile() function prototype to reflect the new implementation, fixing compilation. 2011-09-16 12:36:17 +02:00
antirez db3c2a4fb4 postpone the AOF fsync if policy is everysec and there is a background fsync already going. 2011-09-16 12:35:12 +02:00
antirez 9a35eb22bf Fixed typo preventing compilation 2011-09-16 12:11:48 +02:00
Pieter Noordhuis 3aa4b00970 Failing test related to AOF rewrite buffers 2011-09-16 11:25:05 +02:00
Pieter Noordhuis bdbf3acff5 Enforce upper bound for number of requests in redis-benchmark 2011-09-16 11:25:00 +02:00
Pieter Noordhuis 5f54a5e615 Fix AOF race that may duplicate commands 2011-09-16 11:12:31 +02:00
antirez 4b77700a33 Added aof_backgronud_fsync() function, and use it in the bacground rewrite done handler when the fsync policy is everysec. 2011-09-16 11:08:39 +02:00
antirez 96674b6d95 Fixed a bug in the average latency metering of redis-cli --latency 2011-09-15 19:33:24 +02:00
antirez 43071993e1 Implemented --latency in redis-cli 2011-09-15 19:28:00 +02:00
antirez 9fc1e1b1d4 REDIS_BIO_AOF_FSYNC implemented 2011-09-15 18:25:53 +02:00
antirez b39a4d0b39 fixed a problem in bioOlderJobType() when there are no jobs of the specified type. Now the function returns 0 when this happens. 2011-09-15 18:23:58 +02:00
antirez fbb23ce496 some more comment about bio.c design 2011-09-15 18:23:11 +02:00
antirez 1317b7c233 Remove backgroud jobs from the queue only when the processing was completed, and not just started. 2011-09-15 18:21:27 +02:00
antirez 50be9b97bc Use a different thread for every different type of background job 2011-09-15 18:21:16 +02:00
antirez 91de5421b2 Better comments for bioWaitPendingJobsLE 2011-09-15 18:21:02 +02:00
antirez fde4e4c428 Background I/O library enhanced so that the main thread can query for the number of pending jobs of the specified type. 2011-09-15 18:20:53 +02:00
antirez f190705722 TODO updated with 2.6 things 2011-09-15 17:04:55 +02:00
antirez 7feb90fa37 enable zmalloc thread safety support now that we have the bio.c background thread. 2011-09-14 19:09:48 +02:00
Salvatore Sanfilippo e0ba14557e Merge pull request #77 from Poincare/patch-1 
Added note about Tcl
 2011-09-14 06:27:47 -07:00
antirez 7cfeb8cccf Optimize LRANGE to scan the list starting from the head or the tail in order to traverse the minimal number of elements. Thanks to Didier Spezia for noticing the problem and providing a patch. 2011-09-14 15:10:28 +02:00
antirez 6468a6faad Don't replicate/AOF SAVE 2011-09-14 13:09:25 +02:00
antirez 986630afad use bio.c instead of libeio for closing the AOF file in background. Some comment added and other minor changes. 2011-09-13 18:27:08 +02:00
Pieter Noordhuis b454056d94 Asynchronously close old file after BGREWRITEAOF 2011-09-13 18:17:06 +02:00
antirez b1c892f3f6 test background close operation removed 2011-09-13 16:57:18 +02:00
antirez d47ded66ce make bio_list static as well 2011-09-13 16:56:43 +02:00
antirez b147cb9e1a debug message removed 2011-09-13 16:53:33 +02:00
antirez b7c739b0dd use pthread_cond_signal() to make sure the background thread will process the job. 2011-09-13 16:21:12 +02:00
antirez 8ea2dfd75c typo fixed in bio.c 2011-09-13 16:19:37 +02:00
antirez f67e338d5c Removed no longer useful fields in the redis server structure 2011-09-13 16:17:45 +02:00
antirez 8f61a72fa8 add background jobs initialization to Redis main init function 2011-09-13 16:10:26 +02:00
antirez f81a5f54d8 bio.c typos fixes to make it compile 2011-09-13 16:09:06 +02:00
antirez 02925dd96e bio.o added to redis-server objects 2011-09-13 15:59:48 +02:00
Pieter Noordhuis d1ec6c8b22 Speed up protocol synthesis for AOF 2011-09-13 12:23:50 +02:00
Pieter Noordhuis f990782f4d Re-use AOF buffer when it is small enough 2011-09-13 12:22:54 +02:00
Pieter Noordhuis a57225c2cf Fix indent 2011-09-13 12:22:39 +02:00
antirez 4f06867afe added a newline for code readability 2011-09-13 12:21:54 +02:00
Pieter Noordhuis 297322481d Avoid unnecessary calls to time(3) 2011-09-13 12:18:55 +02:00
antirez 3bc89500d2 Remove the write handler only if there are no longer objects in the output queue AND if the static buffer is empty. This bug was the cause of a possible server-stop-responding-to-client bug under some specific work load. Thanks to Pieter Noordhuis for spotting and fixing it. 2011-09-12 11:06:28 +02:00
antirez 7551e23716 regression test for bug 593 added 2011-09-12 10:53:49 +02:00
antirez 48082cc044 fixed typos in the comments of rpoplpushHandlePush() 2011-09-12 10:04:23 +02:00
Hampus Wessman c47d152c8d Fix crash when chaining brpoplpush with other blocking commands. 2011-09-07 19:08:48 +02:00
Konstantin Tokarev ef9444e0d2 Fixed compilation on Mac/PPC 
Applied patch from Issue 119 provided by 
pahowes@gmail.com
 2011-09-01 22:16:20 +04:00
vambo c74b7c7730 Corrected typo in documentation for R_Zero, R_PosInf, R_NegInf, R_Nan; 2011-08-26 21:40:18 +03:00
Poincare 08c4b019b7 Added note about Tcl 2011-08-25 07:35:33 -03:00
antirez b56567017d Merge remote-tracking branch 'origin/unstable' into unstable 2011-08-09 11:46:21 +02:00
antirez 11aaf52313 Auto rewrite integer overflow fixed. Thanks to Gaurav for finding it (see issue 602 on google code), and Pieter Noordhuis for tracing it. 2011-08-09 11:46:18 +02:00
Konstantin Merenkov be6f639564 [issue 620] don't segfault if number of arguments is less than 1 2011-08-02 17:05:04 +04:00
antirez f65c8f654f Merge remote-tracking branch 'origin/unstable' into unstable 2011-07-28 14:40:47 +02:00
antirez 4bb1ccc6eb Regression for a crash with blocking ops and pipelining 2011-07-28 14:40:06 +02:00
Hampus Wessman 64f201c2aa Fix crash when pipelining several blocking commands. 2011-07-28 13:50:44 +02:00
antirez 3d9704d335 Merge remote branch 'origin/unstable' into unstable 2011-07-28 12:33:17 +02:00
antirez 243b783fdc Maxmemory test 2011-07-28 12:32:52 +02:00
antirez e54484876a fixed assert proc on Redis test 2011-07-28 12:31:44 +02:00
antirez 47cadab42b TODO updated 2011-07-27 16:39:03 +02:00
Pieter Noordhuis bb831c313c Fix adding bulk reply when getcwd fails 2011-07-27 16:22:28 +02:00
Pieter Noordhuis 4d6bf65c42 Wait for Lua to be built before linking redis-server 2011-07-27 14:47:02 +02:00
Pieter Noordhuis 2d7162bb1d HDEL: Abort deleting fields when hash is removed 2011-07-27 12:29:40 +02:00
Bruce Mitchener 5215ab1418 Spelling fixes. 2011-07-26 10:19:02 +07:00
antirez 80e87a461a Close the listening sockets. Apparently this allows faster restarts. 2011-07-22 12:11:40 +02:00
antirez adae85cdae prepareForShutdown() fixed for correctness regarding what to do with AOF and RDB persistence on exit. 2011-07-22 11:52:21 +02:00
antirez d50292d2cc make a scripting test more valgrind friendly 2011-07-15 18:28:24 +02:00
antirez 61fee31999 test that EVALSHA is replicated as EVAL 2011-07-15 17:41:40 +02:00
antirez f790bd028c Added a 'runtest' script that is responsible to check if Tcl is available and run the test. This is invoked from Makefile as well. 2011-07-15 17:20:57 +02:00
antirez 319bb48c01 top level Makefile now just a proxy. Doing make clean inside src now does a full clean including deps, not just Redis source. Thanks to Pieter for the top level Makefile proxy trick. 2011-07-13 19:15:22 +02:00
antirez 5cb4850708 Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-07-13 15:38:33 +02:00
antirez 4dd444bb4a Replicate EVALSHA as EVAL taking a dictionary of sha1 -> script source code. 2011-07-13 15:38:03 +02:00
antirez 70ecddc9f4 scripting.txt file removed, content merged into the TODO file 2011-07-13 12:04:02 +02:00
Salvatore Sanfilippo 706b6e58ba Merge pull request #69 from AquaGeek/patch-1 
Minor spelling fixes
 2011-07-13 03:00:28 -07:00
Tyler Stromberg c984d42fb1 Minor spelling fixes 2011-07-12 18:50:18 -07:00
antirez 331bf3292f when building redis-server use LINK and not CC as prefix in the output 2011-07-12 12:58:32 +02:00
antirez 5f54619494 Fix for a typo left after merging. 2011-07-12 12:43:05 +02:00
antirez 0681c5ad84 master branch merged into scripting. 2011-07-12 12:39:16 +02:00
antirez 34a8b51768 solved a slow positive in the slow log test when the test is executed slowly 2011-07-11 16:04:18 +02:00
antirez bf1379c8df fixed another timing issue when running the test over valgrind or with very slow PCs 2011-07-11 15:58:31 +02:00
antirez 4bab631223 Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-07-11 15:45:06 +02:00
antirez 0717e3aa16 fix a test timing issue when running the test over valgrind 2011-07-11 15:44:38 +02:00
antirez e6fcb5b30d make test depends on redis-check-aof as well. 2011-07-11 14:48:45 +02:00
antirez 6b91a11e3e removed --file argument from test invocation inside Makefile 2011-07-11 13:46:09 +02:00
antirez 4c378d7f6c new test engine valgrind support 2011-07-11 13:41:06 +02:00
antirez c7c16a32a7 cleanup after test 2011-07-11 13:19:42 +02:00
antirez e4715f0011 --help implemented 2011-07-11 13:03:22 +02:00
antirez 121ffc85f4 better recap of failed tests. 2011-07-11 12:56:00 +02:00
antirez 04e2410d2d initial support for failed tests summary at end of the test. 2011-07-11 12:44:55 +02:00
antirez 30cf7be60c the test runs less iterations of slow tests if no --accurate is given. 2011-07-11 12:15:35 +02:00
antirez 524d515f0c implemented --single, --list-tests. --accurante and --help are now recognized but not actually implemented. 2011-07-11 11:59:55 +02:00
antirez 6209797d37 list test further split into smaller units 2011-07-11 11:41:23 +02:00
antirez be9250c8cd list-2 added to the list of tests 2011-07-11 11:32:19 +02:00
antirez e46f78bd26 list test split into smaller parts 2011-07-11 11:30:46 +02:00
antirez 52c863ddac use tclsh8.5 not only in which but also in the actual test execution. 2011-07-11 11:18:46 +02:00
antirez d4edba6f43 explicitly use tclsh8.5 to run the test 2011-07-11 11:17:05 +02:00
antirez daab159932 no longer prints testing... but just shows the test when we get the final result from the testing client. 2011-07-11 00:53:30 +02:00
antirez 569f84aa7c replication test split into three parts in order to improve test execution time. Random fixes and improvements. 2011-07-11 00:46:25 +02:00
antirez 82e5dd35ea display less useless information in parallel test 2011-07-11 00:14:12 +02:00
antirez 3744824c3e some test colorization and some fix 2011-07-11 00:09:56 +02:00
antirez 36e790a040 print execution time of different units at end 2011-07-10 23:57:35 +02:00
antirez 13566085cf Initial implementation of a client-server parallel testing system for Redis in order to speedup execution of the test suite. 2011-07-10 23:25:48 +02:00
antirez 5ab1461f98 The test now gives some more time for Redis to start before of exiting with an error, since starting with valgrind can take a significant amount of time. 2011-07-09 19:23:46 +02:00
antirez 09e2d9eeba Take a pointer to the relevant entry of the command table in the client structure. This is generally a more sounding design, simplifies a few functions prototype, and as a side effect fixes a bug related to the conversion of EXPIRE -1 to DEL: before of this fix Redis tried to convert it into an EXPIREAT in the AOF code, regardless of our rewrite of the command. 2011-07-08 12:59:30 +02:00
antirez 5521fa6a9f Test for regression about: Redis should not try to convert DEL into EXPIREAT for EXPIRE -1 2011-07-08 12:20:30 +02:00
antirez 812ecc8b10 don't process EXPIRE with negative TTL or EXPIREAT with time in the past if we are a slave too (see http://groups.google.com/group/redis-db/browse_thread/thread/5a931fefb88b16d5). Also propagate it as DEL. 2011-07-07 16:24:37 +02:00
Hampus Wessman 040b0ade7d Don't expire keys while loading AOF. 
They will be expired (and a DEL will be logged) after the loading is done
instead.
 2011-07-07 16:08:30 +02:00
Hampus Wessman 72bae0cc75 Add test for incorrect expiration when loading AOF. 2011-07-07 16:08:22 +02:00
antirez cabe03eb75 more valgrind friendly test 2011-07-06 15:22:00 +02:00
antirez 10ba492fa3 Ingore jemalloc make distclean error in Makefile, caused by make clean against a tree that never compiled jemalloc. 2011-07-05 13:27:21 +02:00
antirez 208092ad70 Fixed typo in redis.conf related to slow log parameters 2011-07-03 15:49:49 +02:00
antirez 5a9dd97c7c childrens -> children in INFO output field names. This is a backward incompatible change, but really a minimal one. 2011-07-02 10:52:28 +02:00
antirez 6504634019 no more allocation stats info in INFO, useless now that we have jemalloc. 2011-07-02 10:31:16 +02:00
antirez bbbf79ddb5 Added slowlog.tcl 2011-07-01 15:49:17 +02:00
antirez 17daf16f85 SLOWLOG tests 2011-06-30 18:44:58 +02:00
antirez 2cb6828480 Added an unique ID field to every slow log entry. 2011-06-30 17:36:15 +02:00
antirez de32c37c06 More redis.conf self-documentation. Now even queries that took exactly server.slow_log_slower_than are logged, as this is not exact but is more intuitive for people, and a value of 0 will force every query to be logged. 2011-06-30 15:54:05 +02:00
antirez 35a6044140 slow log configuration implemented 2011-06-30 15:47:15 +02:00
antirez 63d62eb786 hiredis max multi bulk nesting level raised to 7 2011-06-30 13:34:38 +02:00
antirez 404345d8e5 DEUBG SLEEP implemented 2011-06-30 13:31:44 +02:00
antirez daa70b1798 First implementation of the slow log feature 2011-06-30 13:27:32 +02:00
antirez ef67a2fc3c Partial fix for a problem with AOF and BRPOP. We have at least a single evidence of an AOF that ended with a BRPOP call against an empty list, causing later issues at run time. Now the problem is detected while loading the AOF at startup, even if it is still to understand how the corrupted AOF was generated. 2011-06-29 16:11:23 +02:00
antirez 15bc1cc1bc colorized test output shows scripting tests with X 2011-06-25 12:42:55 +02:00
antirez 994ed2bc55 unstable merge conflicts resolved 2011-06-25 12:29:24 +02:00
antirez d5b36c5114 Merge remote branch 'origin/unstable' into unstable 2011-06-25 12:22:59 +02:00
antirez c9d0c3623a diskstore removed 2011-06-25 12:22:03 +02:00
Salvatore Sanfilippo a004de0f15 Merge pull request #63 from djanowski/tcl 
Add warning when `tcl` is not available to run tests
 2011-06-24 09:59:37 -07:00
Damian Janowski b3bc4452d2 Avoid bug reports when Tcl is not installed. 2011-06-24 13:56:06 -03:00
antirez c1c9d551da Fix for bug 561 and other related problems 2011-06-20 17:19:36 +02:00
antirez 634bae94fb useless call removed, thanks to Pieter for spotting this 2011-06-20 16:42:37 +02:00
Pieter Noordhuis b0d1d2684d SETNX regression tests. 2011-06-20 16:42:29 +02:00
antirez f85cd526c1 DB API refactoring. The changes were designed together with Pieter Noordhuis. 2011-06-20 16:42:16 +02:00
antirez 3e2a0bf44a Makefile deps updated 2011-06-20 11:58:54 +02:00
antirez e93d36a638 It is now possible to build with: make 32bit USE_JEMALLOC=yes 2011-06-20 11:54:57 +02:00
antirez 7d9ba15669 Added -ldl when linking against jemalloc, needed on Linux 2011-06-20 11:54:35 +02:00
antirez fec5a664f8 Fixed INFO implementation to use the new allocator name define 2011-06-20 11:54:08 +02:00
antirez 9e62dc17da Makefile modified to build with jemalloc 2011-06-20 11:52:15 +02:00
Pieter Noordhuis b4a79b9ffc Ignore jemalloc build artifacts 2011-06-20 11:38:25 +02:00
antirez 16cd4acdee ZMALLOC related stuff removed from config.h. They are now defined in zmalloc.h 2011-06-20 11:35:29 +02:00
antirez 29d04257b0 forward-ported changes in zmalloc.c/h to support jemalloc build 2011-06-20 11:34:04 +02:00
antirez a78e148b7d jemalloc source added 2011-06-20 11:30:06 +02:00
antirez 07486df6fe new INFO filed master_link_down_since_seconds 2011-06-17 16:16:46 +02:00
antirez 591f29e0f5 Use a safe iterator while saving the DB, since the getExpire() function will access the iterating dictionary. 2011-06-17 15:41:22 +02:00
Pierre Chapuis 45ec3243f8 Untrack and ignore Lua binary files (2) 2011-06-14 18:08:45 +02:00
Pierre Chapuis 6d0014cf58 Untrack and ignore Lua binary files 2011-06-14 18:08:36 +02:00
antirez 0d1650f8a9 Updated to unstable 2011-06-14 18:06:39 +02:00
antirez 55937b7972 Fixed init script bug, thanks to Henrik Westphal 2011-06-14 17:11:44 +02:00
antirez b0d68504e9 Better init script 2011-06-14 13:49:15 +02:00
Salvatore Sanfilippo 4abd096f51 Merge pull request #58 from hampus/fix-autorewrite 
Fix automatic rewrite starting too early
 2011-06-14 01:26:42 -07:00
Hampus Wessman 0b17517c7c Fix automatic rewrite starting too early. 
Slight adjustment in growth calculation.
 2011-06-12 07:27:01 +02:00
antirez 9e40bce3fa different message on BGREWRITEAOF when it is just scheduled and not started. 2011-06-10 18:35:16 +02:00
antirez 2c6cc5e502 minor code aesthetic change 2011-06-10 18:13:57 +02:00
antirez 19b46c9a09 Make sure an automatic rewrite only gets triggered when no other child is in execution. 2011-06-10 15:15:29 +02:00
antirez 0ec932208e CONFIG SET/GET support for new automatic AOF rewrite parameters 2011-06-10 15:14:10 +02:00
antirez c66bf1fa32 correctly set AOF base size field in server structure 2011-06-10 14:56:45 +02:00
antirez d630abcdaf Added new INFO fields related to the new AOF auto rewrite feature 2011-06-10 14:52:47 +02:00
antirez 4f948300bc Fixed bug in AOF rewrite not working because of integer overflow 2011-06-10 14:47:12 +02:00
antirez 73b9e8aedf debugging printf removed 2011-06-10 14:44:06 +02:00
antirez 4ff34b6adb AOF auto rewrite percentage option parsing fixed 2011-06-10 14:43:11 +02:00
antirez 50528f0eea Merge remote branch 'pietern/unstable-bench' into unstable 2011-06-10 14:28:14 +02:00
antirez e3d27a7261 Avoid division by zero issues in the automatically triggered AOF rewrite feature. 2011-06-10 14:25:27 +02:00
antirez e087b6d7a1 Merge remote branch 'origin/unstable' into unstable 2011-06-10 12:39:40 +02:00
antirez b333e23997 automatic AOF rewrite first implementation. Still to be tested. 2011-06-10 12:39:23 +02:00
Salvatore Sanfilippo 00ba9b641c Merge pull request #57 from melo/patch-1 
Small fix in the installation instructions
 2011-06-09 09:50:55 -07:00
Pedro Melo b6c46d17a7 Small fix in the installation instructions 2011-06-09 09:14:18 -07:00
Hampus Wessman db6cea11c4 Fix Makefile colors. 
echo with escape sequences isn't portable. Use printf instead.
 2011-06-09 16:38:18 +02:00
antirez 45029d37cb Two fixes for replication: Slave performs the AOF rewrite at the right point. Non blocking connect also uses readable handler as with old Linux kernels like 2.6.18 on connection refused the writable even is not fired (kernel bug). 2011-06-09 15:39:12 +02:00
antirez 73e718675b Better colorized Makefile implementation, inspired by git makefile 2011-06-08 17:09:18 +02:00
antirez fc879abe4a Merge branch 'unstable' 2011-06-07 17:01:45 +02:00
antirez cf1eefa420 Merge remote branch 'origin/unstable' into unstable 2011-06-05 20:51:49 +02:00
antirez 5daa1f060a no more client-libraries directory 2011-06-05 20:51:45 +02:00
antirez 06e09c9b58 no more html documentation 2011-06-05 20:51:31 +02:00
antirez 936c4ab64b touch less pages in decrRefCount 2011-06-03 17:33:23 +02:00
Pieter Noordhuis 294cd536e0 There is no debug mode in benchmark tool 2011-06-01 09:22:19 -07:00
Pieter Noordhuis 39bf440243 Provide custom command to benchmark against via args 2011-06-01 09:22:19 -07:00
Pieter Noordhuis fc05e8c850 More const's in redis-benchmark.c 2011-06-01 09:22:19 -07:00
antirez b002546bb4 ZREM tests 2011-05-31 20:30:54 +02:00
antirez 3f7b2b1f30 Variadic ZREM 2011-05-31 20:15:18 +02:00
antirez 3738ff5f32 Fix for the variadic version of SREM. Regression test added. 2011-05-31 20:14:29 +02:00
antirez faa2a80f89 disabled development test entry, tests moved in the right place 2011-05-31 18:49:12 +02:00
antirez 3fe40d6e3c Variadic ZADD tests 2011-05-31 18:35:09 +02:00
antirez ef231a7c56 Variadic ZADD 2011-05-31 17:47:34 +02:00
Pieter Noordhuis 632e4c09ac Make replication faster (biggest gain for small number of slaves) 2011-05-30 12:45:07 +02:00
antirez 615e414c5d INFO now contains the time (in usecs) needed to fork() in order to persist. The info is available in the stats section of INFO. 2011-05-29 15:17:29 +02:00
antirez 43222f253a Merge remote branch 'pietern/unstable-replnonblock' into unstable 2011-05-29 14:26:01 +02:00
antirez 18f63d8d51 redis-cli: -i (interval) implemented, to wait the specified number of seconds (decimal digits are allowed) between commands. 2011-05-28 15:41:08 +02:00
antirez 442c748d84 redis-cli: Use the repetiton prefix after a reconnection. 2011-05-28 15:25:48 +02:00
antirez 4d19e3443c redis-cli in REPL mode is now able to send the same command multiple times, prefixing the command with a number as in "10 ping" 2011-05-28 15:13:55 +02:00
antirez a45f9a1a1d redis-cli no longer aborts in repl-mode on error, and retries to reconncet with the server at every command issued if the state is not connected. Also the prompt shows the server we are connected to. 2011-05-28 15:04:12 +02:00
antirez e4f6b8c327 Merge branch 'unstable' 2011-05-25 12:33:25 +02:00
antirez 3bb818df40 Make sure error and status replies emitted by Lua scripts can never have more than a newline, otherwise it is a protocol violation and clients will desync. 2011-05-25 12:32:50 +02:00
antirez 449286a588 Scripting tests added 2011-05-25 12:32:50 +02:00
antirez e927a24610 Fix for wrong error level detection in redis.log() 2011-05-25 12:32:50 +02:00
antirez 288f811fea All commands exported by Redis to Lua scripts are now inside the redis table. Implemented redis.log() function. 2011-05-25 12:32:50 +02:00
antirez 7229d60d03 EVALSHA implemented 2011-05-25 12:32:49 +02:00
antirez 82c6b8257a Now it is possible to return multi bulks of multi bulks from Lua, just returning tables of tables. The new conversion of the Null bulk and multi bulk type is the Lua false type. 2011-05-25 12:32:49 +02:00
antirez da386cdf33 for performance reasons only set the debug hook if the time limit is enabled. 2011-05-25 12:32:49 +02:00
antirez 87c219243c Lua max exec time fixes: redisLog() to log the event moved before the lua_error() call that does a long jump so never returns. Aesthetic fixes in error message. 2011-05-25 12:32:49 +02:00
antirez bcc72bc381 call the Lua hook with minor frequency. It is already enough to call it every 100000 istructions for near millisecond precision. 2011-05-25 12:32:48 +02:00
antirez eeffcf380f Lua scripts max execution time 2011-05-25 12:32:48 +02:00
antirez 8c3402dffa Lua memory usage in INFO 2011-05-25 12:32:48 +02:00
antirez 40531be0e3 call lua_gc() for incremental garbage collection. Likely there is to tune this at some point. 2011-05-25 12:32:48 +02:00
antirez c2a7dd85ae Lua scripting: Honor SELECT command in both ways 2011-05-25 12:32:47 +02:00
antirez 09ab5591fa Fixed a problem in the script engine that crashed the server if the reply was not stored just int the static buffer of the client structure. 2011-05-25 12:32:47 +02:00
antirez 0d916763af Status replies converted in a special way like errors in the Lua <-> Redis conversion code. 2011-05-25 12:32:47 +02:00
antirez e8092bf601 Scripting doc updated 2011-05-25 12:32:47 +02:00
antirez 497fc8775f Eval command table fixed to return the keys arguments of the command. We use zunionInterGetKeys() as it ZUNION/ZINTER have exactly the same semantic with a prefixed number of keys followed by key names. 2011-05-25 12:32:47 +02:00
antirez 171ef6813d TODO updated 2011-05-25 12:32:46 +02:00
antirez 379789cc52 Fixed nul bulk parsing in Redis protocol to Lua type convertion. Check for bad Lua types as arguements of the redis() Lua command. 2011-05-25 12:32:46 +02:00
antirez 67e829dce5 Scripting todo list updated 2011-05-25 12:32:46 +02:00
antirez 10a6da7af4 Multi bulk to Lua table now working 2011-05-25 12:32:46 +02:00
antirez 3791000f15 Fixed arity detection of Redis command executed from Lua script. Error reporting from Lua fixed. More work on the Redis reply to lua conversion code. 2011-05-25 12:32:45 +02:00
antirez 532e0f5ded full conversion from Lua return value to redis reply. Partial conversion from Redis reply to Lua type. 2011-05-25 12:32:45 +02:00
antirez 4ae5b5e163 pupulate the Lua global tables KEYS and ARGV before executing the script 2011-05-25 12:32:45 +02:00
antirez 00b7541b21 function to access to Redis command from Lua renamed from r() to redis() 2011-05-25 12:32:45 +02:00
antirez 7b72272790 when creating not connected clients do not add them into the clients list, otherwise they will be subject to timeouts and other stuff 2011-05-25 12:32:44 +02:00
antirez 7156f43c04 Correctly glue the reply buffer. For now returned as it is to Lua, but will be converted into Lua native type later. 2011-05-25 12:32:44 +02:00
antirez 0f1d64ca57 Lua call of Redis command work in progress: sorry I have to go to the cinema to watch the Source Code movie 2011-05-25 12:32:44 +02:00
antirez 7585836e6e Lua function creation on EVAL, basic Lua return type to Redis protocol convertion done. 2011-05-25 12:32:43 +02:00
antirez 21d3294c70 makefile adapted to link against lua lib and to pass the 32bit flag to Lua building system 2011-05-25 12:32:43 +02:00
antirez efc3408748 Fix for Pub/Sub system, introduced in Redis 2.2.6 with the new copy-on-write safe iterator semantics. In the hope this is the last bug I introduced this way. 2011-05-25 12:32:15 +02:00
antirez 27fee630f5 Comment typo fixed 2011-05-24 10:43:35 +02:00
antirez d200342ac2 Fixed typo in comment 2011-05-24 10:35:58 +02:00
Pieter Noordhuis 890a2ed989 Configurable synchronous I/O timeout 2011-05-22 12:58:18 +02:00
Pieter Noordhuis b075621fb7 Minor changes in non-blocking repl. connect 2011-05-22 12:51:09 +02:00
antirez 6326c3ce73 Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-05-20 11:45:05 +02:00
antirez 206d62710c Print version info before running the test 2011-05-20 11:44:54 +02:00
Pieter Noordhuis a330913999 Non-blocking connect with master 2011-05-19 18:54:57 +02:00
antirez 521ddcce14 Fix for ZUNIONSTORE bug when there is an empty set among input sets. Regression test added. 2011-05-19 17:58:52 +02:00
antirez 38a154a9a6 Fix for ZUNIONSTORE bug when there is an empty set among input sets. Regression test added. 2011-05-19 17:58:29 +02:00
antirez 330c90b018 fmacros in utils.c to avoid warning about strcasecmp() 2011-05-16 17:20:27 +02:00
antirez 19e8458930 Merge branch 'unstable' 2011-05-16 10:04:09 +02:00
antirez cb16b6c389 Fixed misuse of the new iterator semantics in ZUNIONSTORE 2011-05-15 17:28:06 +02:00
antirez d070abe44c Fix for a possible bug related to ZINTER/UNIONSTORE called with the same source set more than one time. 2011-05-15 15:33:01 +02:00
antirez dd1eefa4f3 Fixed SINTER[STORE] problem related to the new copy on write safe iterator 2011-05-15 12:18:00 +02:00
Pieter Noordhuis 041d8e2a5c Fix up rdbWriteRaw to return number of bytes written 2011-05-14 12:47:42 +02:00
Pieter Noordhuis 7271198cf0 Use rio.h functions in aof.c 2011-05-14 12:36:22 +02:00
Pieter Noordhuis fd535c5862 More rioRead() 2011-05-14 12:32:50 +02:00
Pieter Noordhuis 221782ccc6 Move rdbLoad* to top; update comments 2011-05-13 23:24:19 +02:00
Pieter Noordhuis f1d8e4968e Make RDB types/opcodes explicit; load/save object type 2011-05-13 22:14:39 +02:00
Pieter Noordhuis 2e4b0e7727 Abstract file/buffer I/O to support in-memory serialization 2011-05-13 17:31:00 +02:00
antirez 70bc5f7724 replication with expire test modified to produce no or less false failures 2011-05-12 20:21:43 +02:00
antirez 06d22a8a8a Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-05-12 20:21:00 +02:00
antirez af9aed25e4 ZINTERSTORE regressiont test with two sets, intset+hashtable 2011-05-12 20:20:40 +02:00
antirez defb5f66a0 removed assert causing an illegal memory access. This was responsible of crashes during BLPOP and other list blocking operations. 2011-05-11 09:50:57 +02:00
antirez bfe85f7ca9 Merge branch 'unstable' 2011-05-10 14:05:59 +02:00
antirez 6b52ad87c0 RDB version signature bumped to 2, now that direct saving of specially encoded types is implemented 2.2 instances are no longer able to read rdb files produced by 2.4 or unstable. 2011-05-10 11:31:37 +02:00
antirez 4b53e7365c Introduced a safe iterator interface that can be used to iterate while accessing the dictionary at the same time. Now the default interface is consireded unsafe and should be used only with dictNext() 2011-05-10 10:15:50 +02:00
antirez f13cb0d9de DEBUG DIGEST additional lookup needed for VM removed from unstable branch that does not have VM at all 2011-05-10 10:08:01 +02:00
antirez 6901fe7729 Useless statement removed 2011-05-10 10:07:04 +02:00
antirez df541beae3 when Redis fails accepting a new connection reports the error at WARNING and not VERBOSE error level. Thanks to offby1 for proposing this in the Redis mailing list. #backport-candidate 2011-05-07 11:47:34 +02:00
antirez 8905378c56 Cluster TODO updated 2011-05-06 16:23:14 +02:00
antirez 0caa750748 clear importing/exporing state when appropriate on SETSLOT or ADDSLOTS cluster commands 2011-05-06 16:08:10 +02:00
antirez f9cbdcb1a6 CLUSTER SETSLOT command and some refactoring of the cluster command 2011-05-06 15:44:09 +02:00
antirez f384df8302 CLUSTER DELSLOTS 2011-05-06 13:38:27 +02:00
antirez a7b058dae6 Fixed semantics of CLUSTER SETSLOT, SELECT now only denied in cluster mode if selected DB is not 0 so that MIGRATE still works well. 2011-05-05 18:10:02 +02:00
antirez 46834808fe CLUSTER SETSLOT STABLE fixed 2011-05-05 17:56:12 +02:00
antirez 4763ecc9ad missing return caused protocol desync in CLUSTER SETSLOT 2011-05-05 17:52:19 +02:00
antirez 0276e5545c Fixed problem in cluster redirection due to a stupid typo 2011-05-05 17:50:14 +02:00
Pieter Noordhuis 3edbcab95a Extra tests for string2ll 2011-05-05 16:32:22 +02:00
Pieter Noordhuis c040cbd625 Check seplen and len before malloc'ing "tokens" 2011-05-05 16:32:22 +02:00
Pieter Noordhuis bf9fd5ffa2 Check for \n after finding \r 2011-05-05 16:32:22 +02:00
Pieter Noordhuis 32f99c5128 Show memory allocator in INFO output 2011-05-05 16:32:18 +02:00
Pieter Noordhuis edf23aff0e Use string2ll in ziplist code (faster) 2011-05-05 16:26:51 +02:00
Pieter Noordhuis 5d08193126 Tests for string2ll; move isObject* to object.c 2011-05-05 16:26:50 +02:00
Pieter Noordhuis af0e51f2e1 Move code 2011-05-05 16:25:48 +02:00
Pieter Noordhuis 5282562117 More allocators 2011-05-05 16:25:48 +02:00
Pieter Noordhuis ded614f803 Inline sdslen and sdsavail (thanks to @bitbckt) 2011-05-05 16:25:48 +02:00
Pieter Noordhuis 9703b1b3dc Update hiredis 2011-05-05 16:25:48 +02:00
Pieter Noordhuis 5af302011c Use custom string2ll and strchr 2011-05-05 16:25:48 +02:00
Pieter Noordhuis b7d3bf5157 Fix ziplist regression and update stresser 2011-05-05 15:16:51 +02:00
antirez eda827f8b7 cluster import/export of hash slots implemented in the query redirection engine 2011-05-05 11:13:21 +02:00
antirez a5dce40726 if /dev/urandom is not available use rand() to get a random node name 2011-05-04 10:30:22 +02:00
antirez 35845afba0 Better pretty printed makefile output 2011-05-04 10:17:05 +02:00
antirez 0d8f1ac359 initial pretty printing in makefile 2011-05-04 09:32:32 +02:00
antirez 0ba2932298 CLUSTER SETSLOT implemented 2011-05-04 09:31:37 +02:00
antirez 66f2517feb render migrating and importing slots in cluster nodes info 2011-05-02 19:04:33 +02:00
antirez 74a640492f Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-05-01 15:38:59 +02:00
antirez 5cbe90dbfd Fixed a problem with string2ll() 2011-05-01 15:36:47 +02:00
antirez 2f52dac9ef CLUSTER subcommands to set slots in migrating or importing state. Still a work in progress... 2011-04-29 17:34:03 +02:00
antirez 484354ff95 CLUSTER GETKEYSINSLOT implemented 2011-04-29 16:17:58 +02:00
antirez 1eb713a4c1 CLUSTER KEYSLOT command 2011-04-29 14:31:18 +02:00
antirez d37299e3b7 Fixed a bug with replication where SLAVEOF NO ONE caused a slave to close the connection with its slaves 2011-04-29 14:18:16 +02:00
antirez c772d9c6e7 take a hashslot -> keys index, will be used for cluster rehasing 2011-04-28 19:00:33 +02:00
Pieter Noordhuis 45b0f6fb14 Use correct argc/argv for cleanup when loading AOF 2011-04-22 09:44:06 +02:00
Pieter Noordhuis d8b6ae3cd6 Reformat AOF tests 2011-04-22 09:43:26 +02:00
antirez 7e79dd3f4c peak fragmentation ratio removed as it is a confusing field for users and trivial to compute at hand now that there is peak memory information in INFO output 2011-04-21 16:56:31 +02:00
antirez b93fdb7bbb CLIENT KILL implemented 2011-04-21 15:47:47 +02:00
antirez 3cd12b5687 CLIENT LIST implemented 2011-04-21 15:38:02 +02:00
antirez 17b24ff30d save peak memory usage as statistic and show it in INFO. Also a new INFO field was added showing the fragmentation ratio using the peak memory info. 2011-04-21 10:49:52 +02:00
antirez c7ba7b8bbb removed check for zmalloc return NULL in createClient(). The check was misplaced, and zmalloc never returns NULL. 2011-04-20 12:51:03 +02:00
antirez b3a96d454e Variadic SREM 2011-04-19 17:37:03 +02:00
antirez 64a13a36e6 variadic HDEL with tests 2011-04-19 17:07:55 +02:00
antirez 271f087842 Variadic SADD tests 2011-04-15 18:28:25 +02:00
antirez 22f294d24a variadic SADD 2011-04-15 18:08:32 +02:00
antirez 009db67645 addReplyLongLong optimized to return shared objects when the value to reply is 0 or 1 2011-04-15 18:08:24 +02:00
antirez 2e667806ed version bumped to 2.9, that is, 3.0 unstable 2011-04-15 17:44:08 +02:00
antirez 9d0a8656c1 tests for variadic list push 2011-04-15 16:52:07 +02:00
antirez 72dff2c084 test fixed after ascii art banner modified the output of a running server 2011-04-15 16:35:54 +02:00
antirez fb2feae599 variadic LPUSH/RPUSH 2011-04-15 16:35:27 +02:00
antirez 7c0e1b53c4 Merge remote branch 'pietern/unstable-zset' into unstable 2011-04-14 13:31:14 +02:00
antirez d0cfb2be64 redis-trib: functions implementing commands moved at bottom 2011-04-14 09:41:22 +02:00
antirez 8bad0ddd9d fixed bug due to typo in code 2011-04-13 21:26:12 +02:00
antirez 180ba187d3 fixed typo in comment 2011-04-13 20:17:11 +02:00
antirez 9a440ad418 fixed a typo in redis-trib 2011-04-13 18:50:50 +02:00
antirez 3883a381b6 use map! instead of reassigning 2011-04-13 18:46:29 +02:00
antirez 583fc5dd60 hopefully more readable info_string method in redis-trib 2011-04-13 18:40:51 +02:00
antirez 57d83d5673 redis-trib create, first version is know working 2011-04-13 15:58:05 +02:00
antirez 8231b1ef60 TODO updated 2011-04-13 15:39:43 +02:00
antirez 9d89e99c99 no longer useful debugging printf removed 2011-04-13 11:42:05 +02:00
antirez 996d503d1a ASCII ART FTW 2011-04-13 10:58:21 +02:00
antirez a54d9805ee Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-04-13 10:39:06 +02:00
antirez 1087227d41 redis-trib: show list of commands 2011-04-13 10:38:56 +02:00
antirez 4e17be0ef5 TODO update 2011-04-12 19:21:03 +02:00
antirez 744f34d834 redis-trib can now assign hash slots to initialize a cluster. Next step is to join the cluster. 2011-04-12 17:06:33 +02:00
antirez d415f9ff41 Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-04-12 15:36:10 +02:00
antirez e9ee513b6a TODO modified 2011-04-11 21:47:45 +02:00
Salvatore Sanfilippo f29d1fb0b2 assert_empty in redis-trib 2011-04-11 18:26:00 +02:00
Salvatore Sanfilippo 8c4c50906c added known nodes info in CLUSTER INFO 2011-04-11 17:40:35 +02:00
Salvatore Sanfilippo b800a3ab20 a first refactoring of redis-trib.rb 2011-04-11 16:58:47 +02:00
Salvatore Sanfilippo 407798c1e1 Redis-trib initial implementation (currently can not do any actual work) 2011-04-11 16:41:06 +02:00
Salvatore Sanfilippo 1c708b25ee added minimal cluster section in INFO output. This is only useful to check if the instance is or not configured as a cluster node, all the other informations are accessible using the CLUSTER command. 2011-04-11 16:40:37 +02:00
antirez 8d727af8c4 update state when FAIL is cleared from a node 2011-04-08 14:19:52 +02:00
antirez fd7a584f7a do not process node failure messages about yourself 2011-04-08 10:17:41 +02:00
antirez a55c7868c1 fixed lame error in slot assignment 2011-04-07 23:33:18 +02:00
antirez 5a547b27f7 evaluate cluster state after config load. Still bugs in the slot allocation after nodes config load to fix. 2011-04-07 23:23:27 +02:00
antirez 93666e583c master node without slaves rejoin fixed 2011-04-07 23:10:32 +02:00
antirez 152d937b8c when loading config set the ping/pong timestaps to the current value. Even a node in FAIL state can be accepted again if there are no slaves for this node so election was not possible 2011-04-07 23:06:01 +02:00
antirez 1793752d97 node cluster configuration file configurable via redis.conf 2011-04-07 21:34:41 +02:00
antirez 2bc52b2c02 useless newline removed from log message 2011-04-07 19:22:24 +02:00
antirez d01a6bb3f9 fixes to configuration loading and saving. However there is to still fix the logic for reconnection/behavior of nodes after a restart. 2011-04-07 19:04:16 +02:00
antirez 92690d29fe first version of cluster config loading code 2011-04-07 17:46:28 +02:00
antirez 726a39c15f initial cluster config load code 2011-04-07 12:55:02 +02:00
Pieter Noordhuis 02e6006532 Explicitly zero zval since it is stored on the stack 2011-04-06 16:39:22 +02:00
Pieter Noordhuis d4d3a70da2 Update target encoding for sorted set from rdb 2011-04-06 16:38:34 +02:00
Pieter Noordhuis 100ed062c0 Test for ENCODING_SKIPLIST instead of ENCODING_RAW 2011-04-06 16:17:07 +02:00
Pieter Noordhuis 4cc4d1648b Typo 2011-04-06 16:15:15 +02:00
Pieter Noordhuis 7d8c555e92 Merge branch 'unstable' into unstable-zset 
Conflicts:
	src/object.c
 2011-04-06 16:15:01 +02:00
antirez 0b7f6d0913 make sure that OBJECT ENCODING returns skiplist for sorted sets, and not raw, so that once we will merge specially encoded sorted sets everything will make sense. 2011-04-06 15:36:10 +02:00
antirez ece74202bb OBJECT command implemented 2011-04-06 12:19:45 +02:00
antirez f797c7dc17 bad data on RESTORE can no longer crash the server but create a memory leak with some input string 2011-04-05 13:57:28 +02:00
antirez f304f22b22 Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-04-01 19:00:21 +02:00
antirez 626f6b2d64 DUMP implemented, RESTORE and MIGRATE fixed. Use zcalloc() instead of zmalloc()+memset() in sds.c when a new string is created with NULL initialization pointer. 2011-04-01 18:59:28 +02:00
antirez 5b94b8ac5d fixed memory leak introduced with the previous commit. Many thanks to Pieter Noordhuis for spotting it in no time 2011-03-31 19:52:15 +02:00
antirez 25ef31920a Fixed issue #503. MONITOR + QUIT could crash the server, there are actually other interactions that could have the same effect (for instance Pub/Sub). 2011-03-31 16:44:43 +02:00
antirez 6c390c0b23 minor change with no actual effects til the loading node config code is in place 2011-03-30 18:12:51 +02:00
antirez ef21ab960e centralized cluster config file name. Assigned slots in CLUSTER NODES output and in cluster config file. 2011-03-30 17:41:13 +02:00
antirez 4b72c5617f fix to configuration saving on first start 2011-03-30 16:51:28 +02:00
antirez c7c7cfbddc cluster configuration saving 2011-03-30 14:58:19 +02:00
antirez e6f0a7b237 todo list for a first cluster howto guide for early testers 2011-03-29 18:02:34 +02:00
antirez 143d0077ba CLUSTER file updated with warning 2011-03-29 17:52:02 +02:00
antirez ecc9109434 Cluster branch merged to unstable. 2011-03-29 17:51:15 +02:00
antirez b46251d929 TODO modified 2011-03-29 11:04:55 +02:00
antirez cc1cb7e540 Merge remote branch 'origin/getkeys' into unstable 2011-03-29 10:54:38 +02:00
antirez a07bc92692 Preloading messages log level changed from WARNING to DEBUG 2011-03-28 19:03:57 +02:00
antirez 4b61ca460c fixed a bug in RENAME getKeys() function 2011-03-28 18:46:22 +02:00
antirez 6e1b9b58ec bug fixed in zunionstore specific getKeys() implementation 2011-03-28 18:21:06 +02:00
antirez b4b5144694 Fixes to the new preloading / key discovery APIs 2011-03-28 17:54:42 +02:00
antirez 42b2621cdc implemented preload of EXEC using the new getKeys() system 2011-03-28 17:29:26 +02:00
antirez b8513c9398 Merge branch 'unstable' 2011-03-23 23:13:34 +01:00
antirez a3cf904121 minor style change 2011-03-23 23:13:15 +01:00
antirez 9791f0f8ce new preloading implemented, still EXEC not handled correctly, everything to test 2011-03-23 18:09:17 +01:00
antirez e360e3bb7e Fixed sdssplitargs() handling of hex-style escapes. 2011-03-22 22:49:12 +01:00
antirez 03af999cb0 command table altered for the new API to get position of keys arguments in commands 2011-03-22 19:36:09 +01:00
Pieter Noordhuis 04a10b1a6d Test that sorted sets are removed when empty 2011-03-22 09:28:45 +01:00
Pieter Noordhuis 48991620f7 Remove sorted set when empty after ZREMRANGEBY* 2011-03-21 23:54:46 +01:00
antirez 612810af6e sdscatrepr() fixed. Now newlines and other special chars are escaped correctly 2011-03-20 18:24:49 +01:00
antirez 4eddb12156 gitignore updated 2011-03-15 10:47:04 +01:00
Pieter Noordhuis dddf5335f4 Fix DEBUG DIGEST, SORT and AOF rewrite 2011-03-14 13:30:06 +01:00
Pieter Noordhuis 9ec4ea20a7 Test both sorted set encodings for every test 2011-03-14 10:54:37 +01:00
Pieter Noordhuis 69298a05eb Offset should be size_t 2011-03-14 10:53:53 +01:00
Pieter Noordhuis 72690afdd2 Remove unused function 2011-03-14 10:53:14 +01:00
Pieter Noordhuis e53ca04b50 Test for empty inner range when looking for elements in range 2011-03-11 18:18:02 +01:00
Pieter Noordhuis 8588bfa370 Make zzl API unaware of the robj where the ziplist is stored 2011-03-11 17:06:07 +01:00
Pieter Noordhuis df26a0ae0b Encode sorted set after loading from dump 2011-03-10 17:50:13 +01:00
Pieter Noordhuis 255eebe221 Convert encoding of result when in limits 2011-03-10 17:02:05 +01:00
Pieter Noordhuis dba3a153a7 Remove comment 2011-03-10 16:53:20 +01:00
antirez 7226cbd1d9 Merge branch 'unstable' 2011-03-10 16:47:09 +01:00
antirez 0aed6a952a fixed diskstore race condition 2011-03-10 16:46:58 +01:00
antirez 7493d2a032 fixed diskstore race condition 2011-03-10 16:39:19 +01:00
Pieter Noordhuis 56ce42faf1 Generic iterator code for usage in ZUNIONSTORE/ZINTERSTORE 2011-03-10 16:34:52 +01:00
Pieter Noordhuis bbfe232f60 Make zzlLength take a ziplist argument 2011-03-10 16:17:14 +01:00
Pieter Noordhuis d4e07f1714 Add new string to long long function 2011-03-10 16:16:27 +01:00
antirez dc75b1edfb encoding agnostic intsets 2011-03-09 19:14:04 +01:00
antirez f220437450 ziplist are now endianess agnostic 2011-03-09 18:49:59 +01:00
antirez 336c82d583 zipmaps are now endianess agnostic, needed for on disk serialization of zipmaps without convertions layers 2011-03-09 17:31:02 +01:00
antirez b5325132f1 memrev variants only doing the work if the target host is big endian 2011-03-09 17:28:16 +01:00
antirez bcde63785d TODO updated 2011-03-09 16:36:02 +01:00
antirez 48e4621590 Ehm... sorry if we don't support PDP endianess 2011-03-09 16:33:09 +01:00
antirez e12cb14308 endianess conversion API, to be applied to specially encoded data types for arch agnostic encoding. 2011-03-09 16:24:18 +01:00
Pieter Noordhuis cc4c964b33 Fix used function in ZCARD 2011-03-09 16:13:39 +01:00
Pieter Noordhuis a669d5e999 Convert encoding when thresholds overflow 2011-03-09 16:13:06 +01:00
antirez b1a8e3e89e byte ordering detection in config.h 2011-03-09 15:44:21 +01:00
Pieter Noordhuis 3ea204e103 Configurable thresholds for encoded sorted sets 2011-03-09 14:01:57 +01:00
Pieter Noordhuis e12b27acf7 Persistence code for encoded sorted sets 2011-03-09 13:16:38 +01:00
Pieter Noordhuis d1c920c538 Support dual encoding for more commands 2011-03-09 12:37:59 +01:00
Pieter Noordhuis aff255c81d Support dual encoding for ZRANGEBYSCORE et al 2011-03-09 11:29:21 +01:00
Pieter Noordhuis 4c5f0966b2 Helpers to move around in encoded sorted set 2011-03-09 11:06:31 +01:00
Pieter Noordhuis 63b7b7fb34 Support dual encoding for ZREMRANGEBYRANK 2011-03-09 10:30:55 +01:00
Pieter Noordhuis 5d1b4fb698 Support dual encoding for ZRANGE 2011-03-09 00:12:29 +01:00
Pieter Noordhuis 0f23eb3b10 Properly free encoded sorted set 2011-03-08 23:56:59 +01:00
Pieter Noordhuis 4a14dbbac2 Look up and remove elements by range 2011-03-08 23:56:33 +01:00
Pieter Noordhuis 9f9b60f974 Typo 2011-03-08 22:14:46 +01:00
Pieter Noordhuis 8218db3dae Little less obfuscation 2011-03-08 21:36:46 +01:00
Pieter Noordhuis 0b10e10444 Support dual encoding in ZREM 2011-03-08 17:11:15 +01:00
antirez e270a7fa58 removed useless call 2011-03-08 16:52:39 +01:00
Pieter Noordhuis 3ca7532a2d Don't encode element argument when dealing with ziplist 2011-03-08 16:51:41 +01:00
Pieter Noordhuis 21c5b508a4 Initial work for ziplist backed sorted sets 2011-03-08 16:44:22 +01:00
Pieter Noordhuis 9e7cee0ed0 Add function to create ziplist-backed sorted set 2011-03-08 16:08:52 +01:00
Pieter Noordhuis 672b0a1b25 Fast conversion of double when representable as long long 2011-03-08 16:08:11 +01:00
antirez 3a21cb99e5 same final call of two if branches moved outside 2011-03-08 15:55:34 +01:00
antirez e09b51869b useless function removed 2011-03-08 15:10:20 +01:00
antirez c5b6f46132 comment typo fixed 2011-03-08 15:03:51 +01:00
Pieter Noordhuis 45290ad9bb Rename zset range functions 2011-03-08 10:57:53 +01:00
Pieter Noordhuis 8e1b327706 Test for ranges where min > max 2011-03-08 10:57:39 +01:00
Pieter Noordhuis df278b8b0b Compiler should decide on inlining 2011-03-08 10:57:33 +01:00
Pieter Noordhuis 22b9bf1594 Move logic concerned with zset ranges 
This also optimizes ZREVRANGEBYSCORE for pathological cases where a
sorted set contains many elements with the same score. Previously,
it would traverse the list from back to front in such a case.
 2011-03-08 10:57:24 +01:00
Pieter Noordhuis a300477368 Undo rename of function names where something went wrong 2011-03-08 10:57:06 +01:00
antirez c2f816b423 Merge remote branch 'pietern/unstable-sigs' into unstable 2011-03-07 12:22:00 +01:00
Pieter Noordhuis 3f4eef215b Show database number in cli prompt when non-zero 2011-03-06 21:14:40 +01:00
Pieter Noordhuis 96e34b3ce1 Store SELECTed database for reconnect (issue #468) 2011-03-06 20:46:49 +01:00
Pieter Noordhuis 33753a732a Minor memory leak in redis-cli (issue #464) 2011-03-06 20:13:01 +01:00
Pieter Noordhuis ca36b4ab31 Only save history when stdin is a tty (issue #465) 2011-03-06 20:02:33 +01:00
Pieter Noordhuis 633a941028 Save RDB on SIGTERM (see issue #471) 2011-03-06 17:49:40 +01:00
antirez d9fac6c07a possible fix for a failed assertion introduced with the latest diskstore speedup fix 2011-03-04 18:10:43 +01:00
antirez 0b537972f4 Fixed return value of GETRANGE / SUBSTR 2011-03-04 16:22:50 +01:00
antirez 4aec2ec8c4 AOF file descriptor leak fixed 2011-03-04 16:13:54 +01:00
antirez 15db4aa006 TTL command fixed to work reliably with diskstore 2011-03-04 15:49:01 +01:00
antirez 8b108ed307 debugging message moved from REDIS_WARING to REDIS_DEBUG 2011-03-04 15:45:38 +01:00
antirez be14f38de6 Redis manifesto added 2011-03-01 15:20:35 +01:00
antirez 1213abe477 fix type in rdbSaveKeyValuePair() when saving an intset. Don't merge this commit into 2.2 as will not apply cleanly. 2011-02-28 17:55:05 +01:00
antirez 26117e84f0 direct saving of specially encoded types implemented for lists and sets too 2011-02-28 17:53:47 +01:00
antirez 0a04b5f5a0 define name typo fixed 2011-02-28 17:36:12 +01:00
antirez 9ad853ccde added new RDB codes for ziplist encoded lists and intset encodeed sets 2011-02-28 17:06:31 +01:00
antirez 9fa7beb515 save ziplist encoded type as a different type id. Done as separated commit since this is the part that will not merge cleanly in 2.2 2011-02-28 17:06:09 +01:00
antirez f8956ed6d8 convert the zipmap into hash in rdb loading if the zipmap has too many elements 2011-02-28 16:55:34 +01:00
antirez 4d60dea853 RDB/AOF loading times logged with millisecond precision 2011-02-28 15:31:54 +01:00
antirez 1187b0a56e fix for zipmap raw saving in unstable branch 2011-02-28 15:13:49 +01:00
antirez d4fb9f4144 encoded types API to get blob length 2011-02-28 14:48:49 +01:00
antirez 2cc993655f save zipmap encoded hashes as blobs. Work in progress. 2011-02-28 09:56:48 +01:00
antirez 419e1cca74 less delays in diskstore when there are many writes. more work to do. 2011-02-25 19:11:25 +01:00
antirez 900ab8fc7a more disk store fields in INFO 2011-02-25 10:59:57 +01:00
Pieter Noordhuis 6c682e559c Restore argc/argv in EXEC after command is executed 2011-02-23 14:37:22 +01:00
antirez bf45f266b3 fixed client-libraries/README to reflect the new site path instead of google code 2011-02-22 18:22:16 +01:00
antirez 802fd01b7c linenoise updated 2011-02-22 17:46:11 +01:00
antirez 4a67d194e9 fixed color support for test suite. Now colors are shown as long as there is xterm somewhere inside the TERM env var 2011-02-22 16:41:27 +01:00
Paul Gideon Dann df50d707f3 Allow port to be 0 in configuration, in order to allow disabling of TCP. 2011-02-22 12:13:29 +01:00
antirez 29920dceb8 conflict resolved in cherry pick 2011-02-22 00:03:03 +01:00
antirez f96a9f82d8 suppress a Linux warning, for 2.2 sake 2011-02-21 17:51:52 +01:00
antirez 09d2abdca2 check return value of getcwd() 2011-02-21 17:51:28 +01:00
antirez 465b4189e6 CONFIG SET/GET for all the special encoding parameters of sets, lists, hashes 2011-02-17 12:20:59 +01:00
antirez 30318c1ddd SPOP replication/AOF patch ported to unstable branch 2011-02-16 12:41:40 +01:00
antirez 21cda08b3b CONFIG GET/SET dir implemented 2011-02-14 02:51:27 +01:00
antirez 05600eb8a7 fixed two diskstore issues, a quasi-deadlock creating problems with I/O speed and a race condition among threads 2011-02-11 11:16:15 +01:00
antirez 9c104c6886 introduced a new logging function for big messages 2011-02-09 10:10:35 +01:00
antirez a8cc969cd4 Merge remote branch 'pietern/unstable-brpoplpush' into unstable 2011-02-07 15:28:01 +01:00
antirez a53b4c2409 Fix for the previous port 0 patch, thanks to Pieter for noticing the error 2011-02-07 13:43:29 +01:00
antirez 68d6345ddf If port zero is specified, Redis will not listen for TCP connections 2011-02-07 12:52:01 +01:00
Pieter Noordhuis 554a5dd2fc Clarify comment 2011-02-03 12:56:53 +01:00
Pieter Noordhuis d5870d7ac3 Reply with single null bulk for unsuccesful BRPOPLPUSH 2011-02-03 12:56:50 +01:00
antirez 452229b6fb propagate key eviction to slaves and AOF, like for expires, so that replicas and AOFs are always consistent even in maxmemory mode. 2011-01-27 16:52:37 +01:00
antirez be86082be4 Fixed a theoretical non exploitable security bug reported by @chrisrohlf. In theory if we undefine SDS_ABORT_ON_OOM from sds.c AND modify zmalloc.c in order to don't quit on out of memory (but this would break every other part of Redis), on out of memory there is a possible heap overflow. 2011-01-27 10:27:25 +01:00
antirez cc9f0eee71 TODO list updated 2011-01-26 10:42:32 +01:00
antirez d7ed7fd2fc INFO commandstats section reset with config resetstat, a new microseconds per call field, and a few other minor things 2011-01-24 10:56:06 +01:00
antirez 9b45592c69 portability fix in redis benchmark, using zu format specified with size_t instead of lu 2011-01-24 10:05:42 +01:00
antirez 0d808ef275 Now INFO outputs the "INFO default" output if no argument is passed. Also added counts of calls for every command, thanks to an idea by Mike Shaver 2011-01-23 22:14:15 +01:00
antirez 1b085c9f48 INFO output refactoring. Now "INFO section" will report only the specified section 2011-01-23 16:41:07 +01:00
antirez d9cb288c9e profiling information in INFO 2011-01-23 11:46:34 +01:00
antirez 89a1433e69 Fixed issue #435 and at the same time introduced explicit ping in the master-slave channel that will detect a blocked master or a broken even if apparently connected TCP link. 2011-01-20 13:18:23 +01:00
antirez f25d5b5c9c Merge branch 'unstable' of github.com:antirez/redis into unstable 2011-01-20 11:00:25 +01:00
antirez c4df22afb6 Fixed compilation on FreeBSD 2011-01-20 10:59:49 +01:00
antirez a4fdc3c021 a better release note for Redis unstable provided 2011-01-17 12:36:37 +01:00
Pieter Noordhuis 3bcffcbe5b Remove client from list of unblocked clients when it is free'd 2011-01-17 10:04:13 +01:00
antirez b8cfcea424 default hash encoding settings also fixed in redis.conf 2011-01-15 22:31:07 +01:00
antirez 52dc87bb18 lame typo causing bad hashes memory characteristic 2011-01-15 22:24:22 +01:00
antirez 7a1fd61e3d implemented two new INFO fields showing the size of clients max input and output buffers. 2011-01-14 10:20:02 +01:00
antirez 3a73be7524 master-slave replication fixed, it was not listing any key using KEYS command in the slave. 2011-01-14 09:53:57 +01:00
antirez 5b831607e7 redisServer structure fields reordered for clarity 2011-01-14 00:19:19 +01:00
antirez 2d9fdb9d84 TODO modified 2011-01-10 21:45:08 +01:00
antirez 69cb546223 minor aesthetic change in test suite messages 2011-01-09 22:17:08 +01:00
antirez 322ea972fe run both inmemory and diskstore tests. 2011-01-09 19:42:56 +01:00
antirez a5062bbab0 fixed bgsave_in_progress in INFO when BGSAVEing with diskstore enabled, don't DEBUG FLUSHCACHE when bgsave is in progress. 2011-01-09 19:25:34 +01:00
antirez 0a0f83ab2c DEBUG FLUSHCACHE needs to wait that everything was synched on disk 2011-01-09 19:01:44 +01:00
antirez 69bfffb4a7 test adapted to run with diskstore, and a few bugs fixed 2011-01-09 18:25:34 +01:00
antirez 5e1d2d30f7 initial fix of the test suite to run both in in-memory and diskstore mode 2011-01-09 16:49:52 +01:00
antirez 67a1810b32 allocation stats in INFO 2011-01-09 15:56:50 +01:00
antirez 7cca366bdf TODO updated 2011-01-08 02:07:48 +01:00
antirez b39619d8e7 DEBUG OBJECT fixed with diskstore, force loading 2011-01-08 02:06:01 +01:00
antirez f1df1739e3 resolved a problem with delayed loading of keys 2011-01-08 01:42:08 +01:00
antirez cc2750670f blocking SAVE implemented 2011-01-08 00:09:57 +01:00
antirez 43574a72c3 avoid bgsaving temp files 2011-01-07 23:51:31 +01:00
antirez f03fe802b7 diskstore BGSAVE should work now 2011-01-07 23:41:00 +01:00
antirez 5b8ce85378 more work towards diskstore bgsave 2011-01-07 19:31:42 +01:00
antirez 36c17a53b6 source reshaped a bit to play well with a bgsaving thread, still work to do, does not compile. 2011-01-07 18:15:14 +01:00
antirez cb9b35c8ca Merge branch 'unstable' of github.com:antirez/redis-private into unstable 2011-01-07 16:43:51 +01:00
antirez 44898603aa TODO updated 2011-01-07 16:43:36 +01:00
antirez 249ad25f4f BGSAVE work in progress 2011-01-05 18:38:31 +01:00
antirez 51335102ac Merge remote branch 'pietern/writev-unstable' into unstable 2011-01-05 13:51:37 +01:00
antirez f771dc23a0 IO performances greatly improved under high writes load 2011-01-05 12:34:14 +01:00
Pieter Noordhuis 9b1d738f96 Remove newlines for networking errors 2011-01-05 10:51:37 +01:00
Pieter Noordhuis a2b33f2f93 Remove glueoutputbuf from redis.conf 2011-01-05 10:42:37 +01:00
Pieter Noordhuis a510cb0c03 Remove glueoutputbuf option and broken code 2011-01-05 10:39:41 +01:00
antirez 6eaad66373 Merge branch 'master' into unstable 2011-01-04 19:07:15 +01:00
antirez 1190c6cbf7 handle end of space on device in a better way 2011-01-04 19:03:23 +01:00
antirez 418d5eaf50 strictly honour memory limit 2011-01-04 18:53:09 +01:00
antirez 779fa2af7a rename(2) used in diskstore.c for atomic updates of keys 2011-01-04 18:28:04 +01:00
antirez fad97fbeea TODO list updated 2011-01-04 13:20:04 +01:00
antirez 5ab7bbfc27 fixed logging level for debugging message 2011-01-03 17:40:10 +01:00
antirez bafa88c8b7 stupid bug fixed 2011-01-03 17:37:20 +01:00
antirez 9b24d8adbe serious performance enhancement of diskstore 2011-01-03 17:27:47 +01:00
antirez 5d46e370b7 diskstore more fixes 2011-01-03 17:18:37 +01:00
antirez d158dc28f6 comments on top of dscache.c updated 2011-01-03 10:58:59 +01:00
antirez 0b305fcfd9 diskstore FLUSH* fixed 2011-01-03 10:51:45 +01:00
antirez c15a3887e0 diskstore bug fixing and negative cache proper implementation 2011-01-03 10:47:39 +01:00
antirez 120b9ba8f8 FLUSHALL / FLUSHDB for diskstore implemented 2011-01-03 10:17:39 +01:00
antirez d2033feb5c fixed cache-flush-delay configuration parsing 2011-01-02 22:27:26 +01:00
antirez 4942145d72 fixed a bug in diskstore 2011-01-02 21:39:17 +01:00
antirez e37efb0d8b diskstore race condition fixed 2011-01-02 17:42:04 +01:00
antirez 133cf28ee8 minor comment change of dscache.c 2011-01-01 21:44:14 +01:00
antirez 9a3730289f blocking load fixed with the new design 2011-01-01 21:39:48 +01:00
antirez 3be00d7ed6 implemented a different approach to IO scheduling, so object->storage is no longer used, instead there is a queue and hash table of IO tasks to process, and it is always possible to know what are the scheduled and acrtive IO operations against every single key. 2011-01-01 21:35:56 +01:00
antirez aa81e4d5f4 minor changes to doc and comments 2010-12-31 18:23:31 +01:00
antirez d934e1e85b negative caching implemented 2010-12-31 17:32:59 +01:00
antirez c4b64a1395 don't overload the IO job queue if there are alrady too much entries 2010-12-31 16:10:09 +01:00
antirez a440ecf0d3 major bug and a dead lock fixed 2010-12-31 14:30:24 +01:00
antirez 8e6bb67108 brainstorming with myself in dscache.c comments 2010-12-31 02:39:00 +01:00
antirez ad01a25553 blocking load of keys on lookup -- nor tested, nor finished 2010-12-31 00:18:17 +01:00
antirez 4e941ecacb short but important comment added 2010-12-30 19:20:23 +01:00
antirez 3122229228 handled DEL command as a special optimized case for disk store 2010-12-30 19:16:59 +01:00
antirez 8d51fb6a80 diskstore cache bug fixing 2010-12-30 18:37:46 +01:00
antirez 98a9abb66d don't use small shared integer objects when disk store is enabled 2010-12-30 18:17:01 +01:00
antirez 82ef6ebf73 added cache-flush-delay configuration directive. Don't BGSAVE when using diskstore 2010-12-30 17:53:28 +01:00
antirez eea15afe5f fixed bugs on diskstore cache 2010-12-30 17:21:54 +01:00
antirez 1fce320114 filesystem based KV store mostly implemented in diskstore.c 2010-12-30 17:07:06 +01:00
antirez 0b470ebdaa new gitingore 2010-12-30 16:42:11 +01:00
antirez 4ab988238f more work done for diskstore without trying to compile, more work needed to build again. 2010-12-30 16:41:36 +01:00
antirez 1609a1c42d fixed a few bugs in DS store, now it's somewhat about able to reply to 'GET foo' with a bogus value. 2010-12-30 15:55:24 +01:00
antirez 67b0b41c87 disk store logged messages improved 2010-12-29 23:08:18 +01:00
antirez ddbc81af33 diskstore directory structure creation 2010-12-29 23:00:00 +01:00
antirez f63f0928c3 cron part of disk store object cache implemented. Objects are pushed as IO jobs if needed, so that the IO thread will process them. 2010-12-29 22:18:20 +01:00
antirez cea8c5cd75 touched key for WATCH refactored into a more general thing that can be used also for the cache system. Some more changes towards diskstore working. 2010-12-29 19:39:42 +01:00
antirez d021221086 version set to 2.3.0 2010-12-29 17:02:43 +01:00
antirez 5ef640986b more step forwards for disk store to be able to run 2010-12-29 16:58:57 +01:00
antirez 5f6e1183e7 implemented complete job handling 2010-12-29 15:57:27 +01:00
antirez f34a6cd85e still more work for diskstore 2010-12-29 03:57:35 +01:00
antirez f2da3a620c more work converting vm into object cache 2010-12-28 20:03:21 +01:00
antirez f081eaf1c0 removed parts of VM that probably will be of no use with object cache 2010-12-28 18:46:34 +01:00
antirez 16d778780e a lot of code reworked/removed to implement object caching 2010-12-28 18:06:40 +01:00
antirez 697af434fb initial changes needed to turn the current VM code into a cache system. Tons of work to do still. 2010-12-28 15:20:20 +01:00
antirez 33388d4304 added diskstore.c in Makefile and prototypes in redis.h 2010-12-28 14:42:09 +01:00
antirez 52970711cf added diskstore.c, currently just a stub 2010-12-26 18:46:25 +01:00
antirez 04a2ade90d Merge remote branch 'pietern/bench-fix' 2010-12-23 13:18:07 +01:00
Pieter Noordhuis 8ce39260a4 Fix compiler warnings on Solaris 2010-12-23 11:26:11 +00:00
Pieter Noordhuis 89191613f1 limits.h is already included from redis.h 2010-12-23 11:14:05 +00:00
Pieter Noordhuis 65b472acce Solaris 10 doesn't know AF_LOCAL 2010-12-23 11:09:07 +00:00
Pieter Noordhuis ba55932674 Solaris doesn't support -rdynamic 2010-12-23 11:08:50 +00:00
Pieter Noordhuis a1e97d692e Update hiredis to 0.9.2 2010-12-23 11:01:11 +00:00
Pieter Noordhuis 2380388974 Randomize keys and set start time when first write event fires 2010-12-23 11:22:40 +01:00
Pieter Noordhuis 3c49070b35 Find substrings to randomize when the client is created 2010-12-23 11:04:44 +01:00
Pieter Noordhuis d69a483556 Make the MSET benchmark *really* work with 10 keys 2010-12-22 18:39:52 +01:00
Pieter Noordhuis f2f2424e00 Remove code duplication 2010-12-22 18:31:33 +01:00
antirez d3e5e28804 added mkrelease.sh script into utils. gitignore modified accordingly since this script was originally ignored 2010-12-20 15:47:33 +01:00
antirez d841dedb5f Merge branch 'master' of github.com:antirez/redis 2010-12-20 13:14:12 +01:00
antirez 4c22fd3409 TODO udpated 2010-12-20 13:13:46 +01:00
antirez 180596766b help.h updated 2010-12-19 15:19:08 +01:00
antirez f21779ff36 removed some dead code, added evicted keys counter separated from expired keys. 2010-12-19 15:15:08 +01:00
antirez 9d7165e885 overflow detection in INCR family functions 2010-12-19 12:22:12 +01:00
Didier Spezia f474a5bd4e Add wait states to deal with many connections. 2010-12-18 12:11:26 +01:00
Pieter Noordhuis 1cd3c1e08c Use multi-bulk protocol by default in redis-benchmark 2010-12-18 12:10:24 +01:00
Pieter Noordhuis 174df6fe49 Re-use variable data in redis-benchmark 2010-12-16 23:41:58 +01:00
Pieter Noordhuis 53f1d81712 Fix NULL-termination of variable data in redis-benchmark 2010-12-16 23:35:56 +01:00
Pieter Noordhuis 021321e0ef Update hiredis to 0.9.2 2010-12-16 23:32:02 +01:00
antirez 6418b4c790 help.h updated 2010-12-15 17:39:40 +01:00
antirez 59aee5513d added the mandatory Cheers in the release notes 2010-12-15 17:19:01 +01:00
antirez 9fecc4057a release notes, info on 2.0 compatibility 2010-12-15 17:18:15 +01:00
antirez 51716cbdb8 added more info in the release notes 2010-12-15 17:00:54 +01:00
antirez 9a7a07a466 removed a line from release notes that was not true (feature already backported into 2.0) 2010-12-15 16:54:08 +01:00
antirez 0898aa462e typo fixed 2010-12-15 16:51:01 +01:00
antirez 4c4dec7202 release notes added 2010-12-15 16:44:29 +01:00
antirez 401c3e213c bulk transfers limited to 512 MB as this is the new limit of all the redis strings 2010-12-15 16:07:49 +01:00
antirez beb1aab3fd Merge remote branch 'pietern/cli' 2010-12-15 16:03:43 +01:00
Pieter Noordhuis 28c07c7bf8 Specify multi-bulk delimiter via options 2010-12-15 16:02:07 +01:00
antirez 6674e3b910 added new HTML doc pages 2010-12-15 15:59:45 +01:00
Pieter Noordhuis 65add0a311 Improved raw output mode 
Raw output mode is selected by default when STDOUT is not a tty. For
ttys, raw output can be forced using the option "--raw".
 2010-12-15 15:59:06 +01:00
antirez b9f7e9e63c HTML doc updated 2010-12-15 15:49:29 +01:00
antirez 2136a880a7 version bumped to 2.1.8 2010-12-15 15:22:04 +01:00
Pieter Noordhuis f18e059e82 Make redis-cli help a little better 2010-12-15 15:00:47 +01:00
Pieter Noordhuis c392edf531 Show redis-cli version with repository information if present 2010-12-15 14:34:05 +01:00
Pieter Noordhuis c8061392e1 Minor changes to Makefile 2010-12-15 12:48:12 +01:00
antirez 53f0efe28c Merge remote branch 'pietern/string-patches' 2010-12-15 11:52:55 +01:00
Pieter Noordhuis 7d5f5712d9 Update tests for STRLEN 2010-12-15 11:49:39 +01:00
Pieter Noordhuis ad1b4f4f59 Use helper function for string object length 2010-12-15 11:49:04 +01:00
Pieter Noordhuis 1333f98dd2 Use helper functions in APPEND 2010-12-15 11:40:36 +01:00
Pieter Noordhuis 8f8eeffec1 Disable negative offsets for SETRANGE 2010-12-15 11:30:50 +01:00
Pieter Noordhuis e983cf34be Add fuzzy test for SETBIT 2010-12-15 11:20:54 +01:00
antirez 180e07b8c0 fixed test rendering when test failed in non verbose mode 2010-12-15 11:04:04 +01:00
antirez 7982173fef print test names inline in the non verbose output mode 2010-12-15 10:44:36 +01:00
antirez eae9cce166 colorized make test output when the verbose output is disabled (default) 2010-12-15 10:14:34 +01:00
Pieter Noordhuis 30407e1f4f Make SETBIT return original bit value 2010-12-15 00:42:32 +01:00
antirez 6a246b1e7e special encoding limits redefined, and all the config options for special encodings added in the example redis.conf file 2010-12-14 18:23:52 +01:00
antirez cc7c4158bc Merge remote branch 'jonahharris/syslog' 2010-12-14 17:53:28 +01:00
antirez 8e33831b2b Merge remote branch 'pietern/cli-monitor' 2010-12-14 17:48:52 +01:00
antirez 648e965460 removed a test that will never be true fixing the compilation on Linux 2010-12-14 17:46:20 +01:00
antirez 603e616bf4 Merge branch 'master' of github.com:antirez/redis 2010-12-14 17:42:46 +01:00
antirez 57997664ea Merge remote branch 'pietern/strrange' 2010-12-14 17:42:01 +01:00
Pieter Noordhuis 2b2eca1f56 Zero-pad timestamps in MONITOR output 
Original report and fix:
http://code.google.com/p/redis/issues/detail?id=404
 2010-12-14 17:39:34 +01:00
antirez 96b5d05fde initialized a few vars just to avoid warnings, not a real problem. 2010-12-14 16:39:33 +01:00
antirez f858c11d7d Merge remote branch 'pietern/brpoplpush' 2010-12-14 16:26:37 +01:00
Pieter Noordhuis d8f160a848 Add test cases for GETRANGE against integer-encoded strings 2010-12-14 15:35:35 +01:00
Pieter Noordhuis ef11bcccca Refactor and rename SUBSTR to GETRANGE 
SUBSTR is renamed to GETRANGE to have better consistency between command
names (with SETRANGE as its dual). GETRANGE is still aliased as SUBSTR.
 2010-12-14 15:16:29 +01:00
Pieter Noordhuis 9f9e1ceaa0 Add SETRANGE command implementation and tests 2010-12-14 14:20:51 +01:00
antirez 8c304be359 cow friendly HGETALL and variants 2010-12-14 12:10:51 +01:00
Pieter Noordhuis 7ecd4644e7 Don't decode object on STRLEN when not necessary 2010-12-14 10:42:41 +01:00
antirez 5ce3a24dba Linenoise updated to latest version 2010-12-10 19:22:13 +01:00
antirez afd438dfff previouse INCR implementation restored, was actually faster for some reson not fully clear at the moment 2010-12-10 19:15:00 +01:00
antirez d6107fd6f1 Merge remote branch 'pietern/testverbosity' 2010-12-10 17:24:03 +01:00
antirez a15742a41b dont take the fast path for INCR if the resulting integer will fit into a shared integer range 2010-12-10 17:23:38 +01:00
antirez b215a496a4 faster INCR doing far less allocation in common cases 2010-12-10 17:03:38 +01:00
Pieter Noordhuis 6f8a32d5c7 Be less verbose in testing; improve error handling 2010-12-10 16:13:21 +01:00
antirez dd48de748c fixed compilation on Linux 2010-12-10 15:47:01 +01:00
antirez 4b918769a1 command line option in test suite to activate valgrind mode 2010-12-10 15:40:48 +01:00
antirez e9bc56dbed valgrind suppression file added 2010-12-10 15:36:04 +01:00
antirez 3d24304ff9 HGET HMGET are now COW friendly, plus API refactoring and changes needed for the new implementation. 2010-12-10 15:17:55 +01:00
Pieter Noordhuis cc20906390 Change function name to match what it does 2010-12-10 12:16:16 +01:00
Pieter Noordhuis 586500c0ef Typo 2010-12-10 12:06:24 +01:00
Pieter Noordhuis eae33c1c81 Add generic function to grow an sds value 
Move logic concerned with setting a bit in an sds to the SETBIT command
instead of keeping it in sds.c. The function to grow an sds can and will
be reused for a command to set a range within a string value.
 2010-12-10 11:58:21 +01:00
antirez 1b508da7ca SINTER/MEMBERS are now COW friendly, also some refactoring around was needed to get this result. 2010-12-09 23:01:09 +01:00
Jonah H. Harris 8b5db0a8dc removed dup syslog-ident conditional 2010-12-09 11:26:33 -05:00
Pieter Noordhuis 076f88d657 Enforce maximum string value length of 512MB 2010-12-09 17:16:10 +01:00
Jonah H. Harris e1a586ee69 syslog support 2010-12-09 11:10:21 -05:00
Pieter Noordhuis 3c1bf4957e Add commands SETBIT/GETBIT 2010-12-09 16:39:33 +01:00
antirez a5be65f71c COW friendly versions of SPOP and SRANDMEMBER commands, with some change to the set encoding-agnostic API. 2010-12-09 10:21:02 +01:00
antirez d51ebef509 LRANGE converted into a COW friendly command. Some refactoring, comment, and new addReply*() family function added in the process. 2010-12-07 16:33:13 +01:00
Pieter Noordhuis a4ce758155 Don't execute commands for clients when they are unblocked 2010-12-06 16:39:39 +01:00
Pieter Noordhuis ecf9401415 Fix case and indent 2010-12-06 16:04:42 +01:00
Pieter Noordhuis 8a88c368ed Check other blocked clients when value could not be pushed 2010-12-06 16:04:10 +01:00
Pieter Noordhuis ac06fc011d Move code for pushing on a (blocking) RPOPLPUSH 2010-12-06 14:48:58 +01:00
Pieter Noordhuis 5fa95ad763 Rename blpop_blocked_clients to bpop_blocked_clients 2010-12-06 14:05:01 +01:00
Pieter Noordhuis c8a0070a61 Move timeout logic 2010-12-06 13:45:48 +01:00
Pieter Noordhuis bc8ffafeb9 Merge branch 'master' into brpoplpush 2010-12-06 12:31:56 +01:00
antirez bbac56c2f8 added support for ctrl-l and clear command into redis-cli. To clear the screen is a good idea from time to time :). Also linenoise updated to the current version to support this new feature. 2010-12-01 11:18:59 +01:00
antirez ce260f736e minor merge conflicts merging cli-help branch fixed 2010-11-30 11:39:55 +01:00
Michel Martens & Damian Janowski baa14ef913 Fix BRPOPLPUSH behavior for all use cases. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens 8987bf23bf Adhere to conventions. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens e3c51c4b1b Rename bstate to bpop. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens 59bd44d1c8 Remove warning. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens 7c25a43adc Handle BRPOPLPUSH inside a transaction. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens ba3b474111 Refactor code for BRPOPLPUSH. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens 357a841714 Move to struct. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens b2a7fd0cf7 BRPOPLPUSH. 2010-11-29 23:52:07 -03:00
Damian Janowski & Michel Martens 8a979f0390 Fix case in RPOPLPUSH. 2010-11-29 23:52:07 -03:00
Pieter Noordhuis c54afb6d0d Minor update to linenoise 2010-11-29 20:53:57 +01:00
Pieter Noordhuis b2cc45bfbc Refactor and support help for command names with spaces 2010-11-29 20:26:32 +01:00
Pieter Noordhuis 41945ba6ae Use linenoise completion API from redis-cli 2010-11-29 19:27:36 +01:00
Pieter Noordhuis ff4058183b Update linenoise 2010-11-29 19:27:06 +01:00
antirez d8d528e992 quick and dirty fix for hiredis bug creating problem with the new redis-cli connect commmand. Also change prompt when redis-cli is not connected 2010-11-29 12:20:17 +01:00
antirez efcf948c1a new redis-cli command connect 2010-11-29 12:17:55 +01:00
antirez 5b761a3387 gitignore now includes a few more files 2010-11-29 11:14:57 +01:00
Pieter Noordhuis a2a69d5803 Refactor help-related code into redis-cli.c 2010-11-28 21:37:19 +01:00
Pieter Noordhuis 50d0e82d54 Update help.h generator script to output man-style argument list 2010-11-28 17:45:58 +01:00
Pieter Noordhuis 2612e0521f Merge remote branch 'visionmedia/cli-help' into cli-help 2010-11-26 20:46:42 +01:00
Pieter Noordhuis a4e48b417d Don't hardcode make to "make" 2010-11-21 16:44:17 +01:00
Pieter Noordhuis bd70a5f588 Stop using /dev/null to find out the serialized object length 
Now the rdbSave* functions return the number of bytes written (or
required to write) in serializing a Redis object, writing to /dev/null
and using ftell (which doesn't work on FreeBSD) isn't needed anymore.
 2010-11-21 16:31:27 +01:00
Pieter Noordhuis 9a68cf91f0 Wrapper around fwrite to make API consistent 2010-11-21 16:12:25 +01:00
Pieter Noordhuis 8a623a98c3 Return number of bytes written from rdbSave* functions 2010-11-21 16:07:46 +01:00
Tj Holowaychuk 5397f2b596 Added redis-cli interactive help support 
updated via commands.json in redis-doc repo. Currently
use `make src/help.h` to re-generate. The following
are valid from the REPL:

  help
  help [command]
  help [group]
  help groups

ex:

  help sort
  help hash
 2010-11-16 05:50:26 -08:00
antirez c086b85afb added a few more files to gitignore 2010-11-15 15:50:41 +01:00
antirez 9fd01051bf Fix for bug 374, thanks to Jeremy Zawodny for reporting and tracing why it was crashing. 2010-11-12 20:02:20 +01:00
antirez 0c2f75c6d8 volatile-lru maxmemory policy segfault fixed, thanks to Anthony Lauzon for reporting the problem with the patch. Original patch modified a bit in order to avoid the double lookup if the policy is allkeys-lru 2010-11-11 13:19:17 +01:00
antirez 11fd0c422b now redis-cli is able to show the Git SHA1 in the version output 2010-11-08 16:26:02 +01:00
antirez d9d8ccab93 make sure to flush stdout every line read in monitor mode, to play well with redirection to file 2010-11-08 16:14:15 +01:00
antirez 5402c4262e added noeviction policy to redis maxmemory. ZSCORE removed from the list of commands that can't be called when we are low on memory, this command was added in the past for a stupid error. 2010-11-08 16:12:16 +01:00
antirez 240f8dbf3f build redis-server at the end so have a more pleasing to see Make output and the advice to run the test suite at the end. 2010-11-08 13:19:58 +01:00
antirez be98a33b51 fixed compilation with 32bit target 2010-11-08 12:53:36 +01:00
antirez 97e7f8aec3 non blocking loading of the DB / AOF with informations and ETA in INFO output 2010-11-08 11:52:03 +01:00
Pieter Noordhuis 57c9babd81 Update hiredis 2010-11-05 17:24:48 +01:00
Pieter Noordhuis abc3ff4d90 Only cascade clean target to deps 2010-11-05 17:22:16 +01:00
antirez 645e9962cb version bumped to 2.1.7 after merging with aaslave branch for non blocking slaves 2010-11-05 11:00:20 +01:00
antirez 3b5e72d402 Merge branch 'aaslave' 2010-11-05 10:59:49 +01:00
antirez ecc72ca17f version bumped to 2.1.6 2010-11-05 10:57:53 +01:00
antirez 56fce7ce7d Merge remote branch 'pietern/benchmark' 2010-11-05 10:29:59 +01:00
Pieter Noordhuis feecb608ed Use multi-bulk protocol in SET benchmark 2010-11-05 10:22:52 +01:00
antirez 4ebfc45528 config option to select if when replication link with master a slave should or not serve stale data 2010-11-04 19:59:21 +01:00
antirez 12ebe2ac17 replication asynchronous SYNC information in INFO output 2010-11-04 18:50:23 +01:00
antirez f6433915fe more replication info in logs 2010-11-04 18:14:20 +01:00
antirez 26b3366993 non blocking slave replication is now more non blocking than the first implementation... 2010-11-04 18:09:35 +01:00
antirez 62ec599c36 typos and minor stuff fixed in the new non blocking replication code 2010-11-04 17:35:03 +01:00
antirez f4aa600b99 first attempt to non blocking implementation of slave replication and SYNC bulk data download. Never compiled so far... 2010-11-04 17:29:53 +01:00
Pieter Noordhuis 8146e31677 Microsecond latency resolution in redis-benchmark 2010-11-04 16:15:35 +01:00
Pieter Noordhuis ec8f06675a Use hiredis from redis-benchmark 2010-11-04 13:37:05 +01:00
Pieter Noordhuis afc156c2d8 Update hiredis 2010-11-04 13:35:47 +01:00
antirez 4d7e125519 minor test suite bug fixed 2010-11-04 10:48:49 +01:00
Pieter Noordhuis a9b18e54d4 Stop using the freelist robj* cache 2010-11-04 10:09:30 +01:00
antirez 8df3dcada5 top level make clean also clean hiredis and linoise (deps) 2010-11-03 18:14:09 +01:00
Pieter Noordhuis e902b579b4 Import linenoise as dependency for redis-cli 2010-11-03 17:15:23 +01:00
Pieter Noordhuis 339b9dc2d2 Put duration in parenthesis 2010-11-03 17:07:10 +01:00
Pieter Noordhuis cfcd5d6d43 Add proper numbering for multi bulk replies in redis-cli 2010-11-03 17:03:54 +01:00
Pieter Noordhuis 7fc4ce13ed Use hiredis from redis-cli 2010-11-03 16:09:38 +01:00
Pieter Noordhuis 24f753a8b9 Add hiredis dependency for redis-cli, redis-benchmark, etc 2010-11-03 16:03:04 +01:00
antirez 8d3e063a0a added support for command renaming/suppression in redis.conf 2010-11-03 12:14:36 +01:00
antirez 1b1f47c915 command lookup process turned into a much more flexible and probably faster hash table 2010-11-03 11:23:59 +01:00
antirez 297e77c6ab Now it is possible to use quoted strings in the redis.conf file accordingly to the sdssplitargs() function capabilities. 2010-11-03 10:31:19 +01:00
antirez 0a546fc017 Merge remote branch 'pietern/unixsocket' 2010-11-02 23:47:52 +01:00
antirez 539fb43820 RSS information in INFO output 2010-11-02 22:47:10 +01:00
antirez 7d0966a6b7 Do not update the LRU info on key lookup when we have a saving child. With this trivial change the additional memory used while saving with a background child in presence of many read operations is zero. 2010-11-02 18:59:48 +01:00
antirez 3ce014c766 redis-cli in interactive mode now prints the time elapsed of the operation performed took more than half a second. 2010-11-02 18:08:30 +01:00
antirez b5b22da8e6 Revert "Function to compute RSS memory usage in a fast way, suitable to be called inside keys eviction loops" 
This reverts commit 7d47ecd543.
 2010-11-02 12:10:41 +01:00
antirez ca734d17ad Revert "Now maxmemory, VM, and everything else uses the fast RSS memory used estimation instead of raw memory reported by zmalloc(). This means that setting maxmemory to 2GB will really have the effect of using up to 2GB of memory." 
This reverts commit a3e60027e7.
 2010-11-02 12:09:59 +01:00
antirez a3e60027e7 Now maxmemory, VM, and everything else uses the fast RSS memory used estimation instead of raw memory reported by zmalloc(). This means that setting maxmemory to 2GB will really have the effect of using up to 2GB of memory. 2010-11-02 11:50:55 +01:00
antirez 7d47ecd543 Function to compute RSS memory usage in a fast way, suitable to be called inside keys eviction loops 2010-11-02 11:40:35 +01:00
antirez 10c12171cc removed a number of stupid compilation warnings on Linux 2010-11-02 11:15:09 +01:00
antirez 92e282288f zmalloc functions to get RSS and fragmentation refactored into two separated functions 2010-11-02 10:51:09 +01:00
antirez 21dbc6499a merge conflict resolved 2010-10-28 22:59:47 +02:00
Pieter Noordhuis 4794d88f15 Rewrite comment that was no longer valid 2010-10-28 16:59:05 +01:00
Pieter Noordhuis a3a323e0e5 When REDIS_CLOSE_AFTER_REPLY is set, there may never be new replies 2010-10-28 16:52:23 +01:00
Pieter Noordhuis 5e78edb350 Unify two client flags that mean the same 2010-10-28 15:07:45 +01:00
antirez 73abd0a9d2 Merge remote branch 'remotes/pietern/zrevrangebyscore' 2010-10-28 14:12:25 +02:00
antirez 1de98301f6 removed useless spaces from DEBUG OBJECT output 2010-10-27 17:11:17 +02:00
Pieter Noordhuis e584d82fec Return error to client on wrong type for HMGET 2010-10-26 12:33:17 +02:00
antirez 244201f6ba added a missing prototype from syncio.c in redis.h 2010-10-25 10:54:37 +02:00
antirez d08fac3eb9 more generally usable i/o functions moved to syncio.c 2010-10-25 10:53:28 +02:00
antirez 19e61097c5 synchronous I/O networking functions originally used just for replication refactored in a file as generally useful, they are used in the cluster branch for MIGRATE. 2010-10-24 16:22:52 +02:00
Pieter Noordhuis d94ac406ba Exclusively use either tcmalloc or OSX's native malloc_size() 2010-10-23 10:18:48 +02:00
Pieter Noordhuis 7cdc98b630 Don't use prefix when malloc_size() can be called 
Also, use tcmalloc functions explicitly via macros to prevent symbol
lookups to resolve to native malloc/free on OSX.
 2010-10-23 09:59:28 +02:00
antirez da47440d44 Make sure to reset the signal handler and deliver again the original crashing signal when dumping the stack trace. This will allow to dump the core if core dumping is enabled. 2010-10-22 23:30:48 +02:00
Robey Pointer d8a717fb1a if server.saveparamslen is not set, don't save the DB on exit. 2010-10-22 23:17:28 +02:00
antirez fc41345116 README more specific about tcmalloc space benefits 2010-10-22 00:18:26 +02:00
antirez 1a587ff843 Makefile typo fixed for tcmalloc option 2010-10-22 00:16:32 +02:00
antirez 13b3715925 reports if tcmalloc is in used in INFO output 2010-10-22 00:10:17 +02:00
antirez 0a802bd7a0 support for compiling with tcmalloc 2010-10-22 00:06:44 +02:00
antirez 75fcab8c23 version bumped to 2.1.5 2010-10-21 17:55:44 +02:00
Pieter Noordhuis a375b077cc Skip object encoding where it doesn't make sense 2010-10-17 18:13:31 +02:00
Pieter Noordhuis 19408d83a4 Object encoding in hash function is done by a more specific function 2010-10-17 17:46:55 +02:00
Pieter Noordhuis 33aba595b0 Removed unused command flags 2010-10-17 17:31:40 +02:00
Pieter Noordhuis 75b41de8ca Convert objects in the command procs instead of the protocol code 2010-10-17 17:21:41 +02:00
Pieter Noordhuis b19c33d48a Prevent clients from making too large multibulk requests 2010-10-15 19:15:38 +02:00
Pieter Noordhuis ea5b70924d Add benchmark for MSET 2010-10-15 18:17:06 +02:00
antirez 13a49af44b prevent small integer sharing when maxmemory is active. So every object will use a private LRU field and the LRU algorithm can work well 2010-10-15 18:04:05 +02:00
Pieter Noordhuis 1aa608fc68 Change protocol from bulk to inline in redis-benchmark 2010-10-15 17:34:20 +02:00
Pieter Noordhuis 9da6caac4e Don't reset the client when processCommand returns REDIS_ERR 2010-10-15 17:27:05 +02:00
Pieter Noordhuis dc11daf3b5 Change tests to use either the inline or the multibulk protocol 2010-10-15 17:25:20 +02:00
Pieter Noordhuis 5b12b47df9 Show output of leaks command on a leak 2010-10-15 15:56:16 +02:00
Pieter Noordhuis 00cf82c0bd Change tcl client to only use the multibulk protocol 2010-10-15 15:50:29 +02:00
Pieter Noordhuis cd8788f26d Refactor request parsing code for efficiency 2010-10-15 15:44:55 +02:00
Pieter Noordhuis 5a4f9f27e7 Add tests for OK on QUIT 2010-10-15 12:54:53 +02:00
antirez 9f8ded8ced CONFIG RESETSTAT no longer resets the server uptime. Now keyspace hits/misses are reset as well. 2010-10-15 12:29:05 +02:00
antirez 95506e4611 minor aesthetic change 2010-10-15 12:22:48 +02:00
antirez 53eeeaff08 added keyspace_hits and keyspace_misses fields in INFO output 2010-10-15 12:19:21 +02:00
antirez a36879293d maxmemory-samples implemented in CONFIG command and configuration file 2010-10-15 11:57:38 +02:00
antirez 670bf2fd36 Don't increment dirty on expireIfNeeded() as natural expires are not considered database changes. This will avoid useless read-only commands in the AOF file as a result of, for instance, a GET operation triggering an expirIfNeeded() call resulting in an expired key removed. 2010-10-15 11:29:03 +02:00
antirez b33ef40105 Merge remote branch 'pietern/ziplist-regression' 2010-10-14 21:23:01 +02:00
antirez 165346ca29 implemented different algorithms for maxmemory 2010-10-14 21:22:21 +02:00
Pieter Noordhuis 306c6a02e3 Replace ziplist stresser and fix regression 2010-10-14 21:11:42 +02:00
antirez ef59a8bc9e Object approximated LRU algorithm enhanced / fixed / refactored. This is used for the VM currently but will soon be used for maxmemory expiring. 2010-10-14 13:52:58 +02:00
Pieter Noordhuis 7236fdb22f Return error when min and/or max in the sorted set range spec is not a double 2010-10-13 21:59:24 +02:00
Pieter Noordhuis 91504b6cbe Make ZREMRANGEBYSCORE accept the same range spec as ZRANGEBYSCORE 
This allows to use inclusive/exclusive bounds for min and max when
deleting a range of scores from a sorted set.
 2010-10-13 21:43:58 +02:00
Pieter Noordhuis 26f3388d27 Merge branch 'master' into zrevrangebyscore 2010-10-13 20:29:50 +02:00
Pieter Noordhuis b04ce2a35c Merge master with resolved conflict in src/redis-cli.c 2010-10-13 18:55:46 +02:00
Pieter Noordhuis 4fe83b554a sockaddr_un.sun_path appears to never hold anything after accept() 2010-10-13 18:50:07 +02:00
Pieter Noordhuis ab17b909fe Use different accept handlers for TCP and unix socket connections 2010-10-13 18:34:24 +02:00
Pieter Noordhuis 893819801d Remove disabling TCP with port -1 2010-10-13 17:18:58 +02:00
Pieter Noordhuis 5d10923f7b Rename variable sockpath to unixsocket 2010-10-13 17:17:56 +02:00
Pieter Noordhuis 704bd093be Move creating socket/bind+listen on socket to separate functions 
Thanks to tav (http://github.com/tav) for original code.
 2010-10-13 16:47:22 +02:00
Pieter Noordhuis 941c9fa285 Return OK on QUIT 2010-10-13 11:25:40 +02:00
Pieter Noordhuis 9f1ae9abee Allow to specify which specific test files to run 2010-10-13 09:26:44 +02:00
antirez b4f2e412d0 free memory if the maxmemory parameter is reduced via CONFIG SET 2010-10-11 16:46:21 +02:00
antirez 1dd10ca233 maxmemory fixed, we now try to release memory just before we check for the memory limit. Before fixing there was code between the attempt to free memory and the check for memory limits, and this code could result into allocations going again after the memory limit. 2010-10-11 13:05:09 +02:00
antirez 144a5e72f2 fixed an alignment problem with time_t is 32 bit, long is 64 bit, and arch is sparc or any other where unaligned accesses will result to sigbus 2010-10-07 16:21:35 +02:00
antirez fdc0bde935 minor typo fixed, reported by Thomas Bassetto 2010-10-07 12:49:14 +02:00
antirez 4610b0332c intset stress testing added, ziplist stress testing relocated in a more appropriate place 2010-09-24 11:15:06 +02:00
antirez ef27ba988b explicit regression test for a ziplist bug added 2010-09-24 10:37:00 +02:00
antirez 1a06bf93c4 ziplist implementation fuzzy tests 2010-09-24 10:30:15 +02:00
antirez e43505d6e9 Merge remote branch 'pietern/ziplist-fix' 2010-09-24 01:15:16 +02:00
Pieter Noordhuis b0d605c1d6 Add regression test and fix for >255 byte string entries 2010-09-23 22:04:19 +02:00
antirez 30d31cc8bb Contributing file added 2010-09-23 18:24:47 +02:00
antirez 963238f713 more tests for sds.c 2010-09-23 16:39:02 +02:00
antirez 136cf53f22 minimal C test framework + a first example sds.c tests 2010-09-23 16:05:17 +02:00
Pieter Noordhuis 56e52b69fe Update rdb.c to properly work with new memory strategy for sorted sets 2010-09-22 18:07:52 +02:00
antirez 50a9fad5d5 two leaks fixed 2010-09-22 17:49:04 +02:00
antirez beb7756dcb error generation format reverted to the new style after merge 2010-09-22 16:10:13 +02:00
antirez b882056c93 Merge remote branch 'pietern/zset-mem' 2010-09-22 16:09:33 +02:00
antirez 5ca2f0c498 preventive conflict resolution to merge pietern/zset-mem 2010-09-22 16:09:30 +02:00
antirez 5171777bf1 Merge remote branch 'pietern/ziplist-eff' 2010-09-22 12:59:25 +02:00
antirez a4f3f93b90 new parsing code bugfixing 2010-09-17 16:05:01 +02:00
antirez 34a719d250 try to parse the request in a smarter way to gain speed... work in progress 2010-09-17 15:26:07 +02:00
Pieter Noordhuis d433ebc681 Finished code for sorted set memory efficiency 2010-09-16 15:42:36 +02:00
Pieter Noordhuis 25bb8a4452 Add ZREVRANGEBYSCORE and refactor Z*RANGEBYSCORE 2010-09-16 14:38:07 +02:00
Pieter Noordhuis 192fc3376a Merge branch 'zset-mem' into zrevrangebyscore 2010-09-16 14:32:30 +02:00
antirez 2b00385d51 Added used CPU statistics in INFO output, obtained via getrusage() 2010-09-16 13:28:58 +02:00
antirez 4c2e506a39 modified a bit addReply() to play better with copy on write now that we have a static buffer. Changed the name of a function from _ensureFileEvent() to _installWriteEvent(). 2010-09-16 13:08:40 +02:00
antirez 83f39c7ab2 Merge remote branch 'pietern/networking-perf' 2010-09-16 12:02:18 +02:00
Pieter Noordhuis f335779240 Static buffer in client struct has a constant size 2010-09-16 11:59:53 +02:00
antirez 89f9f83769 Merge remote branch 'pietern/networking-perf' 2010-09-16 11:38:40 +02:00
antirez 3856f14759 This should fix Issue 332: when there is a background process saving we still allow the hash tables to grow, but only when a critical treshold is reached. Formerly we prevented the resize at all triggering pathological O(N) behavior. Also there is a fix for the statistics in INFO about the number of keys expired 2010-09-15 14:09:41 +02:00
antirez 412e457c27 fixed typo in the latest commit 2010-09-14 15:18:18 +02:00
antirez 1d18f50458 Advertise the existence of redis-check dump --fix when logging an error about corrupted AOF file 2010-09-14 15:09:37 +02:00
Pedro Melo 0997b4119d Fixed missed use of INSTALL_TOP 
Thanks to sylr@github

Signed-off-by: Pedro Melo <melo@simplicidade.org>
 2010-09-13 16:50:57 +01:00
Pedro Melo e984050fb9 Make sure INSTALL_TOP exists before we install to it 
Signed-off-by: Pedro Melo <melo@simplicidade.org>
 2010-09-13 16:11:55 +01:00
Pedro Melo e13865033d Rename INSTALL_TOP to PREFIX; update documentation 
Signed-off-by: Pedro Melo <melo@simplicidade.org>
 2010-09-13 16:09:11 +01:00
Pieter Noordhuis 3c23ee1ba2 Fix another test that sometimes returned the swapped object instead of encoding 2010-09-13 16:59:46 +02:00
antirez bc63407be6 redis-cli does no longer try to auto detect if it is used inside a pipe. To read last argument from stdandard input there is to use the -x option. This will make it playing better inside cron scripts and in general when stdin is hacked. 2010-09-09 16:38:10 +02:00
antirez 7f9a4db3c0 Fix for the init script provided with Redis, thanks to Rowan. This fixes issue 316 2010-09-09 10:24:56 +02:00
antirez da14590bd9 Fix re-enabled again, I forgot to check if VM was enabled before calling handleClientsBlockedOnSwappedKey() 2010-09-08 13:47:28 +02:00
antirez 155fb4b45e latest fix reverted, there is some problem reported by the CI test 2010-09-08 13:45:51 +02:00
antirez 7f00cd2264 Fixed a race condition in VM happening when a key was deleted while there was a client waiting for this key to be resumed from swap to memory. The client would hang forever. 2010-09-08 13:26:16 +02:00
Pieter Noordhuis efc5d4cc0d Fix test that sometimes returned the swapped object instead of encoding 2010-09-07 11:49:33 +02:00
Pieter Noordhuis 106bd87a3c Fix bug where the client is not present in server.clients when free'ing it 
When creating the readable event results in an error (this happens when
the server hits OS limits), the client was not added to the list of
clients when freeClient was called. This results in an assertion error.
It is better to check this condition first and free the client
immediately when this condition occurs. Port of 00a90feb.
 2010-09-07 10:25:34 +02:00
Pieter Noordhuis 84403fe7c1 Allow a random seed argument for the ziplist test binary 2010-09-07 00:08:42 +02:00
Pieter Noordhuis 169d2ef1e0 Fix updating the prevlen field of consecutive entries 
In the condition where the prevlen field of the next entry on insert
and delete operations needs more bytes to be properly encoded, the next
entry also needs to be updated with a new prevlen. This patch makes sure
that this effect cascades throughout the ziplist.
 2010-09-07 00:04:57 +02:00
antirez abe18d0e00 Fix for solaris compilation bug Issue 325 2010-09-06 10:12:44 +02:00
Pieter Noordhuis 9e83ac06ef Merge branch 'master' into networking-perf 
Resolved conflict in src/db.c and changed adding an error to the reply
in blockingPopGenericCommand to use the new API.
 2010-09-03 16:44:50 +02:00
antirez 556bdfbab9 added some comment and changed coding style for fix for 237 2010-09-03 10:24:18 +02:00
Anko painting b435f64510 fix for issue 237 2010-09-02 21:13:27 -07:00
Pieter Noordhuis 49128f0b9d Fix bug in gluing a deferred multi bulk length to the next reply chunk 2010-09-02 23:34:41 +02:00
Pieter Noordhuis 3ab203762f Use specialized function to add status and error replies 2010-09-02 23:33:06 +02:00
Pieter Noordhuis 60361e5aac Add sds function that can be called with va_list 2010-09-02 21:00:15 +02:00
Pieter Noordhuis 36c19d03e0 Changed reply buildup internals 2010-09-02 19:52:04 +02:00
Pieter Noordhuis 4a7893ca9c Removed unneeded function 2010-09-02 19:52:04 +02:00
Pieter Noordhuis b70d355521 Use existing reply functions where possible 2010-09-02 19:52:04 +02:00
Pieter Noordhuis cd76bb651d Free the sds in addReplySds when it cannot be added to the reply 2010-09-02 19:52:04 +02:00
Pieter Noordhuis 2403fc9fde Intialize bufpos in the fake client 2010-09-02 19:52:04 +02:00
Pieter Noordhuis 0537e7bf80 Use specialized function to add multi bulk reply length 2010-09-02 12:51:14 +02:00
antirez 73db2acc37 memory fragmentation reporting in INFO also added for Mac OS X 2010-09-02 10:57:58 +02:00
antirez eddb388ef9 memory fragmentation ratio in INFO output 2010-09-02 10:34:39 +02:00
antirez a047bf52a4 fixed a few harmless warnings complining on Linux 2010-09-01 18:31:30 +02:00
antirez dbebd395eb Version is now 2.1.4 -- AKA 2.2-alpha1 2010-08-31 18:34:34 +02:00
antirez ec96ef47a6 Merge remote branch 'pietern/issue-300' 2010-08-31 17:39:06 +02:00
Pieter Noordhuis f85202c3dc Fix compilation errors and add warning for 32-bit platforms 2010-08-31 13:06:26 +02:00
antirez d320764706 We finally have an half decent README! (Issue 277) 2010-08-31 11:42:52 +02:00
antirez f7f12a606c resolved conflict merging pietern/bpop-timeout 2010-08-31 11:23:12 +02:00
antirez fc09b9f449 Merge remote branch 'pietern/issue-300' 2010-08-31 11:19:11 +02:00
antirez 08f55b786b faster server starting in Redis tests 2010-08-31 11:17:06 +02:00
Pieter Noordhuis 7b30cc3a7b Fix issue 300 by upgrading variable types to 64-bit 2010-08-31 10:21:35 +02:00
Pieter Noordhuis 1eb13e4913 Fix set tests to make sets have a deterministic encoding 2010-08-31 09:37:35 +02:00
Pieter Noordhuis 57b0738011 Don't build a reply when replaying the AOF 2010-08-30 16:51:39 +02:00
Pieter Noordhuis b301c1fc2b Wrapper for adding unknown multi bulk length to reply list 2010-08-30 16:39:14 +02:00
Pieter Noordhuis 834ef78e27 Refactor reply buildup for speed on large multi bulk replies 2010-08-30 16:39:08 +02:00
antirez fb92ecece7 BLPOP inside MULTI/EXEC block no longer crashes, instead if the list is empty the behavior is like if the timeout is reached. This fixes Issue 285 2010-08-30 16:31:03 +02:00
antirez 8079656a8e Now redis-cli replies to help showing some basic usage information (Issue 291) 2010-08-30 15:57:03 +02:00
antirez 93b2a7718e It is now possible to use authentication and DB selection options at the same time in redis-cli (Issue 298) 2010-08-30 15:36:13 +02:00
antirez e0e1c19520 Fixed MONITOR mode and Issue 296 2010-08-30 11:51:45 +02:00
antirez 8fedd04dcc Makefile deps updated 2010-08-30 11:37:17 +02:00
Pieter Noordhuis ed0dd55402 Show the current throughput while benchmarking 2010-08-30 11:25:02 +02:00
Pieter Noordhuis 36babc1e31 Refactor reply parsing code in redis-benchmark for efficiency 2010-08-30 11:14:54 +02:00
antirez 2f6b31c3bb Fix for a race in BGSAVE that may result in some data not being saved as soon as possible (when the configured saving triggers should fire). Also known as Issue 313, more details there in the google code issue. 2010-08-30 10:32:32 +02:00
antirez e5f257c2b2 fix for the prev fix 2010-08-27 17:06:36 +02:00
antirez c1ae36aea8 Fix for bug 312, yet to verify in a couple of minutes... 2010-08-27 17:04:26 +02:00
antirez 09252fc4f3 Fixed another instace of the Issue 173 2010-08-27 12:46:10 +02:00
antirez 357d36733d Fixed segfault in freeMemoryIfNeeded due to the fact that keys are now sds strings and not objects in the main hash table, thanks to Anthony Lauzon for spotting the bug and providing a patch. 2010-08-27 11:01:03 +02:00
antirez 2df84b7269 intset loading bug fixed 2010-08-26 19:10:40 +02:00
antirez ec7e138926 test for intset integer encodability test and some small refactoring 2010-08-26 18:47:03 +02:00
antirez 23c64fe50d translated a few long logn into int64_t for correctness and to avoid compilation warnings as well 2010-08-26 18:11:26 +02:00
antirez 588cd980e9 redis-cli tests commented out since there is a problem with the CI server and this tests, will be activated again once the problem is fixed 2010-08-26 17:52:02 +02:00
antirez 452ccf7a41 SORT stress testing against bigger aggregate values 2010-08-26 17:29:13 +02:00
antirez e4ecc93119 Version is now 2.1.3 2010-08-26 16:58:02 +02:00
antirez 4e0d34089a Merge remote branch 'pietern/redis-cli-fix' 2010-08-26 16:19:06 +02:00
Pieter Noordhuis 94364d53b4 Verify that the blocking pop timeout value is a non-negative integer 2010-08-26 14:05:14 +02:00
antirez e59a64b8d3 forgot the traceleaks var in tests set to 1, reverted to the default, 0 2010-08-26 13:18:56 +02:00
Pieter Noordhuis 2929ca9786 Fix parenthesis error on decrementing *argc 2010-08-26 13:18:44 +02:00
antirez bad7d097e9 fixed a memory leak in the new Set code 2010-08-26 13:18:24 +02:00
antirez 5f19e8a4a5 computeObjectSwappability is now able to compute swappability for intset encoded sets 2010-08-26 12:28:53 +02:00
Pieter Noordhuis 740eee1cc6 Fix type that was not renamed and compiler warning 2010-08-26 12:13:51 +02:00
antirez 674492bceb removed a duplicated ERRNO checking that is useless at all 2010-08-26 12:10:16 +02:00
antirez acc75bfd4f Merge remote branch 'pietern/intset-split' 2010-08-26 12:04:24 +02:00
Pieter Noordhuis f9d5c4e33c Make the function intsetUpgrade self-contained 2010-08-26 11:22:58 +02:00
Pieter Noordhuis 76864d5626 Expand macro's to functions for readability 2010-08-26 11:06:30 +02:00
Pieter Noordhuis b4b62c34db Use fstat to detect if stdin was redirected 2010-08-25 14:48:50 +02:00
Pieter Noordhuis f791d66e20 Make helper functions simpler 2010-08-25 14:15:41 +02:00
Pieter Noordhuis f9b252613b Comments in redis-cli tests 2010-08-25 14:08:32 +02:00
Pieter Noordhuis ae77016e57 Add a newline to tty output after every reply 2010-08-25 13:39:11 +02:00
Pieter Noordhuis 5d15b5207d Re-introduce the interactive field so we can reconnect in interactive mode 2010-08-25 13:09:22 +02:00
Pieter Noordhuis 4b93e5e267 Merge master and move argument splitting patch to sds.c 2010-08-25 13:08:43 +02:00
antirez c0b3d42372 redis-cli now supports automatically reconnection in interactive mode 2010-08-24 18:39:34 +02:00
antirez b37ca6edb1 Issue 179 fixed, now redis-cli is able to parse correctly multi bulk replies with elements that are errors 2010-08-24 18:08:09 +02:00
antirez 695fe87456 The pid file is now created only after the server is correctly initialied. It is also removed on sigterm and when the stack trace is produced after a sigbus or a sigsegv. This two changes should fix the Issue 175 2010-08-24 17:09:25 +02:00
antirez b91d605a35 slave now detect lost connection during SYNC, fixing Issue 173 2010-08-24 16:25:00 +02:00
antirez 778b2210a9 slave with attached slaves now close the conection to all the slaves when the connection to the master is lost. Now a slave without a connected link to the master will refuse SYNC from other slaves. Enhanced the replication error reporting. All this will fix Issue 156 2010-08-24 16:04:13 +02:00
antirez e452436a07 BLPOPping clients are no longer subject to connection timeouts, fixing issues 155 2010-08-24 12:10:59 +02:00
antirez 01daeecee7 added tests for invalid bulk argument 2010-08-24 11:49:05 +02:00
antirez a679185aa5 sanity check for the bulk argument in protocol parsing code, fixing issue 146 2010-08-24 11:45:05 +02:00
antirez e193873025 changed the comments on top of redis-copy.rb to reflect what the program really does 2010-08-24 10:10:01 +02:00
antirez c91abdcd07 Fixed overflow detection in argument to long convertion function in general, and in expire/ttl pairs specifically, addressing issue 54 2010-08-23 17:06:38 +02:00
Pieter Noordhuis cb72d0f155 Rename iterator to setTypeIterator for consistency 2010-08-21 11:38:24 +02:00
Pieter Noordhuis 029e5577ff Make SORT use the hybrid set accessors to allow sorting intsets 2010-08-21 11:20:41 +02:00
Pieter Noordhuis 2b9a59471f SORT tests with hash table encoded set as input 2010-08-21 11:03:56 +02:00
Pieter Noordhuis ced6709cb9 Make SORT tests use both ziplists and linked lists as input 2010-08-21 11:02:22 +02:00
Pieter Noordhuis 5d4f3a8c85 Move SORT tests around 2010-08-21 10:55:53 +02:00
Pieter Noordhuis a53ebb4c8e Don't abort test suite when the server block has a return value 2010-08-21 10:54:31 +02:00
Pieter Noordhuis 87c74dfaa8 Check if stroll return value was clamped 2010-08-20 13:42:42 +02:00
Pieter Noordhuis aaada3f962 Merge branch 'master' into intset-split 
Conflicts:
	src/Makefile
	src/t_set.c
 2010-08-20 12:40:55 +02:00
Pieter Noordhuis c470538142 Make ziplist schema more efficient for strings with length > 15 2010-08-13 19:29:22 +02:00
antirez cbce517145 redis cli argument splitting is general and is now moved into the sds.c lib 2010-08-05 11:36:39 +02:00
Pieter Noordhuis abb731e5b8 Deprecate starting interactive mode using the -i flag 2010-08-04 18:36:03 +02:00
Pieter Noordhuis cf0c6b78f1 Set tty before going into interactive mode to get non-pretty output when 
the commands are read from stdin.
 2010-08-04 18:16:39 +02:00
Pieter Noordhuis 3a51bff035 Change output format for non-tty redis-cli execution 2010-08-04 17:46:56 +02:00
Pieter Noordhuis 123a10f7a5 Let the output mode depend on having a tty or not 2010-08-04 17:16:05 +02:00
Pieter Noordhuis 07242c0ccf Tests for redis-cli in non-interactive mode 
Minor change in redis-cli output for the (multi-)bulk response but this
will be fixed in the next commit.
 2010-08-04 17:02:13 +02:00
Pieter Noordhuis 0439d792c4 Add tests for quotation in an interactive redis-cli session 
Patched redis-cli to abort on unexpected quotation. This caused
redis-cli to get into an infinite, memory-consuming loop.
 2010-08-04 16:15:33 +02:00
Pieter Noordhuis f2dd4769dd Tests for the interactive mode of redis-cli 
Changed redis-cli to output the raw response for a bulk reply when it is
run in interactive mode instead of checking isatty.
 2010-08-04 15:28:03 +02:00
Pieter Noordhuis 69ef89f2cf Reference zset score in zskiplistNode from dict entries 
This avoids the extra allocation of sizeof(double) for storing the score
of a zset entry in the hash table. Saves sizeof(double) + malloc
overhead = approx. 16 bytes per zset entry.
 2010-08-03 20:49:53 +02:00
Pieter Noordhuis 2159782b51 Use flexible array in zskiplistNode to reduce memory usage 2010-08-03 19:22:09 +02:00
antirez 1fb4e8def7 PERSIST: a fix and some basic test 2010-08-03 14:25:22 +02:00
antirez a539d29ac5 PERSIST command implemented 2010-08-03 14:19:20 +02:00
antirez 6146329f1f replication test with expires 2010-08-03 13:38:39 +02:00
Pieter Noordhuis a5639e7dd9 Change initialization to allow listening on both a port and socket 2010-08-03 13:33:12 +02:00
antirez 23d7298b15 Merge remote branch 'pietern/test-encoding-fix' into writeonexpire 2010-08-03 13:09:27 +02:00
antirez 2c572622fb no longer passing tests due to the new write-on-volatile semantics modified/removed 2010-08-03 13:08:32 +02:00
antirez 0cf5b7b57c allow to set a new EXPIRE of an existing volatile key 2010-08-03 12:26:30 +02:00
antirez c25a5d3b10 memory leak removed from expire propagation code 2010-08-02 21:37:39 +02:00
antirez bcf2995c98 support for write operations against expiring keys, by master-controlled expiring in replication and AOF synthesizing DEL operations 2010-08-02 18:13:39 +02:00
Pieter Noordhuis 7e91f971f7 Add support for domain sockets to redis-cli 2010-08-01 23:06:00 +02:00
Pieter Noordhuis c61e69257a Support for Redis to listen on a Unix socket 2010-08-01 22:55:24 +02:00
Pieter Noordhuis 6825491928 Fix assertion function on value encoding 2010-08-01 11:20:26 +02:00
antirez 0c7a9dec65 Merge remote branch 'pietern/list-test' 2010-07-31 13:55:06 +02:00
Pieter Noordhuis 673e1fb7e4 Change getDoubleFromObject to fail on NaN. 
Return an error when the resulting value is not a number (NaN). Fix
ZUNIONSTORE/ZINTERSTORE to clean up when a weight argument is not a
double value.
 2010-07-29 23:05:01 +02:00
Pieter Noordhuis d9e28bcf00 Fix ZUNIONSTORE/ZINTERSTORE to never store a NaN score. 
When +inf and -inf are added, the result is NaN. We don't want NaN
scores in a sorted set, so agreed on the result of this operation being
zero.
 2010-07-29 23:03:11 +02:00
Pieter Noordhuis 715c801a07 Use a large value to consistently trigger a list encoding, 
even when the list is swapped out and in again.
 2010-07-29 13:56:35 +02:00
Pieter Noordhuis 86d392498b ensure the value is swapped in before testing its encoding 2010-07-29 13:31:24 +02:00
antirez cbf7e1070a fix of the fix for the replication bug 2010-07-28 18:56:52 +02:00
antirez 8c1420ff2a Fixed a replication bug in ZINTERSTORE. 
In order to trigger the bug what's needed is to call ZINTERSTORE
resulting into an empty set created, bug against a key that already
existed. The command was not propagated, so the replica ended with the
key that the master removed. Sequence of command to reproduce:

redis-cli hset 446 34 905
redis-cli hset 446 393 911
redis-cli zadd 966 0.085412045980529885 652
redis-cli zadd 645 0.25081839284432045 280
redis-cli zinterstore 446 2 966 645
 2010-07-28 18:42:02 +02:00
antirez a0573260b0 better random dataset creation function in test. master-slave replication test now is able to save the two datasets in CSV when an inconsistency is detected. 2010-07-28 14:08:46 +02:00
antirez 6171250871 fixed a ziplist bug about encoding of integer values overflowing 64 bit 2010-07-27 15:26:08 +02:00
antirez db0c43a70c removed test code having bad effects... 2010-07-27 14:46:39 +02:00
antirez dd3f505ff5 Consistency test improved 2010-07-27 14:42:11 +02:00
antirez f99e660b44 malloc definition with deprecated attribute was duplicated, one removed 2010-07-27 14:30:02 +02:00
antirez 80091bbaac STRLEN command implemented 2010-07-27 10:09:26 +02:00
antirez e0be2289e9 hash table example commented out in dict.c 2010-07-27 10:00:38 +02:00
antirez b3aa6d712e use the function deprecated attribute if compiling with GCC to get warnings for malloc/free usages. We always want to use our zmalloc/zfree versions for memory usage tracking 2010-07-27 09:36:42 +02:00
Benjamin Kramer 399f2f401c Add zcalloc and use it where appropriate 
calloc is more effecient than malloc+memset when the system uses mmap to
allocate memory. mmap always returns zeroed memory so the memset can be
avoided.  The threshold to use mmap is 16k in osx libc and 128k in bsd
libc and glibc. The kernel can lazily allocate the pages, this reduces
memory usage when we have a page table or hash table that is mostly
empty.

This change is most visible when you start a new redis instance with vm
enabled.  You'll see no increased memory usage no matter how big your
page table is.
 2010-07-25 00:11:20 +02:00
Benjamin Kramer d9dd352b36 Remove _dictAlloc and friends 
zmalloc calls abort() so _dictPanic will never be called.
 2010-07-24 23:10:42 +02:00
Benjamin Kramer b1e0bd4b9b Reduce code duplication 2010-07-24 22:37:01 +02:00
Pieter Noordhuis e39c8b5047 exit with non-zero status when there are failed tests 2010-07-23 13:08:35 +02:00
antirez 230729617d don't open/close log file if log level is not matched 2010-07-22 23:31:40 +02:00
Pieter Noordhuis c8a10631d1 fix rare condition where 'key' would already be destroyed while is was needed later on 2010-07-22 16:06:27 +02:00
antirez 2f996f0217 defensive programming: set o->ptr to NULL before freeing objects 2010-07-22 15:48:57 +02:00
antirez e002ec6801 other shared objects where created in the I/O thread in createStringObjectFromLongLong. Fixed as well. 2010-07-22 14:48:45 +02:00
antirez cdbea20afb minor typo fixed in a comment 2010-07-22 13:12:24 +02:00
antirez 0e5441d816 don't use object sharing inside I/O threads, as a fix for a well known instability of VM introduced with the new object sharing code 2010-07-22 13:08:02 +02:00
antirez 1a71fb9669 vm_blocked_clients count fixed in INFO, thanks to Pietern Noordhuis 2010-07-21 13:16:26 +02:00
antirez 2cffe2993b TODO list modified, trivial change to source code 2010-07-16 23:56:18 +02:00
antirez 5b4bff9c17 WATCH is now affected only when write commands actually modify the key content 2010-07-12 12:01:15 +02:00
antirez e51a74aa40 fmacro included in linenoise.c 2010-07-09 10:51:41 +02:00
antirez 99628c1af8 redis-cli history saved across sessions 2010-07-07 18:44:53 +02:00
antirez 185cabda45 redis-cli is now able to report version information using -v 2010-07-06 19:17:09 +02:00
antirez 443d1e9efe Make install fixed using a dummy taget 2010-07-06 19:10:20 +02:00
antirez acc0185493 make install target, finally ;) 2010-07-06 19:07:16 +02:00
antirez f26dde8ca9 top level Makefile added, so you do not need to cd src 2010-07-06 18:54:54 +02:00
antirez b056ca39f2 improved random dataset creation in test: del, sunionstore, zunionstore 2010-07-06 18:30:38 +02:00
antirez 8b654e54c4 First implementation of a replication consistency test 2010-07-06 17:24:00 +02:00
Pieter Noordhuis d0a4e24e32 merged code from 184d74ab, 4774a53b, f483ce5f to new file structure 2010-07-05 15:16:33 -04:00
antirez 70a214c46d INSTALL file added BETATESTING.txt removed 2010-07-05 20:37:20 +02:00
antirez d06a5b23c8 Fixed compilation on *BSD systems 2010-07-05 20:14:48 +02:00
antirez b67d234563 Fixed a crash loading the AOF file containing MULTI/EXEC, a result of WATCH implementation. Test needed... 2010-07-05 20:06:54 +02:00
antirez b7a8daef60 WATCH will now consider touched keys target of EXPIRE command after the WATCH is performed, but not before 2010-07-05 19:38:12 +02:00
Pieter Noordhuis 2767f1c0c6 fix aof and digest code to work with dual set encoding 2010-07-02 20:42:20 +02:00
Pieter Noordhuis 96ffb2fe97 merged intset code into the split files 2010-07-02 19:57:12 +02:00
antirez 3688d7f308 Compilation fixed on Linux after the source code split 2010-07-01 21:13:38 +02:00
antirez d3b958c3fc Fixed MONITOR output for consistency: now integer encoded values are also formatted like this: "3932" 2010-07-01 20:22:46 +02:00
antirez 5bd09cd4c5 Fix the AOF fix of the latest commit ;) 2010-07-01 20:18:48 +02:00
antirez daf2049d0d fixed error code checking for *write operations and return value in AOF rewriting function 2010-07-01 20:13:33 +02:00
antirez af4e866dbb Merged with split 2010-07-01 16:55:12 +02:00
antirez 0f49d6b049 minor aesthetic change 2010-07-01 15:14:25 +02:00
antirez bb8716b6dc TODO updated 2010-07-01 14:52:01 +02:00
antirez 2c24c22039 Version is now 2.1.2 2010-07-01 14:47:26 +02:00
antirez 24110a4d7d Make log target fixed 2010-07-01 14:45:37 +02:00
antirez fb829ec6a2 gitignore modified 2010-07-01 14:41:03 +02:00
antirez e2641e09cc redis.c split into many different C files. 
networking related stuff moved into networking.c

moved more code

more work on layout of source code

SDS instantaneuos memory saving. By Pieter and Salvatore at VMware ;)

cleanly compiling again after the first split, now splitting it in more C files

moving more things around... work in progress

split replication code

splitting more

Sets split

Hash split

replication split

even more splitting

more splitting

minor change
 2010-07-01 14:38:51 +02:00
antirez b8b8501d70 Merge remote branch 'pietern/zfixes' 2010-06-25 00:23:38 +02:00
Pieter Noordhuis f483ce5ffe fix unexpected behavior on an out of range end index for LRANGE and LTRIM 2010-06-24 15:12:42 -07:00
Pieter Noordhuis c2ff0e90b8 more pub/sub tests 2010-06-16 11:03:23 +02:00
Pieter Noordhuis 4589a823fd initial basic pub/sub tests 2010-06-16 11:01:42 +02:00
Pieter Noordhuis 676740a960 fix BLPOP/BRPOP to use the wrapped function for list length 2010-06-15 21:23:24 +02:00
Pieter Noordhuis 5eedc9c65e tests for BLPOP/BRPOP via an option in the tcl client that defers reading the reply 2010-06-15 21:23:18 +02:00
Pieter Noordhuis 4774a53b24 fix behavior for out-of-range negative end index on ZREMRANGEBYRANK 2010-06-15 16:21:42 +02:00
Pieter Noordhuis 184d74abc6 more tests for zrange and zrevrange; fix behavior for out-of-range negative end index 2010-06-15 15:40:28 +02:00
antirez d5096a28d7 TODO updated 2010-06-14 17:59:03 +02:00
antirez d52e588869 Merge branch 'ltrim-tests' of git://github.com/pietern/redis 2010-06-14 10:22:38 +02:00
Pieter Noordhuis 7e79de541a rename "list" to "linkedlist" to be more verbose 2010-06-14 10:21:23 +02:00
antirez 7d04fc7563 allow running the test suite against an external Redis instance, without auto spawning 2010-06-14 10:19:45 +02:00
Pieter Noordhuis 08b5920750 change ltrim tests to cover all min/max cases and add stronger stresser 2010-06-14 09:45:34 +02:00
antirez 3042fb0537 Fixed deps in makefile and mkreleasehdr.sh script to really take advantage of the new trick to avoid recompilation of redis.c on git sha1 or dirty status change 2010-06-13 23:51:43 +02:00
antirez 7c4fc71c15 hopefully faster recompiling with a trick 2010-06-13 23:45:14 +02:00
Pieter Noordhuis 400aea2b13 expand the dictionary of the target set to the right size when converting from intset 2010-06-13 21:52:07 +02:00
Pieter Noordhuis 273f616930 make sure sets have the right encoding when loaded from rdb 2010-06-13 21:42:04 +02:00
Pieter Noordhuis ab37269c38 use max number of intset entries in tests and make SUNION/SINTER/SDIFF tests use less entries 2010-06-13 21:37:46 +02:00
antirez bb039e853d fixed a bug in rdbLoadObject abount specially encoded objects 2010-06-13 21:09:51 +02:00
Pieter Noordhuis 70ff3511bc configure maximum number of entries in an intset 2010-06-13 15:21:25 +02:00
Pieter Noordhuis a8dca69bb3 use raw strings when loading a hash from the rdb into a zipmap 2010-06-13 15:07:53 +02:00
Pieter Noordhuis b978abbf02 small refactor of SMOVE and tests for SMOVE on sets with different encoding 2010-06-13 11:16:18 +02:00
Pieter Noordhuis d0b58d5300 intset encoding for sets, refactored set tests to test both encodings 2010-06-12 22:25:22 +02:00
antirez f6475c7250 Merge branch 'expire' of git://github.com/pietern/redis 2010-06-12 16:26:04 +02:00
antirez ac9b8cfe57 Merge branch 'lists' of git://github.com/pietern/redis 2010-06-11 20:27:56 +02:00
Pieter Noordhuis e24d93762f intset housekeeping 2010-06-11 19:22:27 +02:00
Pieter Noordhuis 35cabcb505 wrapper functions for the set type to support multiple encodings 2010-06-11 18:35:57 +02:00
Pieter Noordhuis 7d288d6547 LPUSHX, RPUSHX, LINSERT only work on non-empty lists, so there are no clients waiting for a push 2010-06-11 17:35:48 +02:00
Pieter Noordhuis 23d3a5feef make LINSERT return -1 when the value could not be inserted 2010-06-11 17:34:23 +02:00
Pieter Noordhuis 3ab98cef4e move logic for obvious corner cases to intsetSearch 2010-06-11 15:07:24 +02:00
Pieter Noordhuis 144b0094c3 initial implementation for the intset 2010-06-11 15:07:24 +02:00
Pieter Noordhuis 70b4b320ae check if the list encoding needs to be changed on LPUSHX, RPUSHX, LINSERT 2010-06-11 14:52:35 +02:00
Pieter Noordhuis 244b873b0c make sure the value to insert is string encoded 2010-06-11 13:27:21 +02:00
Pieter Noordhuis bcfb387694 rename vars, move arguments, add comments 2010-06-11 13:15:28 +02:00
Pieter Noordhuis 1240552da9 always iterate from head to tail on LINSERT 2010-06-11 12:03:15 +02:00
Pieter Noordhuis 279d7e67cf use REDIS_TAIL to insert AFTER an entry and REDIS_HEAD to insert BEFORE an entry 2010-06-11 11:53:16 +02:00
Pieter Noordhuis 0e1684bcd0 move listTypeInsert to be grouped with other wrapper functions 2010-06-11 10:52:09 +02:00
Robey Pointer dedff272f6 squashed merge from robey/twitter3: LINSERT BEFORE|AFTER, LPUSHX, RPUSHX 2010-06-11 10:09:46 +02:00
Pieter Noordhuis 306974f5d7 remove pop function and the sds dependency; can be implemented using get+delete 2010-06-09 11:36:58 +02:00
Pieter Noordhuis 4e16d8b312 compute swappability for ziplist encoded lists 2010-06-07 21:53:21 +02:00
Pieter Noordhuis 829137b9a0 reuse the sds from the main dictionary in the expiration dictionary 2010-06-07 20:44:12 +02:00
antirez 056c19c6f2 TODO updated 2010-06-07 18:30:18 +02:00
antirez b785b2bf66 encode integers while loading an hash 2010-06-07 18:07:46 +02:00
antirez 65cc766581 Merge branch 'lists' of git://github.com/pietern/redis 2010-06-05 11:18:58 +02:00
Pieter Noordhuis cd627d4e78 fixed two leaks for the dual encoded lists 2010-06-05 11:16:50 +02:00
antirez 6b0253fabd TODO updated 2010-06-04 21:19:28 +02:00
antirez a26452263f DISCSARD now unwatches all keys, as it should 2010-06-04 21:19:06 +02:00
Pieter Noordhuis ab193fe452 generated tests for different encodings to avoid test code duplication 2010-06-04 17:05:54 +02:00
Pieter Noordhuis d4507ec615 refactor list tests to test both encodings; implemented assert functions 2010-06-04 16:31:27 +02:00
Pieter Noordhuis d1578a33ee renamed hash wrapper functions to match wrapper function naming convention: "<type>Type<func>" 2010-06-04 11:52:39 +02:00
antirez 422cf21f15 Merge branch 'lists' of git://github.com/pietern/redis 2010-06-04 11:36:37 +02:00
antirez 4dc1218c58 Merge branch 'smallkeys' 2010-06-04 11:35:43 +02:00
Pieter Noordhuis dda20542ab safety assert in listTypeNext 2010-06-04 11:32:33 +02:00
Pieter Noordhuis 003f0840ff renamed list wrapper functions to be more verbose 2010-06-04 11:04:02 +02:00
Pieter Noordhuis d0686e070d add thresholds for converting a ziplist to a real list 2010-06-04 10:57:31 +02:00
Pieter Noordhuis 846d8b3ea5 merge antirez/smallkeys 2010-06-04 10:10:50 +02:00
antirez 22194a7ffe test restored 2010-06-03 18:32:52 +02:00
antirez 4c8f23700b memory leak introduced in the latest big changes fixed 2010-06-03 18:22:15 +02:00
antirez 44262c58a4 Fixed VM bugs introduced with the top level keys as sds strings changes 2010-06-03 18:15:56 +02:00
antirez 0924181364 top level keys are no longer redis objects but sds strings. There are still a few bugs to fix when VM is enabled 2010-06-03 18:02:54 +02:00
Pieter Noordhuis 178d690372 update Makefile to include ziplist.o 2010-06-03 00:50:33 +02:00
Pieter Noordhuis 74e0f445a8 use ziplists in SORT STORE until the thresholds are determined 2010-06-03 00:48:52 +02:00
antirez 9fcfd6b651 Merge branch 'testsuite' of git://github.com/pietern/redis 2010-06-03 00:31:41 +02:00
antirez a7159fe817 Merge branch 'testsuite' of git://github.com/pietern/redis into smallkeys 2010-06-03 00:31:15 +02:00
Pieter Noordhuis 5a9fcb87ca tag memory leak check on kill server as "leaks" 2010-06-03 00:27:09 +02:00
Pieter Noordhuis afbf59145a tag test with sleep() as slow 2010-06-03 00:26:39 +02:00
Pieter Noordhuis f6fa411d6c make sure the config it returned when called without code 2010-06-03 00:25:32 +02:00
Pieter Noordhuis 6b6f101c27 tag more slow tests 2010-06-03 00:16:10 +02:00
Pieter Noordhuis 5713f06b33 change how arguments are passed from the AOF tests 2010-06-03 00:16:02 +02:00
Pieter Noordhuis 7a6ae0a2b2 scope res variable outside test 2010-06-03 00:06:58 +02:00
Pieter Noordhuis 7f7499eeac tags for existing tests 2010-06-02 23:22:25 +02:00
Pieter Noordhuis 73bd6c583b pass tags to filter and match via arguments 2010-06-02 23:22:20 +02:00
Pieter Noordhuis 6e0e5bedd9 basic support to tag tests 2010-06-02 22:53:22 +02:00
Pieter Noordhuis 9e5d2e8bd6 changed how server.tcl accepts options to support more directives without requiring more arguments to the proc 2010-06-02 22:23:52 +02:00
Pieter Noordhuis 38273a9ed6 removed obsolete code 2010-06-02 21:53:15 +02:00
Pieter Noordhuis 436f18b618 catch exceptions in the server proc, to be able to kill the entire chain of running servers 2010-06-02 21:53:10 +02:00
antirez 1edbae8678 Merge branch 'master' into smallkeys 2010-06-02 15:19:50 +02:00
antirez a89b7013ff smarter swapout policy on AOF too 2010-06-02 15:18:22 +02:00
antirez 7e02fe32d3 better swapout policy while loading RDB file 2010-06-02 15:13:53 +02:00
antirez 169dd6b7d3 minor code comment change 2010-06-02 15:00:02 +02:00
Pieter Noordhuis e1f93d4b2c use integer types from stdint.h to be more verbose on the size in bytes of encoded elements. update list length to use 2 bytes instead of 1. 2010-06-01 18:55:37 +02:00
Pieter Noordhuis ffc1585267 added stress test for heavy i/o in ziplists 2010-06-01 18:51:23 +02:00
Pieter Noordhuis b84186ff2d fix signedness errors in ziplist testing code 2010-06-01 17:57:09 +02:00
antirez 612e4de8ce minor code movements and free object pull restored to 1 million 2010-06-01 17:08:43 +02:00
antirez 356f923913 TODO updated with syslog plans for 2.2 2010-06-01 15:01:16 +02:00
antirez dbc289aed1 Debug message was printing stuff that are sometimes not initialized/valid 2010-06-01 14:22:21 +02:00
antirez 4e538759c4 Merge branch 'smallkeys' of github.com:antirez/redis into smallkeys 2010-06-01 14:18:38 +02:00
antirez a4798f733d fixed a few comments 2010-06-01 14:18:30 +02:00
antirez e4ed181d40 fixed bugs introduced in the rewrite of the new VM engine 2010-06-01 14:15:46 +02:00
Pieter Noordhuis 6ddc908ab6 support rewriting the AOF with dual list encoding 2010-05-31 23:49:16 +02:00
Pieter Noordhuis 9eaef89fbc small refactor of fwrite* commands for AOF rewrite to allow writing a bulk long long 2010-05-31 23:41:04 +02:00
Pieter Noordhuis dc845730a9 use list wrapper functions in computing the dataset digest 2010-05-31 23:35:51 +02:00
Pieter Noordhuis d71b98650f ziplistNext should work as expected when called with a pointer to ZIP_END 2010-05-31 23:35:21 +02:00
Pieter Noordhuis a03611e133 update SORT to work with the dual list encoding 2010-05-31 23:22:00 +02:00
Pieter Noordhuis 1cd92e7f04 function to create a new ziplist encoded list 2010-05-31 23:10:05 +02:00
antirez 7dd8e7cfdf fixed missing incrRefCount 2010-05-31 22:56:53 +02:00
Pieter Noordhuis 23f964946b support rdb saving/loading with dual list encoding 2010-05-31 22:26:32 +02:00
Pieter Noordhuis b6eb970394 fixed signedness and disambiguate variable names 2010-05-31 22:26:32 +02:00
Pieter Noordhuis 2796f6da7b added rdb save function to directly save long long values 2010-05-31 22:26:25 +02:00
Pieter Noordhuis 0f62e1775d update RPOPLPUSH to support dual encoding 2010-05-31 20:47:39 +02:00
Pieter Noordhuis be02a7c0d6 update list iteration semantic to work as expected (i.e. "while(lNext(..))") 2010-05-31 20:25:31 +02:00
Pieter Noordhuis 6a8e35ad92 ziplistDelete no longer needs a direction now ziplistPrev is fixed 2010-05-31 20:18:35 +02:00
Pieter Noordhuis 8632fb3040 ziplistPrev should return the tail when the argument is ZIP_END 2010-05-31 20:17:21 +02:00
antirez 560db612e1 first step of VM rewrite. blocking VM tests passing, more work needed in the async side 2010-05-31 20:10:18 +02:00
antirez 752203d86d Merge branch 'no-appendfsync-on-rewrite' 2010-05-31 10:08:14 +02:00
Pieter Noordhuis 3fbf9001ce fix LREM to remove *all* occurances when a zero argument is given 2010-05-30 03:25:14 +02:00
Pieter Noordhuis bd8db0ada8 fixed LINDEX to always return bulk response 2010-05-30 03:25:04 +02:00
Pieter Noordhuis dbaa41c618 the tail offset must be an integer pointer to hold a 32-bit offset 2010-05-30 03:06:12 +02:00
Pieter Noordhuis d2ee16abec update LREM to support dual encoding via extra iteration primitives 2010-05-30 02:17:36 +02:00
Pieter Noordhuis 9ae6b0be4a support dual encoding in LTRIM 2010-05-30 02:16:12 +02:00
Pieter Noordhuis a6dd455b09 update LRANGE to use basic iteration code to support dual encoding 2010-05-30 02:15:34 +02:00
Pieter Noordhuis 697bd5673f inline support for dual encoding in the LINDEX and LSET commands 2010-05-30 02:11:59 +02:00
Pieter Noordhuis d72562f7ba generic pop and length function for ziplist encoding 2010-05-30 01:51:55 +02:00
Pieter Noordhuis c7d9d662a4 generic push function that supports the dual encoding 2010-05-30 01:46:59 +02:00
Pieter Noordhuis 0f3dfa87bc change delete function to accept a direction argument, so "p" can be properly updated 2010-05-30 01:39:41 +02:00
Pieter Noordhuis 033fb554be expose extra functionality from ziplist.c 2010-05-30 01:38:56 +02:00
Pieter Noordhuis 177a0a0b0d code style consistency fixes 2010-05-30 01:37:03 +02:00
Pieter Noordhuis c03206fdf1 ziplistIndex now accepts negative indices 2010-05-29 22:17:16 +02:00
Pieter Noordhuis a24ba809d1 fix compile warnings 2010-05-29 21:10:17 +02:00
Pieter Noordhuis 7b1f85c0a2 use simpler encoding for the length of the previous entry 2010-05-29 21:10:17 +02:00
Pieter Noordhuis 1ce81fa59b replace functions to get pointers to head and tail by macros 2010-05-29 21:10:17 +02:00
Pieter Noordhuis 6435c76772 function to insert an element at an arbitrary position in the list 2010-05-29 21:10:17 +02:00
Pieter Noordhuis 0c0d056412 extract a generic delete function that can be used in pop and delete(range) 2010-05-29 21:10:17 +02:00
Pieter Noordhuis bb57b965c4 use the entry struct in zipRawEntryLength 2010-05-29 21:10:17 +02:00
Pieter Noordhuis 03e52931dd rename argument names to s* to disambiguate from e* 2010-05-29 21:10:17 +02:00
Pieter Noordhuis c8d9e7f4c0 change ziplistRepr to use the entry struct 2010-05-29 21:10:17 +02:00
Pieter Noordhuis d593c48869 modify compare function to check if the encoding is equal before comparing 2010-05-29 21:10:16 +02:00
Pieter Noordhuis a5456b2cb6 use a struct to retrieve all details for an entry 2010-05-29 21:10:16 +02:00
Pieter Noordhuis dcb9cf4e82 initial implementation for making the ziplist doubly linked 2010-05-29 21:10:16 +02:00
Pieter Noordhuis fc2c0f7a6c fix some warnings 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 4812cf2873 add function to retrieve ziplist size in bytes 2010-05-29 21:10:16 +02:00
Pieter Noordhuis c4aace9003 fix compare function of ziplist to only load integer from ziplist when it is encoded as integer 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 6205b46387 add function to retrieve length of ziplist 2010-05-29 21:10:16 +02:00
Pieter Noordhuis aa549962a4 re-introduce ZIP_BIGLEN for clarity 2010-05-29 21:10:16 +02:00
Pieter Noordhuis c7019f7019 added header ziplist.h 2010-05-29 21:10:16 +02:00
Pieter Noordhuis c09c2c3b04 code to compare strings with entries in ziplist, regardless of their encoding 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 75d8978e7a updated iteration code to work well with different encodings 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 37fff074a2 move code from zip.c to ziplist.c 2010-05-29 21:10:16 +02:00
Pieter Noordhuis af5f66fb98 partial revert of c80df5 because ziplist functions are starting to divert too much from zipmap functions 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 29b14d5fac initial work for integer encoding in ziplists 2010-05-29 21:10:16 +02:00
Pieter Noordhuis f6eb17477a move length housekeeping to a macro 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 0f10458c4f allow entries to be deleted in place when iterating over a ziplist 2010-05-29 21:10:16 +02:00
Pieter Noordhuis 924727d905 allow pointer to be stored to current element when iterating over ziplist 2010-05-29 21:10:16 +02:00
Pieter Noordhuis ba5b4bde21 rename ziplistDelete to ziplistDeleteRange 2010-05-29 21:10:15 +02:00
Pieter Noordhuis 779deb60f9 code to delete an inner range from the ziplist 2010-05-29 21:10:15 +02:00
Pieter Noordhuis 33c1269efc check if *value is non-NULL before setting it 2010-05-29 21:10:15 +02:00
Pieter Noordhuis 335d16bc0f change iteration code to avoid allocating a new sds for each traversed entry 2010-05-29 21:10:15 +02:00
Pieter Noordhuis 08253bf42b code to iterate over a ziplist 2010-05-29 21:10:15 +02:00
Pieter Noordhuis 11ac6ff613 implementation for a ziplist with push and pop support 2010-05-29 21:10:15 +02:00
Pieter Noordhuis 3d04d29e7f extracted general methods to zip.c for reuse in other zip* structures 2010-05-29 21:10:15 +02:00
antirez d55d5c5dd3 command table size calculated with sizeof 2010-05-28 23:44:32 +02:00
Pieter Noordhuis 1a132bbc9c use qsort and bsearch to lookup commands in O(log(N)) instead of O(N) 2010-05-28 22:14:10 +02:00
antirez 3caf15e6da Merge branch 'cli-stdin' of git://github.com/pietern/redis 2010-05-28 20:26:01 +02:00
antirez 5fc9229c34 Fixed ZINCR Nan bugs leading to server crash and added tests 2010-05-28 12:24:47 +02:00
antirez d5d23dabd7 redis.conf new features the new option, a minor typo preventing the compilation fixed 2010-05-28 10:48:04 +02:00
antirez b0bd87f60c don't fsync after a rewrite if appendfsync is set to no. use aof_fsycn instead of fsync where appropriate 2010-05-28 10:40:53 +02:00
antirez 38db91713c added new option no-appendfsync-on-rewrite to avoid blocking on fsync() in the main thread while a background process is doing big I/O 2010-05-28 10:35:23 +02:00
antirez 8a3b0d2d9a Added Git sha1 and dirty status in redis-server -v output 2010-05-28 00:23:27 +02:00
antirez ea2c08e922 changed the message in the Makefile with the new command like to run the test suite 2010-05-28 00:20:12 +02:00
Vincent Palmer ca3f830b32 Fixed typo. 2010-05-27 16:15:09 +01:00
antirez 4ea93ad3ac new multi/exec tests 2010-05-27 16:08:44 +02:00
Pieter Noordhuis a2f4f8711a build command outside while loop 2010-05-26 18:22:05 +02:00
Pieter Noordhuis 37dc9e5abb require the flag "-c" to be used for redis-cli to read the last argument from stdin 2010-05-26 18:18:37 +02:00
antirez 90fdc82697 Merge branch 'master' into nested-multi 2010-05-26 17:55:28 +02:00
antirez 1ad4d31631 Fix EXEC bug that was leaving the client in dirty status when used with WATCH 2010-05-26 17:50:29 +02:00
antirez 6531c94d28 raise error on nested MULTI and WATCH inside multi 2010-05-26 16:47:37 +02:00
Josiah Carlson bc000c1db0 allow regular sets to be passed to zunionstore/zinterstore 2010-05-26 16:07:04 +02:00
antirez 9005896c0a Version is now 2.1.1 2010-05-25 22:42:47 +02:00
antirez b167f87705 RENAME is now WATCH-aware 2010-05-25 21:26:10 +02:00
antirez cef34df02a TODO updated 2010-05-25 20:28:22 +02:00
antirez 9b30e1a207 WATCH is now able to detect keys removed by FLUSHALL and FLUSHDB 2010-05-25 19:30:24 +02:00
antirez c20c189db5 WATCH tests 2010-05-25 14:04:46 +02:00
antirez 87c68815c3 minor bug fixed in WATCH 2010-05-25 13:21:23 +02:00
antirez 37ab76c99d WATCH for MULTI/EXEC (CAS alike concurrency) 2010-05-25 13:15:11 +02:00
antirez 75a190ca72 gitignore updated 2010-05-25 10:06:37 +02:00
antirez fda32cb601 Master is now already unfreezed, unstable, and ready to hacking sessions! 2010-05-21 15:48:39 +02:00
544 changed files with 123528 additions and 25855 deletions
Expand all files Collapse all files
.gitignore
+15 -8
View File
@@ -1,15 +1,22 @@
00-RELEASENOTES
.*.swp
*.o
*.rdb
*.log
redis-cli
redis-server
redis-benchmark
redis-check-dump
redis-check-aof
redis-*
doc-tools
mkrelease.sh
release
myredis.conf
misc/*
src/release.h
appendonly.aof
SHORT_TERM_TODO
release.h
src/transfer.sh
src/configs
redis.ds
src/redis.conf
deps/lua/src/lua
deps/lua/src/luac
deps/lua/src/liblua.a
.make-*
.prerequisites
*.dSYM
00-RELEASENOTES
+297
View File
@@ -0,0 +1,297 @@
Redis 2.6 release notes
=======================
** IMPORTANT ** Check the 'Migrating from 2.4 to 2.6' section at the end of
this file for information about what changed between 2.4 and
2.6 and how this may affect your application.
--------------------------------------------------------------------------------
Upgrade urgency levels:
LOW: No need to upgrade unless there are new features you want to use.
MODERATE: Program an upgrade of the server, but it's not urgent.
HIGH: There is a critical bug that may affect a subset of users. Upgrade!
CRITICAL: There is a critical bug affecting MOST USERS. Upgrade ASAP.
--------------------------------------------------------------------------------
--[ Redis 2.6.6 ]
UPGRADE URGENCY: CRITICAL if you experienced one more more crashes.
MODERATE if Redis is running fine for you.
* [BUGFIX] Jemalloc updated to 3.2.0.
--[ Redis 2.6.5 ]
UPGRADE URGENCY: MODERATE
Warning: this release of Redis introduces a different behavior in MULTI/EXEC
handling of errors. This was done because the new behavior is safer
compared to the old one, and should not break any code targeting
Redis 2.6 in a critical way.
For more information check http://redis.io/topics/transactions
* [IMPROVED] RDB/AOF childern now log amount of additional memory used
because of copy on write.
* [BUGFIX] MIGRATE non critical fixes (see commits for details).
* [BUGFIX] MULTI/EXEC: now EXEC aborts on errors before EXEC.
* [BUGFIX] Fix integer overflow in zunionInterGenericCommand resulting
into Z[INTER|UNION][STORE] commands to crash under extremely
unlikely conditions (almost impossible in real world).
* [BUGFIX] EVALSHA is now case insensitive (and will not crash).
--[ Redis 2.6.4 ]
UPGRADE URGENCY: LOW
* [IMPROVED] BSD license and copyright notice added to every .c and .h file.
--[ Redis 2.6.2 ]
UPGRADE URGENCY: LOW
* [BUGFIX] The compilation fix for RHLE5 in 2.6.1 was broken. Fixed.
* [IMPROVED] Linenoise updated, now supports Ctrl+w.
---[ Redis 2.6.1 ]
UPGRADE URGENCY: LOW
* [BUGFIX] Compilation on Linux < 2.6.17 or glibc < 2.6 fixed (RHLE5 & co).
---[ Redis 2.6.0 ]
UPGRADE URGENCY: HIGH
* [BUGFIX] Allow AUTH when server is in -BUSY state because of a slow script.
* [BUGFIX] MULTI/EXEC flow now makes sense when observed in MONITOR
* [BUGFIX] SCRIPT KILL now uses different error prefixes for different errors.
* [BUGFIX] Default memory limit for 32bit archs lowered from 3.5 to 3 GB.
* [BUGFIX] redis-check-dump is now compatible with RDB files generated by 2.6.
* [IMPROVED] New field in INFO: slave_read_only.
---[ Redis 2.5.14 (2.6 Release Candidate 8) ]
* [BUGFIX] Fixed compilation on FreeBSD.
* [IMPROVED] SRANDMEMBER <count> that returns multiple random elements.
* [IMPROVED] Sentinel backported to 2.6. It will be taken in sync with 2.8.
* [IMPROVED] Helper function for scripting to return errors and status replies.
* [IMPROVED] SORT by nosort [ASC|DESC] uses sorted set elements ordering.
* [BUGFIX] Better resistence to system clock skew.
* [IMPROVED] Warn the user when the configured maxmemory seems odd.
* [BUGFIX] Hashing function is now murmurhash2 for security purposes.
* [IMPROVED] Install script no longer uses a template but redis.conf itself.
---[ Redis 2.5.13 (2.6 Release Candidate 7) ]
UPGRADE URGENCY: HIGH
* [BUGFIX] Theoretical bug in ziplist fixed.
* [BUGFIX] Better out of memory handling (Log produced in log file).
* [BUGFIX] Incrementally flush RDB file on slave side while performing the
first synchronization with the master. This makes Redis less
blocking in environments where disk I/O is slow.
* [BUGFIX] Don't crash with Lua's redis.call() without arguments.
* [BUGFIX] Don't crash after a big number of Lua calls on 32 bit systems
because of a failed assertion.
* [BUGFIX] Fix SORT behaviour when called from scripting.
* [BUGFIX] Adjust slave PING period accordingly to REDIS_HZ define.
* [BUGFIX] BITCOUNT: fix crash on overflowing arguments.
* [BUGFIX] Return an error when SELECT argument is not an integer.
* [BUGFIX] Blocking operations on lists were completely reimplemented for
correctness. Now blocking list ops and pushes originated from
Lua scripts will play well together and will be replicated
and transmitted to the AOF correctly.
* [IMPROVED] Send async PING before starting replication to avoid blocking if
master allows us to connect but it is actually not able to reply.
* [IMPROVED] Support slave-priority for Redis Sentinel.
* [IMPROVED] Hiredis library updated.
---[ Redis 2.5.12 (2.6 Release Candidate 6) ]
UPGRADE URGENCY: MODERATE.
* [BUGFIX] Fixed a timing attack on AUTH (Issue #560).
* [BUGFIX] Don't assume that "char" is signed.
* [BUGFIX] Check that we have connection before enabling pipe mode.
* [BUGFIX] Use the optimized version of the function to convert a double to
its string representation. Compilation was disabled because of
a typo in the #if statement.
* [IMPROVED} REPLCONF internal command introduced, now INFO shows slaves with
correct port numbers. This makes 2.5.12 Redis Sentinel compatible.
* [IMPROVED] Truncate short write from the AOF for a cleaner restart. On short
writes (for instance out of space) Redis will now try to remove
the half-written data so that the next restart will work without
the need for the "redis-check-aof" utility.
* [IMPROVED] New in INFO: aof_last_bgrewrite_status
* [IMPROVED] Allow Pub/Sub in contexts where other commands are blocked.
* [BUGFIX] mark fd as writable when EPOLLERR or EPOLLHUP is returned by
epoll_wait.
---[ Redis 2.5.11 (2.6 Release Candidate 5) ]
UPGRADE URGENCY: HIGH.
* [BUGFIX] Fixed Hash corruption when loading an RDB file generated by
previous versions of Redis that encoded hashes using
a different ziplist encoding format for small integers.
All the fileds that are integers in the range 0-255 may not
be recognized, or duplicated un updates, causing a crash
when the ziplist is converted to a real hash. (Issue #547).
* [BUGFIX] Fixed the count of memory used by output buffers in the
setDeferredMultiBulkLength() function.
---[ Redis 2.5.10 (2.6 Release Candidate 4) ]
UPGRADE URGENCY: HIGH.
* [BUGFIX] Allow PREFIX to be overwritten on "make install".
* [BUGFIX] Run the test with just one client if the computer is slow.
* [BUGFIX] Event port support in our event driven libray.
* [BUGFIX] Jemalloc updated to 3.0.0. This fixes a possibly AOF rewrite issue.
See https://github.com/antirez/redis/issues/504 for info.
* [BUGFIX] Fixed issue #516: ZINTERSTORE / ZUNIONSTORE with mixed sets/zsets.
* [BUGFIX] Set fd to writable when poll(2) detects POLLERR or POLLHUP event.
* [BUGFIX] Fixed RESTORE hash failure (Issue #532).
* [IMPROVED] Allow an AOF rewrite buffer > 2GB (Related to issue #504).
* [IMPROVED] Server cron function frequency is now configurable (REDIS_HZ).
* [IMPROVED] Better, less blocking expired keys collection algorithm.
* [FEATURE] New commands: BITOP and BITCOUNT.
* [FEATURE] redis-cli --pipe for mass import.
What's new in Redis 2.5.9 (aka 2.6 Release Candidate 3)
=======================================================
UPGRADE URGENCY: critical, upgrade ASAP.
* [BUGFIX] Fix for issue #500 (https://github.com/antirez/redis/pull/500).
Redis 2.6-RC1 and RC2 may corrupt ziplist-encoded sorted sets
produced by Redis 2.4.x.
* [BUGFIX] Fixed several bugs in init.d script.
* [BUGFIX] syncio.c functions modified for speed and correctness. On osx
(and possibly other BSD-based systems) the slave would block on
replication to send the SYNC command when the master was not
available. This is fixed now, but was not affecting Linux installs.
* Now when slave-serve-stale-data is set to yes and the master is down, instead
of reporting a generic error Redis replies with -MASTERDOWN.
What's new in Redis 2.5.8 (aka 2.6 Release Candidate 2)
=======================================================
UPGRADE URGENCY: high for all the users of the KEYS command, otherwise low.
* [BUGFIX] Fix for KEYS command: if the DB contains keys with expires the KEYS
command may return the wrong output, having duplicated or missing
keys. See issue #487 and #488 on github for details.
What's new in Redis 2.5.7 (aka 2.6 Release Candidate 1)
=======================================================
UPGRADE URGENCY: upgrade not recommended because this is an RC release.
* This is the first release candidate for Redis 2.6. We are not aware of
bugs, but part of this code is young and was never tested in production
environments, so handle with care.
An overview of new features and changes in Redis 2.6.x
======================================================
* Server side Lua scripting, see http://redis.io/commands/eval
* Virtual Memory removed (was deprecated in 2.4)
* Hardcoded limits about max number of clients removed.
* AOF low level semantics is generally more sane, and especially when used
in slaves.
* Milliseconds resolution expires, also added new commands with milliseconds
precision (PEXPIRE, PTTL, ...).
* Better memory usage for "small" lists, ziplists and hashes when fields or
values contain small integers.
* Read only slaves.
* New bit opeations: BITCOUNT and BITOP commands.
* Clients max output buffer soft and hard limits. You can specifiy different
limits for different classes of clients (normal,pubsub,slave).
* More incremental (less blocking) expired keys collection algorithm, in
practical terms this means that Redis is more responsive when a very
big number of keys expire about at the same time.
* AOF is now able to rewrite aggregate data types using variadic commands,
often producing an AOF that is faster to save, load, and is smaller in size.
* Every redis.conf directive is now accepted as a command line option for the
redis-server binary, with the same name and number of arguments.
* Hash table seed randomization for protection against collisions attacks.
* Performances improved when writing large objects to Redis.
* Integrated memory test, see redis-server --test-memory.
* INCRBYFLOAT and HINCRBYFLOAT commands.
* New DUMP, RESTORE, MIGRATE commands (back ported from Redis Cluster to 2.6).
* CRC64 checksump in RDB files.
* Better MONITOR output and behavior (now commands are logged before execution).
* "Software Watchdog" feature to debug latency issues.
* Significant parts of the core refactored or rewritten. New internal APIs
and core changes allowed to develop Redis Cluster on top of the new code,
however for 2.6 all the cluster code was removed, and will be released with
Redis 3.0 when it is more complete and stable.
* Redis ASCII art logo added at startup.
* Crash report on memory violation or failed asserts improved significantly
to make debugging of hard to catch bugs simpler.
* redis-benchmark improvements: ability to run selected tests,
CSV output, faster, better help.
* redis-cli improvements: --eval for comfortable development of Lua scripts.
* SHUTDOWN now supports two optional arguments: "SAVE" and "NOSAVE".
* INFO output split into sections, the command is now able to just show
pecific sections.
* New statistics about how many time a command was called, and how much
execution time it used (INFO commandstats).
* More predictable SORT behavior in edge cases.
* Better support for big endian and *BSD systems.
* Build system improved.
Migrating from 2.4 to 2.6
=========================
Redis 2.4 is mostly a strict subset of 2.6. However there are a few things
that you should be aware of:
* You can't use .rdb and AOF files generated with 2.6 into a 2.4 instance.
* 2.6 slaves can be attached to 2.4 masters, but not the contrary, and only
for the time needed to perform the version upgrade.
There are also a few API differences, that are unlikely to cause problems,
but it is better to keep them in mind:
* SORT now will refuse to sort in numerical mode elements that can't be parsed
as numbers.
* EXPIREs now all have millisecond resolution (but this is very unlikely to
break code that was not conceived exploting the previous resolution error
in some way.)
* INFO output is a bit different now, and contains empty lines and comments
starting with '#'. All the major clients should be already fixed to work
with the new INFO format.
* Slaves are only read-only by default (but you can change this easily
setting the "slave-read-only" configuration option to "no" editing your
redis.conf or using CONFIG SET.
The following INFO fields were renamed for consistency:
changes_since_last_save -> rdb_changes_since_last_save
bgsave_in_progress -> rdb_bgsave_in_progress
last_save_time -> rdb_last_save_time
last_bgsave_status -> rdb_last_bgsave_status
bgrewriteaof_in_progress -> aof_rewrite_in_progress
bgrewriteaof_scheduled -> aof_rewrite_scheduled
The following redis.conf and CONFIG GET / SET parameters changed:
* hash-max-zipmap-entries, now replaced by hash-max-ziplist-entries
* hash-max-zipmap-value, now replaced by hash-max-ziplist-value
* glueoutputbuf option was now completely removed (was deprecated)
--------------------------------------------------------------------------------
Credits: Where not specified the implementation and design are done by
Salvatore Sanfilippo and Pieter Noordhuis. Thanks to VMware for making all
this possible. Also many thanks to all the other contributors and the amazing
community we have.
See commit messages for more credits.
Cheers,
Salvatore
BETATESTING.txt
-9
View File
@@ -1,9 +0,0 @@
This is a stable release, for beta testing make sure to download the latest source code from Git:
git clone git://github.com/antirez/redis.git
It's also possibe to download the latest source code as a tarball:
http://github.com/antirez/redis/tree/master
(use the download button)
BUGS
+1 -1
View File
@@ -1 +1 @@
Plese check http://code.google.com/p/redis/issues/list
Plese check https://github.com/antirez/redis/issues
CONTRIBUTING
+33
View File
@@ -0,0 +1,33 @@
Note: by contributing code to the Redis project in any form, including sending
a pull request via Github, a code fragment or patch via private email or
public discussion groups, you agree to release your code under the terms
of the BSD license that you can find in the COPYING file included in the Redis
source distribution.
# IMPORTANT: HOW TO USE REDIS GITHUB ISSUES
* Github issues SHOULD ONLY BE USED to report bugs, and for DETAILED feature
requests. Everything else belongs to the Redis Google Group.
PLEASE DO NOT POST GENERAL QUESTIONS that are not about bugs or suspected
bugs in the Github issues system. We'll be very happy to help you and provide
all the support in the Redis Google Group.
Redis Google Group address:
https://groups.google.com/forum/?fromgroups#!forum/redis-db
# How to provide a patch for a new feature
1. Drop a message to the Redis Google Group with a proposal of semantics/API.
2. If in steps 1 you get an acknowledge from the project leaders, use the
following procedure to submit a patch:
a. Fork Redis on github ( http://help.github.com/fork-a-repo/ )
b. Create a topic branch (git checkout -b my_branch)
c. Push to your branch (git push origin my_branch)
d. Initiate a pull request on github ( http://help.github.com/send-pull-requests/ )
e. Done :)
Thanks!
COPYING
+1 -1
View File
@@ -1,4 +1,4 @@
Copyright (c) 2006-2009, Salvatore Sanfilippo
Copyright (c) 2006-2012, Salvatore Sanfilippo
All rights reserved.
Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
Changelog
-865
View File
@@ -1,865 +0,0 @@
2010-05-21 redis version is now 1.3.14 (aka 2.0.0 RC1) (antirez)
2010-05-21 html doc updated (antirez)
2010-05-21 by default test with valgrind does not show full leak info (antirez)
2010-05-21 minor fix for the skiplist code, resulting in a false positive with valgrind, and in general into a useless small allocation (antirez)
2010-05-21 Merge branch 'master' of git@github.com:antirez/redis (antirez)
2010-05-21 tests suite initial support for valgrind, fixed the old test suite until the new one is able to target a specific host/port (antirez)
2010-05-20 Don't exit with error in tests temp file cleanup if there are no files to clean (antirez)
2010-05-20 fix memory leak on 32-bit builds (Pieter Noordhuis)
2010-05-20 Merge branch 'master' of github.com:antirez/redis (antirez)
2010-05-20 Fix for DEBUG DIGEST (antirez)
2010-05-20 Merge branch 'test_vm' of git://github.com/pietern/redis (antirez)
2010-05-20 code to enable running tests with the vm enabled (Pieter Noordhuis)
2010-05-20 minor change to shutdown (antirez)
2010-05-20 shutdown on SIGTERM (antirez)
2010-05-20 Merge http://github.com/ngmoco/redis (antirez)
2010-05-20 fix compile error on solaris (Pieter Noordhuis)
2010-05-20 added regression for zipmap bug (antirez)
2010-05-20 fix lookup of keys with length larger than ZIPMAP_BIGLEN (Pieter Noordhuis)
2010-05-19 TODO updated (antirez)
2010-05-19 initial tests for AOF (and small changes to server.tcl to support these) (Pieter Noordhuis)
2010-05-19 Merge branch 'master' into integration (Pieter Noordhuis)
2010-05-19 Fix for 'CONFIG SET appendonly no' (antirez)
2010-05-19 It's now possible to turn off and on the AOF via CONFIG (antirez)
2010-05-18 git hash 00000000 in reelase.h when git is not found enabled again after some shell scripting fix that is now compatible with most shells (antirez)
2010-05-18 build fixed when simpler shells are used to create release.h (antirez)
2010-05-18 use git diff when generating release.h to check for dirty status (antirez)
2010-05-18 Solaris fixes (antirez)
2010-05-18 html doc rebuild (antirez)
2010-05-18 buliding of release.h moved into an external script. Avoided recompialtion of redis.c if git sha1 is the same as the previous one (antirez)
2010-05-17 create release.h in make process and add this information to INFO listing (Pieter Noordhuis)
2010-05-16 Redis version is now 1.3.12 (antirez)
2010-05-16 redis version is now 1.3.11 (antirez)
2010-05-16 random refactoring and speedups (antirez)
2010-05-16 faster INCR with very little efforts... (antirez)
2010-05-15 print warnings in redis log when a test raises an exception (very likely to be caused by something like a failed assertion) (Pieter Noordhuis)
2010-05-15 Merge branch 'redis-cli-fix' of http://github.com/tizoc/redis (antirez)
2010-05-15 added pid info to the check memory leaks test, so that those tests don't appear to be duplicated (antirez)
2010-05-15 Merge branch 'integration' of git://github.com/pietern/redis (antirez)
2010-05-14 more endianess detection fix for SHA1 (antirez)
2010-05-14 fixed a warning seen with some GCC version under Linux (antirez)
2010-05-14 initial rough integration test for replication (Pieter Noordhuis)
2010-05-14 store entire server object on the stack instead of just the client (Pieter Noordhuis)
2010-05-14 proc to retrieve values from INFO properties (Pieter Noordhuis)
2010-05-14 one more fix for endianess detection (antirez)
2010-05-14 Fixed sha1.c compilation on Linux, due to endianess detection lameness (antirez)
2010-05-14 ZUNION,ZINTER -> ZUNIONSTORE,ZINTERSTORE (antirez)
2010-05-14 minor fixes to the new test suite, html doc updated (antirez)
2010-05-14 wait for redis-server to be settled and ready for connections (Pieter Noordhuis)
2010-05-14 fix cleaning up tmp folder (Pieter Noordhuis)
2010-05-14 update makefile to use the new test suite (Pieter Noordhuis)
2010-05-14 check for memory leaks before killing a server (Pieter Noordhuis)
2010-05-14 extract code to kill a server to a separate proc (Pieter Noordhuis)
2010-05-14 start servers on different ports to prevent conflicts (Pieter Noordhuis)
2010-05-14 use DEBUG DIGEST in new test suite (Pieter Noordhuis)
2010-05-14 split test suite into multiple files; runs redis-server in isolation (Pieter Noordhuis)
2010-05-14 use DEBUG DIGEST in the test instead of a function that was doing a similar work, but in a much slower and buggy way (antirez)
2010-05-14 Don't rely on cliReadReply being able to return on shutdown (Bruno Deferrari)
2010-05-14 If command is a shutdown, ignore errors on reply (Bruno Deferrari)
2010-05-14 DEBUG DIGEST implemented, in order to improve the ability to test persistence and replication consistency (antirez)
2010-05-13 Add SIGTERM shutdown handling. (Ashley Martens)
2010-05-13 makefile deps updated (antirez)
2010-05-13 conflicts resolved (antirez)
2010-05-13 feed SETEX as SET and EXPIREAT to AOF (Pieter Noordhuis)
2010-05-13 very strong speedup in saving time performance when there are many integers in the dataset. Instead of decoding the object before to pass them to the rdbSaveObject layer we check asap if the object is integer encoded and can be written on disk as an integer. (antirez)
2010-05-13 include limits.h otherwise no double precison macros (antirez)
2010-05-13 explicitly checks with ifdefs if our floating point and long long assumptions are verified (antirez)
2010-05-13 Yet another version of the double saving code, with comments explaining what's happening there (antirez)
2010-05-12 added overflow check in the double -> long long conversion trick to avoid integer overflows. I think this was not needed in practical terms, but it is safer (antirez)
2010-05-12 use withscores when performing the dataset digest (antirez)
2010-05-12 If a float can be casted to a long long without rounding loss, we can use the integer conversion function to write the score on disk. This is a seriuous speedup (antirez)
2010-05-12 fixed compilation warnings in the AOF sanity check tool (antirez)
2010-05-12 Merge branch 'vm-speedup' (antirez)
2010-05-11 fix to return error when calling INCR on a non-string type (Pieter Noordhuis)
2010-05-11 load objects encoded from disk directly without useless conversion (antirez)
2010-05-11 fixed a problem leading to crashes, as keys can't be currently specially encoded, so we can't encode integers at object loading time... For now this can be fixed passing a few flags, or later can be fixed allowing encoded keys as well (antirez)
2010-05-11 long long to string conversion speedup applied in other places as well. Still the code has bugs, fixing right now... (antirez)
2010-05-11 hand written code to turn a long long into a string -> very big speed win (antirez)
2010-05-11 added specialized function to compare string objects for perfect match that is optimized for this task (antirez)
2010-05-11 better use of encoding inforamtion in dictEncObjKeyCompare (antirez)
2010-05-10 CONFIG now can change appendfsync policy at run time (antirez)
2010-05-10 CONFIG command now supports hot modification of RDB saving parameters. (antirez)
2010-05-10 while loading the rdb file don't add the key to the dictionary at all if it's already expired, instead of removing it just after the insertion. (antirez)
2010-05-10 Merge branch 'check-aof' of git://github.com/pietern/redis (antirez)
2010-05-08 minor changes to improve code readability (antirez)
2010-05-08 swap objects out directly while loading an RDB file if we detect we can't stay in the vm max memory limits anyway (antirez)
2010-05-07 change command names no longer used to zunion/zinter (Pieter Noordhuis)
2010-05-07 DEBUG POPULATE command for fast creation of test databases (antirez)
2010-05-07 update TODO (Pieter Noordhuis)
2010-05-07 swap arguments in blockClientOnSwappedKeys to be consistent (Pieter Noordhuis)
2010-05-07 added function that preloads all keys needed to execute a MULTI/EXEC block (Pieter Noordhuis)
2010-05-07 add sanity check to zunionInterBlockClientOnSwappedKeys, as the number of keys used is provided as argument to the function (Pieter Noordhuis)
2010-05-07 make prototype of custom function to preload keys from the vm match the prototype of waitForMultipleSwappedKeys (Pieter Noordhuis)
2010-05-07 extract preloading of multiple keys according to the command prototype to a separate function (Pieter Noordhuis)
2010-05-07 make append only filename configurable (Pieter Noordhuis)
2010-05-07 don't load value from VM for EXISTS (Pieter Noordhuis)
2010-05-07 swap file name pid expansion removed. Not suited for mission critical software... (antirez)
2010-05-07 Swap file is now locked (antirez)
2010-05-06 Merge branch 'master' into aof-speedup (antirez)
2010-05-06 log error and quit when the AOF contains an unfinished MULTI (antirez)
2010-05-06 log error and quit when the AOF contains an unfinished MULTI (Pieter Noordhuis)
2010-05-06 Merge branch 'master' into check-aof (Pieter Noordhuis)
2010-05-06 hincrby should report an error when called against a hash key that doesn't contain an integer (Pieter Noordhuis)
2010-05-06 AOF writes are now accumulated into a buffer and flushed into disk just before re-entering the event loop. A lot less writes but still this guarantees that AOF is written before the client gets a positive reply about a write operation, as no reply is trasnmitted before re-entering into the event loop. (antirez)
2010-05-06 clarified a few messages in redis.conf (antirez)
2010-05-05 ask for confirmation before AOF is truncated (Pieter Noordhuis)
2010-05-05 str can be free'd outside readString (Pieter Noordhuis)
2010-05-05 moved argument parsing around (Pieter Noordhuis)
2010-05-05 ignore redis-check-aof binary (Pieter Noordhuis)
2010-05-05 allow AOF to be fixed by truncating to the portion of the file that is valid (Pieter Noordhuis)
2010-05-05 tool to check if AOF is valid (Pieter Noordhuis)
2010-05-02 included fmacros.h in linenose.c to avoid compilation warnings on Linux (antirez)
2010-05-02 compilation fix for mac os x (antirez)
2010-05-02 Merge branch 'master' of git@github.com:antirez/redis (antirez)
2010-05-02 On Linux now fdatasync() is used insetad of fsync() in order to flush the AOF file kernel buffers (antirez)
2010-04-30 More tests for APPEND and tests for SUBSTR (antirez)
2010-04-30 linenoise.c updated, now redis-cli can be used in a pipe (antirez)
2010-04-29 redis-cli minor fix (less segfault is better) (antirez)
2010-04-29 New MONITOR output format with timestamp, every command in a single line, string representations (antirez)
2010-04-29 redis-cli INFO output format is now raw again (antirez)
2010-04-29 Added more information about slave election in Redis Cluster alternative doc (antirez)
2010-04-29 Redis cluster version 2 (antirez)
2010-04-27 Fixed a redis-cli bug, was using free instead of zfree call (antirez)
2010-04-27 AOF is now rewritten on slave after SYNC with master. Thanks to @_km for finding this bug and any others' (antirez)
2010-04-27 redis-cli is now using only the new protocol (antirez)
2010-04-27 Minimal support for subscribe/psubscribe in redis-cli (antirez)
2010-04-26 don't output the newline when stdout is not a tty (antirez)
2010-04-26 redis-cli now is able to also output the string representation instead of the raw string. Much better for debugging (antirez)
2010-04-26 Initial support for quoted strings in redis-cli (antirez)
2010-04-23 SETEX implemented (antirez)
2010-04-23 Pub/Sub API change: now messages received via pattern matching have a different message type and an additional field representing the original pattern the message matched (antirez)
2010-04-22 typo fixed, reloaded (antirez)
2010-04-22 typo fixed (antirez)
2010-04-22 REDIS-CLUSTER doc updated (antirez)
2010-04-22 Virtual memory design document removed, no longer needed as we have a full specification and implementation (antirez)
2010-04-22 new units for bytes specification (antirez)
2010-04-22 Now in redis.conf it is possible to specify units where appropriate instead of amounts of bytes, like 2Gi or 4M and so forth (antirez)
2010-04-21 binary safe keys ready implementation of RANDOMKEYS (antirez)
2010-04-21 Now that's the right 1.3.10 (antirez)
2010-04-21 Revert "fsync always now uses O_DIRECT on Linux" (antirez)
2010-04-21 Revert "define __USE_GNU to get O_DIRECT" (antirez)
2010-04-21 Merge branch 'master' of github.com:antirez/redis (antirez)
2010-04-21 Revert "version 1.3.10" (antirez)
2010-04-21 version 1.3.10 (antirez)
2010-04-20 define __USE_GNU to get O_DIRECT (antirez)
2010-04-20 fsync always now uses O_DIRECT on Linux (antirez)
2010-04-20 More precise memory used guesswork in zmalloc.c (antirez)
2010-04-19 Fix for MULTI/EXEC and Replication/AOF: now the block is correctly sent as MULTI/..writing operations../EXEC. Ok for slaves but more work needed for the AOF as it should be a write-all-or-nothing business (antirez)
2010-04-19 running the test using tcl8.5 directly instead of tclsh that too often it's a symlink to 8.4 (antirez)
2010-04-19 Added package require Tcl 8.5 in redis.tcl so it will show a clear error when the test suit is attempted to run under 8.4 (antirez)
2010-04-18 Fix for a SORT bug introduced with commit 16fa22f1, regression test added (antirez)
2010-04-18 Guru mediation -> meditation (antirez)
2010-04-16 check eptr inline (Pieter Noordhuis)
2010-04-16 refactor code that retrieves value from object or replies to client (Pieter Noordhuis)
2010-04-17 Merge branch 'hash' of git://github.com/pietern/redis (antirez)
2010-04-17 redisAssert(0) => redisPanic("something meaningful") (antirez)
2010-04-17 make sure that the resulting value in hincrby is encoded when possible (Pieter Noordhuis)
2010-04-17 increment dirty counter after hmset (Pieter Noordhuis)
2010-04-17 strip tryObjectEncoding from hashSet, to enable the arguments being encoded in-place (Pieter Noordhuis)
2010-04-17 Added support for Guru Mediation, and raising a guru mediation if refCount <= 0 but decrRefCount is called against such an object (antirez)
2010-04-16 fix small error and memory leaks in SORT (Pieter Noordhuis)
2010-04-16 SORT/GET test added (antirez)
2010-04-16 Added tests for GET/BY against hashes fields (antirez)
2010-04-16 Merge branch 'hash-refactor' of git://github.com/pietern/redis (antirez)
2010-04-16 check object type in lookupKeyByPattern (Pieter Noordhuis)
2010-04-16 make sortCommand aware that lookupKeyByPattern always increased the refcount of the returned value (Pieter Noordhuis)
2010-04-16 revert 0c390a to stop using tricks with o->refcount (Pieter Noordhuis)
2010-04-16 store the hash iterator on the heap instead of the stack (Pieter Noordhuis)
2010-04-16 drop inline directive (Pieter Noordhuis)
2010-04-16 rename hashReplace to hashSet (Pieter Noordhuis)
2010-04-16 added dictFetchValue() to dict.c to make hash table API a bit less verbose in the common cases (antirez)
2010-04-03 Don't set expire to keys with ttl=0, remove them immediately. (antirez)
2010-04-15 make sure that cmpobj is in decoded form when sorting by ALPHA (this solves edge case from previous commit where (!sortby && alpha) == 1) (Pieter Noordhuis)
2010-04-15 enable hash dereference in SORT on BY and GET (Pieter Noordhuis)
2010-04-15 use shared replies for hset (Pieter Noordhuis)
2010-04-15 set refcount of string objects retrieved from zipmaps to 0, so we don't have to touch the refcount of the objects inside dicts (Pieter Noordhuis)
2010-04-15 added HSETNX (Pieter Noordhuis)
2010-04-14 refactor of hash commands to use specialized api that abstracts zipmap and dict apis (Pieter Noordhuis)
2010-04-13 move retrieval of long up to prevent an empty hash from being created (Pieter Noordhuis)
2010-04-15 more advanced leaks detection in test redis (antirez)
2010-04-15 ability to select port/host from make test (antirez)
2010-04-15 Active rehashing (antirez)
2010-04-15 Incrementally rehahsing hash table! Thanks to Derek Collison and Pieter Noordhuis for feedbacks/help (antirez)
2010-04-14 Does not allow commands other than Pub/Sub commands when there is at least one pattern (antirez)
2010-04-13 Fixed a tiny memory leak when loading the configuration file. (Alex McHale)
2010-04-13 Merge branch 'hmget' of git://github.com/pietern/redis (antirez)
2010-03-29 Validate numeric inputs. (Alex McHale)
2010-03-24 Remove trailing whitespace. (Alex McHale)
2010-04-12 Now all the commands returning a multi bulk reply against non existing keys will return an empty multi bulk, not a nil one (antirez)
2010-04-12 implemented HMGET (Pieter Noordhuis)
2010-04-12 implemented HMSET (Pieter Noordhuis)
2010-04-12 Sharing of small integer objects: may save a lot of memory with datasets having many of this (antirez)
2010-04-10 dict.c fixed to play well with enabling/disabling of the hash table (antirez)
2010-04-09 removed a no longer true assert in the VM code (antirez)
2010-04-09 shareobjects feautres killed - no gains most of the time, but VM complexities (antirez)
2010-04-09 use directly the real key object in VM I/O jobs to match by pointer, and to handle different keys with the same name living in different DBs, but being at the same moment in the IO job queues (antirez)
2010-04-08 last change reverted as it was unstable... more testing needed (antirez)
2010-04-08 Prevent hash table resize while there are active child processes in order to play well with copy on write (antirez)
2010-04-08 Merge branch 'issue_218' of git://github.com/pietern/redis (antirez)
2010-04-08 -1 not needed... (antirez)
2010-04-08 Skiplist theoretical fix (antirez)
2010-04-07 Now when a child is terminated by a signal, the signal number is logged as well (antirez)
2010-04-07 First version of evented Redis Tcl client, that will be used for BLPOP and Pub/Sub tests (antirez)
2010-04-05 use long long reply type for HINCRBY (Pieter Noordhuis)
2010-04-05 last argument is never encoded for HINCRBY (Pieter Noordhuis)
2010-04-02 Now PUBLISH commands are replicated to slaves (antirez)
2010-04-01 use the right object when cleaning up after zunion/zinter (fixes issue 216) (Pieter Noordhuis)
2010-04-01 Merge branch 'zipmap' of git://github.com/pietern/redis (antirez)
2010-04-01 reduce code complexity because zipmapLen now is O(1) (Pieter Noordhuis)
2010-04-01 update the zipmap entry in-place instead of appending it (Pieter Noordhuis)
2010-04-01 updated zipmap documentation to match the implementation (Pieter Noordhuis)
2010-04-01 allow 4 free trailing bytes for each value (Pieter Noordhuis)
2010-04-01 Pub/Sub pattern matching capabilities (antirez)
2010-04-01 use function to determine length of a single entry (Pieter Noordhuis)
2010-03-31 Deny EXEC under out of memory (antirez)
2010-03-29 No timeouts nor other commands for clients in a Pub/Sub context (antirez)
2010-03-29 free hash table entries about no longer active classes, so that PUBSUB can be abused with millions of different classes (antirez)
2010-03-29 Fixed a refcount stuff leading to PUBSUB crashes (antirez)
2010-03-29 fmacros added to linenoise, avoiding all the nice warnings... (antirez)
2010-03-29 First pubsub fix (antirez)
2010-03-29 PUBSUB implemented (antirez)
2010-03-29 Redis version is now 1.3.8 (antirez)
2010-03-28 removed references in code to ZIPMAP_EMPTY (Pieter Noordhuis)
2010-03-28 use first byte of zipmap to store length (Pieter Noordhuis)
2010-03-28 implemented strategy that doesn't use free blocks in zipmaps (Pieter Noordhuis)
2010-03-26 Merge branch 'hincrby' of git://github.com/pietern/redis (antirez)
2010-03-26 removed unnecessary refcount increase that caused the HINCRBY memleak (Pieter Noordhuis)
2010-03-26 implements HINCRBY and tests (todo: find and fix small memleak) (Pieter Noordhuis)
2010-03-26 Removed a useless if spotted by Pieter Noordhuis (antirez)
2010-03-26 Fixed a critical replication bug: binary values issued with the multi bulk protocol caused a protocol desync with slaves. (antirez)
2010-03-24 Fixed the reply about denied write commands under maxmemory reached condition: now the error will no longer lead to a client-server protocol desync (antirez)
2010-03-24 CONFIG command implemened -- just a start but already useful (antirez)
2010-03-24 redis-cli prompt is now redis> (antirez)
2010-03-23 with --help states that you can use - as config file name to feed config via stdin (antirez)
2010-03-23 New INFO field: expired_keys (antirez)
2010-03-23 the Cron timer function is now called 10 times per second instead of 1 time per second to make Redis more responsibe to BGSAVE and to delete expired keys more incrementally (antirez)
2010-03-23 Use linenoise for line editing on redis-cli. (Michel Martens)
2010-03-23 Fix authentication for redis-cli on non-interactive mode. (Michel Martens)
2010-03-23 key deletion on empty value fix + some refactoring (antirez)
2010-03-23 Empty value trigger key removal in all the operations (antirez)
2010-03-22 Merged gnrfan patches fixing issues 191, 193, 194 (antirez)
2010-03-22 Merge branch 'issue_193' of git://github.com/gnrfan/redis (antirez)
2010-03-22 Merge branch 'issue_191' of git://github.com/gnrfan/redis (antirez)
2010-03-22 Redis master version is now 1.3.7 (antirez)
2010-03-19 support for include directive in config parser (Jeremy Zawodny)
2010-03-19 Removed a stupid overriding of config values due to a wrong cut&paste (antirez)
2010-03-19 VM hash type swappability implemented. Handling of failed pthread_create() call. (antirez)
2010-03-19 Solving issue #191 on Google Code: -v and --version should print the version of Redis (Antonio Ognio)
2010-03-19 Solves issue #194 on Google Code: --help parameter to redis-srver prints the usage message (Antonio Ognio)
2010-03-19 Fixing issue 193 (Antonio Ognio)
2010-03-18 increment server.dirty on HDEL (antirez)
2010-03-18 Redis 1.3.6 (antirez)
2010-03-18 test-redis.tcl dataset digest function Hash support (antirez)
2010-03-18 zipmap fix for large values (antirez)
2010-03-18 Optimization fixed and re-activated (antirez)
2010-03-18 reverted an optimization that makes Redis not stable (antirez)
2010-03-18 Fixed redis-cli auth code (antirez)
2010-03-17 HDEL fix, an optimization for comparison of objects in hash table lookups when they are integer encoding (antirez)
2010-03-17 Version is now 1.3.5 (antirez)
2010-03-17 Merged Pietern patch for VM key args helper function. Fixed an obvious bug in the redis-cli passwd auth stuff (antirez)
2010-03-17 Merge branch 'aggregates' of git://github.com/pietern/redis (antirez)
2010-03-17 Added Authentication to redis-cli.c using -a switch Update usage fixed Makefile to delete redis-check-dump during make clean (root)
2010-03-17 HEXISTS and tests implemented (antirez)
2010-03-17 More hash tests (antirez)
2010-03-17 better HSET test (antirez)
2010-03-17 Fixed a bug in HSET, a memory leak, and a theoretical bug in dict.c (antirez)
2010-03-17 More Hash tests (antirez)
2010-03-13 added preloading keys from VM when using ZINTER or ZUNION (Pieter Noordhuis)
2010-03-13 added explicit AGGREGATE [SUM|MIN|MAX] option to ZUNION/ZINTER (Pieter Noordhuis)
2010-03-16 HGET fix for integer encoded field against zipmap encoded hash (antirez)
2010-03-16 zrevrank support in redis-cli (antirez)
2010-03-16 HKEYS / HVALS / HGETALL (antirez)
2010-03-16 Solved a memory leak with Hashes (antirez)
2010-03-15 pretty big refactoring (antirez)
2010-03-15 An interesting refactoring + more expressive internal API (antirez)
2010-03-15 Fixed the same problem in ZREVRANK (antirez)
2010-03-15 Fixed a ZRANK bug (antirez)
2010-03-15 zipmap to hash conversion in HSET (antirez)
2010-03-14 max zipmap entries and max zipmap value parameters added into INFO output (antirez)
2010-03-14 HDEL and some improvement in DEBUG OBJECT command (antirez)
2010-03-14 Append only file support for hashes (antirez)
2010-03-13 utility to check rdb files for unprocessable opcodes (Pieter Noordhuis)
2010-03-12 A minor fix and a few debug messages removed (antirez)
2010-03-12 Applied the replication bug patch provided by Jeremy Zawodny, removing temp file collision after the slave got the dump.rdb file in the SYNC stage (antirez)
2010-03-11 Fix for HGET against non Hash type, debug messages used to understand a bit better a corrupted rdb file (antirez)
2010-03-09 fix: use zmalloc instead of malloc (Pieter Noordhuis)
2010-03-09 Merged zsetops branch from Pietern (antirez)
2010-03-09 Merged ZREMBYRANK from Pietern (antirez)
2010-03-09 Merged ZREVRANK from Pietern (antirez)
2010-03-09 use a struct to store both a dict and its weight for ZUNION and ZINTER, so qsort can be applied (Pieter Noordhuis)
2010-03-09 Hash auto conversion from zipmap to hash table, type fixed for hashes, hash loading from disk (antirez)
2010-03-09 replaced ZMERGE by ZUNION and ZINTER. note: key preloading by the VM does not yet work (Pieter Noordhuis)
2010-03-08 Hashes saving / fixes (antirez)
2010-03-08 use ZMERGE as starting point (Pieter Noordhuis)
2010-03-07 HSET fixes, now the new pointer is stored back in the object pointer field (antirez)
2010-03-07 added ZREVRANK (Pieter Noordhuis)
2010-03-06 Fix for replicaiton with over 2GB dump file initial SYNC stage (antirez)
2010-03-06 first implementation of HSET/HSET. More work needed (antirez)
2010-03-05 zipmaps functions to get, iterate, test for existence. Initial works for Hash data type (antirez)
2010-03-04 redis-benchmark now implements Set commands benchmarks (antirez)
2010-03-04 zipmap iteration code (antirez)
2010-03-04 moved code to delete a single node from a zset to a separate function (Pieter Noordhuis)
2010-03-04 rename zslDeleteRange to zslDeleteRangeByScore (to differentiate between deleting using score or rank) (Pieter Noordhuis)
2010-03-04 use 1-based rank across zsl*Rank functions consistently (Pieter Noordhuis)
2010-03-04 implemented ZREMBYRANK (Pieter Noordhuis)
2010-03-04 A fix for initialization of augmented skip lists (antirez)
2010-03-04 A fix for an invalid access when VM is disabled (antirez)
2010-03-04 Merge branch 'zsl-get-rank' of git://github.com/pietern/redis (antirez)
2010-03-04 redis-cli now runs in interactive mode if no command is provided (antirez)
2010-03-04 merged memory reduction patch (Pieter Noordhuis)
2010-03-04 Now list push commands return the length of the new list, thanks to Gustavo Picon (antirez)
2010-03-04 first check if starting point is trivial (head or tail) before applying log(N) search (Pieter Noordhuis)
2010-03-04 use rank to find starting point for ZRANGE and ZREVRANGE (Pieter Noordhuis)
2010-03-04 lookup rank of a zset entry in a different function (Pieter Noordhuis)
2010-03-04 SUBSTR fix for integer encoded vals (antirez)
2010-03-04 fix ZRANK (realize that rank is 1-based due to the skip list header) (Pieter Noordhuis)
2010-03-03 initial implementation of SUBSTR (antirez)
2010-03-03 TODO updated (antirez)
2010-03-03 fpurge call removed from redis-cli (antirez)
2010-03-03 ZRANK stress tester (antirez)
2010-03-03 use less memory as element->span[0] will always be 1; any level 0 skip list is essentially a linked list (Pieter Noordhuis)
2010-03-03 rank is very unlikely to overflow integer range (Pieter Noordhuis)
2010-03-03 x->backward never equals zsl->header (Pieter Noordhuis)
2010-03-03 initial implementation for augmented zsets and the zrank command (Pieter Noordhuis)
2010-03-03 zipampDel() implemented (antirez)
2010-03-03 added quit and exit commands to redis-cli in order to quit the interactive mode (antirez)
2010-03-03 Merge remote branch 'djanowski/interactive' (antirez)
2010-03-02 Add support for MULTI/EXEC. (Damian Janowski & Michel Martens)
2010-03-02 Remove trailing newline in interactive mode. (Damian Janowski & Michel Martens)
2010-03-02 minor fix for a Linux warning (antirez)
2010-03-02 Add interactive mode to redis-cli. (Michel Martens & Damian Janowski)
2010-03-02 Better to increment the version minor number when a VM bug is fixed... it will be simpler to understand what's going on when users will report problems with the INFO trace. (antirez)
2010-03-02 Fixed a subtle VM bug... was not flushing the buffer so the child process read truncated data (antirez)
2010-03-01 KEYS now returns a multi bulk reply (antirez)
2010-02-27 Add DISCARD command to discard queued MULTI commands. (antirez)
2010-03-01 Swappability bug due to a typo fixed thanks to code review by Felix Geisendörfer @felixge (antirez)
2010-02-28 minor fixes for zipmap.c (antirez)
2010-02-27 first zipmap fix of a long sequence in the days to come ;) (antirez)
2010-02-27 initial zipmap.c implementation (antirez)
2010-02-27 Bug #169 fixed (BLOP/BRPOP interrupted connections are not cleared from the queue) (antirez)
2010-02-22 Fixed 32bit make target to work on Linux out of the box (antirez)
2010-02-19 A problem with replication with multiple slaves connectiong to a single master fixed. It was due to a typo, and reported on github by the user micmac. Also the copyright year fixed from many files. (antirez)
2010-02-10 Saner VM defaults for redis.conf (antirez)
2010-02-09 VM now is able to block clients on swapped keys for all the commands (antirez)
2010-02-07 ZCOUNT and ZRANGEBYSCORE new tests (antirez)
2010-02-07 ZRANGEBYSCORE now supports open intervals, prefixing double values with a open paren. Added ZCOUNT that can count the elements inside an interval of scores, this supports open intervals too (antirez)
2010-02-07 WITHSCORES in ZRANGEBYSCORE thanks to Sam Hendley (antirez)
2010-02-06 Added "withscores" option to zrangebyscore command. Based on withscores support in zrange function, ugliest part was the argument parsing to handle using it with the limit option. (Sam Hendley)
2010-02-06 DEBUG OBJECT provide info about serialized object length even when VM is disabled (antirez)
2010-02-06 multi bulk requests in redis-benchmark, default fsync policy changed to everysec, added a prefix character for DEBUG logs (antirez)
2010-02-04 APPEND tests (antirez)
2010-02-04 APPEND command (antirez)
2010-02-02 Faster version of the function hashing possibly encoded objects, leading to a general speed gain when working with Sets of integers (antirez)
2010-02-02 faster Set loading time from .rdb file resizing the hash table to the right size before loading elements (antirez)
2010-02-02 Log time taken to load the DB at startup, in seconds (antirez)
2010-01-31 Fixed VM corruption due to child fclosing the VM file directly or indirectly calling exit(), now replaced with _exit() in all the sensible places. Masked a few signals from IO threads. (antirez)
2010-01-28 loading side of the threaded VM (antirez)
2010-01-26 TODO cahnges (antirez)
2010-01-23 Fixed memory human style memory reporting, removed server.usedmemory, now zmalloc_used_memory() is used always. (antirez)
2010-01-22 VM tuning thanks to redis-stat vmstat. Now it performs much better under high load (antirez)
2010-01-21 Changelog updated (antirez)
2010-01-21 REDIS_MAX_COMPLETED_JOBS_PROCESSED is now in percentage, not number of jobs. Moved a debugging message a few lines forward as it was called where a few logged parameters where invalid, leading to a crash (antirez)
2010-01-20 fixed a deadlock caused by too much finished processes in queue so that I/O clients writing to the wirte side of the pipe used to awake the main thread where blocking. Then a BGSAVE started waiting for the last active thread to finish, condition impossible because all the I/O threads where blocking on threads. Takes this as a note to myself... (antirez)
2010-01-20 ae.c event loop does no longer support exception notifications, as they are fully pointless. Also a theoretical bug that never happens in practice fixed. (antirez)
2010-01-19 commercial tools stuff removed from the Redis makefile. cotools are now migrated into a different repos (antirez)
2010-01-19 removed a bug in the function to cancel an I/O job (antirez)
2010-01-17 static symbols update (antirez)
2010-01-16 removed support for REDIS_HELGRIND_FRIENDLY since Helgrind 3.5.0 is friendly enough even with many threads created and destroyed (antirez)
2010-01-15 now redis-cli understands -h (antirez)
2010-01-15 Create swap file only if not exists (antirez)
2010-01-15 I hate warnings (antirez)
2010-01-15 fixed a minor memory leak in configuration file parsing (antirez)
2010-01-15 minor fix (antirez)
2010-01-15 support for named VM swap file. Fixed a few important interaction issues between the background saving processes and IO threads (antirez)
2010-01-15 fix for the just added new test (antirez)
2010-01-15 useless debugging messages removed (antirez)
2010-01-15 new test added (antirez)
2010-01-15 thread safe zmalloc used memory counter (antirez)
2010-01-15 A define to make Redis more helgrind friendly (antirez)
2010-01-15 removed a few races from threaded VM (antirez)
2010-01-14 Fixed a never experienced, theoretical bug that can actually happen in practice. Basically when a thread is working on a I/O Job we need to wait it to finish before to cancel the Job in vmCancelThreadedIOJob(), otherwise the thread may mess with an object that is being manipulated by the main thread as well. (antirez)
2010-01-14 Set the new threads stack size to a LZF friendly amount (antirez)
2010-01-13 access to already freed job structure fixed by statements reoredering (antirez)
2010-01-13 removed a useless debugging message (antirez)
2010-01-13 Wait zero active threads condition before to fork() for BGSAVE or BGREWRITEAOF (antirez)
2010-01-13 list API is now thread safe (antirez)
2010-01-13 minor TODO and debugging info changes (antirez)
2010-01-12 support for blocking VM in config file (antirez)
2010-01-12 more non blocking VM changes (antirez)
2010-01-12 fix for test #11 (antirez)
2010-01-12 a few more stuff in INFO about VM. Test #11 changed a bit in order to be less lame (antirez)
2010-01-12 Added a define to configure how many completed IO jobs the handler should process at every call. (antirez)
2010-01-11 Fixed a bug in the IO Job canceling funtion (antirez)
2010-01-11 more steps towards a working non blocking VM (antirez)
2010-01-11 converted random printfs in debug logs (antirez)
2010-01-11 removed a bug introduced with non blocking VM (antirez)
2010-01-11 a few non blocking VM bugs fixed (antirez)
2010-01-11 More work on non-blocking VM. Should work in a few days (antirez)
2010-01-11 More threaded I/O VM work + Redis init script (antirez)
2010-01-10 more work on VM threaded I/O. Still nothing of usable (antirez)
2010-01-09 non-blocking VM data structures, just a start (antirez)
2010-01-08 used_memory_human added to INFO output. Human readable amount of memory used. (antirez)
2010-01-07 Now DEBUG OBJECT plays well with swapped out objects (antirez)
2010-01-07 fflush VM swap file after object swapping (antirez)
2010-01-07 added the fmacros to enable support for fseeko() lseeko() with 64bit off_t (antirez)
2010-01-07 VM now swaps objects out while loading datasets not fitting into vm-max-memory bytes of RAM (antirez)
2010-01-07 added process id information in INFO (antirez)
2010-01-06 vm-enabled set to no by default in redis.conf (antirez)
2010-01-06 a new default redis.conf (antirez)
2010-01-06 VM stats in INFO command (antirez)
2010-01-06 Introduced a new log verbosity level, so now DEBUG is really for debugging. Refactored a bit maxmemory. When virtual memory is short in RAM free the objects freelist as well as swapping things out. (antirez)
2010-01-05 fixed a bug in bgsave when VM is off but still it was testing for obj->storage field (antirez)
2010-01-05 converted a few calls to assert() => redisAssert() to print stack trace (antirez)
2010-01-05 BGREWRITEAOF now works with swapping on (antirez)
2010-01-05 A first fix for SET key overwrite (antirez)
2010-01-05 SAVE now works with VM (antirez)
2010-01-05 swapping algorithm a bit more aggressive under low memory (antirez)
2010-01-05 basic VM mostly working! (antirez)
2010-01-05 New object field (one of the unused bytes) to hold the type of the swapped out value object in key objects (antirez)
2010-01-05 VM internals bugfixes, set 1 (antirez)
2010-01-05 load key from swap on key lookup (antirez)
2010-01-05 more object-level VM primitives (antirez)
2010-01-05 Redis objects swapping / loading (antirez)
2010-01-05 rdbLoadObject() as a separated function to load objects from disk. Dropped support for RDB version 0, I guess no longer has this legacy DBs around (antirez)
2010-01-04 VM low level pages handling (antirez)
2010-01-04 vm swap file creation, and some basic configuration (antirez)
2010-01-04 version marked 1.3.2 (antirez)
2010-01-04 saving code refactored a bit, added a function returning the number of bytes an object will use on disk (antirez)
2010-01-02 Now the PUSH side of RPOPLPUSH is able to unblock clients blocked on BLPOP (antirez)
2010-01-02 Version is now 1.3.1 (antirez)
2010-01-02 New vararg BLPOP able to block against multiple keys (antirez)
2009-12-29 fixed a problem with BLPOP timeout of zero, now it blocks forever (antirez)
2009-12-29 BLPOP timeouts implemented (antirez)
2009-12-29 first working implementation of BLPOP and BRPOP, still everything is to test well (antirez)
2009-12-29 a few more fixes, still broken (antirez)
2009-12-29 First fix, still broken (antirez)
2009-12-29 minor fix for Linux 64 bit (antirez)
2009-12-29 not yet working BLPOP implementation (antirez)
2009-12-27 AOFSYNC removed, got a better idea... (antirez)
2009-12-27 AOFSYNC command implemented (antirez)
2009-12-27 Version changed to 1.3.0, welcome to the new unstable (antirez)
2009-12-27 Now MULTI returns +OK as well (antirez)
2009-12-27 MULTI/EXEC first implementation (antirez)
2009-12-24 Fixed a minor bug in GETSET, now the SET part is not performed if the GET fails because the key does not contain a string value (antirez)
2009-12-23 html doc readded (antirez)
2009-12-23 ZRANGE WITHSCORES test added (antirez)
2009-12-23 version is now 1.1.94 (antirez)
2009-12-23 Add the command name in the unknown command error message. (antirez)
2009-12-22 ZRANGE, ZREVRANGE now support WITHSCORES options (antirez)
2009-12-22 html docs update (ZINCRBY added) (antirez)
2009-12-18 TODO list update (antirez)
2009-12-18 the pipelining test was ran against DB 1 for error, now it runs on DB 9 like all the other tests (antirez)
2009-12-18 still more tests (antirez)
2009-12-18 SORT STORE test added (antirez)
2009-12-18 Now SORT returns an empty bulk reply if the key does not exist (antirez)
2009-12-18 modified a bit the ZREVRANGE test to cover a few lines of code more (antirez)
2009-12-18 SHUTDOWN now does the right thing when append only is on, that is, fsync instead to save the snapshot. (antirez)
2009-12-18 Added a missing server.dirty increment in a non critical place, added more tests (antirez)
2009-12-18 LTRIM stress testing test added (antirez)
2009-12-18 LTRIM now returns +OK against non existing keys. More tests in test-redis.tcl (antirez)
2009-12-18 added sdstoupper() declaration in sds.h (antirez)
2009-12-18 Fixed sds.c bug #124 (antirez)
2009-12-16 LZF compression re-enabled by default, but with INIT_HTAB set to 0 to avoid the very costly memset initialization. Note that with this option set valgrind will output some false positive about lzf_c.c (antirez)
2009-12-16 lzf compression switched off by default now, with config file option to enable it in redis.conf (antirez)
2009-12-16 Regression for epoll bug in redis-test.tcl, version is now 1.1.93 (antirez)
2009-12-16 Fixed a lame epoll issue (antirez)
2009-12-15 html doc updated (antirez)
2009-12-15 version is now 1.1.92 (antirez)
2009-12-15 Two important fixes to append only file: zero length values and expires. A pretty neat new test to check consistency of randomly build datasets against snapshotting and AOF. (antirez)
2009-12-15 debug loadaof implemented in order to add more consistency tests in test-redis.tcl (antirez)
2009-12-15 Added a new test able to stress a lot the snapshotting engine (antirez)
2009-12-15 Unified handling of empty queries with normal queries. (antirez)
2009-12-15 Fixed some subtle bug in the command processing code almost impossible to spot in the real world, thanks to gcov (antirez)
2009-12-15 Regression test for SINTERSTORE added (antirez)
2009-12-15 Fixed issue #121 (antirez)
2009-12-14 a few more tests and ability to run a specific test in test-redis.tcl (antirez)
2009-12-13 Changed the reply of BGSAVE and BGREWRITEAOF from +OK to a more meaningful message that makes the user aware of an operation that just started and is not yet finished. (antirez)
2009-12-13 Set the master->slave logical client as authenticated on creation, so that if the slave requires a password replication works anyway (antirez)
2009-12-13 TODO update (antirez)
2009-12-12 bgrewriteaof_in_progress added to INFO (antirez)
2009-12-12 TODO list modified. What's planned for 1.4 is now written in the stone ;) (antirez)
2009-12-12 better handling of non blocking connect on redis-benchmark: EPIPE on read does not print an error message now (antirez)
2009-12-11 some change to redis-sha1.rb utility to make it more robust against non-meaningful changes in the dataset (antirez)
2009-12-10 redis-sha1.rb utility updated (antirez)
2009-12-10 a bit more verbose -ERR wrong number o arguments error, now gives info about the command name causing the error (antirez)
2009-12-10 TODO change and minor SETNX optimization (antirez)
2009-12-06 in rdbLoadDoubleValue now the buffer is nul terminated correctly. Thanks valgrind. (antirez)
2009-12-06 printf format warnings fixed by casting (antirez)
2009-12-06 Regression tests for SETNX and MSETNX bugs added (antirez)
2009-12-06 SETNX and MSETNX now respect the delete-on-write operation of EXPIREing keys (antirez)
2009-12-06 Fixed daemonization when using kqueue/kevent. Now the server initialization is performed *after* the daemonization (antirez)
2009-12-05 more HTML doc changes (antirez)
2009-12-05 HTML doc update (antirez)
2009-12-05 a few redis-cli format specified fixed (antirez)
2009-12-05 use __attribute__ format in sdscatprintf() when the compiler is GCC. Fixed format bugs resulting from the new warnings. (antirez)
2009-12-01 TODO update (antirez)
2009-12-01 compilation problem on 64bit mac os x 10.5 possibly fixed (antirez)
2009-12-01 virtual memory design doc typos (antirez)
2009-12-01 design documents added to the project (antirez)
2009-11-30 Fixed issued #85 (getDecodedObject: Assertion 1 != 1 failed. While sorting a set), added a smarter assert() function to dump the stacktrace, provided a macro to initalize Redis objects on the stack to avoid this kind of bugs. (antirez)
2009-11-30 fixed a subtle bug in redis-cli not having visible effects (antirez)
2009-11-29 TODO updated (antirez)
2009-11-29 Version chagned to 1.100, also known as the first first 2.0 beta version (antirez)
2009-11-29 more tests in test-redis.tcl, some minor fix (antirez)
2009-11-29 SORT support for sorted sets (antirez)
2009-11-28 Implemented LIMIT option in ZRANGEBYSCORE. We now enter feature-freeze (antirez)
2009-11-28 Changelog updated (antirez)
2009-11-28 html doc updated (antirez)
2009-11-28 enable kqueue/kevent only for Mac OS X 10.6.x as it seems that 10.5.x has a broken implementation of this syscalls. (antirez)
2009-11-28 TODO updated (antirez)
2009-11-28 ZRANGEBYSCORE fuzzy test (antirez)
2009-11-28 ZRANGEBYSCORE memory leak fixed, ZRANGEBYSCORE initial test added (antirez)
2009-11-28 INFO refactored. Stack trace on memory corruption now dumps the same information as the INFO command (antirez)
2009-11-28 ifdefs added to use kevent on Free Open and Net BSD as well. INFO and ae.c modified in order to report the multiplexing API in use (antirez)
2009-11-28 Enabled object encoding for multiple keys in MSET. Added a test for memory leaks in test-redis.tcl when running on Mac OS X (antirez)
2009-11-28 Merge branch 'kqueue' of git://github.com/mallipeddi/redis (antirez)
2009-11-28 Changes to TODO list, commented a function in redis.c (antirez)
2009-11-28 Added support for kqueue. (Harish Mallipeddi)
2009-11-27 TODO updated (antirez)
2009-11-26 zero length bulk data reading fixed in loadAppendOnlyFile() (antirez)
2009-11-26 append only file fixes (antirez)
2009-11-26 log rebuilding, random refactoring, work in progress please wait for an OK commit before to use this version (antirez)
2009-11-24 DEBUG RELOAD implemented, and test-redis.tcl modified to use it to check for persistence consistency. (antirez)
2009-11-24 Redis version set to 1.07 (antirez)
2009-11-24 sorted sets saving fixed (antirez)
2009-11-24 minor TODO change (antirez)
2009-11-24 minor fix to avoid a false valgrind warning. (antirez)
2009-11-23 epoll support enabled by default for Linux builds (antirez)
2009-11-23 epoll module for ae.c implemented. Some more testing needed (antirez)
2009-11-23 commented the HAVE_EPOLL test in config.h to allow compilation under Linux now that the epoll module is still missing (antirez)
2009-11-23 ae_select module added (antirez)
2009-11-23 ae.c now supports multiple polling API modules, even if only ae_select.c is implemented currently. Also adding and removing an event is now O(1). (antirez)
2009-11-23 ae.c initial refactoring for epoll implementation (antirez)
2009-11-21 version incremented up to 1.06 (antirez)
2009-11-21 TODO aesthetic changes (antirez)
2009-11-21 TODO updated with plans up to 1.5 (antirez)
2009-11-21 SRANDMEMBER test (antirez)
2009-11-21 Fixed a SORT memory leak that should never happen in practice (antirez)
2009-11-21 SORT GET # implemented, with a test (antirez)
2009-11-21 EXPIREAT test (antirez)
2009-11-20 EXPIRE tests (antirez)
2009-11-20 more RPOPLPUSH tests (antirez)
2009-11-20 RPOPLPUSH tests added (antirez)
2009-11-20 ZINCRBY return value fixed (antirez)
2009-11-20 ZINCRSCOREBY => ZINCRBY (antirez)
2009-11-19 ZINCRSCOREBY implemented (antirez)
2009-11-19 writev() finally uncommented again (antirez)
2009-11-19 redis-benchmark hopefully last bug with multi bulk reply fixed (antirez)
2009-11-19 debug mode in redis-bench (antirez)
2009-11-19 Use writev(2) if glue output buffers is disabled (antirez)
2009-11-19 benchmark.c fixes (antirez)
2009-11-18 more experiments with long replies, glue output buffer, and writev. (antirez)
2009-11-18 benchmarking with different number of LRANGE elements. Ability to change the glue output buffer limit by #define (antirez)
2009-11-18 more writev tests/work (antirez)
2009-11-18 redis-benchmark multi bulk reply support hopefully fixed (antirez)
2009-11-17 support for writev implemented but currently ifdef-ed in order to understan why I can't see the improvements expected. Btw code provided by Stefano Barbato (antirez)
2009-11-17 multi-bulk reply support for redis-bench, and as a result LRANGE is not tested, providing some number for the tuning of multi-bulk requests performances server-side (antirez)
2009-11-12 Solaris fix thanks to Alan Harder (antirez)
2009-11-12 Merge git://github.com/ianxm/redis (antirez)
2009-11-12 ZSCORE fixed, now returns NULL on missing key or missing element (antirez)
2009-11-12 Redis test will not fail the SAVE test even if a background save is in progress (antirez)
2009-11-12 LPOPPUSH renamed into RPOPLPUSH (antirez)
2009-11-11 can select db num (ian)
2009-11-11 Workaround for test-redis.tcl and Tcl 8.4.x about ZSCORE test (antirez)
2009-11-11 Removed a long time warning compiling with recent GCC on Linux (antirez)
2009-11-11 TODO updated (antirez)
2009-11-11 LPUSHPOP first implementation (antirez)
2009-11-10 Tcl script, make target, and redis.c changes to build the static symbol table automagically (antirez)
2009-11-10 Implemented a much better lazy expiring algorithm for EXPIRE (antirez)
2009-11-10 Fixed issue 92 in redis: redis-cli (nil) return value lacks CR/LF (antirez)
2009-11-10 Minor TODO change with new expiring algorithm description. New expiring algorithm moved since it'll go in 1.1 (antirez)
2009-11-04 redis-test is now a better Redis citizen, testing everything against DB 9 and 10 and only if this DBs are empty. (antirez)
2009-11-04 fixed a refcounting bug with SORT ... STORE leading to random crashes (root)
2009-11-04 masterauth option merged, thanks to Anthony Lauzon (antirez)
2009-11-03 ZSets double to string serialization fixed (antirez)
2009-11-03 client-libraries directory readded (antirez)
2009-11-03 redis.tcl put at toplevel since it's uesd for the test-redis.tcl script (antirez)
2009-11-03 client libs removed from Redis git (antirez)
2009-11-03 redis-cli now accepts a -r (repeat) switch. Still there is a memory leaks to fix (antirez)
2009-11-01 TODO updated again (antirez)
2009-11-01 TODO updated (antirez)
2009-11-01 redis-cli now makes clear when the returned string is an integer (antirez)
2009-11-01 SORT STORE option (antirez)
2009-11-01 now Redis prints DB stats just after the startup without to wait a second for the first report (antirez)
2009-11-01 another fix for append only mode, now read-only operations are not appended (antirez)
2009-11-01 appendfsync parsing in config file fixed. If you benchmarked Redis against different appendfsync options is time to try again ;) (antirez)
2009-11-01 append only file loading fixed (antirez)
2009-11-01 first version of append only file loading -- STILL BROKEN don't use it (antirez)
2009-10-31 Fixed Issue 83:Using TYPE on a zset results in a malformed response from the Redis server (antirez)
2009-10-31 Fixed compilation on Linux (antirez)
2009-10-30 append only mode is now able to translate EXPIRE into EXPIREAT transparently (antirez)
2009-10-30 appendfsync is now set to NO by default (antirez)
2009-10-30 support for appendonly mode no, always, everysec (antirez)
2009-10-30 first fix for append only mode (antirez)
2009-10-30 Initial implementation of append-only mode. Loading still not implemented. (antirez)
2009-10-30 EXPIRE behaviour changed a bit, a negative TTL or an EXPIREAT with unix time in the past will now delete the key. It seems saner to me than doing nothing. (antirez)
2009-10-30 EXPIREAT implemented, will be useful for the append-only mode (antirez)
2009-10-29 Fixed Issue 74 (ERR just returned on invalid password), now the error message is -ERR invalid password. (antirez)
2009-10-29 Fixed issue 72 (SLAVEOF shutdowns redis-server on malformed reply) (antirez)
2009-10-29 Fixed issue 77 (Incorrect time in log files) thanks to youwantalex (antirez)
2009-10-29 Fixed Issue 76 (redis-server crashes when it can't connect to MASTER and client connects to SLAVE) (antirez)
2009-10-29 ZREMRANGEBYSCORE implemented. Remove a range of elements with score between min and max (antirez)
2009-10-28 TODO changes and mostly theoretical minor skiplist change (antirez)
2009-10-28 ZLEN renamed ZCARD for consistency with SCARD (antirez)
2009-10-27 TODO reworked to reflect the real roadmap (antirez)
2009-10-27 Fix for 'make 32bit' (antirez)
2009-10-27 a fix for the solaris fix itself ;) (antirez)
2009-10-27 More Solaris fixes (antirez)
2009-10-27 A lot of ZSETs tests implemented, and a bug fixed thanks to this new tests (antirez)
2009-10-27 zmalloc Solaris fixes thanks to Alan Harder (antirez)
2009-10-27 ZSCORE implemented (antirez)
2009-10-26 fix for ZRANGEBYSCORE (antirez)
2009-10-26 ZRANGEBYSCORE implemented. Redis got range queries! (antirez)
2009-10-26 A trivial change makes the new implementation O(log(N)) instead of O(log(N))+O(M) when there are M repeated scores! (antirez)
2009-10-26 ZSET now saved on disk like any other type (antirez)
2009-10-26 double serialization routines implemented (antirez)
2009-10-26 ZSETs random fixes. Now the implementation appears to be pretty stable (antirez)
2009-10-26 another leak fixed. Can't find more for now, but still a bug in ZSETs to fix (antirez)
2009-10-26 ZSETs memory leak #1 solved, another one missing (antirez)
2009-10-26 Fix for skiplists backward link (antirez)
2009-10-26 Merged Solaris patches provided by Alan Harder (antirez)
2009-10-26 backward support to skiplists for ZREVRANGE, still broken, committing since I've to merge the Solaris patches (antirez)
2009-10-26 TODO updated (antirez)
2009-10-26 ZREM implemented (antirez)
2009-10-24 fix for ZADD in score update mode (antirez)
2009-10-24 some work on ZADD against existing element (score update), still broken... (antirez)
2009-10-23 zrange now starts to work. zadd still does not support update and will crash or leak or b000mmmmm (antirez)
2009-10-23 zrange initial hack (not working for now) (antirez)
2009-10-23 first skiplist fix, courtesy of valgrind (antirez)
2009-10-23 zset symbols added to stack trace code. ZSets will simply crash at the moment (antirez)
2009-10-23 more work on ZSETs and a new make target called 32bit to build i386 binaries on mac os x leopard (antirez)
2009-10-23 initial skiplist implementation. Most memory checks removed and zmalloc() modified to fail with an error message and abort. Anyway Redis is not designed to recover from out of memory conditions. (antirez)
2009-10-23 Fixed compilation in mac os x snow leopard when compiling a 32 bit binary. (antirez)
2009-10-22 version incremented to 1.050 to distinguish from 1.001 stable and next stable versions with minor fixes (antirez)
2009-10-21 TODO updated (antirez)
2009-10-21 SRANDMEMBER added (antirez)
2009-10-20 Imporant bug leading to data corruption fixed (NOT affecting stable distribution), Tcl client lib MSET/MSETNX implementation fixed, Added new tests for MSET and MSETNX in test-redis.tcl (antirez)
2009-10-17 added multi-bulk protocol support to redis-cli and support for MSET and MSETNX (antirez)
2009-10-17 MSET fixed, was not able to replace keys already set for a stupid bug (antirez)
2009-10-16 some dead code removed (antirez)
2009-10-16 multi bulk input protocol fixed (antirez)
2009-10-16 MSET and MSETNX commands implemented (antirez)
2009-10-07 undoed all the sds hacking that lead just to random bugs and no memory saving ;) (antirez)
2009-10-07 initial multi-bulk query protocol, this will allow MSET and other interesting features. (antirez)
2009-10-03 benchmark now outputs the right command line to shorten the TIME_WAIT interval on Mac OS X when keep alive is set (antirez)
2009-10-02 Issue 69 fixed. Object integer encoding now works with replication and MONITORing again. (antirez)
2009-09-18 LREM fixed, used to crash since the new object integer encoding is on the stage (antirez)
2009-09-17 maxmemory didn't worked in 64 systems for values > 4GB since it used to be an unsigned int. Fixed (antirez)
2009-09-10 incremented version number to 1.001, AKA Redis edge is no longer stable... (antirez)
2009-09-10 in-memory specialized object encoding (for now 32 signed integers only) (antirez)
2009-09-03 Latest doc changes for 1.0 (antirez)
2009-09-03 Redis 1.0.0 release (antirez)
2009-09-02 Redis version pushed to 1.0 (antirez)
2009-09-02 Ruby client lib updated to the latest git version (antirez)
2009-09-02 update-scala-client script added (antirez)
2009-09-02 Scala client added thanks to Alejanro Crosa (antirez)
2009-09-02 QuickStart added (antirez)
2009-09-01 Fixed crash with only space and newline as command (issue 61), thanks to a guy having as nick "fixxxerrr" (antirez)
2009-08-11 TODO list modified (antirez)
2009-07-24 more snow leopard related fixes (for 32bit systems) (antirez)
2009-07-24 fixed compilation with Snow Leopard, thanks to Lon Baker for providing SSH access to Snow Leopard box (antirez)
2009-07-22 Fixed NetBSD compile problems (antirez)
2009-07-17 now the size of the shared pool can be really modified via config, also the number of objects in the sharing pool is logged when the log level is set to debug. Thanks to Aman Gupta (antirez)
2009-07-05 added utils/redis-copy.rb, a script that is able to copy data from one Redis server to another one on the fly. (antirez)
2009-07-04 Applied three different patches thanks to Chris Lamb, one to fix compilation and get the IP register value on Linux IA64 and other systems. One in order to log the overcommit problem on the logs instead of the standard output when Redis is demonized. The latest in order to suggest a more consistent way in order to switch to 1 the memory overcommit Linux feature. (antirez)
2009-07-03 bugfix: EXPIRE now propagates to the Slave. (antirez)
2009-06-16 Redis version modified to 0.900 (antirez)
2009-06-16 update-ruby-client script already points to ezmobius repo (antirez)
2009-06-16 client libraries updated (antirez)
2009-06-16 Redis release candidate 1 (antirez)
2009-06-16 Better handling of background saving process killed or crashed (antirez)
2009-06-14 number of keys info in INFO command thanks to Diego Rosario Brogna (antirez)
2009-06-14 SPOP documented (antirez)
2009-06-14 Clojure library thanks to Ragnar Dahlén (antirez)
2009-06-10 It is now possible to specify - as config file name to read it from stdin (antirez)
2009-06-10 sync with jodosha redis-rb (antirez)
2009-06-10 Redis-rb sync (antirez)
2009-06-10 max inline request raised again to 1024*1024*256 bytes (antirez)
2009-06-10 max bytes in an inline command raised to 1024*1024 bytes, in order to allow for very large MGETs and still protect from client crashes (antirez)
2009-06-08 SPOP implemented. Hash table resizing for Sets and Expires too. Changed the resize policy to play better with RANDOMKEY and SPOP. (antirez)
2009-06-07 some minor changes to the backtrace code (antirez)
2009-06-07 enable backtrace capabilities only for Linux and MacOSX (antirez)
2009-06-07 Dump a backtrace on sigsegv/sigbus, original coded thanks to Diego Rosario Brogna, modified in order to work on different OSes and to enhance reliability (antirez)
2009-06-06 Merge git://github.com/dierbro/redis (antirez)
2009-06-06 add more output (hrothgar)
2009-06-06 store static function pointer for a useful stack trace (hrothgar)
2009-06-06 TODO updated (antirez)
2009-06-06 Makefile dependencies updated (antirez)
2009-06-05 Avoid a busy loop while sending very large replies against very fast links, this allows to be more responsive with other clients even under a KEY * against the loopback interface (antirez)
2009-06-05 Kill the background saving process before performing SHUTDOWN to avoid races (antirez)
2009-06-05 LREM now returns :0 for non existing keys (antirez)
2009-06-05 - put some order in code - better output (hrothgar)
2009-06-05 added config.h for #ifdef business isolation, added fstat64 for Mac OS X (antirez)
2009-06-04 remove die() :-) (hrothgar)
2009-06-04 add compile options to debug (hrothgar)
2009-06-04 initial commit print stack trace (hrothgar)
2009-06-04 initial commit print stack trace (hrothgar)
2009-06-04 macosx specific zmalloc.c, uses malloc_size function in order to avoid to waste memory and time to put an additional header (antirez)
2009-06-04 DEBUG OBJECT implemented (antirez)
2009-06-04 backtrace support removed: unreliable stack trace :( (antirez)
2009-06-04 initial backtrace dumping on sigsegv/sigbus + debug command (antirez)
2009-06-03 Python lib updated (antirez)
2009-06-03 shareobjectspoolsize implemented in reds.conf, in order to control the pool size when object sharing is on (antirez)
2009-05-30 Erlang client updated (antirez)
2009-05-30 Python client library updated (antirez)
2009-05-29 Redis-rb minor bool convertion fix (antirez)
2009-05-29 ruby library client is not Redis-rb merged with RubyRedis "engine" by Brian McKinney (antirez)
2009-05-28 __P completely removed from pqsort.c/h (antirez)
2009-05-28 another minor fix for Solaris boxes (antirez)
2009-05-28 minor fix for Solaris boxes (antirez)
2009-05-28 minor fix for Solaris boxes (antirez)
2009-05-27 maxmemory implemented (antirez)
2009-05-26 Redis git version modified to 0.101 in order to distinguish that from the latest tar.gz via INFO ;) (antirez)
2009-05-26 Redis 0.100 released (antirez)
2009-05-26 client libraries synched in git (antirez)
2009-05-26 ignore gcc warning about write() return code not checked. It is esplicitily this way since the "max number of clients reached" is a best-effort error (antirez)
2009-05-26 max bytes of a received command enlarged from 1k to 16k (antirez)
2009-05-26 RubyRedis: set TCP_NODELAY TCP socket option to to disable the neagle algorithm. Makes a huge difference under some OS, notably Linux (antirez)
2009-05-25 maxclients implemented, see redis.conf for details (antirez)
2009-05-25 INFO command now reports replication info (antirez)
2009-05-25 minor fix to RubyRedis about bulk commands sent without arguments (antirez)
2009-05-24 Warns if using the default config (antirez)
2009-05-24 Issue with redis-client used in scripts solved, now to check if the latest argument must come from standard input we do not check that stdin is or not a tty but the command arity (antirez)
2009-05-23 RubyRedis: now sets are returned as arrays again, and not as Set objects (antirez)
2009-05-23 SLAVEOF command documented (antirez)
2009-05-23 SLAVEOF command implemented for replication remote control (antirez)
2009-05-22 Fix: no connection timeout for the master! (antirez)
2009-05-22 replication slave timeout when receiving the initial bulk data set to 3600 seconds, now that replication is non-blocking the server must save the db before to start the async replication and this can take a lot of time with huge datasets (antirez)
2009-05-22 README tutorial now reflects the new proto (antirez)
2009-05-22 critical bug about glueoutputbuffers=yes fixed. Under load and with pipelining and clients disconnecting on the middle of the chat with the server, Redis could block. Now it's ok (antirez)
2009-05-22 TTL command doc added (antirez)
2009-05-22 TTL command implemented (antirez)
2009-05-22 S*STORE now return the cardinality of the resulting set (antirez)
2009-05-22 rubyredis more compatible with Redis-rb (antirez)
2009-05-21 minor indentation fix (antirez)
2009-05-21 timeout support and Redis-rb compatibility aliases implemented in RubyRedis (antirez)
2009-05-21 RubyRedis info postprocessor rewritten in a more functional way (antirez)
2009-05-21 dead code removed from RubyRedis (antirez)
2009-05-21 command postprocessing implemented into RubyRedis (antirez)
2009-05-20 Automagically reconnection of RubyRedis (antirez)
2009-05-20 RubyRedis: Array alike operators implemented (antirez)
2009-05-20 random testing code removed (antirez)
2009-05-20 RubyRedis DB selection forced at object creation (antirez)
2009-05-20 Initial version of an alternative Ruby client added (antirez)
2009-05-20 SDIFF / SDIFFSTORE added to doc (antirez)
2009-05-20 Aman Gupta changes merged (antirez)
2009-05-20 Merge git://github.com/tmm1/redis (antirez)
2009-05-19 Allow timeout=0 config to disable client timeouts (Aman Gupta)
2009-05-19 Partial qsort implemented in SORT command, only when both BY and LIMIT is used. minor fix for a warning compiling under Linux. (antirez)
2009-05-19 psort.c/h added. This is a partial qsort implementation that Redis will use when SORT+LIMIT is requested (antirez)
2009-05-17 Fix SINTER/UNIONSTORE to allow for &=/|= style operations (i.e. SINTERSTORE set1 set1 set2) (Aman Gupta)
2009-05-17 Optimize SDIFF to return as soon as the result set is empty (Aman Gupta)
2009-05-17 SDIFF/SDIFFSTORE implemnted unifying it with the implementation of SUNION/SUNIONSTORE (antirez)
2009-05-11 timestamp in log lines (antirez)
2009-05-11 Python client updated pushing from Ludo's repository (antirez)
2009-05-11 disconnect when we cannot read from the socket (Ludovico Magnocavallo)
2009-05-11 benchmark utility now supports random keys (antirez)
2009-05-10 minor doc changes (antirez)
2009-05-09 added tests for vararg DEL (antirez)
2009-05-09 DEL is now a vararg, IMPORTANT: memory leak fixed in loading DB code (antirez)
2009-05-09 doc changes (antirez)
2009-05-09 CPP client added thanks to Brian Hammond (antirez)
2009-05-06 Infinite number of arguments for MGET and all the other commands (antirez)
2009-05-04 Warns if /proc/sys/vm/overcommit_memory is set to 0 on Linux. Also make sure to don't resize the hash tables while the child process is saving in order to avoid copy-on-write of memory pages (antirez)
2009-04-30 zmalloc fix, return NULL or real malloc failure (antirez)
2009-04-30 more fixes for dict.c and the 150 million keys limit (antirez)
2009-04-30 dict.c modified to be able to handle more than 150,000,000 keys (antirez)
2009-04-29 fuzz stresser implemented in redis-test (antirez)
2009-04-29 fixed for HT resize check 32bits overflow (antirez)
2009-04-29 Check for fork() failure in background saving (antirez)
2009-04-29 fix for the LZF off-by-one bug added (antirez)
2009-04-28 print bytes used at exit on SHUTDOWN (antirez)
2009-04-28 SMOVE test added (antirez)
2009-04-28 SMOVE command implemented (antirez)
2009-04-28 less CPU usage in command parsing, case insensitive config directives (antirez)
2009-04-28 GETSET command doc added (antirez)
2009-04-28 GETSET tests (antirez)
2009-04-28 GETSET implemented (antirez)
2009-04-27 ability to specify a different file name for the DB (antirez)
2009-04-27 log file parsing code improved a bit (antirez)
2009-04-27 bgsave_in_progress field in INFO output (antirez)
2009-04-27 INCRBY/DECRBY now support 64bit increments, with tests (antirez)
2009-04-23 RANDOMKEY regression test added (antirez)
2009-04-23 dictGetRandomKey bug fixed, RANDOMKEY will not block the server anymore (antirez)
2009-04-22 FLUSHALL/FLUSHDB no longer sync on disk. Just increment the dirty counter by the number of elements removed, that will probably trigger a background saving operation (antirez)
2009-04-21 forgot to comment testing code in PHP lib. Now it is ok (antirez)
2009-04-21 PHP client ported to PHP5 and fixed (antirez)
2009-04-21 doc update (antirez)
2009-04-20 Non blocking replication (finally!). C-side linked lists API improved. (antirez)
2009-04-19 SUNION, SUNIONSTORE, Initial work on non blocking replication (antirez)
2009-04-10 Redis 0.091 released (antirez)
2009-04-10 SINTER/SINTERSTORE/SLEMENTS fix: misisng keys are now not errors, but just like empty sets (antirez)
2009-04-09 doc changes (antirez)
2009-04-08 TODO changes, minor change to default redis.conf (antirez)
2009-04-08 html doc updated (antirez)
2009-04-08 library clients update scripts (antirez)
2009-04-08 Ruby client updated (antirez)
2009-04-08 Lua client updated (antirez)
2009-04-08 Changelog updated (antirez)
2009-04-08 Merge git://github.com/ludoo/redis (antirez)
2009-04-08 add expire command to the php lib (Ludovico Magnocavallo)
2009-04-08 fix decode bug, add flush and info commands (Ludovico Magnocavallo)
2009-04-07 Rearrange redisObject struct to reduce memory usage in 64bit environments (as recommended http://groups.google.com/group/redis-db/msg/68f5a743f8f4e287) (Bob Potter)
2009-04-07 ruby19 compat: use each_line on string (Bob Potter)
2009-04-07 64bit fixes for usedmemory (Bob Potter)
2009-04-08 RANDOMKEY issue 26 fixed, generic test + regression added (antirez)
2009-04-06 Don't accept SAVE if BGSAVE is in progress (antirez)
2009-04-06 add expire command to the python lib (Ludovico Magnocavallo)
2009-04-03 persistent EXPIRE (antirez)
2009-04-03 dirty increment was missing in two points. TODO updated (antirez)
2009-04-02 LZF configured to initalize the HT in order to be determinsitic and play well with valgrind (antirez)
2009-04-02 fix select test (Ludovico Magnocavallo)
2009-04-02 fix trailing cr+nl in values (Ludovico Magnocavallo)
2009-04-02 compression/decompression of large values on disk now working (antirez)
2009-04-02 disable LZF compression since it's not able to load the DB for now, the load part is missing (antirez)
2009-04-02 new LZF files added (antirez)
2009-04-02 Fixed issue 23 about AUTH (antirez)
2009-04-02 Issue 22 fixed (antirez)
2009-04-01 non-lazy expired keys purging implemented (antirez)
2009-04-01 fastlz dependence removed (antirez)
2009-04-01 Initial implementation of EXPIRE (antirez)
2009-03-30 TODO updated (antirez)
2009-03-30 changelog added (antirez)
2009-03-28 redis-sha1 utility added (antirez)
2009-03-28 Integer encoding implemented in dump file. Doc updated (antirez)
2009-03-27 feature macros defined to play well with C99 (antirez)
2009-03-27 feature macros defined to play well with C99 (antirez)
2009-03-27 now Redis is C99-ok (antirez)
2009-03-27 IMPORTANT FIX: new dump format implementation was broken. Now it's ok but tests for the 32-bit case values are needed (antirez)
2009-03-27 ANSI-C compatibility changes (antirez)
2009-03-27 Ruby client library updated. Important changes in this new version! (antirez)
2009-03-26 Lua client added thanks to Daniele Alessandri (antirez)
2009-03-26 Lua client added thanks to Daniele Alessandri (antirez)
2009-03-26 AUTH merged from Brian Hammond fork, reworked a bit to fix minor problems (antirez)
2009-03-25 Adds AUTH command. (Brian Hammond)
2009-03-25 Nasty bug of the new DB format fixed, objects sharing implemented (antirez)
2009-03-25 doc update (antirez)
2009-03-25 Erlang client synched with Valentiono's repo (antirez)
2009-03-25 New file dump format, perl client library added (antirez)
2009-03-25 New protocol fix for LREM (antirez)
2009-03-24 two typos fixed (antirez)
2009-03-24 Now the Redis test uses the proper Tcl client library (antirez)
2009-03-24 Tcl client library (antirez)
2009-03-24 redis-benchmark sync with the new protocol (antirez)
2009-03-24 git mess :) (Ludovico Magnocavallo)
2009-03-24 sync python client to the new protocol (Ludovico Magnocavallo)
2009-03-24 protocol fix in SORT reply with null elements (antirez)
2009-03-24 protocol doc changed (antirez)
2009-03-24 Server replies now in the new format, test-redis.tcl and redis-cli modified accordingly (antirez)
2009-03-24 Python client library updated, thanks to Ludo! (antirez)
2009-03-24 random tested mode for test-redis.tcl, minor other stuff, version switched to 0.8 (antirez)
2009-03-23 Now MONITOR/SYNC cannot be issued multiple times (antirez)
2009-03-23 MONITOR command implemented. (antirez)
2009-03-23 lucsky changes imported. pid file path can now be configured, redis-cli fixes (antirez)
2009-03-23 Merge git://github.com/lucsky/redis (antirez)
2009-03-23 another missing free->zfree replacement fixed. Thanks to Ludo (antirez)
2009-03-23 Fixed redis-cli readLine loop to correctly handle EOF. (Luc Heinrich)
2009-03-23 Display the port on server startup. (Luc Heinrich)
2009-03-23 Allow to specify the pid file from the config file. (Luc Heinrich)
2009-03-23 Added gitignore file. (Luc Heinrich)
2009-03-22 MGET tests added (antirez)
2009-03-22 doc changes (antirez)
2009-03-22 added doc for MGET (antirez)
2009-03-22 redis-cli now checks the arity of vararg commnads (antirez)
2009-03-22 INFO fixed, MGET implemented, redis-cli implements INFO/MGET (antirez)
2009-03-22 first commit (antirez)
INSTALL
+1
View File
@@ -0,0 +1 @@
See README
MANIFESTO
+20
View File
@@ -0,0 +1,20 @@
[Note: this is the Redis manifesto, for general information about
installing and running Redis read the README file instead.]
Redis Manifesto
===============
1 - A DSL for Abstract Data Types. Redis is a DSL (Domain Specific Language) that manipulates abstract data types and implemented as a TCP daemon. Commands manipulate a key space where keys are binary-safe strings and values are different kinds of abstract data types. Every data type represents an abstract version of a fundamental data structure. For instance Redis Lists are an abstract representation of linked lists. In Redis, the essence of a data type isn't just the kind of operations that the data types support, but also the space and time complexity of the data type and the operations performed upon it.
2 - Memory storage is #1. The Redis data set, composed of defined key-value pairs, is primarily stored in the computer's memory. The amount of memory in all kinds of computers, including entry-level servers, is increasing significantly each year. Memory is fast, and allows Redis to have very predictable performance. Datasets composed of 10k or 40 millions keys will perform similarly. Complex data types like Redis Sorted Sets are easy to implement and manipulate in memory with good performance, making Redis very simple. Redis will continue to explore alternative options (where data can be optionally stored on disk, say) but the main goal of the project remains the development of an in-memory database.
3 - Fundamental data structures for a fundamental API. The Redis API is a direct consequence of fundamental data structures. APIs can often be arbitrary but not an API that resembles the nature of fundamental data structures. If we ever meet intelligent life forms from another part of the universe, they'll likely know, understand and recognize the same basic data structures we have in our computer science books. Redis will avoid intermediate layers in API, so that the complexity is obvious and more complex operations can be performed as the sum of the basic operations.
4 - Code is like a poem; it's not just something we write to reach some practical result. Sometimes people that are far from the Redis philosophy suggest using other code written by other authors (frequently in other languages) in order to implement something Redis currently lacks. But to us this is like if Shakespeare decided to end Enrico IV using the Paradiso from the Divina Commedia. Is using any external code a bad idea? Not at all. Like in "One Thousand and One Nights" smaller self contained stories are embedded in a bigger story, we'll be happy to use beautiful self contained libraries when needed. At the same time, when writing the Redis story we're trying to write smaller stories that will fit in to other code.
5 - We're against complexity. We believe designing systems is a fight against complexity. We'll accept to fight the complexity when it's worthwhile but we'll try hard to recognize when a small feature is not worth 1000s of lines of code. Most of the time the best way to fight complexity is by not creating it at all.
6 - Two levels of API. The Redis API has two levels: 1) a subset of the API fits naturally into a distributed version of Redis and 2) a more complex API that supports multi-key operations. Both are useful if used judiciously but there's no way to make the more complex multi-keys API distributed in an opaque way without violating our other principles. We don't want to provide the illusion of something that will work magically when actually it can't in all cases. Instead we'll provide commands to quickly migrate keys from one instance to another to perform multi-key operations and expose the tradeoffs to the user.
7 - We optimize for joy. We believe writing code is a lot of hard work, and the only way it can be worth is by enjoying it. When there is no longer joy in writing code, the best thing to do is stop. To prevent this, we'll avoid taking paths that will make Redis less of a joy to develop.
Makefile
+7 -111
View File
@@ -1,115 +1,11 @@
# Redis Makefile
# Copyright (C) 2009 Salvatore Sanfilippo <antirez at gmail dot com>
# This file is released under the BSD license, see the COPYING file
# Top level makefile, the real shit is at src/Makefile
release_hdr := $(shell sh -c './mkreleasehdr.sh')
uname_S := $(shell sh -c 'uname -s 2>/dev/null || echo not')
OPTIMIZATION?=-O2
ifeq ($(uname_S),SunOS)
CFLAGS?= -std=c99 -pedantic $(OPTIMIZATION) -Wall -W -D__EXTENSIONS__ -D_XPG6
CCLINK?= -ldl -lnsl -lsocket -lm -lpthread
else
CFLAGS?= -std=c99 -pedantic $(OPTIMIZATION) -Wall -W $(ARCH) $(PROF)
CCLINK?= -lm -pthread
endif
CCOPT= $(CFLAGS) $(CCLINK) $(ARCH) $(PROF)
DEBUG?= -g -rdynamic -ggdb
default: all
OBJ = adlist.o ae.o anet.o dict.o redis.o sds.o zmalloc.o lzf_c.o lzf_d.o pqsort.o zipmap.o sha1.o
BENCHOBJ = ae.o anet.o redis-benchmark.o sds.o adlist.o zmalloc.o
CLIOBJ = anet.o sds.o adlist.o redis-cli.o zmalloc.o linenoise.o
CHECKDUMPOBJ = redis-check-dump.o lzf_c.o lzf_d.o
CHECKAOFOBJ = redis-check-aof.o
.DEFAULT:
cd src && $(MAKE) $@
PRGNAME = redis-server
BENCHPRGNAME = redis-benchmark
CLIPRGNAME = redis-cli
CHECKDUMPPRGNAME = redis-check-dump
CHECKAOFPRGNAME = redis-check-aof
install:
cd src && $(MAKE) $@
all: redis-server redis-benchmark redis-cli redis-check-dump redis-check-aof
# Deps (use make dep to generate this)
adlist.o: adlist.c adlist.h zmalloc.h
ae.o: ae.c ae.h zmalloc.h config.h ae_kqueue.c
ae_epoll.o: ae_epoll.c
ae_kqueue.o: ae_kqueue.c
ae_select.o: ae_select.c
anet.o: anet.c fmacros.h anet.h
dict.o: dict.c fmacros.h dict.h zmalloc.h
dict2.o: dict2.c fmacros.h dict2.h zmalloc.h
linenoise.o: linenoise.c fmacros.h
lzf_c.o: lzf_c.c lzfP.h
lzf_d.o: lzf_d.c lzfP.h
pqsort.o: pqsort.c
printraw.o: printraw.c
redis-benchmark.o: redis-benchmark.c fmacros.h ae.h anet.h sds.h adlist.h \
zmalloc.h
redis-check-aof.o: redis-check-aof.c fmacros.h config.h
redis-check-dump.o: redis-check-dump.c lzf.h
redis-cli.o: redis-cli.c fmacros.h anet.h sds.h adlist.h zmalloc.h \
linenoise.h
redis.o: redis.c fmacros.h config.h redis.h ae.h sds.h anet.h dict.h \
adlist.h zmalloc.h lzf.h pqsort.h zipmap.h staticsymbols.h sha1.h
sds.o: sds.c sds.h zmalloc.h
zipmap.o: zipmap.c zmalloc.h
zmalloc.o: zmalloc.c config.h
redis-server: $(OBJ)
$(CC) -o $(PRGNAME) $(CCOPT) $(DEBUG) $(OBJ)
@echo ""
@echo "Hint: To run the test-redis.tcl script is a good idea."
@echo "Launch the redis server with ./redis-server, then in another"
@echo "terminal window enter this directory and run 'make test'."
@echo ""
redis-benchmark: $(BENCHOBJ)
$(CC) -o $(BENCHPRGNAME) $(CCOPT) $(DEBUG) $(BENCHOBJ)
redis-cli: $(CLIOBJ)
$(CC) -o $(CLIPRGNAME) $(CCOPT) $(DEBUG) $(CLIOBJ)
redis-check-dump: $(CHECKDUMPOBJ)
$(CC) -o $(CHECKDUMPPRGNAME) $(CCOPT) $(DEBUG) $(CHECKDUMPOBJ)
redis-check-aof: $(CHECKAOFOBJ)
$(CC) -o $(CHECKAOFPRGNAME) $(CCOPT) $(DEBUG) $(CHECKAOFOBJ)
.c.o:
$(CC) -c $(CFLAGS) $(DEBUG) $(COMPILE_TIME) $<
clean:
rm -rf $(PRGNAME) $(BENCHPRGNAME) $(CLIPRGNAME) $(CHECKDUMPPRGNAME) $(CHECKAOFPRGNAME) *.o *.gcda *.gcno *.gcov
dep:
$(CC) -MM *.c
staticsymbols:
tclsh utils/build-static-symbols.tcl > staticsymbols.h
test:
tclsh8.5 tests/test_helper.tcl
bench:
./redis-benchmark
log:
git log '--pretty=format:%ad %s (%cn)' --date=short > Changelog
32bit:
@echo ""
@echo "WARNING: if it fails under Linux you probably need to install libc6-dev-i386"
@echo ""
make ARCH="-m32"
gprof:
make PROF="-pg"
gcov:
make PROF="-fprofile-arcs -ftest-coverage"
noopt:
make OPTIMIZATION=""
32bitgprof:
make PROF="-pg" ARCH="-arch i386"
.PHONY: install
README
+143 -1
View File
@@ -1 +1,143 @@
Check the 'doc' directory. doc/README.html is a good starting point :)
Where to find complete Redis documentation?
-------------------------------------------
This README is just a fast "quick start" document. You can find more detailed
documentation at http://redis.io
Building Redis
--------------
Redis can be compiled and used on Linux, OSX, OpenBSD, NetBSD, FreeBSD.
We support big endian and little endian architectures.
It may compile on Solaris derived systems (for instance SmartOS) but our
support for this platform is "best effort" and Redis is not guaranteed to
work as well as in Linux, OSX, and *BSD there.
It is as simple as:
% make
You can run a 32 bit Redis binary using:
% make 32bit
After building Redis is a good idea to test it, using:
% make test
NOTE: if after building Redis with a 32 bit target you need to rebuild it
with a 64 bit target you need to perform a "make clean" in the root
directory of the Redis distribution.
Allocator
---------
Selecting a non-default memory allocator when building Redis is done by setting
the `MALLOC` environment variable. Redis is compiled and linked against libc
malloc by default, with the exception of jemalloc being the default on Linux
systems. This default was picked because jemalloc has proven to have fewer
fragmentation problems than libc malloc.
To force compiling against libc malloc, use:
% make MALLOC=libc
To compile against jemalloc on Mac OS X systems, use:
% make MALLOC=jemalloc
Verbose build
-------------
Redis will build with a user friendly colorized output by default.
If you want to see a more verbose output use the following:
% make V=1
Running Redis
-------------
To run Redis with the default configuration just type:
% cd src
% ./redis-server
If you want to provide your redis.conf, you have to run it using an additional
parameter (the path of the configuration file):
% cd src
% ./redis-server /path/to/redis.conf
It is possible to alter the Redis configuration passing parameters directly
as options using the command line. Examples:
% ./redis-server --port 9999 --slaveof 127.0.0.1 6379
% ./redis-server /etc/redis/6379.conf --loglevel debug
All the options in redis.conf are also supported as options using the command
line, with exactly the same name.
Playing with Redis
------------------
You can use redis-cli to play with Redis. Start a redis-server instance,
then in another terminal try the following:
% cd src
% ./redis-cli
redis> ping
PONG
redis> set foo bar
OK
redis> get foo
"bar"
redis> incr mycounter
(integer) 1
redis> incr mycounter
(integer) 2
redis>
You can find the list of all the available commands here:
http://redis.io/commands
Installing Redis
-----------------
In order to install Redis binaries into /usr/local/bin just use:
% make install
You can use "make PREFIX=/some/other/directory install" if you wish to use a
different destination.
Make install will just install binaries in your system, but will not configure
init scripts and configuration files in the appropriate place. This is not
needed if you want just to play a bit with Redis, but if you are installing
it the proper way for a production system, we have a script doing this
for Ubuntu and Debian systems:
% cd utils
% ./install_server
The script will ask you a few questions and will setup everything you need
to run Redis properly as a background daemon that will start again on
system reboots.
You'll be able to stop and start Redis using the script named
/etc/init.d/redis_<portnumber>, for instance /etc/init.d/redis_6379.
Code contributions
---
Note: by contributing code to the Redis project in any form, including sending
a pull request via Github, a code fragment or patch via private email or
public discussion groups, you agree to release your code under the terms
of the BSD license that you can find in the COPYING file included in the Redis
source distribution.
Please see the CONTRIBUTING file in this source distribution for more
information.
Enjoy!
TODO
-66
View File
@@ -1,66 +0,0 @@
Redis TODO and Roadmap
----------------------
VERSION 2.2 TODO (Optimizations and latency)
============================================
* Lower the CPU usage.
* Lower the RAM usage everywhere possible.
* Specially encoded Sets (like Hashes).
* Implement an UDP interface for low-latency operations.
* What about a special coding that is about storing the "rdb" serialized format instead of the actual value? This can be used when we have LRU in order to super-compress data into memory, for data not accessed frequetly. It's a VM-alike strategy but fully in memory, may reduce the space to hold some dataset in an impressive way. Trivial to implement.
* Another idea: LRU does not need to be super precise right? Maybe it's a good idea to just put into the skiplist implementing the LRU just the pointer to the key without evne incr/decr business, nor the need to remove the pointer when the key is deleted. There is to think more about that.
VERSION 2.x TODO
================
* BRPOPLPUSH
* Save dataset / fsync() on SIGTERM
* Change the implementation of ZCOUNT to use the augmented skiplist in order to be much faster.
Virtual Memory optimizations:
* Use multiple open FDs against the VM file, one for thread.
* Check what happens performance-wise if instead of creating threads again and again the same threads are reused forever. Note: this requires a way to disable this clients in the child, but waiting for empty new jobs queue can be enough.
* Implement LEN, PEEK, POKE, SETBIT, GETBIT
OTHER IMPORTANT THINGS THAT WILL BE ADDED BUT I'M NOT SURE WHEN
===============================================================
BIG ONES:
* Specially encoded memory-saving integer sets.
* A command to export a JSON dump (there should be mostly working patch needing major reworking).
* Specially encoded sets of integers (this includes a big refactoring providing an higher level layer for Sets manipulation)
SMALL ONES:
* If sizeof(double) == sizeof(void*) we could store the double value of sorted sets directly in place of the pointer instead of allocating it in the heap.
* Delete on writes against expire policy should only happen after argument parsing for commands doing their own arg parsing stuff.
* Give errors when incrementing a key that does not look like an integer, when providing as a sorted set score something can't be parsed as a double, and so forth.
* MSADD (n keys) (n values). See this thread in the Redis google group: http://groups.google.com/group/redis-db/browse_thread/thread/e766d84eb375cd41
* Don't save empty lists / sets / zsets on disk with snapshotting.
* Remove keys when a list / set / zset reaches length of 0.
* An option to exec a command slave-side if the master connection is lost: even cooler: if the script returns "0" the slave elects itself as master, otherwise continue trying to reconnect.
* PING the master from time to time to check if it's gone.
THE "MAYBE" TODO LIST: things that may or may not get implemented
=================================================================
Most of this can be seen just as proposals, the fact they are in this list
it's not a guarantee they'll ever get implemented ;)
* Move dict.c from hash table to skip list, in order to avoid the blocking resize operation needed for the hash table.
* FORK command (fork()s executing the commands received by the current
client in the new process). Hint: large SORTs can use more cores,
copy-on-write will avoid memory problems.
* DUP command? DUP srckey dstkey, creates an exact clone of srckey value in dstkey.
* SORT: Don't copy the list into a vector when BY argument is constant.
* Write the hash table size of every db in the dump, so that Redis can resize the hash table just one time when loading a big DB.
* LOCK / TRYLOCK / UNLOCK as described many times in the google group
* Replication automated tests
* Byte Array type (BA prefixed commands): BASETBIT BAGETBIT BASETU8 U16 U32 U64 S8 S16 S32 S64, ability to atomically INCRBY all the base types. BARANGE to get a range of bytes as a bulk value, BASETRANGE to set a range of bytes.
* zmalloc() should avoid to add a private header for archs where there is some other kind of libc-specific way to get the size of a malloced block. Already done for Mac OS X.
* Read-only mode.
* Pattern-matching replication.
* Add an option to relax the delete-expiring-keys-on-write semantic *denying* replication and AOF when this is on? Can be handy sometimes, when using Redis for non persistent state, but can create problems. For instance should rename and move also "move" the timeouts? How does this affect other commands?
* Multiple BY in SORT.
client-libraries/README
-11
View File
@@ -1,11 +0,0 @@
An updated list of client libraries for Redis can be found here:
http://code.google.com/p/redis
All the links are in the front page.
Most mainstream (and less mainstream) languages are supported.
If you plan to write a new client library for a missing language, please
get in touch with the other developers writing a message to the Redis google
group that is located here: http://groups.google.com/group/redis-db
config.h
-45
View File
@@ -1,45 +0,0 @@
#ifndef __CONFIG_H
#define __CONFIG_H
#ifdef __APPLE__
#include <AvailabilityMacros.h>
#endif
/* test for malloc_size() */
#ifdef __APPLE__
#include <malloc/malloc.h>
#define HAVE_MALLOC_SIZE 1
#define redis_malloc_size(p) malloc_size(p)
#endif
/* define redis_fstat to fstat or fstat64() */
#if defined(__APPLE__) && !defined(MAC_OS_X_VERSION_10_6)
#define redis_fstat fstat64
#define redis_stat stat64
#else
#define redis_fstat fstat
#define redis_stat stat
#endif
/* test for backtrace() */
#if defined(__APPLE__) || defined(__linux__)
#define HAVE_BACKTRACE 1
#endif
/* test for polling API */
#ifdef __linux__
#define HAVE_EPOLL 1
#endif
#if (defined(__APPLE__) && defined(MAC_OS_X_VERSION_10_6)) || defined(__FreeBSD__) || defined(__OpenBSD__) || defined (__NetBSD__)
#define HAVE_KQUEUE 1
#endif
/* define aof_fsync to fdatasync() in Linux and fsync() for all the rest */
#ifdef __linux__
#define aof_fsync fdatasync
#else
#define aof_fsync fsync
#endif
#endif
deps/Makefile
Vendored
+78
View File
@@ -0,0 +1,78 @@
# Redis dependency Makefile
uname_S:= $(shell sh -c 'uname -s 2>/dev/null || echo not')
CCCOLOR="\033[34m"
LINKCOLOR="\033[34;1m"
SRCCOLOR="\033[33m"
BINCOLOR="\033[37;1m"
MAKECOLOR="\033[32;1m"
ENDCOLOR="\033[0m"
default:
@echo "Explicit target required"
.PHONY: default
# Prerequisites target
.make-prerequisites:
@touch $@
# Clean everything when CFLAGS is different
ifneq ($(shell sh -c '[ -f .make-cflags ] && cat .make-cflags || echo none'), $(CFLAGS))
.make-cflags: distclean
-(echo "$(CFLAGS)" > .make-cflags)
.make-prerequisites: .make-cflags
endif
# Clean everything when LDFLAGS is different
ifneq ($(shell sh -c '[ -f .make-ldflags ] && cat .make-ldflags || echo none'), $(LDFLAGS))
.make-ldflags: distclean
-(echo "$(LDFLAGS)" > .make-ldflags)
.make-prerequisites: .make-ldflags
endif
distclean:
-(cd hiredis && $(MAKE) clean) > /dev/null || true
-(cd linenoise && $(MAKE) clean) > /dev/null || true
-(cd lua && $(MAKE) clean) > /dev/null || true
-(cd jemalloc && [ -f Makefile ] && $(MAKE) distclean) > /dev/null || true
-(rm -f .make-*)
.PHONY: distclean
hiredis: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd hiredis && $(MAKE) static
.PHONY: hiredis
linenoise: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd linenoise && $(MAKE)
.PHONY: linenoise
ifeq ($(uname_S),SunOS)
# Make isinf() available
LUA_CFLAGS= -D__C99FEATURES__=1
endif
LUA_CFLAGS+= -O2 -Wall -DLUA_ANSI $(CFLAGS)
LUA_LDFLAGS+= $(LDFLAGS)
lua: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd lua/src && $(MAKE) all CFLAGS="$(LUA_CFLAGS)" MYLDFLAGS="$(LUA_LDFLAGS)"
.PHONY: lua
JEMALLOC_CFLAGS= -std=gnu99 -Wall -pipe -g3 -O3 -funroll-loops $(CFLAGS)
JEMALLOC_LDFLAGS= $(LDFLAGS)
jemalloc: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd jemalloc && ./configure --with-jemalloc-prefix=je_ --enable-cc-silence CFLAGS="$(JEMALLOC_CFLAGS)" LDFLAGS="$(JEMALLOC_LDFLAGS)"
cd jemalloc && $(MAKE) CFLAGS="$(JEMALLOC_CFLAGS)" LDFLAGS="$(JEMALLOC_LDFLAGS)" lib/libjemalloc.a
.PHONY: jemalloc
deps/hiredis/.gitignore
+6
View File
@@ -0,0 +1,6 @@
/hiredis-test
/hiredis-example*
/*.o
/*.so
/*.dylib
/*.a
deps/hiredis/CHANGELOG.md
Vendored
+16
View File
@@ -0,0 +1,16 @@
### 0.10.1
* Makefile overhaul. Important to check out if you override one or more
variables using environment variables or via arguments to the "make" tool.
* Issue #45: Fix potential memory leak for a multi bulk reply with 0 elements
being created by the default reply object functions.
* Issue #43: Don't crash in an asynchronous context when Redis returns an error
reply after the connection has been made (this happens when the maximum
number of connections is reached).
### 0.10.0
* See commit log.
deps/hiredis/COPYING
Vendored
+29
View File
@@ -0,0 +1,29 @@
Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
* Neither the name of Redis nor the names of its contributors may be used
to endorse or promote products derived from this software without specific
prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
deps/hiredis/Makefile
Vendored
+148
View File
@@ -0,0 +1,148 @@
# Hiredis Makefile
# Copyright (C) 2010-2011 Salvatore Sanfilippo <antirez at gmail dot com>
# Copyright (C) 2010-2011 Pieter Noordhuis <pcnoordhuis at gmail dot com>
# This file is released under the BSD license, see the COPYING file
OBJ=net.o hiredis.o sds.o async.o
BINS=hiredis-example hiredis-test
LIBNAME=libhiredis
HIREDIS_MAJOR=0
HIREDIS_MINOR=10
# Fallback to gcc when $CC is not in $PATH.
CC:=$(shell sh -c 'type $(CC) >/dev/null 2>/dev/null && echo $(CC) || echo gcc')
OPTIMIZATION?=-O3
WARNINGS=-Wall -W -Wstrict-prototypes -Wwrite-strings
DEBUG?= -g -ggdb
REAL_CFLAGS=$(OPTIMIZATION) -fPIC $(CFLAGS) $(WARNINGS) $(DEBUG) $(ARCH)
REAL_LDFLAGS=$(LDFLAGS) $(ARCH)
DYLIBSUFFIX=so
STLIBSUFFIX=a
DYLIB_MINOR_NAME=$(LIBNAME).$(DYLIBSUFFIX).$(HIREDIS_MAJOR).$(HIREDIS_MINOR)
DYLIB_MAJOR_NAME=$(LIBNAME).$(DYLIBSUFFIX).$(HIREDIS_MAJOR)
DYLIBNAME=$(LIBNAME).$(DYLIBSUFFIX)
DYLIB_MAKE_CMD=$(CC) -shared -Wl,-soname,$(DYLIB_MINOR_NAME) -o $(DYLIBNAME) $(LDFLAGS)
STLIBNAME=$(LIBNAME).$(STLIBSUFFIX)
STLIB_MAKE_CMD=ar rcs $(STLIBNAME)
# Platform-specific overrides
uname_S := $(shell sh -c 'uname -s 2>/dev/null || echo not')
ifeq ($(uname_S),SunOS)
REAL_LDFLAGS+= -ldl -lnsl -lsocket
DYLIB_MAKE_CMD=$(CC) -G -o $(DYLIBNAME) -h $(DYLIB_MINOR_NAME) $(LDFLAGS)
INSTALL= cp -r
endif
ifeq ($(uname_S),Darwin)
DYLIBSUFFIX=dylib
DYLIB_MINOR_NAME=$(LIBNAME).$(HIREDIS_MAJOR).$(HIREDIS_MINOR).$(DYLIBSUFFIX)
DYLIB_MAJOR_NAME=$(LIBNAME).$(HIREDIS_MAJOR).$(DYLIBSUFFIX)
DYLIB_MAKE_CMD=$(CC) -shared -Wl,-install_name,$(DYLIB_MINOR_NAME) -o $(DYLIBNAME) $(LDFLAGS)
endif
all: $(DYLIBNAME) $(BINS)
# Deps (use make dep to generate this)
net.o: net.c fmacros.h net.h hiredis.h
async.o: async.c async.h hiredis.h sds.h dict.c dict.h
example.o: example.c hiredis.h
hiredis.o: hiredis.c fmacros.h hiredis.h net.h sds.h
sds.o: sds.c sds.h
test.o: test.c hiredis.h
$(DYLIBNAME): $(OBJ)
$(DYLIB_MAKE_CMD) $(OBJ)
$(STLIBNAME): $(OBJ)
$(STLIB_MAKE_CMD) $(OBJ)
dynamic: $(DYLIBNAME)
static: $(STLIBNAME)
# Binaries:
hiredis-example-libevent: example-libevent.c adapters/libevent.h $(STLIBNAME)
$(CC) -o $@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -levent example-libevent.c $(STLIBNAME)
hiredis-example-libev: example-libev.c adapters/libev.h $(STLIBNAME)
$(CC) -o $@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -lev example-libev.c $(STLIBNAME)
ifndef AE_DIR
hiredis-example-ae:
@echo "Please specify AE_DIR (e.g. <redis repository>/src)"
@false
else
hiredis-example-ae: example-ae.c adapters/ae.h $(STLIBNAME)
$(CC) -o $@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -I$(AE_DIR) $(AE_DIR)/ae.o $(AE_DIR)/zmalloc.o example-ae.c $(STLIBNAME)
endif
hiredis-%: %.o $(STLIBNAME)
$(CC) -o $@ $(REAL_LDFLAGS) $< $(STLIBNAME)
test: hiredis-test
./hiredis-test
check: hiredis-test
echo \
"daemonize yes\n" \
"pidfile /tmp/hiredis-test-redis.pid\n" \
"port 56379\n" \
"bind 127.0.0.1\n" \
"unixsocket /tmp/hiredis-test-redis.sock" \
| redis-server -
./hiredis-test -h 127.0.0.1 -p 56379 -s /tmp/hiredis-test-redis.sock || \
( kill `cat /tmp/hiredis-test-redis.pid` && false )
kill `cat /tmp/hiredis-test-redis.pid`
.c.o:
$(CC) -std=c99 -pedantic -c $(REAL_CFLAGS) $<
clean:
rm -rf $(DYLIBNAME) $(STLIBNAME) $(BINS) hiredis-example* *.o *.gcda *.gcno *.gcov
dep:
$(CC) -MM *.c
# Installation related variables and target
PREFIX?=/usr/local
INCLUDE_PATH?=include/hiredis
LIBRARY_PATH?=lib
INSTALL_INCLUDE_PATH= $(PREFIX)/$(INCLUDE_PATH)
INSTALL_LIBRARY_PATH= $(PREFIX)/$(LIBRARY_PATH)
ifeq ($(uname_S),SunOS)
INSTALL?= cp -r
endif
INSTALL?= cp -a
install: $(DYLIBNAME) $(STLIBNAME)
mkdir -p $(INSTALL_INCLUDE_PATH) $(INSTALL_LIBRARY_PATH)
$(INSTALL) hiredis.h async.h adapters $(INSTALL_INCLUDE_PATH)
$(INSTALL) $(DYLIBNAME) $(INSTALL_LIBRARY_PATH)/$(DYLIB_MINOR_NAME)
cd $(INSTALL_LIBRARY_PATH) && ln -sf $(DYLIB_MINOR_NAME) $(DYLIB_MAJOR_NAME)
cd $(INSTALL_LIBRARY_PATH) && ln -sf $(DYLIB_MAJOR_NAME) $(DYLIBNAME)
$(INSTALL) $(STLIBNAME) $(INSTALL_LIBRARY_PATH)
32bit:
@echo ""
@echo "WARNING: if this fails under Linux you probably need to install libc6-dev-i386"
@echo ""
$(MAKE) CFLAGS="-m32" LDFLAGS="-m32"
gprof:
$(MAKE) CFLAGS="-pg" LDFLAGS="-pg"
gcov:
$(MAKE) CFLAGS="-fprofile-arcs -ftest-coverage" LDFLAGS="-fprofile-arcs"
coverage: gcov
make check
mkdir -p tmp/lcov
lcov -d . -c -o tmp/lcov/hiredis.info
genhtml --legend -o tmp/lcov/report tmp/lcov/hiredis.info
noopt:
$(MAKE) OPTIMIZATION=""
.PHONY: all test check clean dep install 32bit gprof gcov noopt
deps/hiredis/README.md
Vendored
+379
View File
@@ -0,0 +1,379 @@
# HIREDIS
Hiredis is a minimalistic C client library for the [Redis](http://redis.io/) database.
It is minimalistic because it just adds minimal support for the protocol, but
at the same time it uses an high level printf-alike API in order to make it
much higher level than otherwise suggested by its minimal code base and the
lack of explicit bindings for every Redis command.
Apart from supporting sending commands and receiving replies, it comes with
a reply parser that is decoupled from the I/O layer. It
is a stream parser designed for easy reusability, which can for instance be used
in higher level language bindings for efficient reply parsing.
Hiredis only supports the binary-safe Redis protocol, so you can use it with any
Redis version >= 1.2.0.
The library comes with multiple APIs. There is the
*synchronous API*, the *asynchronous API* and the *reply parsing API*.
## UPGRADING
Version 0.9.0 is a major overhaul of hiredis in every aspect. However, upgrading existing
code using hiredis should not be a big pain. The key thing to keep in mind when
upgrading is that hiredis >= 0.9.0 uses a `redisContext*` to keep state, in contrast to
the stateless 0.0.1 that only has a file descriptor to work with.
## Synchronous API
To consume the synchronous API, there are only a few function calls that need to be introduced:
redisContext *redisConnect(const char *ip, int port);
void *redisCommand(redisContext *c, const char *format, ...);
void freeReplyObject(void *reply);
### Connecting
The function `redisConnect` is used to create a so-called `redisContext`. The
context is where Hiredis holds state for a connection. The `redisContext`
struct has an integer `err` field that is non-zero when an the connection is in
an error state. The field `errstr` will contain a string with a description of
the error. More information on errors can be found in the **Errors** section.
After trying to connect to Redis using `redisConnect` you should
check the `err` field to see if establishing the connection was successful:
redisContext *c = redisConnect("127.0.0.1", 6379);
if (c->err) {
printf("Error: %s\n", c->errstr);
// handle error
}
### Sending commands
There are several ways to issue commands to Redis. The first that will be introduced is
`redisCommand`. This function takes a format similar to printf. In the simplest form,
it is used like this:
reply = redisCommand(context, "SET foo bar");
The specifier `%s` interpolates a string in the command, and uses `strlen` to
determine the length of the string:
reply = redisCommand(context, "SET foo %s", value);
When you need to pass binary safe strings in a command, the `%b` specifier can be
used. Together with a pointer to the string, it requires a `size_t` length argument
of the string:
reply = redisCommand(context, "SET foo %b", value, valuelen);
Internally, Hiredis splits the command in different arguments and will
convert it to the protocol used to communicate with Redis.
One or more spaces separates arguments, so you can use the specifiers
anywhere in an argument:
reply = redisCommand(context, "SET key:%s %s", myid, value);
### Using replies
The return value of `redisCommand` holds a reply when the command was
successfully executed. When an error occurs, the return value is `NULL` and
the `err` field in the context will be set (see section on **Errors**).
Once an error is returned the context cannot be reused and you should set up
a new connection.
The standard replies that `redisCommand` are of the type `redisReply`. The
`type` field in the `redisReply` should be used to test what kind of reply
was received:
* **`REDIS_REPLY_STATUS`**:
* The command replied with a status reply. The status string can be accessed using `reply->str`.
The length of this string can be accessed using `reply->len`.
* **`REDIS_REPLY_ERROR`**:
* The command replied with an error. The error string can be accessed identical to `REDIS_REPLY_STATUS`.
* **`REDIS_REPLY_INTEGER`**:
* The command replied with an integer. The integer value can be accessed using the
`reply->integer` field of type `long long`.
* **`REDIS_REPLY_NIL`**:
* The command replied with a **nil** object. There is no data to access.
* **`REDIS_REPLY_STRING`**:
* A bulk (string) reply. The value of the reply can be accessed using `reply->str`.
The length of this string can be accessed using `reply->len`.
* **`REDIS_REPLY_ARRAY`**:
* A multi bulk reply. The number of elements in the multi bulk reply is stored in
`reply->elements`. Every element in the multi bulk reply is a `redisReply` object as well
and can be accessed via `reply->element[..index..]`.
Redis may reply with nested arrays but this is fully supported.
Replies should be freed using the `freeReplyObject()` function.
Note that this function will take care of freeing sub-replies objects
contained in arrays and nested arrays, so there is no need for the user to
free the sub replies (it is actually harmful and will corrupt the memory).
**Important:** the current version of hiredis (0.10.0) free's replies when the
asynchronous API is used. This means you should not call `freeReplyObject` when
you use this API. The reply is cleaned up by hiredis _after_ the callback
returns. This behavior will probably change in future releases, so make sure to
keep an eye on the changelog when upgrading (see issue #39).
### Cleaning up
To disconnect and free the context the following function can be used:
void redisFree(redisContext *c);
This function immediately closes the socket and then free's the allocations done in
creating the context.
### Sending commands (cont'd)
Together with `redisCommand`, the function `redisCommandArgv` can be used to issue commands.
It has the following prototype:
void *redisCommandArgv(redisContext *c, int argc, const char **argv, const size_t *argvlen);
It takes the number of arguments `argc`, an array of strings `argv` and the lengths of the
arguments `argvlen`. For convenience, `argvlen` may be set to `NULL` and the function will
use `strlen(3)` on every argument to determine its length. Obviously, when any of the arguments
need to be binary safe, the entire array of lengths `argvlen` should be provided.
The return value has the same semantic as `redisCommand`.
### Pipelining
To explain how Hiredis supports pipelining in a blocking connection, there needs to be
understanding of the internal execution flow.
When any of the functions in the `redisCommand` family is called, Hiredis first formats the
command according to the Redis protocol. The formatted command is then put in the output buffer
of the context. This output buffer is dynamic, so it can hold any number of commands.
After the command is put in the output buffer, `redisGetReply` is called. This function has the
following two execution paths:
1. The input buffer is non-empty:
* Try to parse a single reply from the input buffer and return it
* If no reply could be parsed, continue at *2*
2. The input buffer is empty:
* Write the **entire** output buffer to the socket
* Read from the socket until a single reply could be parsed
The function `redisGetReply` is exported as part of the Hiredis API and can be used when a reply
is expected on the socket. To pipeline commands, the only things that needs to be done is
filling up the output buffer. For this cause, two commands can be used that are identical
to the `redisCommand` family, apart from not returning a reply:
void redisAppendCommand(redisContext *c, const char *format, ...);
void redisAppendCommandArgv(redisContext *c, int argc, const char **argv, const size_t *argvlen);
After calling either function one or more times, `redisGetReply` can be used to receive the
subsequent replies. The return value for this function is either `REDIS_OK` or `REDIS_ERR`, where
the latter means an error occurred while reading a reply. Just as with the other commands,
the `err` field in the context can be used to find out what the cause of this error is.
The following examples shows a simple pipeline (resulting in only a single call to `write(2)` and
a single call to `read(2)`):
redisReply *reply;
redisAppendCommand(context,"SET foo bar");
redisAppendCommand(context,"GET foo");
redisGetReply(context,&reply); // reply for SET
freeReplyObject(reply);
redisGetReply(context,&reply); // reply for GET
freeReplyObject(reply);
This API can also be used to implement a blocking subscriber:
reply = redisCommand(context,"SUBSCRIBE foo");
freeReplyObject(reply);
while(redisGetReply(context,&reply) == REDIS_OK) {
// consume message
freeReplyObject(reply);
}
### Errors
When a function call is not successful, depending on the function either `NULL` or `REDIS_ERR` is
returned. The `err` field inside the context will be non-zero and set to one of the
following constants:
* **`REDIS_ERR_IO`**:
There was an I/O error while creating the connection, trying to write
to the socket or read from the socket. If you included `errno.h` in your
application, you can use the global `errno` variable to find out what is
wrong.
* **`REDIS_ERR_EOF`**:
The server closed the connection which resulted in an empty read.
* **`REDIS_ERR_PROTOCOL`**:
There was an error while parsing the protocol.
* **`REDIS_ERR_OTHER`**:
Any other error. Currently, it is only used when a specified hostname to connect
to cannot be resolved.
In every case, the `errstr` field in the context will be set to hold a string representation
of the error.
## Asynchronous API
Hiredis comes with an asynchronous API that works easily with any event library.
Examples are bundled that show using Hiredis with [libev](http://software.schmorp.de/pkg/libev.html)
and [libevent](http://monkey.org/~provos/libevent/).
### Connecting
The function `redisAsyncConnect` can be used to establish a non-blocking connection to
Redis. It returns a pointer to the newly created `redisAsyncContext` struct. The `err` field
should be checked after creation to see if there were errors creating the connection.
Because the connection that will be created is non-blocking, the kernel is not able to
instantly return if the specified host and port is able to accept a connection.
redisAsyncContext *c = redisAsyncConnect("127.0.0.1", 6379);
if (c->err) {
printf("Error: %s\n", c->errstr);
// handle error
}
The asynchronous context can hold a disconnect callback function that is called when the
connection is disconnected (either because of an error or per user request). This function should
have the following prototype:
void(const redisAsyncContext *c, int status);
On a disconnect, the `status` argument is set to `REDIS_OK` when disconnection was initiated by the
user, or `REDIS_ERR` when the disconnection was caused by an error. When it is `REDIS_ERR`, the `err`
field in the context can be accessed to find out the cause of the error.
The context object is always free'd after the disconnect callback fired. When a reconnect is needed,
the disconnect callback is a good point to do so.
Setting the disconnect callback can only be done once per context. For subsequent calls it will
return `REDIS_ERR`. The function to set the disconnect callback has the following prototype:
int redisAsyncSetDisconnectCallback(redisAsyncContext *ac, redisDisconnectCallback *fn);
### Sending commands and their callbacks
In an asynchronous context, commands are automatically pipelined due to the nature of an event loop.
Therefore, unlike the synchronous API, there is only a single way to send commands.
Because commands are sent to Redis asynchronously, issuing a command requires a callback function
that is called when the reply is received. Reply callbacks should have the following prototype:
void(redisAsyncContext *c, void *reply, void *privdata);
The `privdata` argument can be used to curry arbitrary data to the callback from the point where
the command is initially queued for execution.
The functions that can be used to issue commands in an asynchronous context are:
int redisAsyncCommand(
redisAsyncContext *ac, redisCallbackFn *fn, void *privdata,
const char *format, ...);
int redisAsyncCommandArgv(
redisAsyncContext *ac, redisCallbackFn *fn, void *privdata,
int argc, const char **argv, const size_t *argvlen);
Both functions work like their blocking counterparts. The return value is `REDIS_OK` when the command
was successfully added to the output buffer and `REDIS_ERR` otherwise. Example: when the connection
is being disconnected per user-request, no new commands may be added to the output buffer and `REDIS_ERR` is
returned on calls to the `redisAsyncCommand` family.
If the reply for a command with a `NULL` callback is read, it is immediately free'd. When the callback
for a command is non-`NULL`, the memory is free'd immediately following the callback: the reply is only
valid for the duration of the callback.
All pending callbacks are called with a `NULL` reply when the context encountered an error.
### Disconnecting
An asynchronous connection can be terminated using:
void redisAsyncDisconnect(redisAsyncContext *ac);
When this function is called, the connection is **not** immediately terminated. Instead, new
commands are no longer accepted and the connection is only terminated when all pending commands
have been written to the socket, their respective replies have been read and their respective
callbacks have been executed. After this, the disconnection callback is executed with the
`REDIS_OK` status and the context object is free'd.
### Hooking it up to event library *X*
There are a few hooks that need to be set on the context object after it is created.
See the `adapters/` directory for bindings to *libev* and *libevent*.
## Reply parsing API
Hiredis comes with a reply parsing API that makes it easy for writing higher
level language bindings.
The reply parsing API consists of the following functions:
redisReader *redisReaderCreate(void);
void redisReaderFree(redisReader *reader);
int redisReaderFeed(redisReader *reader, const char *buf, size_t len);
int redisReaderGetReply(redisReader *reader, void **reply);
The same set of functions are used internally by hiredis when creating a
normal Redis context, the above API just exposes it to the user for a direct
usage.
### Usage
The function `redisReaderCreate` creates a `redisReader` structure that holds a
buffer with unparsed data and state for the protocol parser.
Incoming data -- most likely from a socket -- can be placed in the internal
buffer of the `redisReader` using `redisReaderFeed`. This function will make a
copy of the buffer pointed to by `buf` for `len` bytes. This data is parsed
when `redisReaderGetReply` is called. This function returns an integer status
and a reply object (as described above) via `void **reply`. The returned status
can be either `REDIS_OK` or `REDIS_ERR`, where the latter means something went
wrong (either a protocol error, or an out of memory error).
### Customizing replies
The function `redisReaderGetReply` creates `redisReply` and makes the function
argument `reply` point to the created `redisReply` variable. For instance, if
the response of type `REDIS_REPLY_STATUS` then the `str` field of `redisReply`
will hold the status as a vanilla C string. However, the functions that are
responsible for creating instances of the `redisReply` can be customized by
setting the `fn` field on the `redisReader` struct. This should be done
immediately after creating the `redisReader`.
For example, [hiredis-rb](https://github.com/pietern/hiredis-rb/blob/master/ext/hiredis_ext/reader.c)
uses customized reply object functions to create Ruby objects.
### Reader max buffer
Both when using the Reader API directly or when using it indirectly via a
normal Redis context, the redisReader structure uses a buffer in order to
accumulate data from the server.
Usually this buffer is destroyed when it is empty and is larger than 16
kb in order to avoid wasting memory in unused buffers
However when working with very big payloads destroying the buffer may slow
down performances considerably, so it is possible to modify the max size of
an idle buffer changing the value of the `maxbuf` field of the reader structure
to the desired value. The special value of 0 means that there is no maximum
value for an idle buffer, so the buffer will never get freed.
For instance if you have a normal Redis context you can set the maximum idle
buffer to zero (unlimited) just with:
context->reader->maxbuf = 0;
This should be done only in order to maximize performances when working with
large payloads. The context should be set back to `REDIS_READER_MAX_BUF` again
as soon as possible in order to prevent allocation of useless memory.
## AUTHORS
Hiredis was written by Salvatore Sanfilippo (antirez at gmail) and
Pieter Noordhuis (pcnoordhuis at gmail) and is released under the BSD license.
deps/hiredis/adapters/ae.h
Vendored
+97
View File
@@ -0,0 +1,97 @@
#ifndef __HIREDIS_AE_H__
#define __HIREDIS_AE_H__
#include <sys/types.h>
#include <ae.h>
#include "../hiredis.h"
#include "../async.h"
typedef struct redisAeEvents {
redisAsyncContext *context;
aeEventLoop *loop;
int fd;
int reading, writing;
} redisAeEvents;
static void redisAeReadEvent(aeEventLoop *el, int fd, void *privdata, int mask) {
((void)el); ((void)fd); ((void)mask);
redisAeEvents *e = (redisAeEvents*)privdata;
redisAsyncHandleRead(e->context);
}
static void redisAeWriteEvent(aeEventLoop *el, int fd, void *privdata, int mask) {
((void)el); ((void)fd); ((void)mask);
redisAeEvents *e = (redisAeEvents*)privdata;
redisAsyncHandleWrite(e->context);
}
static void redisAeAddRead(void *privdata) {
redisAeEvents *e = (redisAeEvents*)privdata;
aeEventLoop *loop = e->loop;
if (!e->reading) {
e->reading = 1;
aeCreateFileEvent(loop,e->fd,AE_READABLE,redisAeReadEvent,e);
}
}
static void redisAeDelRead(void *privdata) {
redisAeEvents *e = (redisAeEvents*)privdata;
aeEventLoop *loop = e->loop;
if (e->reading) {
e->reading = 0;
aeDeleteFileEvent(loop,e->fd,AE_READABLE);
}
}
static void redisAeAddWrite(void *privdata) {
redisAeEvents *e = (redisAeEvents*)privdata;
aeEventLoop *loop = e->loop;
if (!e->writing) {
e->writing = 1;
aeCreateFileEvent(loop,e->fd,AE_WRITABLE,redisAeWriteEvent,e);
}
}
static void redisAeDelWrite(void *privdata) {
redisAeEvents *e = (redisAeEvents*)privdata;
aeEventLoop *loop = e->loop;
if (e->writing) {
e->writing = 0;
aeDeleteFileEvent(loop,e->fd,AE_WRITABLE);
}
}
static void redisAeCleanup(void *privdata) {
redisAeEvents *e = (redisAeEvents*)privdata;
redisAeDelRead(privdata);
redisAeDelWrite(privdata);
free(e);
}
static int redisAeAttach(aeEventLoop *loop, redisAsyncContext *ac) {
redisContext *c = &(ac->c);
redisAeEvents *e;
/* Nothing should be attached when something is already attached */
if (ac->ev.data != NULL)
return REDIS_ERR;
/* Create container for context and r/w events */
e = (redisAeEvents*)malloc(sizeof(*e));
e->context = ac;
e->loop = loop;
e->fd = c->fd;
e->reading = e->writing = 0;
/* Register functions to start/stop listening for events */
ac->ev.addRead = redisAeAddRead;
ac->ev.delRead = redisAeDelRead;
ac->ev.addWrite = redisAeAddWrite;
ac->ev.delWrite = redisAeDelWrite;
ac->ev.cleanup = redisAeCleanup;
ac->ev.data = e;
return REDIS_OK;
}
#endif
deps/hiredis/adapters/libev.h
Vendored
+117
View File
@@ -0,0 +1,117 @@
#ifndef __HIREDIS_LIBEV_H__
#define __HIREDIS_LIBEV_H__
#include <stdlib.h>
#include <sys/types.h>
#include <ev.h>
#include "../hiredis.h"
#include "../async.h"
typedef struct redisLibevEvents {
redisAsyncContext *context;
struct ev_loop *loop;
int reading, writing;
ev_io rev, wev;
} redisLibevEvents;
static void redisLibevReadEvent(EV_P_ ev_io *watcher, int revents) {
#if EV_MULTIPLICITY
((void)loop);
#endif
((void)revents);
redisLibevEvents *e = (redisLibevEvents*)watcher->data;
redisAsyncHandleRead(e->context);
}
static void redisLibevWriteEvent(EV_P_ ev_io *watcher, int revents) {
#if EV_MULTIPLICITY
((void)loop);
#endif
((void)revents);
redisLibevEvents *e = (redisLibevEvents*)watcher->data;
redisAsyncHandleWrite(e->context);
}
static void redisLibevAddRead(void *privdata) {
redisLibevEvents *e = (redisLibevEvents*)privdata;
struct ev_loop *loop = e->loop;
((void)loop);
if (!e->reading) {
e->reading = 1;
ev_io_start(EV_A_ &e->rev);
}
}
static void redisLibevDelRead(void *privdata) {
redisLibevEvents *e = (redisLibevEvents*)privdata;
struct ev_loop *loop = e->loop;
((void)loop);
if (e->reading) {
e->reading = 0;
ev_io_stop(EV_A_ &e->rev);
}
}
static void redisLibevAddWrite(void *privdata) {
redisLibevEvents *e = (redisLibevEvents*)privdata;
struct ev_loop *loop = e->loop;
((void)loop);
if (!e->writing) {
e->writing = 1;
ev_io_start(EV_A_ &e->wev);
}
}
static void redisLibevDelWrite(void *privdata) {
redisLibevEvents *e = (redisLibevEvents*)privdata;
struct ev_loop *loop = e->loop;
((void)loop);
if (e->writing) {
e->writing = 0;
ev_io_stop(EV_A_ &e->wev);
}
}
static void redisLibevCleanup(void *privdata) {
redisLibevEvents *e = (redisLibevEvents*)privdata;
redisLibevDelRead(privdata);
redisLibevDelWrite(privdata);
free(e);
}
static int redisLibevAttach(EV_P_ redisAsyncContext *ac) {
redisContext *c = &(ac->c);
redisLibevEvents *e;
/* Nothing should be attached when something is already attached */
if (ac->ev.data != NULL)
return REDIS_ERR;
/* Create container for context and r/w events */
e = (redisLibevEvents*)malloc(sizeof(*e));
e->context = ac;
#if EV_MULTIPLICITY
e->loop = loop;
#else
e->loop = NULL;
#endif
e->reading = e->writing = 0;
e->rev.data = e;
e->wev.data = e;
/* Register functions to start/stop listening for events */
ac->ev.addRead = redisLibevAddRead;
ac->ev.delRead = redisLibevDelRead;
ac->ev.addWrite = redisLibevAddWrite;
ac->ev.delWrite = redisLibevDelWrite;
ac->ev.cleanup = redisLibevCleanup;
ac->ev.data = e;
/* Initialize read/write events */
ev_io_init(&e->rev,redisLibevReadEvent,c->fd,EV_READ);
ev_io_init(&e->wev,redisLibevWriteEvent,c->fd,EV_WRITE);
return REDIS_OK;
}
#endif
deps/hiredis/adapters/libevent.h
Vendored
+78
View File
@@ -0,0 +1,78 @@
#ifndef __HIREDIS_LIBEVENT_H__
#define __HIREDIS_LIBEVENT_H__
#include <event.h>
#include "../hiredis.h"
#include "../async.h"
typedef struct redisLibeventEvents {
redisAsyncContext *context;
struct event rev, wev;
} redisLibeventEvents;
static void redisLibeventReadEvent(int fd, short event, void *arg) {
((void)fd); ((void)event);
redisLibeventEvents *e = (redisLibeventEvents*)arg;
redisAsyncHandleRead(e->context);
}
static void redisLibeventWriteEvent(int fd, short event, void *arg) {
((void)fd); ((void)event);
redisLibeventEvents *e = (redisLibeventEvents*)arg;
redisAsyncHandleWrite(e->context);
}
static void redisLibeventAddRead(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_add(&e->rev,NULL);
}
static void redisLibeventDelRead(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_del(&e->rev);
}
static void redisLibeventAddWrite(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_add(&e->wev,NULL);
}
static void redisLibeventDelWrite(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_del(&e->wev);
}
static void redisLibeventCleanup(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_del(&e->rev);
event_del(&e->wev);
free(e);
}
static int redisLibeventAttach(redisAsyncContext *ac, struct event_base *base) {
redisContext *c = &(ac->c);
redisLibeventEvents *e;
/* Nothing should be attached when something is already attached */
if (ac->ev.data != NULL)
return REDIS_ERR;
/* Create container for context and r/w events */
e = (redisLibeventEvents*)malloc(sizeof(*e));
e->context = ac;
/* Register functions to start/stop listening for events */
ac->ev.addRead = redisLibeventAddRead;
ac->ev.delRead = redisLibeventDelRead;
ac->ev.addWrite = redisLibeventAddWrite;
ac->ev.delWrite = redisLibeventDelWrite;
ac->ev.cleanup = redisLibeventCleanup;
ac->ev.data = e;
/* Initialize and install read/write events */
event_set(&e->rev,c->fd,EV_READ,redisLibeventReadEvent,e);
event_set(&e->wev,c->fd,EV_WRITE,redisLibeventWriteEvent,e);
event_base_set(base,&e->rev);
event_base_set(base,&e->wev);
return REDIS_OK;
}
#endif
deps/hiredis/async.c
Vendored
+622
View File
@@ -0,0 +1,622 @@
/*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "fmacros.h"
#include <stdlib.h>
#include <string.h>
#include <strings.h>
#include <assert.h>
#include <ctype.h>
#include <errno.h>
#include "async.h"
#include "net.h"
#include "dict.c"
#include "sds.h"
#define _EL_ADD_READ(ctx) do { \
if ((ctx)->ev.addRead) (ctx)->ev.addRead((ctx)->ev.data); \
} while(0)
#define _EL_DEL_READ(ctx) do { \
if ((ctx)->ev.delRead) (ctx)->ev.delRead((ctx)->ev.data); \
} while(0)
#define _EL_ADD_WRITE(ctx) do { \
if ((ctx)->ev.addWrite) (ctx)->ev.addWrite((ctx)->ev.data); \
} while(0)
#define _EL_DEL_WRITE(ctx) do { \
if ((ctx)->ev.delWrite) (ctx)->ev.delWrite((ctx)->ev.data); \
} while(0)
#define _EL_CLEANUP(ctx) do { \
if ((ctx)->ev.cleanup) (ctx)->ev.cleanup((ctx)->ev.data); \
} while(0);
/* Forward declaration of function in hiredis.c */
void __redisAppendCommand(redisContext *c, char *cmd, size_t len);
/* Functions managing dictionary of callbacks for pub/sub. */
static unsigned int callbackHash(const void *key) {
return dictGenHashFunction((unsigned char*)key,sdslen((char*)key));
}
static void *callbackValDup(void *privdata, const void *src) {
((void) privdata);
redisCallback *dup = malloc(sizeof(*dup));
memcpy(dup,src,sizeof(*dup));
return dup;
}
static int callbackKeyCompare(void *privdata, const void *key1, const void *key2) {
int l1, l2;
((void) privdata);
l1 = sdslen((sds)key1);
l2 = sdslen((sds)key2);
if (l1 != l2) return 0;
return memcmp(key1,key2,l1) == 0;
}
static void callbackKeyDestructor(void *privdata, void *key) {
((void) privdata);
sdsfree((sds)key);
}
static void callbackValDestructor(void *privdata, void *val) {
((void) privdata);
free(val);
}
static dictType callbackDict = {
callbackHash,
NULL,
callbackValDup,
callbackKeyCompare,
callbackKeyDestructor,
callbackValDestructor
};
static redisAsyncContext *redisAsyncInitialize(redisContext *c) {
redisAsyncContext *ac = realloc(c,sizeof(redisAsyncContext));
c = &(ac->c);
/* The regular connect functions will always set the flag REDIS_CONNECTED.
* For the async API, we want to wait until the first write event is
* received up before setting this flag, so reset it here. */
c->flags &= ~REDIS_CONNECTED;
ac->err = 0;
ac->errstr = NULL;
ac->data = NULL;
ac->ev.data = NULL;
ac->ev.addRead = NULL;
ac->ev.delRead = NULL;
ac->ev.addWrite = NULL;
ac->ev.delWrite = NULL;
ac->ev.cleanup = NULL;
ac->onConnect = NULL;
ac->onDisconnect = NULL;
ac->replies.head = NULL;
ac->replies.tail = NULL;
ac->sub.invalid.head = NULL;
ac->sub.invalid.tail = NULL;
ac->sub.channels = dictCreate(&callbackDict,NULL);
ac->sub.patterns = dictCreate(&callbackDict,NULL);
return ac;
}
/* We want the error field to be accessible directly instead of requiring
* an indirection to the redisContext struct. */
static void __redisAsyncCopyError(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
ac->err = c->err;
ac->errstr = c->errstr;
}
redisAsyncContext *redisAsyncConnect(const char *ip, int port) {
redisContext *c = redisConnectNonBlock(ip,port);
redisAsyncContext *ac = redisAsyncInitialize(c);
__redisAsyncCopyError(ac);
return ac;
}
redisAsyncContext *redisAsyncConnectUnix(const char *path) {
redisContext *c = redisConnectUnixNonBlock(path);
redisAsyncContext *ac = redisAsyncInitialize(c);
__redisAsyncCopyError(ac);
return ac;
}
int redisAsyncSetConnectCallback(redisAsyncContext *ac, redisConnectCallback *fn) {
if (ac->onConnect == NULL) {
ac->onConnect = fn;
/* The common way to detect an established connection is to wait for
* the first write event to be fired. This assumes the related event
* library functions are already set. */
_EL_ADD_WRITE(ac);
return REDIS_OK;
}
return REDIS_ERR;
}
int redisAsyncSetDisconnectCallback(redisAsyncContext *ac, redisDisconnectCallback *fn) {
if (ac->onDisconnect == NULL) {
ac->onDisconnect = fn;
return REDIS_OK;
}
return REDIS_ERR;
}
/* Helper functions to push/shift callbacks */
static int __redisPushCallback(redisCallbackList *list, redisCallback *source) {
redisCallback *cb;
/* Copy callback from stack to heap */
cb = malloc(sizeof(*cb));
if (source != NULL) {
memcpy(cb,source,sizeof(*cb));
cb->next = NULL;
}
/* Store callback in list */
if (list->head == NULL)
list->head = cb;
if (list->tail != NULL)
list->tail->next = cb;
list->tail = cb;
return REDIS_OK;
}
static int __redisShiftCallback(redisCallbackList *list, redisCallback *target) {
redisCallback *cb = list->head;
if (cb != NULL) {
list->head = cb->next;
if (cb == list->tail)
list->tail = NULL;
/* Copy callback from heap to stack */
if (target != NULL)
memcpy(target,cb,sizeof(*cb));
free(cb);
return REDIS_OK;
}
return REDIS_ERR;
}
static void __redisRunCallback(redisAsyncContext *ac, redisCallback *cb, redisReply *reply) {
redisContext *c = &(ac->c);
if (cb->fn != NULL) {
c->flags |= REDIS_IN_CALLBACK;
cb->fn(ac,reply,cb->privdata);
c->flags &= ~REDIS_IN_CALLBACK;
}
}
/* Helper function to free the context. */
static void __redisAsyncFree(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
redisCallback cb;
dictIterator *it;
dictEntry *de;
/* Execute pending callbacks with NULL reply. */
while (__redisShiftCallback(&ac->replies,&cb) == REDIS_OK)
__redisRunCallback(ac,&cb,NULL);
/* Execute callbacks for invalid commands */
while (__redisShiftCallback(&ac->sub.invalid,&cb) == REDIS_OK)
__redisRunCallback(ac,&cb,NULL);
/* Run subscription callbacks callbacks with NULL reply */
it = dictGetIterator(ac->sub.channels);
while ((de = dictNext(it)) != NULL)
__redisRunCallback(ac,dictGetEntryVal(de),NULL);
dictReleaseIterator(it);
dictRelease(ac->sub.channels);
it = dictGetIterator(ac->sub.patterns);
while ((de = dictNext(it)) != NULL)
__redisRunCallback(ac,dictGetEntryVal(de),NULL);
dictReleaseIterator(it);
dictRelease(ac->sub.patterns);
/* Signal event lib to clean up */
_EL_CLEANUP(ac);
/* Execute disconnect callback. When redisAsyncFree() initiated destroying
* this context, the status will always be REDIS_OK. */
if (ac->onDisconnect && (c->flags & REDIS_CONNECTED)) {
if (c->flags & REDIS_FREEING) {
ac->onDisconnect(ac,REDIS_OK);
} else {
ac->onDisconnect(ac,(ac->err == 0) ? REDIS_OK : REDIS_ERR);
}
}
/* Cleanup self */
redisFree(c);
}
/* Free the async context. When this function is called from a callback,
* control needs to be returned to redisProcessCallbacks() before actual
* free'ing. To do so, a flag is set on the context which is picked up by
* redisProcessCallbacks(). Otherwise, the context is immediately free'd. */
void redisAsyncFree(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
c->flags |= REDIS_FREEING;
if (!(c->flags & REDIS_IN_CALLBACK))
__redisAsyncFree(ac);
}
/* Helper function to make the disconnect happen and clean up. */
static void __redisAsyncDisconnect(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
/* Make sure error is accessible if there is any */
__redisAsyncCopyError(ac);
if (ac->err == 0) {
/* For clean disconnects, there should be no pending callbacks. */
assert(__redisShiftCallback(&ac->replies,NULL) == REDIS_ERR);
} else {
/* Disconnection is caused by an error, make sure that pending
* callbacks cannot call new commands. */
c->flags |= REDIS_DISCONNECTING;
}
/* For non-clean disconnects, __redisAsyncFree() will execute pending
* callbacks with a NULL-reply. */
__redisAsyncFree(ac);
}
/* Tries to do a clean disconnect from Redis, meaning it stops new commands
* from being issued, but tries to flush the output buffer and execute
* callbacks for all remaining replies. When this function is called from a
* callback, there might be more replies and we can safely defer disconnecting
* to redisProcessCallbacks(). Otherwise, we can only disconnect immediately
* when there are no pending callbacks. */
void redisAsyncDisconnect(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
c->flags |= REDIS_DISCONNECTING;
if (!(c->flags & REDIS_IN_CALLBACK) && ac->replies.head == NULL)
__redisAsyncDisconnect(ac);
}
static int __redisGetSubscribeCallback(redisAsyncContext *ac, redisReply *reply, redisCallback *dstcb) {
redisContext *c = &(ac->c);
dict *callbacks;
dictEntry *de;
int pvariant;
char *stype;
sds sname;
/* Custom reply functions are not supported for pub/sub. This will fail
* very hard when they are used... */
if (reply->type == REDIS_REPLY_ARRAY) {
assert(reply->elements >= 2);
assert(reply->element[0]->type == REDIS_REPLY_STRING);
stype = reply->element[0]->str;
pvariant = (tolower(stype[0]) == 'p') ? 1 : 0;
if (pvariant)
callbacks = ac->sub.patterns;
else
callbacks = ac->sub.channels;
/* Locate the right callback */
assert(reply->element[1]->type == REDIS_REPLY_STRING);
sname = sdsnewlen(reply->element[1]->str,reply->element[1]->len);
de = dictFind(callbacks,sname);
if (de != NULL) {
memcpy(dstcb,dictGetEntryVal(de),sizeof(*dstcb));
/* If this is an unsubscribe message, remove it. */
if (strcasecmp(stype+pvariant,"unsubscribe") == 0) {
dictDelete(callbacks,sname);
/* If this was the last unsubscribe message, revert to
* non-subscribe mode. */
assert(reply->element[2]->type == REDIS_REPLY_INTEGER);
if (reply->element[2]->integer == 0)
c->flags &= ~REDIS_SUBSCRIBED;
}
}
sdsfree(sname);
} else {
/* Shift callback for invalid commands. */
__redisShiftCallback(&ac->sub.invalid,dstcb);
}
return REDIS_OK;
}
void redisProcessCallbacks(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
redisCallback cb;
void *reply = NULL;
int status;
while((status = redisGetReply(c,&reply)) == REDIS_OK) {
if (reply == NULL) {
/* When the connection is being disconnected and there are
* no more replies, this is the cue to really disconnect. */
if (c->flags & REDIS_DISCONNECTING && sdslen(c->obuf) == 0) {
__redisAsyncDisconnect(ac);
return;
}
/* If monitor mode, repush callback */
if(c->flags & REDIS_MONITORING) {
__redisPushCallback(&ac->replies,&cb);
}
/* When the connection is not being disconnected, simply stop
* trying to get replies and wait for the next loop tick. */
break;
}
/* Even if the context is subscribed, pending regular callbacks will
* get a reply before pub/sub messages arrive. */
if (__redisShiftCallback(&ac->replies,&cb) != REDIS_OK) {
/*
* A spontaneous reply in a not-subscribed context can be the error
* reply that is sent when a new connection exceeds the maximum
* number of allowed connections on the server side.
*
* This is seen as an error instead of a regular reply because the
* server closes the connection after sending it.
*
* To prevent the error from being overwritten by an EOF error the
* connection is closed here. See issue #43.
*
* Another possibility is that the server is loading its dataset.
* In this case we also want to close the connection, and have the
* user wait until the server is ready to take our request.
*/
if (((redisReply*)reply)->type == REDIS_REPLY_ERROR) {
c->err = REDIS_ERR_OTHER;
snprintf(c->errstr,sizeof(c->errstr),"%s",((redisReply*)reply)->str);
__redisAsyncDisconnect(ac);
return;
}
/* No more regular callbacks and no errors, the context *must* be subscribed or monitoring. */
assert((c->flags & REDIS_SUBSCRIBED || c->flags & REDIS_MONITORING));
if(c->flags & REDIS_SUBSCRIBED)
__redisGetSubscribeCallback(ac,reply,&cb);
}
if (cb.fn != NULL) {
__redisRunCallback(ac,&cb,reply);
c->reader->fn->freeObject(reply);
/* Proceed with free'ing when redisAsyncFree() was called. */
if (c->flags & REDIS_FREEING) {
__redisAsyncFree(ac);
return;
}
} else {
/* No callback for this reply. This can either be a NULL callback,
* or there were no callbacks to begin with. Either way, don't
* abort with an error, but simply ignore it because the client
* doesn't know what the server will spit out over the wire. */
c->reader->fn->freeObject(reply);
}
}
/* Disconnect when there was an error reading the reply */
if (status != REDIS_OK)
__redisAsyncDisconnect(ac);
}
/* Internal helper function to detect socket status the first time a read or
* write event fires. When connecting was not succesful, the connect callback
* is called with a REDIS_ERR status and the context is free'd. */
static int __redisAsyncHandleConnect(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
if (redisCheckSocketError(c,c->fd) == REDIS_ERR) {
/* Try again later when connect(2) is still in progress. */
if (errno == EINPROGRESS)
return REDIS_OK;
if (ac->onConnect) ac->onConnect(ac,REDIS_ERR);
__redisAsyncDisconnect(ac);
return REDIS_ERR;
}
/* Mark context as connected. */
c->flags |= REDIS_CONNECTED;
if (ac->onConnect) ac->onConnect(ac,REDIS_OK);
return REDIS_OK;
}
/* This function should be called when the socket is readable.
* It processes all replies that can be read and executes their callbacks.
*/
void redisAsyncHandleRead(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
if (!(c->flags & REDIS_CONNECTED)) {
/* Abort connect was not successful. */
if (__redisAsyncHandleConnect(ac) != REDIS_OK)
return;
/* Try again later when the context is still not connected. */
if (!(c->flags & REDIS_CONNECTED))
return;
}
if (redisBufferRead(c) == REDIS_ERR) {
__redisAsyncDisconnect(ac);
} else {
/* Always re-schedule reads */
_EL_ADD_READ(ac);
redisProcessCallbacks(ac);
}
}
void redisAsyncHandleWrite(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
int done = 0;
if (!(c->flags & REDIS_CONNECTED)) {
/* Abort connect was not successful. */
if (__redisAsyncHandleConnect(ac) != REDIS_OK)
return;
/* Try again later when the context is still not connected. */
if (!(c->flags & REDIS_CONNECTED))
return;
}
if (redisBufferWrite(c,&done) == REDIS_ERR) {
__redisAsyncDisconnect(ac);
} else {
/* Continue writing when not done, stop writing otherwise */
if (!done)
_EL_ADD_WRITE(ac);
else
_EL_DEL_WRITE(ac);
/* Always schedule reads after writes */
_EL_ADD_READ(ac);
}
}
/* Sets a pointer to the first argument and its length starting at p. Returns
* the number of bytes to skip to get to the following argument. */
static char *nextArgument(char *start, char **str, size_t *len) {
char *p = start;
if (p[0] != '$') {
p = strchr(p,'$');
if (p == NULL) return NULL;
}
*len = (int)strtol(p+1,NULL,10);
p = strchr(p,'\r');
assert(p);
*str = p+2;
return p+2+(*len)+2;
}
/* Helper function for the redisAsyncCommand* family of functions. Writes a
* formatted command to the output buffer and registers the provided callback
* function with the context. */
static int __redisAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, char *cmd, size_t len) {
redisContext *c = &(ac->c);
redisCallback cb;
int pvariant, hasnext;
char *cstr, *astr;
size_t clen, alen;
char *p;
sds sname;
/* Don't accept new commands when the connection is about to be closed. */
if (c->flags & (REDIS_DISCONNECTING | REDIS_FREEING)) return REDIS_ERR;
/* Setup callback */
cb.fn = fn;
cb.privdata = privdata;
/* Find out which command will be appended. */
p = nextArgument(cmd,&cstr,&clen);
assert(p != NULL);
hasnext = (p[0] == '$');
pvariant = (tolower(cstr[0]) == 'p') ? 1 : 0;
cstr += pvariant;
clen -= pvariant;
if (hasnext && strncasecmp(cstr,"subscribe\r\n",11) == 0) {
c->flags |= REDIS_SUBSCRIBED;
/* Add every channel/pattern to the list of subscription callbacks. */
while ((p = nextArgument(p,&astr,&alen)) != NULL) {
sname = sdsnewlen(astr,alen);
if (pvariant)
dictReplace(ac->sub.patterns,sname,&cb);
else
dictReplace(ac->sub.channels,sname,&cb);
}
} else if (strncasecmp(cstr,"unsubscribe\r\n",13) == 0) {
/* It is only useful to call (P)UNSUBSCRIBE when the context is
* subscribed to one or more channels or patterns. */
if (!(c->flags & REDIS_SUBSCRIBED)) return REDIS_ERR;
/* (P)UNSUBSCRIBE does not have its own response: every channel or
* pattern that is unsubscribed will receive a message. This means we
* should not append a callback function for this command. */
} else if(strncasecmp(cstr,"monitor\r\n",9) == 0) {
/* Set monitor flag and push callback */
c->flags |= REDIS_MONITORING;
__redisPushCallback(&ac->replies,&cb);
} else {
if (c->flags & REDIS_SUBSCRIBED)
/* This will likely result in an error reply, but it needs to be
* received and passed to the callback. */
__redisPushCallback(&ac->sub.invalid,&cb);
else
__redisPushCallback(&ac->replies,&cb);
}
__redisAppendCommand(c,cmd,len);
/* Always schedule a write when the write buffer is non-empty */
_EL_ADD_WRITE(ac);
return REDIS_OK;
}
int redisvAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *format, va_list ap) {
char *cmd;
int len;
int status;
len = redisvFormatCommand(&cmd,format,ap);
status = __redisAsyncCommand(ac,fn,privdata,cmd,len);
free(cmd);
return status;
}
int redisAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *format, ...) {
va_list ap;
int status;
va_start(ap,format);
status = redisvAsyncCommand(ac,fn,privdata,format,ap);
va_end(ap);
return status;
}
int redisAsyncCommandArgv(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, int argc, const char **argv, const size_t *argvlen) {
char *cmd;
int len;
int status;
len = redisFormatCommandArgv(&cmd,argc,argv,argvlen);
status = __redisAsyncCommand(ac,fn,privdata,cmd,len);
free(cmd);
return status;
}
deps/hiredis/async.h
Vendored
+125
View File
@@ -0,0 +1,125 @@
/*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef __HIREDIS_ASYNC_H
#define __HIREDIS_ASYNC_H
#include "hiredis.h"
#ifdef __cplusplus
extern "C" {
#endif
struct redisAsyncContext; /* need forward declaration of redisAsyncContext */
struct dict; /* dictionary header is included in async.c */
/* Reply callback prototype and container */
typedef void (redisCallbackFn)(struct redisAsyncContext*, void*, void*);
typedef struct redisCallback {
struct redisCallback *next; /* simple singly linked list */
redisCallbackFn *fn;
void *privdata;
} redisCallback;
/* List of callbacks for either regular replies or pub/sub */
typedef struct redisCallbackList {
redisCallback *head, *tail;
} redisCallbackList;
/* Connection callback prototypes */
typedef void (redisDisconnectCallback)(const struct redisAsyncContext*, int status);
typedef void (redisConnectCallback)(const struct redisAsyncContext*, int status);
/* Context for an async connection to Redis */
typedef struct redisAsyncContext {
/* Hold the regular context, so it can be realloc'ed. */
redisContext c;
/* Setup error flags so they can be used directly. */
int err;
char *errstr;
/* Not used by hiredis */
void *data;
/* Event library data and hooks */
struct {
void *data;
/* Hooks that are called when the library expects to start
* reading/writing. These functions should be idempotent. */
void (*addRead)(void *privdata);
void (*delRead)(void *privdata);
void (*addWrite)(void *privdata);
void (*delWrite)(void *privdata);
void (*cleanup)(void *privdata);
} ev;
/* Called when either the connection is terminated due to an error or per
* user request. The status is set accordingly (REDIS_OK, REDIS_ERR). */
redisDisconnectCallback *onDisconnect;
/* Called when the first write event was received. */
redisConnectCallback *onConnect;
/* Regular command callbacks */
redisCallbackList replies;
/* Subscription callbacks */
struct {
redisCallbackList invalid;
struct dict *channels;
struct dict *patterns;
} sub;
} redisAsyncContext;
/* Functions that proxy to hiredis */
redisAsyncContext *redisAsyncConnect(const char *ip, int port);
redisAsyncContext *redisAsyncConnectUnix(const char *path);
int redisAsyncSetConnectCallback(redisAsyncContext *ac, redisConnectCallback *fn);
int redisAsyncSetDisconnectCallback(redisAsyncContext *ac, redisDisconnectCallback *fn);
void redisAsyncDisconnect(redisAsyncContext *ac);
void redisAsyncFree(redisAsyncContext *ac);
/* Handle read/write events */
void redisAsyncHandleRead(redisAsyncContext *ac);
void redisAsyncHandleWrite(redisAsyncContext *ac);
/* Command functions for an async context. Write the command to the
* output buffer and register the provided callback. */
int redisvAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *format, va_list ap);
int redisAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *format, ...);
int redisAsyncCommandArgv(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, int argc, const char **argv, const size_t *argvlen);
#ifdef __cplusplus
}
#endif
#endif
deps/hiredis/dict.c
Vendored
+338
View File
@@ -0,0 +1,338 @@
/* Hash table implementation.
*
* This file implements in memory hash tables with insert/del/replace/find/
* get-random-element operations. Hash tables will auto resize if needed
* tables of power of two in size are used, collisions are handled by
* chaining. See the source code for more information... :)
*
* Copyright (c) 2006-2010, Salvatore Sanfilippo <antirez at gmail dot com>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "fmacros.h"
#include <stdlib.h>
#include <assert.h>
#include <limits.h>
#include "dict.h"
/* -------------------------- private prototypes ---------------------------- */
static int _dictExpandIfNeeded(dict *ht);
static unsigned long _dictNextPower(unsigned long size);
static int _dictKeyIndex(dict *ht, const void *key);
static int _dictInit(dict *ht, dictType *type, void *privDataPtr);
/* -------------------------- hash functions -------------------------------- */
/* Generic hash function (a popular one from Bernstein).
* I tested a few and this was the best. */
static unsigned int dictGenHashFunction(const unsigned char *buf, int len) {
unsigned int hash = 5381;
while (len--)
hash = ((hash << 5) + hash) + (*buf++); /* hash * 33 + c */
return hash;
}
/* ----------------------------- API implementation ------------------------- */
/* Reset an hashtable already initialized with ht_init().
* NOTE: This function should only called by ht_destroy(). */
static void _dictReset(dict *ht) {
ht->table = NULL;
ht->size = 0;
ht->sizemask = 0;
ht->used = 0;
}
/* Create a new hash table */
static dict *dictCreate(dictType *type, void *privDataPtr) {
dict *ht = malloc(sizeof(*ht));
_dictInit(ht,type,privDataPtr);
return ht;
}
/* Initialize the hash table */
static int _dictInit(dict *ht, dictType *type, void *privDataPtr) {
_dictReset(ht);
ht->type = type;
ht->privdata = privDataPtr;
return DICT_OK;
}
/* Expand or create the hashtable */
static int dictExpand(dict *ht, unsigned long size) {
dict n; /* the new hashtable */
unsigned long realsize = _dictNextPower(size), i;
/* the size is invalid if it is smaller than the number of
* elements already inside the hashtable */
if (ht->used > size)
return DICT_ERR;
_dictInit(&n, ht->type, ht->privdata);
n.size = realsize;
n.sizemask = realsize-1;
n.table = calloc(realsize,sizeof(dictEntry*));
/* Copy all the elements from the old to the new table:
* note that if the old hash table is empty ht->size is zero,
* so dictExpand just creates an hash table. */
n.used = ht->used;
for (i = 0; i < ht->size && ht->used > 0; i++) {
dictEntry *he, *nextHe;
if (ht->table[i] == NULL) continue;
/* For each hash entry on this slot... */
he = ht->table[i];
while(he) {
unsigned int h;
nextHe = he->next;
/* Get the new element index */
h = dictHashKey(ht, he->key) & n.sizemask;
he->next = n.table[h];
n.table[h] = he;
ht->used--;
/* Pass to the next element */
he = nextHe;
}
}
assert(ht->used == 0);
free(ht->table);
/* Remap the new hashtable in the old */
*ht = n;
return DICT_OK;
}
/* Add an element to the target hash table */
static int dictAdd(dict *ht, void *key, void *val) {
int index;
dictEntry *entry;
/* Get the index of the new element, or -1 if
* the element already exists. */
if ((index = _dictKeyIndex(ht, key)) == -1)
return DICT_ERR;
/* Allocates the memory and stores key */
entry = malloc(sizeof(*entry));
entry->next = ht->table[index];
ht->table[index] = entry;
/* Set the hash entry fields. */
dictSetHashKey(ht, entry, key);
dictSetHashVal(ht, entry, val);
ht->used++;
return DICT_OK;
}
/* Add an element, discarding the old if the key already exists.
* Return 1 if the key was added from scratch, 0 if there was already an
* element with such key and dictReplace() just performed a value update
* operation. */
static int dictReplace(dict *ht, void *key, void *val) {
dictEntry *entry, auxentry;
/* Try to add the element. If the key
* does not exists dictAdd will suceed. */
if (dictAdd(ht, key, val) == DICT_OK)
return 1;
/* It already exists, get the entry */
entry = dictFind(ht, key);
/* Free the old value and set the new one */
/* Set the new value and free the old one. Note that it is important
* to do that in this order, as the value may just be exactly the same
* as the previous one. In this context, think to reference counting,
* you want to increment (set), and then decrement (free), and not the
* reverse. */
auxentry = *entry;
dictSetHashVal(ht, entry, val);
dictFreeEntryVal(ht, &auxentry);
return 0;
}
/* Search and remove an element */
static int dictDelete(dict *ht, const void *key) {
unsigned int h;
dictEntry *de, *prevde;
if (ht->size == 0)
return DICT_ERR;
h = dictHashKey(ht, key) & ht->sizemask;
de = ht->table[h];
prevde = NULL;
while(de) {
if (dictCompareHashKeys(ht,key,de->key)) {
/* Unlink the element from the list */
if (prevde)
prevde->next = de->next;
else
ht->table[h] = de->next;
dictFreeEntryKey(ht,de);
dictFreeEntryVal(ht,de);
free(de);
ht->used--;
return DICT_OK;
}
prevde = de;
de = de->next;
}
return DICT_ERR; /* not found */
}
/* Destroy an entire hash table */
static int _dictClear(dict *ht) {
unsigned long i;
/* Free all the elements */
for (i = 0; i < ht->size && ht->used > 0; i++) {
dictEntry *he, *nextHe;
if ((he = ht->table[i]) == NULL) continue;
while(he) {
nextHe = he->next;
dictFreeEntryKey(ht, he);
dictFreeEntryVal(ht, he);
free(he);
ht->used--;
he = nextHe;
}
}
/* Free the table and the allocated cache structure */
free(ht->table);
/* Re-initialize the table */
_dictReset(ht);
return DICT_OK; /* never fails */
}
/* Clear & Release the hash table */
static void dictRelease(dict *ht) {
_dictClear(ht);
free(ht);
}
static dictEntry *dictFind(dict *ht, const void *key) {
dictEntry *he;
unsigned int h;
if (ht->size == 0) return NULL;
h = dictHashKey(ht, key) & ht->sizemask;
he = ht->table[h];
while(he) {
if (dictCompareHashKeys(ht, key, he->key))
return he;
he = he->next;
}
return NULL;
}
static dictIterator *dictGetIterator(dict *ht) {
dictIterator *iter = malloc(sizeof(*iter));
iter->ht = ht;
iter->index = -1;
iter->entry = NULL;
iter->nextEntry = NULL;
return iter;
}
static dictEntry *dictNext(dictIterator *iter) {
while (1) {
if (iter->entry == NULL) {
iter->index++;
if (iter->index >=
(signed)iter->ht->size) break;
iter->entry = iter->ht->table[iter->index];
} else {
iter->entry = iter->nextEntry;
}
if (iter->entry) {
/* We need to save the 'next' here, the iterator user
* may delete the entry we are returning. */
iter->nextEntry = iter->entry->next;
return iter->entry;
}
}
return NULL;
}
static void dictReleaseIterator(dictIterator *iter) {
free(iter);
}
/* ------------------------- private functions ------------------------------ */
/* Expand the hash table if needed */
static int _dictExpandIfNeeded(dict *ht) {
/* If the hash table is empty expand it to the intial size,
* if the table is "full" dobule its size. */
if (ht->size == 0)
return dictExpand(ht, DICT_HT_INITIAL_SIZE);
if (ht->used == ht->size)
return dictExpand(ht, ht->size*2);
return DICT_OK;
}
/* Our hash table capability is a power of two */
static unsigned long _dictNextPower(unsigned long size) {
unsigned long i = DICT_HT_INITIAL_SIZE;
if (size >= LONG_MAX) return LONG_MAX;
while(1) {
if (i >= size)
return i;
i *= 2;
}
}
/* Returns the index of a free slot that can be populated with
* an hash entry for the given 'key'.
* If the key already exists, -1 is returned. */
static int _dictKeyIndex(dict *ht, const void *key) {
unsigned int h;
dictEntry *he;
/* Expand the hashtable if needed */
if (_dictExpandIfNeeded(ht) == DICT_ERR)
return -1;
/* Compute the key hash value */
h = dictHashKey(ht, key) & ht->sizemask;
/* Search if this slot does not already contain the given key */
he = ht->table[h];
while(he) {
if (dictCompareHashKeys(ht, key, he->key))
return -1;
he = he->next;
}
return h;
}
deps/hiredis/dict.h
Vendored
+126
View File
@@ -0,0 +1,126 @@
/* Hash table implementation.
*
* This file implements in memory hash tables with insert/del/replace/find/
* get-random-element operations. Hash tables will auto resize if needed
* tables of power of two in size are used, collisions are handled by
* chaining. See the source code for more information... :)
*
* Copyright (c) 2006-2010, Salvatore Sanfilippo <antirez at gmail dot com>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef __DICT_H
#define __DICT_H
#define DICT_OK 0
#define DICT_ERR 1
/* Unused arguments generate annoying warnings... */
#define DICT_NOTUSED(V) ((void) V)
typedef struct dictEntry {
void *key;
void *val;
struct dictEntry *next;
} dictEntry;
typedef struct dictType {
unsigned int (*hashFunction)(const void *key);
void *(*keyDup)(void *privdata, const void *key);
void *(*valDup)(void *privdata, const void *obj);
int (*keyCompare)(void *privdata, const void *key1, const void *key2);
void (*keyDestructor)(void *privdata, void *key);
void (*valDestructor)(void *privdata, void *obj);
} dictType;
typedef struct dict {
dictEntry **table;
dictType *type;
unsigned long size;
unsigned long sizemask;
unsigned long used;
void *privdata;
} dict;
typedef struct dictIterator {
dict *ht;
int index;
dictEntry *entry, *nextEntry;
} dictIterator;
/* This is the initial size of every hash table */
#define DICT_HT_INITIAL_SIZE 4
/* ------------------------------- Macros ------------------------------------*/
#define dictFreeEntryVal(ht, entry) \
if ((ht)->type->valDestructor) \
(ht)->type->valDestructor((ht)->privdata, (entry)->val)
#define dictSetHashVal(ht, entry, _val_) do { \
if ((ht)->type->valDup) \
entry->val = (ht)->type->valDup((ht)->privdata, _val_); \
else \
entry->val = (_val_); \
} while(0)
#define dictFreeEntryKey(ht, entry) \
if ((ht)->type->keyDestructor) \
(ht)->type->keyDestructor((ht)->privdata, (entry)->key)
#define dictSetHashKey(ht, entry, _key_) do { \
if ((ht)->type->keyDup) \
entry->key = (ht)->type->keyDup((ht)->privdata, _key_); \
else \
entry->key = (_key_); \
} while(0)
#define dictCompareHashKeys(ht, key1, key2) \
(((ht)->type->keyCompare) ? \
(ht)->type->keyCompare((ht)->privdata, key1, key2) : \
(key1) == (key2))
#define dictHashKey(ht, key) (ht)->type->hashFunction(key)
#define dictGetEntryKey(he) ((he)->key)
#define dictGetEntryVal(he) ((he)->val)
#define dictSlots(ht) ((ht)->size)
#define dictSize(ht) ((ht)->used)
/* API */
static unsigned int dictGenHashFunction(const unsigned char *buf, int len);
static dict *dictCreate(dictType *type, void *privDataPtr);
static int dictExpand(dict *ht, unsigned long size);
static int dictAdd(dict *ht, void *key, void *val);
static int dictReplace(dict *ht, void *key, void *val);
static int dictDelete(dict *ht, const void *key);
static void dictRelease(dict *ht);
static dictEntry * dictFind(dict *ht, const void *key);
static dictIterator *dictGetIterator(dict *ht);
static dictEntry *dictNext(dictIterator *iter);
static void dictReleaseIterator(dictIterator *iter);
#endif /* __DICT_H */
deps/hiredis/example-ae.c
Vendored
+56
View File
@@ -0,0 +1,56 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include "hiredis.h"
#include "async.h"
#include "adapters/ae.h"
/* Put event loop in the global scope, so it can be explicitly stopped */
static aeEventLoop *loop;
void getCallback(redisAsyncContext *c, void *r, void *privdata) {
redisReply *reply = r;
if (reply == NULL) return;
printf("argv[%s]: %s\n", (char*)privdata, reply->str);
/* Disconnect after receiving the reply to GET */
redisAsyncDisconnect(c);
}
void connectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
return;
}
printf("Connected...\n");
}
void disconnectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
return;
}
printf("Disconnected...\n");
}
int main (int argc, char **argv) {
signal(SIGPIPE, SIG_IGN);
redisAsyncContext *c = redisAsyncConnect("127.0.0.1", 6379);
if (c->err) {
/* Let *c leak for now... */
printf("Error: %s\n", c->errstr);
return 1;
}
loop = aeCreateEventLoop();
redisAeAttach(loop, c);
redisAsyncSetConnectCallback(c,connectCallback);
redisAsyncSetDisconnectCallback(c,disconnectCallback);
redisAsyncCommand(c, NULL, NULL, "SET key %b", argv[argc-1], strlen(argv[argc-1]));
redisAsyncCommand(c, getCallback, (char*)"end-1", "GET key");
aeMain(loop);
return 0;
}
deps/hiredis/example-libev.c
Vendored
+51
View File
@@ -0,0 +1,51 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include "hiredis.h"
#include "async.h"
#include "adapters/libev.h"
void getCallback(redisAsyncContext *c, void *r, void *privdata) {
redisReply *reply = r;
if (reply == NULL) return;
printf("argv[%s]: %s\n", (char*)privdata, reply->str);
/* Disconnect after receiving the reply to GET */
redisAsyncDisconnect(c);
}
void connectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
return;
}
printf("Connected...\n");
}
void disconnectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
return;
}
printf("Disconnected...\n");
}
int main (int argc, char **argv) {
signal(SIGPIPE, SIG_IGN);
redisAsyncContext *c = redisAsyncConnect("127.0.0.1", 6379);
if (c->err) {
/* Let *c leak for now... */
printf("Error: %s\n", c->errstr);
return 1;
}
redisLibevAttach(EV_DEFAULT_ c);
redisAsyncSetConnectCallback(c,connectCallback);
redisAsyncSetDisconnectCallback(c,disconnectCallback);
redisAsyncCommand(c, NULL, NULL, "SET key %b", argv[argc-1], strlen(argv[argc-1]));
redisAsyncCommand(c, getCallback, (char*)"end-1", "GET key");
ev_loop(EV_DEFAULT_ 0);
return 0;
}
deps/hiredis/example-libevent.c
Vendored
+52
View File
@@ -0,0 +1,52 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include "hiredis.h"
#include "async.h"
#include "adapters/libevent.h"
void getCallback(redisAsyncContext *c, void *r, void *privdata) {
redisReply *reply = r;
if (reply == NULL) return;
printf("argv[%s]: %s\n", (char*)privdata, reply->str);
/* Disconnect after receiving the reply to GET */
redisAsyncDisconnect(c);
}
void connectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
return;
}
printf("Connected...\n");
}
void disconnectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
return;
}
printf("Disconnected...\n");
}
int main (int argc, char **argv) {
signal(SIGPIPE, SIG_IGN);
struct event_base *base = event_base_new();
redisAsyncContext *c = redisAsyncConnect("127.0.0.1", 6379);
if (c->err) {
/* Let *c leak for now... */
printf("Error: %s\n", c->errstr);
return 1;
}
redisLibeventAttach(c,base);
redisAsyncSetConnectCallback(c,connectCallback);
redisAsyncSetDisconnectCallback(c,disconnectCallback);
redisAsyncCommand(c, NULL, NULL, "SET key %b", argv[argc-1], strlen(argv[argc-1]));
redisAsyncCommand(c, getCallback, (char*)"end-1", "GET key");
event_base_dispatch(base);
return 0;
}
deps/hiredis/example.c
Vendored
+68
View File
@@ -0,0 +1,68 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "hiredis.h"
int main(void) {
unsigned int j;
redisContext *c;
redisReply *reply;
struct timeval timeout = { 1, 500000 }; // 1.5 seconds
c = redisConnectWithTimeout((char*)"127.0.0.2", 6379, timeout);
if (c->err) {
printf("Connection error: %s\n", c->errstr);
exit(1);
}
/* PING server */
reply = redisCommand(c,"PING");
printf("PING: %s\n", reply->str);
freeReplyObject(reply);
/* Set a key */
reply = redisCommand(c,"SET %s %s", "foo", "hello world");
printf("SET: %s\n", reply->str);
freeReplyObject(reply);
/* Set a key using binary safe API */
reply = redisCommand(c,"SET %b %b", "bar", 3, "hello", 5);
printf("SET (binary API): %s\n", reply->str);
freeReplyObject(reply);
/* Try a GET and two INCR */
reply = redisCommand(c,"GET foo");
printf("GET foo: %s\n", reply->str);
freeReplyObject(reply);
reply = redisCommand(c,"INCR counter");
printf("INCR counter: %lld\n", reply->integer);
freeReplyObject(reply);
/* again ... */
reply = redisCommand(c,"INCR counter");
printf("INCR counter: %lld\n", reply->integer);
freeReplyObject(reply);
/* Create a list of numbers, from 0 to 9 */
reply = redisCommand(c,"DEL mylist");
freeReplyObject(reply);
for (j = 0; j < 10; j++) {
char buf[64];
snprintf(buf,64,"%d",j);
reply = redisCommand(c,"LPUSH mylist element-%s", buf);
freeReplyObject(reply);
}
/* Let's check what we have inside the list */
reply = redisCommand(c,"LRANGE mylist 0 -1");
if (reply->type == REDIS_REPLY_ARRAY) {
for (j = 0; j < reply->elements; j++) {
printf("%u) %s\n", j, reply->element[j]->str);
}
}
freeReplyObject(reply);
return 0;
}
deps/hiredis/fmacros.h
Vendored
+16
View File
@@ -0,0 +1,16 @@
#ifndef __HIREDIS_FMACRO_H
#define __HIREDIS_FMACRO_H
#if !defined(_BSD_SOURCE)
#define _BSD_SOURCE
#endif
#if defined(__sun__)
#define _POSIX_C_SOURCE 200112L
#elif defined(__linux__)
#define _XOPEN_SOURCE 600
#else
#define _XOPEN_SOURCE
#endif
#endif
deps/hiredis/hiredis.c
Vendored
+1285
View File
File diff suppressed because it is too large Load Diff
deps/hiredis/hiredis.h
Vendored
+210
View File
@@ -0,0 +1,210 @@
/*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef __HIREDIS_H
#define __HIREDIS_H
#include <stdio.h> /* for size_t */
#include <stdarg.h> /* for va_list */
#include <sys/time.h> /* for struct timeval */
#define HIREDIS_MAJOR 0
#define HIREDIS_MINOR 10
#define HIREDIS_PATCH 1
#define REDIS_ERR -1
#define REDIS_OK 0
/* When an error occurs, the err flag in a context is set to hold the type of
* error that occured. REDIS_ERR_IO means there was an I/O error and you
* should use the "errno" variable to find out what is wrong.
* For other values, the "errstr" field will hold a description. */
#define REDIS_ERR_IO 1 /* Error in read or write */
#define REDIS_ERR_EOF 3 /* End of file */
#define REDIS_ERR_PROTOCOL 4 /* Protocol error */
#define REDIS_ERR_OOM 5 /* Out of memory */
#define REDIS_ERR_OTHER 2 /* Everything else... */
/* Connection type can be blocking or non-blocking and is set in the
* least significant bit of the flags field in redisContext. */
#define REDIS_BLOCK 0x1
/* Connection may be disconnected before being free'd. The second bit
* in the flags field is set when the context is connected. */
#define REDIS_CONNECTED 0x2
/* The async API might try to disconnect cleanly and flush the output
* buffer and read all subsequent replies before disconnecting.
* This flag means no new commands can come in and the connection
* should be terminated once all replies have been read. */
#define REDIS_DISCONNECTING 0x4
/* Flag specific to the async API which means that the context should be clean
* up as soon as possible. */
#define REDIS_FREEING 0x8
/* Flag that is set when an async callback is executed. */
#define REDIS_IN_CALLBACK 0x10
/* Flag that is set when the async context has one or more subscriptions. */
#define REDIS_SUBSCRIBED 0x20
/* Flag that is set when monitor mode is active */
#define REDIS_MONITORING 0x40
#define REDIS_REPLY_STRING 1
#define REDIS_REPLY_ARRAY 2
#define REDIS_REPLY_INTEGER 3
#define REDIS_REPLY_NIL 4
#define REDIS_REPLY_STATUS 5
#define REDIS_REPLY_ERROR 6
#define REDIS_READER_MAX_BUF (1024*16) /* Default max unused reader buffer. */
#ifdef __cplusplus
extern "C" {
#endif
/* This is the reply object returned by redisCommand() */
typedef struct redisReply {
int type; /* REDIS_REPLY_* */
long long integer; /* The integer when type is REDIS_REPLY_INTEGER */
int len; /* Length of string */
char *str; /* Used for both REDIS_REPLY_ERROR and REDIS_REPLY_STRING */
size_t elements; /* number of elements, for REDIS_REPLY_ARRAY */
struct redisReply **element; /* elements vector for REDIS_REPLY_ARRAY */
} redisReply;
typedef struct redisReadTask {
int type;
int elements; /* number of elements in multibulk container */
int idx; /* index in parent (array) object */
void *obj; /* holds user-generated value for a read task */
struct redisReadTask *parent; /* parent task */
void *privdata; /* user-settable arbitrary field */
} redisReadTask;
typedef struct redisReplyObjectFunctions {
void *(*createString)(const redisReadTask*, char*, size_t);
void *(*createArray)(const redisReadTask*, int);
void *(*createInteger)(const redisReadTask*, long long);
void *(*createNil)(const redisReadTask*);
void (*freeObject)(void*);
} redisReplyObjectFunctions;
/* State for the protocol parser */
typedef struct redisReader {
int err; /* Error flags, 0 when there is no error */
char errstr[128]; /* String representation of error when applicable */
char *buf; /* Read buffer */
size_t pos; /* Buffer cursor */
size_t len; /* Buffer length */
size_t maxbuf; /* Max length of unused buffer */
redisReadTask rstack[9];
int ridx; /* Index of current read task */
void *reply; /* Temporary reply pointer */
redisReplyObjectFunctions *fn;
void *privdata;
} redisReader;
/* Public API for the protocol parser. */
redisReader *redisReaderCreate(void);
void redisReaderFree(redisReader *r);
int redisReaderFeed(redisReader *r, const char *buf, size_t len);
int redisReaderGetReply(redisReader *r, void **reply);
/* Backwards compatibility, can be removed on big version bump. */
#define redisReplyReaderCreate redisReaderCreate
#define redisReplyReaderFree redisReaderFree
#define redisReplyReaderFeed redisReaderFeed
#define redisReplyReaderGetReply redisReaderGetReply
#define redisReplyReaderSetPrivdata(_r, _p) (int)(((redisReader*)(_r))->privdata = (_p))
#define redisReplyReaderGetObject(_r) (((redisReader*)(_r))->reply)
#define redisReplyReaderGetError(_r) (((redisReader*)(_r))->errstr)
/* Function to free the reply objects hiredis returns by default. */
void freeReplyObject(void *reply);
/* Functions to format a command according to the protocol. */
int redisvFormatCommand(char **target, const char *format, va_list ap);
int redisFormatCommand(char **target, const char *format, ...);
int redisFormatCommandArgv(char **target, int argc, const char **argv, const size_t *argvlen);
/* Context for a connection to Redis */
typedef struct redisContext {
int err; /* Error flags, 0 when there is no error */
char errstr[128]; /* String representation of error when applicable */
int fd;
int flags;
char *obuf; /* Write buffer */
redisReader *reader; /* Protocol reader */
} redisContext;
redisContext *redisConnect(const char *ip, int port);
redisContext *redisConnectWithTimeout(const char *ip, int port, struct timeval tv);
redisContext *redisConnectNonBlock(const char *ip, int port);
redisContext *redisConnectUnix(const char *path);
redisContext *redisConnectUnixWithTimeout(const char *path, struct timeval tv);
redisContext *redisConnectUnixNonBlock(const char *path);
int redisSetTimeout(redisContext *c, struct timeval tv);
void redisFree(redisContext *c);
int redisBufferRead(redisContext *c);
int redisBufferWrite(redisContext *c, int *done);
/* In a blocking context, this function first checks if there are unconsumed
* replies to return and returns one if so. Otherwise, it flushes the output
* buffer to the socket and reads until it has a reply. In a non-blocking
* context, it will return unconsumed replies until there are no more. */
int redisGetReply(redisContext *c, void **reply);
int redisGetReplyFromReader(redisContext *c, void **reply);
/* Write a command to the output buffer. Use these functions in blocking mode
* to get a pipeline of commands. */
int redisvAppendCommand(redisContext *c, const char *format, va_list ap);
int redisAppendCommand(redisContext *c, const char *format, ...);
int redisAppendCommandArgv(redisContext *c, int argc, const char **argv, const size_t *argvlen);
/* Issue a command to Redis. In a blocking context, it is identical to calling
* redisAppendCommand, followed by redisGetReply. The function will return
* NULL if there was an error in performing the request, otherwise it will
* return the reply. In a non-blocking context, it is identical to calling
* only redisAppendCommand and will always return NULL. */
void *redisvCommand(redisContext *c, const char *format, va_list ap);
void *redisCommand(redisContext *c, const char *format, ...);
void *redisCommandArgv(redisContext *c, int argc, const char **argv, const size_t *argvlen);
#ifdef __cplusplus
}
#endif
#endif
deps/hiredis/net.c
Vendored
+291
View File
@@ -0,0 +1,291 @@
/* Extracted from anet.c to work properly with Hiredis error reporting.
*
* Copyright (c) 2006-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "fmacros.h"
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/select.h>
#include <sys/un.h>
#include <netinet/in.h>
#include <netinet/tcp.h>
#include <arpa/inet.h>
#include <unistd.h>
#include <fcntl.h>
#include <string.h>
#include <netdb.h>
#include <errno.h>
#include <stdarg.h>
#include <stdio.h>
#include <poll.h>
#include <limits.h>
#include "net.h"
#include "sds.h"
/* Defined in hiredis.c */
void __redisSetError(redisContext *c, int type, const char *str);
static void __redisSetErrorFromErrno(redisContext *c, int type, const char *prefix) {
char buf[128];
size_t len = 0;
if (prefix != NULL)
len = snprintf(buf,sizeof(buf),"%s: ",prefix);
strerror_r(errno,buf+len,sizeof(buf)-len);
__redisSetError(c,type,buf);
}
static int redisSetReuseAddr(redisContext *c, int fd) {
int on = 1;
if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(on)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
return REDIS_ERR;
}
return REDIS_OK;
}
static int redisCreateSocket(redisContext *c, int type) {
int s;
if ((s = socket(type, SOCK_STREAM, 0)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
return REDIS_ERR;
}
if (type == AF_INET) {
if (redisSetReuseAddr(c,s) == REDIS_ERR) {
return REDIS_ERR;
}
}
return s;
}
static int redisSetBlocking(redisContext *c, int fd, int blocking) {
int flags;
/* Set the socket nonblocking.
* Note that fcntl(2) for F_GETFL and F_SETFL can't be
* interrupted by a signal. */
if ((flags = fcntl(fd, F_GETFL)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"fcntl(F_GETFL)");
close(fd);
return REDIS_ERR;
}
if (blocking)
flags &= ~O_NONBLOCK;
else
flags |= O_NONBLOCK;
if (fcntl(fd, F_SETFL, flags) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"fcntl(F_SETFL)");
close(fd);
return REDIS_ERR;
}
return REDIS_OK;
}
static int redisSetTcpNoDelay(redisContext *c, int fd) {
int yes = 1;
if (setsockopt(fd, IPPROTO_TCP, TCP_NODELAY, &yes, sizeof(yes)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(TCP_NODELAY)");
close(fd);
return REDIS_ERR;
}
return REDIS_OK;
}
#define __MAX_MSEC (((LONG_MAX) - 999) / 1000)
static int redisContextWaitReady(redisContext *c, int fd, const struct timeval *timeout) {
struct pollfd wfd[1];
long msec;
msec = -1;
wfd[0].fd = fd;
wfd[0].events = POLLOUT;
/* Only use timeout when not NULL. */
if (timeout != NULL) {
if (timeout->tv_usec > 1000000 || timeout->tv_sec > __MAX_MSEC) {
close(fd);
return REDIS_ERR;
}
msec = (timeout->tv_sec * 1000) + ((timeout->tv_usec + 999) / 1000);
if (msec < 0 || msec > INT_MAX) {
msec = INT_MAX;
}
}
if (errno == EINPROGRESS) {
int res;
if ((res = poll(wfd, 1, msec)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "poll(2)");
close(fd);
return REDIS_ERR;
} else if (res == 0) {
errno = ETIMEDOUT;
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
return REDIS_ERR;
}
if (redisCheckSocketError(c, fd) != REDIS_OK)
return REDIS_ERR;
return REDIS_OK;
}
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
return REDIS_ERR;
}
int redisCheckSocketError(redisContext *c, int fd) {
int err = 0;
socklen_t errlen = sizeof(err);
if (getsockopt(fd, SOL_SOCKET, SO_ERROR, &err, &errlen) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"getsockopt(SO_ERROR)");
close(fd);
return REDIS_ERR;
}
if (err) {
errno = err;
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
return REDIS_ERR;
}
return REDIS_OK;
}
int redisContextSetTimeout(redisContext *c, struct timeval tv) {
if (setsockopt(c->fd,SOL_SOCKET,SO_RCVTIMEO,&tv,sizeof(tv)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(SO_RCVTIMEO)");
return REDIS_ERR;
}
if (setsockopt(c->fd,SOL_SOCKET,SO_SNDTIMEO,&tv,sizeof(tv)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(SO_SNDTIMEO)");
return REDIS_ERR;
}
return REDIS_OK;
}
int redisContextConnectTcp(redisContext *c, const char *addr, int port, struct timeval *timeout) {
int s, rv;
char _port[6]; /* strlen("65535"); */
struct addrinfo hints, *servinfo, *p;
int blocking = (c->flags & REDIS_BLOCK);
snprintf(_port, 6, "%d", port);
memset(&hints,0,sizeof(hints));
hints.ai_family = AF_INET;
hints.ai_socktype = SOCK_STREAM;
if ((rv = getaddrinfo(addr,_port,&hints,&servinfo)) != 0) {
__redisSetError(c,REDIS_ERR_OTHER,gai_strerror(rv));
return REDIS_ERR;
}
for (p = servinfo; p != NULL; p = p->ai_next) {
if ((s = socket(p->ai_family,p->ai_socktype,p->ai_protocol)) == -1)
continue;
if (redisSetBlocking(c,s,0) != REDIS_OK)
goto error;
if (connect(s,p->ai_addr,p->ai_addrlen) == -1) {
if (errno == EHOSTUNREACH) {
close(s);
continue;
} else if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
} else {
if (redisContextWaitReady(c,s,timeout) != REDIS_OK)
goto error;
}
}
if (blocking && redisSetBlocking(c,s,1) != REDIS_OK)
goto error;
if (redisSetTcpNoDelay(c,s) != REDIS_OK)
goto error;
c->fd = s;
c->flags |= REDIS_CONNECTED;
rv = REDIS_OK;
goto end;
}
if (p == NULL) {
char buf[128];
snprintf(buf,sizeof(buf),"Can't create socket: %s",strerror(errno));
__redisSetError(c,REDIS_ERR_OTHER,buf);
goto error;
}
error:
rv = REDIS_ERR;
end:
freeaddrinfo(servinfo);
return rv; // Need to return REDIS_OK if alright
}
int redisContextConnectUnix(redisContext *c, const char *path, struct timeval *timeout) {
int s;
int blocking = (c->flags & REDIS_BLOCK);
struct sockaddr_un sa;
if ((s = redisCreateSocket(c,AF_LOCAL)) < 0)
return REDIS_ERR;
if (redisSetBlocking(c,s,0) != REDIS_OK)
return REDIS_ERR;
sa.sun_family = AF_LOCAL;
strncpy(sa.sun_path,path,sizeof(sa.sun_path)-1);
if (connect(s, (struct sockaddr*)&sa, sizeof(sa)) == -1) {
if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
} else {
if (redisContextWaitReady(c,s,timeout) != REDIS_OK)
return REDIS_ERR;
}
}
/* Reset socket to be blocking after connect(2). */
if (blocking && redisSetBlocking(c,s,1) != REDIS_OK)
return REDIS_ERR;
c->fd = s;
c->flags |= REDIS_CONNECTED;
return REDIS_OK;
}
deps/hiredis/net.h
Vendored
+47
View File
@@ -0,0 +1,47 @@
/* Extracted from anet.c to work properly with Hiredis error reporting.
*
* Copyright (c) 2006-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef __NET_H
#define __NET_H
#include "hiredis.h"
#if defined(__sun)
#define AF_LOCAL AF_UNIX
#endif
int redisCheckSocketError(redisContext *c, int fd);
int redisContextSetTimeout(redisContext *c, struct timeval tv);
int redisContextConnectTcp(redisContext *c, const char *addr, int port, struct timeval *timeout);
int redisContextConnectUnix(redisContext *c, const char *path, struct timeval *timeout);
#endif
deps/hiredis/sds.c
Vendored
+605
View File
@@ -0,0 +1,605 @@
/* SDSLib, A C dynamic strings library
*
* Copyright (c) 2006-2010, Salvatore Sanfilippo <antirez at gmail dot com>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include "sds.h"
#ifdef SDS_ABORT_ON_OOM
static void sdsOomAbort(void) {
fprintf(stderr,"SDS: Out Of Memory (SDS_ABORT_ON_OOM defined)\n");
abort();
}
#endif
sds sdsnewlen(const void *init, size_t initlen) {
struct sdshdr *sh;
sh = malloc(sizeof(struct sdshdr)+initlen+1);
#ifdef SDS_ABORT_ON_OOM
if (sh == NULL) sdsOomAbort();
#else
if (sh == NULL) return NULL;
#endif
sh->len = initlen;
sh->free = 0;
if (initlen) {
if (init) memcpy(sh->buf, init, initlen);
else memset(sh->buf,0,initlen);
}
sh->buf[initlen] = '\0';
return (char*)sh->buf;
}
sds sdsempty(void) {
return sdsnewlen("",0);
}
sds sdsnew(const char *init) {
size_t initlen = (init == NULL) ? 0 : strlen(init);
return sdsnewlen(init, initlen);
}
sds sdsdup(const sds s) {
return sdsnewlen(s, sdslen(s));
}
void sdsfree(sds s) {
if (s == NULL) return;
free(s-sizeof(struct sdshdr));
}
void sdsupdatelen(sds s) {
struct sdshdr *sh = (void*) (s-(sizeof(struct sdshdr)));
int reallen = strlen(s);
sh->free += (sh->len-reallen);
sh->len = reallen;
}
static sds sdsMakeRoomFor(sds s, size_t addlen) {
struct sdshdr *sh, *newsh;
size_t free = sdsavail(s);
size_t len, newlen;
if (free >= addlen) return s;
len = sdslen(s);
sh = (void*) (s-(sizeof(struct sdshdr)));
newlen = (len+addlen)*2;
newsh = realloc(sh, sizeof(struct sdshdr)+newlen+1);
#ifdef SDS_ABORT_ON_OOM
if (newsh == NULL) sdsOomAbort();
#else
if (newsh == NULL) return NULL;
#endif
newsh->free = newlen - len;
return newsh->buf;
}
/* Grow the sds to have the specified length. Bytes that were not part of
* the original length of the sds will be set to zero. */
sds sdsgrowzero(sds s, size_t len) {
struct sdshdr *sh = (void*)(s-(sizeof(struct sdshdr)));
size_t totlen, curlen = sh->len;
if (len <= curlen) return s;
s = sdsMakeRoomFor(s,len-curlen);
if (s == NULL) return NULL;
/* Make sure added region doesn't contain garbage */
sh = (void*)(s-(sizeof(struct sdshdr)));
memset(s+curlen,0,(len-curlen+1)); /* also set trailing \0 byte */
totlen = sh->len+sh->free;
sh->len = len;
sh->free = totlen-sh->len;
return s;
}
sds sdscatlen(sds s, const void *t, size_t len) {
struct sdshdr *sh;
size_t curlen = sdslen(s);
s = sdsMakeRoomFor(s,len);
if (s == NULL) return NULL;
sh = (void*) (s-(sizeof(struct sdshdr)));
memcpy(s+curlen, t, len);
sh->len = curlen+len;
sh->free = sh->free-len;
s[curlen+len] = '\0';
return s;
}
sds sdscat(sds s, const char *t) {
return sdscatlen(s, t, strlen(t));
}
sds sdscpylen(sds s, char *t, size_t len) {
struct sdshdr *sh = (void*) (s-(sizeof(struct sdshdr)));
size_t totlen = sh->free+sh->len;
if (totlen < len) {
s = sdsMakeRoomFor(s,len-sh->len);
if (s == NULL) return NULL;
sh = (void*) (s-(sizeof(struct sdshdr)));
totlen = sh->free+sh->len;
}
memcpy(s, t, len);
s[len] = '\0';
sh->len = len;
sh->free = totlen-len;
return s;
}
sds sdscpy(sds s, char *t) {
return sdscpylen(s, t, strlen(t));
}
sds sdscatvprintf(sds s, const char *fmt, va_list ap) {
va_list cpy;
char *buf, *t;
size_t buflen = 16;
while(1) {
buf = malloc(buflen);
#ifdef SDS_ABORT_ON_OOM
if (buf == NULL) sdsOomAbort();
#else
if (buf == NULL) return NULL;
#endif
buf[buflen-2] = '\0';
va_copy(cpy,ap);
vsnprintf(buf, buflen, fmt, cpy);
if (buf[buflen-2] != '\0') {
free(buf);
buflen *= 2;
continue;
}
break;
}
t = sdscat(s, buf);
free(buf);
return t;
}
sds sdscatprintf(sds s, const char *fmt, ...) {
va_list ap;
char *t;
va_start(ap, fmt);
t = sdscatvprintf(s,fmt,ap);
va_end(ap);
return t;
}
sds sdstrim(sds s, const char *cset) {
struct sdshdr *sh = (void*) (s-(sizeof(struct sdshdr)));
char *start, *end, *sp, *ep;
size_t len;
sp = start = s;
ep = end = s+sdslen(s)-1;
while(sp <= end && strchr(cset, *sp)) sp++;
while(ep > start && strchr(cset, *ep)) ep--;
len = (sp > ep) ? 0 : ((ep-sp)+1);
if (sh->buf != sp) memmove(sh->buf, sp, len);
sh->buf[len] = '\0';
sh->free = sh->free+(sh->len-len);
sh->len = len;
return s;
}
sds sdsrange(sds s, int start, int end) {
struct sdshdr *sh = (void*) (s-(sizeof(struct sdshdr)));
size_t newlen, len = sdslen(s);
if (len == 0) return s;
if (start < 0) {
start = len+start;
if (start < 0) start = 0;
}
if (end < 0) {
end = len+end;
if (end < 0) end = 0;
}
newlen = (start > end) ? 0 : (end-start)+1;
if (newlen != 0) {
if (start >= (signed)len) {
newlen = 0;
} else if (end >= (signed)len) {
end = len-1;
newlen = (start > end) ? 0 : (end-start)+1;
}
} else {
start = 0;
}
if (start && newlen) memmove(sh->buf, sh->buf+start, newlen);
sh->buf[newlen] = 0;
sh->free = sh->free+(sh->len-newlen);
sh->len = newlen;
return s;
}
void sdstolower(sds s) {
int len = sdslen(s), j;
for (j = 0; j < len; j++) s[j] = tolower(s[j]);
}
void sdstoupper(sds s) {
int len = sdslen(s), j;
for (j = 0; j < len; j++) s[j] = toupper(s[j]);
}
int sdscmp(sds s1, sds s2) {
size_t l1, l2, minlen;
int cmp;
l1 = sdslen(s1);
l2 = sdslen(s2);
minlen = (l1 < l2) ? l1 : l2;
cmp = memcmp(s1,s2,minlen);
if (cmp == 0) return l1-l2;
return cmp;
}
/* Split 's' with separator in 'sep'. An array
* of sds strings is returned. *count will be set
* by reference to the number of tokens returned.
*
* On out of memory, zero length string, zero length
* separator, NULL is returned.
*
* Note that 'sep' is able to split a string using
* a multi-character separator. For example
* sdssplit("foo_-_bar","_-_"); will return two
* elements "foo" and "bar".
*
* This version of the function is binary-safe but
* requires length arguments. sdssplit() is just the
* same function but for zero-terminated strings.
*/
sds *sdssplitlen(char *s, int len, char *sep, int seplen, int *count) {
int elements = 0, slots = 5, start = 0, j;
sds *tokens = malloc(sizeof(sds)*slots);
#ifdef SDS_ABORT_ON_OOM
if (tokens == NULL) sdsOomAbort();
#endif
if (seplen < 1 || len < 0 || tokens == NULL) return NULL;
if (len == 0) {
*count = 0;
return tokens;
}
for (j = 0; j < (len-(seplen-1)); j++) {
/* make sure there is room for the next element and the final one */
if (slots < elements+2) {
sds *newtokens;
slots *= 2;
newtokens = realloc(tokens,sizeof(sds)*slots);
if (newtokens == NULL) {
#ifdef SDS_ABORT_ON_OOM
sdsOomAbort();
#else
goto cleanup;
#endif
}
tokens = newtokens;
}
/* search the separator */
if ((seplen == 1 && *(s+j) == sep[0]) || (memcmp(s+j,sep,seplen) == 0)) {
tokens[elements] = sdsnewlen(s+start,j-start);
if (tokens[elements] == NULL) {
#ifdef SDS_ABORT_ON_OOM
sdsOomAbort();
#else
goto cleanup;
#endif
}
elements++;
start = j+seplen;
j = j+seplen-1; /* skip the separator */
}
}
/* Add the final element. We are sure there is room in the tokens array. */
tokens[elements] = sdsnewlen(s+start,len-start);
if (tokens[elements] == NULL) {
#ifdef SDS_ABORT_ON_OOM
sdsOomAbort();
#else
goto cleanup;
#endif
}
elements++;
*count = elements;
return tokens;
#ifndef SDS_ABORT_ON_OOM
cleanup:
{
int i;
for (i = 0; i < elements; i++) sdsfree(tokens[i]);
free(tokens);
return NULL;
}
#endif
}
void sdsfreesplitres(sds *tokens, int count) {
if (!tokens) return;
while(count--)
sdsfree(tokens[count]);
free(tokens);
}
sds sdsfromlonglong(long long value) {
char buf[32], *p;
unsigned long long v;
v = (value < 0) ? -value : value;
p = buf+31; /* point to the last character */
do {
*p-- = '0'+(v%10);
v /= 10;
} while(v);
if (value < 0) *p-- = '-';
p++;
return sdsnewlen(p,32-(p-buf));
}
sds sdscatrepr(sds s, char *p, size_t len) {
s = sdscatlen(s,"\"",1);
if (s == NULL) return NULL;
while(len--) {
switch(*p) {
case '\\':
case '"':
s = sdscatprintf(s,"\\%c",*p);
break;
case '\n': s = sdscatlen(s,"\\n",2); break;
case '\r': s = sdscatlen(s,"\\r",2); break;
case '\t': s = sdscatlen(s,"\\t",2); break;
case '\a': s = sdscatlen(s,"\\a",2); break;
case '\b': s = sdscatlen(s,"\\b",2); break;
default:
if (isprint(*p))
s = sdscatprintf(s,"%c",*p);
else
s = sdscatprintf(s,"\\x%02x",(unsigned char)*p);
break;
}
p++;
if (s == NULL) return NULL;
}
return sdscatlen(s,"\"",1);
}
/* Split a line into arguments, where every argument can be in the
* following programming-language REPL-alike form:
*
* foo bar "newline are supported\n" and "\xff\x00otherstuff"
*
* The number of arguments is stored into *argc, and an array
* of sds is returned. The caller should sdsfree() all the returned
* strings and finally free() the array itself.
*
* Note that sdscatrepr() is able to convert back a string into
* a quoted string in the same format sdssplitargs() is able to parse.
*/
sds *sdssplitargs(char *line, int *argc) {
char *p = line;
char *current = NULL;
char **vector = NULL, **_vector = NULL;
*argc = 0;
while(1) {
/* skip blanks */
while(*p && isspace(*p)) p++;
if (*p) {
/* get a token */
int inq=0; /* set to 1 if we are in "quotes" */
int done=0;
if (current == NULL) {
current = sdsempty();
if (current == NULL) goto err;
}
while(!done) {
if (inq) {
if (*p == '\\' && *(p+1)) {
char c;
p++;
switch(*p) {
case 'n': c = '\n'; break;
case 'r': c = '\r'; break;
case 't': c = '\t'; break;
case 'b': c = '\b'; break;
case 'a': c = '\a'; break;
default: c = *p; break;
}
current = sdscatlen(current,&c,1);
} else if (*p == '"') {
/* closing quote must be followed by a space */
if (*(p+1) && !isspace(*(p+1))) goto err;
done=1;
} else if (!*p) {
/* unterminated quotes */
goto err;
} else {
current = sdscatlen(current,p,1);
}
} else {
switch(*p) {
case ' ':
case '\n':
case '\r':
case '\t':
case '\0':
done=1;
break;
case '"':
inq=1;
break;
default:
current = sdscatlen(current,p,1);
break;
}
}
if (*p) p++;
if (current == NULL) goto err;
}
/* add the token to the vector */
_vector = realloc(vector,((*argc)+1)*sizeof(char*));
if (_vector == NULL) goto err;
vector = _vector;
vector[*argc] = current;
(*argc)++;
current = NULL;
} else {
return vector;
}
}
err:
while((*argc)--)
sdsfree(vector[*argc]);
if (vector != NULL) free(vector);
if (current != NULL) sdsfree(current);
return NULL;
}
#ifdef SDS_TEST_MAIN
#include <stdio.h>
int __failed_tests = 0;
int __test_num = 0;
#define test_cond(descr,_c) do { \
__test_num++; printf("%d - %s: ", __test_num, descr); \
if(_c) printf("PASSED\n"); else {printf("FAILED\n"); __failed_tests++;} \
} while(0);
#define test_report() do { \
printf("%d tests, %d passed, %d failed\n", __test_num, \
__test_num-__failed_tests, __failed_tests); \
if (__failed_tests) { \
printf("=== WARNING === We have failed tests here...\n"); \
} \
} while(0);
int main(void) {
{
sds x = sdsnew("foo"), y;
test_cond("Create a string and obtain the length",
sdslen(x) == 3 && memcmp(x,"foo\0",4) == 0)
sdsfree(x);
x = sdsnewlen("foo",2);
test_cond("Create a string with specified length",
sdslen(x) == 2 && memcmp(x,"fo\0",3) == 0)
x = sdscat(x,"bar");
test_cond("Strings concatenation",
sdslen(x) == 5 && memcmp(x,"fobar\0",6) == 0);
x = sdscpy(x,"a");
test_cond("sdscpy() against an originally longer string",
sdslen(x) == 1 && memcmp(x,"a\0",2) == 0)
x = sdscpy(x,"xyzxxxxxxxxxxyyyyyyyyyykkkkkkkkkk");
test_cond("sdscpy() against an originally shorter string",
sdslen(x) == 33 &&
memcmp(x,"xyzxxxxxxxxxxyyyyyyyyyykkkkkkkkkk\0",33) == 0)
sdsfree(x);
x = sdscatprintf(sdsempty(),"%d",123);
test_cond("sdscatprintf() seems working in the base case",
sdslen(x) == 3 && memcmp(x,"123\0",4) ==0)
sdsfree(x);
x = sdstrim(sdsnew("xxciaoyyy"),"xy");
test_cond("sdstrim() correctly trims characters",
sdslen(x) == 4 && memcmp(x,"ciao\0",5) == 0)
y = sdsrange(sdsdup(x),1,1);
test_cond("sdsrange(...,1,1)",
sdslen(y) == 1 && memcmp(y,"i\0",2) == 0)
sdsfree(y);
y = sdsrange(sdsdup(x),1,-1);
test_cond("sdsrange(...,1,-1)",
sdslen(y) == 3 && memcmp(y,"iao\0",4) == 0)
sdsfree(y);
y = sdsrange(sdsdup(x),-2,-1);
test_cond("sdsrange(...,-2,-1)",
sdslen(y) == 2 && memcmp(y,"ao\0",3) == 0)
sdsfree(y);
y = sdsrange(sdsdup(x),2,1);
test_cond("sdsrange(...,2,1)",
sdslen(y) == 0 && memcmp(y,"\0",1) == 0)
sdsfree(y);
y = sdsrange(sdsdup(x),1,100);
test_cond("sdsrange(...,1,100)",
sdslen(y) == 3 && memcmp(y,"iao\0",4) == 0)
sdsfree(y);
y = sdsrange(sdsdup(x),100,100);
test_cond("sdsrange(...,100,100)",
sdslen(y) == 0 && memcmp(y,"\0",1) == 0)
sdsfree(y);
sdsfree(x);
x = sdsnew("foo");
y = sdsnew("foa");
test_cond("sdscmp(foo,foa)", sdscmp(x,y) > 0)
sdsfree(y);
sdsfree(x);
x = sdsnew("bar");
y = sdsnew("bar");
test_cond("sdscmp(bar,bar)", sdscmp(x,y) == 0)
sdsfree(y);
sdsfree(x);
x = sdsnew("aar");
y = sdsnew("bar");
test_cond("sdscmp(bar,bar)", sdscmp(x,y) < 0)
}
test_report()
}
#endif
sds.h → deps/hiredis/sds.h
Vendored
+21 -6
View File
@@ -32,27 +32,40 @@
#define __SDS_H
#include <sys/types.h>
#include <stdarg.h>
typedef char *sds;
struct sdshdr {
long len;
long free;
int len;
int free;
char buf[];
};
static inline size_t sdslen(const sds s) {
struct sdshdr *sh = (void*)(s-(sizeof(struct sdshdr)));
return sh->len;
}
static inline size_t sdsavail(const sds s) {
struct sdshdr *sh = (void*)(s-(sizeof(struct sdshdr)));
return sh->free;
}
sds sdsnewlen(const void *init, size_t initlen);
sds sdsnew(const char *init);
sds sdsempty();
sds sdsempty(void);
size_t sdslen(const sds s);
sds sdsdup(const sds s);
void sdsfree(sds s);
size_t sdsavail(sds s);
sds sdscatlen(sds s, void *t, size_t len);
sds sdscat(sds s, char *t);
sds sdsgrowzero(sds s, size_t len);
sds sdscatlen(sds s, const void *t, size_t len);
sds sdscat(sds s, const char *t);
sds sdscpylen(sds s, char *t, size_t len);
sds sdscpy(sds s, char *t);
sds sdscatvprintf(sds s, const char *fmt, va_list ap);
#ifdef __GNUC__
sds sdscatprintf(sds s, const char *fmt, ...)
__attribute__((format(printf, 2, 3)));
@@ -61,7 +74,7 @@ sds sdscatprintf(sds s, const char *fmt, ...);
#endif
sds sdstrim(sds s, const char *cset);
sds sdsrange(sds s, long start, long end);
sds sdsrange(sds s, int start, int end);
void sdsupdatelen(sds s);
int sdscmp(sds s1, sds s2);
sds *sdssplitlen(char *s, int len, char *sep, int seplen, int *count);
@@ -69,5 +82,7 @@ void sdsfreesplitres(sds *tokens, int count);
void sdstolower(sds s);
void sdstoupper(sds s);
sds sdsfromlonglong(long long value);
sds sdscatrepr(sds s, char *p, size_t len);
sds *sdssplitargs(char *line, int *argc);
#endif
deps/hiredis/test.c
Vendored
+654
View File
@@ -0,0 +1,654 @@
#include "fmacros.h"
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <strings.h>
#include <sys/time.h>
#include <assert.h>
#include <unistd.h>
#include <signal.h>
#include <errno.h>
#include "hiredis.h"
enum connection_type {
CONN_TCP,
CONN_UNIX
};
struct config {
enum connection_type type;
struct {
const char *host;
int port;
} tcp;
struct {
const char *path;
} unix;
};
/* The following lines make up our testing "framework" :) */
static int tests = 0, fails = 0;
#define test(_s) { printf("#%02d ", ++tests); printf(_s); }
#define test_cond(_c) if(_c) printf("\033[0;32mPASSED\033[0;0m\n"); else {printf("\033[0;31mFAILED\033[0;0m\n"); fails++;}
static long long usec(void) {
struct timeval tv;
gettimeofday(&tv,NULL);
return (((long long)tv.tv_sec)*1000000)+tv.tv_usec;
}
static redisContext *select_database(redisContext *c) {
redisReply *reply;
/* Switch to DB 9 for testing, now that we know we can chat. */
reply = redisCommand(c,"SELECT 9");
assert(reply != NULL);
freeReplyObject(reply);
/* Make sure the DB is emtpy */
reply = redisCommand(c,"DBSIZE");
assert(reply != NULL);
if (reply->type == REDIS_REPLY_INTEGER && reply->integer == 0) {
/* Awesome, DB 9 is empty and we can continue. */
freeReplyObject(reply);
} else {
printf("Database #9 is not empty, test can not continue\n");
exit(1);
}
return c;
}
static void disconnect(redisContext *c) {
redisReply *reply;
/* Make sure we're on DB 9. */
reply = redisCommand(c,"SELECT 9");
assert(reply != NULL);
freeReplyObject(reply);
reply = redisCommand(c,"FLUSHDB");
assert(reply != NULL);
freeReplyObject(reply);
/* Free the context as well. */
redisFree(c);
}
static redisContext *connect(struct config config) {
redisContext *c = NULL;
if (config.type == CONN_TCP) {
c = redisConnect(config.tcp.host, config.tcp.port);
} else if (config.type == CONN_UNIX) {
c = redisConnectUnix(config.unix.path);
} else {
assert(NULL);
}
if (c->err) {
printf("Connection error: %s\n", c->errstr);
exit(1);
}
return select_database(c);
}
static void test_format_commands(void) {
char *cmd;
int len;
test("Format command without interpolation: ");
len = redisFormatCommand(&cmd,"SET foo bar");
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(3+2));
free(cmd);
test("Format command with %%s string interpolation: ");
len = redisFormatCommand(&cmd,"SET %s %s","foo","bar");
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(3+2));
free(cmd);
test("Format command with %%s and an empty string: ");
len = redisFormatCommand(&cmd,"SET %s %s","foo","");
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$0\r\n\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(0+2));
free(cmd);
test("Format command with an empty string in between proper interpolations: ");
len = redisFormatCommand(&cmd,"SET %s %s","","foo");
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$0\r\n\r\n$3\r\nfoo\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(0+2)+4+(3+2));
free(cmd);
test("Format command with %%b string interpolation: ");
len = redisFormatCommand(&cmd,"SET %b %b","foo",3,"b\0r",3);
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$3\r\nb\0r\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(3+2));
free(cmd);
test("Format command with %%b and an empty string: ");
len = redisFormatCommand(&cmd,"SET %b %b","foo",3,"",0);
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$0\r\n\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(0+2));
free(cmd);
test("Format command with literal %%: ");
len = redisFormatCommand(&cmd,"SET %% %%");
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$1\r\n%\r\n$1\r\n%\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(1+2)+4+(1+2));
free(cmd);
/* Vararg width depends on the type. These tests make sure that the
* width is correctly determined using the format and subsequent varargs
* can correctly be interpolated. */
#define INTEGER_WIDTH_TEST(fmt, type) do { \
type value = 123; \
test("Format command with printf-delegation (" #type "): "); \
len = redisFormatCommand(&cmd,"key:%08" fmt " str:%s", value, "hello"); \
test_cond(strncmp(cmd,"*2\r\n$12\r\nkey:00000123\r\n$9\r\nstr:hello\r\n",len) == 0 && \
len == 4+5+(12+2)+4+(9+2)); \
free(cmd); \
} while(0)
#define FLOAT_WIDTH_TEST(type) do { \
type value = 123.0; \
test("Format command with printf-delegation (" #type "): "); \
len = redisFormatCommand(&cmd,"key:%08.3f str:%s", value, "hello"); \
test_cond(strncmp(cmd,"*2\r\n$12\r\nkey:0123.000\r\n$9\r\nstr:hello\r\n",len) == 0 && \
len == 4+5+(12+2)+4+(9+2)); \
free(cmd); \
} while(0)
INTEGER_WIDTH_TEST("d", int);
INTEGER_WIDTH_TEST("hhd", char);
INTEGER_WIDTH_TEST("hd", short);
INTEGER_WIDTH_TEST("ld", long);
INTEGER_WIDTH_TEST("lld", long long);
INTEGER_WIDTH_TEST("u", unsigned int);
INTEGER_WIDTH_TEST("hhu", unsigned char);
INTEGER_WIDTH_TEST("hu", unsigned short);
INTEGER_WIDTH_TEST("lu", unsigned long);
INTEGER_WIDTH_TEST("llu", unsigned long long);
FLOAT_WIDTH_TEST(float);
FLOAT_WIDTH_TEST(double);
test("Format command with invalid printf format: ");
len = redisFormatCommand(&cmd,"key:%08p %b",(void*)1234,"foo",3);
test_cond(len == -1);
const char *argv[3];
argv[0] = "SET";
argv[1] = "foo\0xxx";
argv[2] = "bar";
size_t lens[3] = { 3, 7, 3 };
int argc = 3;
test("Format command by passing argc/argv without lengths: ");
len = redisFormatCommandArgv(&cmd,argc,argv,NULL);
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(3+2));
free(cmd);
test("Format command by passing argc/argv with lengths: ");
len = redisFormatCommandArgv(&cmd,argc,argv,lens);
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$7\r\nfoo\0xxx\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(7+2)+4+(3+2));
free(cmd);
}
static void test_reply_reader(void) {
redisReader *reader;
void *reply;
int ret;
test("Error handling in reply parser: ");
reader = redisReaderCreate();
redisReaderFeed(reader,(char*)"@foo\r\n",6);
ret = redisReaderGetReply(reader,NULL);
test_cond(ret == REDIS_ERR &&
strcasecmp(reader->errstr,"Protocol error, got \"@\" as reply type byte") == 0);
redisReaderFree(reader);
/* when the reply already contains multiple items, they must be free'd
* on an error. valgrind will bark when this doesn't happen. */
test("Memory cleanup in reply parser: ");
reader = redisReaderCreate();
redisReaderFeed(reader,(char*)"*2\r\n",4);
redisReaderFeed(reader,(char*)"$5\r\nhello\r\n",11);
redisReaderFeed(reader,(char*)"@foo\r\n",6);
ret = redisReaderGetReply(reader,NULL);
test_cond(ret == REDIS_ERR &&
strcasecmp(reader->errstr,"Protocol error, got \"@\" as reply type byte") == 0);
redisReaderFree(reader);
test("Set error on nested multi bulks with depth > 2: ");
reader = redisReaderCreate();
redisReaderFeed(reader,(char*)"*1\r\n",4);
redisReaderFeed(reader,(char*)"*1\r\n",4);
redisReaderFeed(reader,(char*)"*1\r\n",4);
redisReaderFeed(reader,(char*)"*1\r\n",4);
ret = redisReaderGetReply(reader,NULL);
test_cond(ret == REDIS_ERR &&
strncasecmp(reader->errstr,"No support for",14) == 0);
redisReaderFree(reader);
test("Works with NULL functions for reply: ");
reader = redisReaderCreate();
reader->fn = NULL;
redisReaderFeed(reader,(char*)"+OK\r\n",5);
ret = redisReaderGetReply(reader,&reply);
test_cond(ret == REDIS_OK && reply == (void*)REDIS_REPLY_STATUS);
redisReaderFree(reader);
test("Works when a single newline (\\r\\n) covers two calls to feed: ");
reader = redisReaderCreate();
reader->fn = NULL;
redisReaderFeed(reader,(char*)"+OK\r",4);
ret = redisReaderGetReply(reader,&reply);
assert(ret == REDIS_OK && reply == NULL);
redisReaderFeed(reader,(char*)"\n",1);
ret = redisReaderGetReply(reader,&reply);
test_cond(ret == REDIS_OK && reply == (void*)REDIS_REPLY_STATUS);
redisReaderFree(reader);
test("Don't reset state after protocol error: ");
reader = redisReaderCreate();
reader->fn = NULL;
redisReaderFeed(reader,(char*)"x",1);
ret = redisReaderGetReply(reader,&reply);
assert(ret == REDIS_ERR);
ret = redisReaderGetReply(reader,&reply);
test_cond(ret == REDIS_ERR && reply == NULL);
redisReaderFree(reader);
/* Regression test for issue #45 on GitHub. */
test("Don't do empty allocation for empty multi bulk: ");
reader = redisReaderCreate();
redisReaderFeed(reader,(char*)"*0\r\n",4);
ret = redisReaderGetReply(reader,&reply);
test_cond(ret == REDIS_OK &&
((redisReply*)reply)->type == REDIS_REPLY_ARRAY &&
((redisReply*)reply)->elements == 0);
freeReplyObject(reply);
redisReaderFree(reader);
}
static void test_blocking_connection_errors(void) {
redisContext *c;
test("Returns error when host cannot be resolved: ");
c = redisConnect((char*)"idontexist.local", 6379);
test_cond(c->err == REDIS_ERR_OTHER &&
(strcmp(c->errstr,"Name or service not known") == 0 ||
strcmp(c->errstr,"Can't resolve: idontexist.local") == 0));
redisFree(c);
test("Returns error when the port is not open: ");
c = redisConnect((char*)"localhost", 1);
test_cond(c->err == REDIS_ERR_IO &&
strcmp(c->errstr,"Connection refused") == 0);
redisFree(c);
test("Returns error when the unix socket path doesn't accept connections: ");
c = redisConnectUnix((char*)"/tmp/idontexist.sock");
test_cond(c->err == REDIS_ERR_IO); /* Don't care about the message... */
redisFree(c);
}
static void test_blocking_connection(struct config config) {
redisContext *c;
redisReply *reply;
c = connect(config);
test("Is able to deliver commands: ");
reply = redisCommand(c,"PING");
test_cond(reply->type == REDIS_REPLY_STATUS &&
strcasecmp(reply->str,"pong") == 0)
freeReplyObject(reply);
test("Is a able to send commands verbatim: ");
reply = redisCommand(c,"SET foo bar");
test_cond (reply->type == REDIS_REPLY_STATUS &&
strcasecmp(reply->str,"ok") == 0)
freeReplyObject(reply);
test("%%s String interpolation works: ");
reply = redisCommand(c,"SET %s %s","foo","hello world");
freeReplyObject(reply);
reply = redisCommand(c,"GET foo");
test_cond(reply->type == REDIS_REPLY_STRING &&
strcmp(reply->str,"hello world") == 0);
freeReplyObject(reply);
test("%%b String interpolation works: ");
reply = redisCommand(c,"SET %b %b","foo",3,"hello\x00world",11);
freeReplyObject(reply);
reply = redisCommand(c,"GET foo");
test_cond(reply->type == REDIS_REPLY_STRING &&
memcmp(reply->str,"hello\x00world",11) == 0)
test("Binary reply length is correct: ");
test_cond(reply->len == 11)
freeReplyObject(reply);
test("Can parse nil replies: ");
reply = redisCommand(c,"GET nokey");
test_cond(reply->type == REDIS_REPLY_NIL)
freeReplyObject(reply);
/* test 7 */
test("Can parse integer replies: ");
reply = redisCommand(c,"INCR mycounter");
test_cond(reply->type == REDIS_REPLY_INTEGER && reply->integer == 1)
freeReplyObject(reply);
test("Can parse multi bulk replies: ");
freeReplyObject(redisCommand(c,"LPUSH mylist foo"));
freeReplyObject(redisCommand(c,"LPUSH mylist bar"));
reply = redisCommand(c,"LRANGE mylist 0 -1");
test_cond(reply->type == REDIS_REPLY_ARRAY &&
reply->elements == 2 &&
!memcmp(reply->element[0]->str,"bar",3) &&
!memcmp(reply->element[1]->str,"foo",3))
freeReplyObject(reply);
/* m/e with multi bulk reply *before* other reply.
* specifically test ordering of reply items to parse. */
test("Can handle nested multi bulk replies: ");
freeReplyObject(redisCommand(c,"MULTI"));
freeReplyObject(redisCommand(c,"LRANGE mylist 0 -1"));
freeReplyObject(redisCommand(c,"PING"));
reply = (redisCommand(c,"EXEC"));
test_cond(reply->type == REDIS_REPLY_ARRAY &&
reply->elements == 2 &&
reply->element[0]->type == REDIS_REPLY_ARRAY &&
reply->element[0]->elements == 2 &&
!memcmp(reply->element[0]->element[0]->str,"bar",3) &&
!memcmp(reply->element[0]->element[1]->str,"foo",3) &&
reply->element[1]->type == REDIS_REPLY_STATUS &&
strcasecmp(reply->element[1]->str,"pong") == 0);
freeReplyObject(reply);
disconnect(c);
}
static void test_blocking_io_errors(struct config config) {
redisContext *c;
redisReply *reply;
void *_reply;
int major, minor;
/* Connect to target given by config. */
c = connect(config);
{
/* Find out Redis version to determine the path for the next test */
const char *field = "redis_version:";
char *p, *eptr;
reply = redisCommand(c,"INFO");
p = strstr(reply->str,field);
major = strtol(p+strlen(field),&eptr,10);
p = eptr+1; /* char next to the first "." */
minor = strtol(p,&eptr,10);
freeReplyObject(reply);
}
test("Returns I/O error when the connection is lost: ");
reply = redisCommand(c,"QUIT");
if (major >= 2 && minor > 0) {
/* > 2.0 returns OK on QUIT and read() should be issued once more
* to know the descriptor is at EOF. */
test_cond(strcasecmp(reply->str,"OK") == 0 &&
redisGetReply(c,&_reply) == REDIS_ERR);
freeReplyObject(reply);
} else {
test_cond(reply == NULL);
}
/* On 2.0, QUIT will cause the connection to be closed immediately and
* the read(2) for the reply on QUIT will set the error to EOF.
* On >2.0, QUIT will return with OK and another read(2) needed to be
* issued to find out the socket was closed by the server. In both
* conditions, the error will be set to EOF. */
assert(c->err == REDIS_ERR_EOF &&
strcmp(c->errstr,"Server closed the connection") == 0);
redisFree(c);
c = connect(config);
test("Returns I/O error on socket timeout: ");
struct timeval tv = { 0, 1000 };
assert(redisSetTimeout(c,tv) == REDIS_OK);
test_cond(redisGetReply(c,&_reply) == REDIS_ERR &&
c->err == REDIS_ERR_IO && errno == EAGAIN);
redisFree(c);
}
static void test_throughput(struct config config) {
redisContext *c = connect(config);
redisReply **replies;
int i, num;
long long t1, t2;
test("Throughput:\n");
for (i = 0; i < 500; i++)
freeReplyObject(redisCommand(c,"LPUSH mylist foo"));
num = 1000;
replies = malloc(sizeof(redisReply*)*num);
t1 = usec();
for (i = 0; i < num; i++) {
replies[i] = redisCommand(c,"PING");
assert(replies[i] != NULL && replies[i]->type == REDIS_REPLY_STATUS);
}
t2 = usec();
for (i = 0; i < num; i++) freeReplyObject(replies[i]);
free(replies);
printf("\t(%dx PING: %.3fs)\n", num, (t2-t1)/1000000.0);
replies = malloc(sizeof(redisReply*)*num);
t1 = usec();
for (i = 0; i < num; i++) {
replies[i] = redisCommand(c,"LRANGE mylist 0 499");
assert(replies[i] != NULL && replies[i]->type == REDIS_REPLY_ARRAY);
assert(replies[i] != NULL && replies[i]->elements == 500);
}
t2 = usec();
for (i = 0; i < num; i++) freeReplyObject(replies[i]);
free(replies);
printf("\t(%dx LRANGE with 500 elements: %.3fs)\n", num, (t2-t1)/1000000.0);
num = 10000;
replies = malloc(sizeof(redisReply*)*num);
for (i = 0; i < num; i++)
redisAppendCommand(c,"PING");
t1 = usec();
for (i = 0; i < num; i++) {
assert(redisGetReply(c, (void*)&replies[i]) == REDIS_OK);
assert(replies[i] != NULL && replies[i]->type == REDIS_REPLY_STATUS);
}
t2 = usec();
for (i = 0; i < num; i++) freeReplyObject(replies[i]);
free(replies);
printf("\t(%dx PING (pipelined): %.3fs)\n", num, (t2-t1)/1000000.0);
replies = malloc(sizeof(redisReply*)*num);
for (i = 0; i < num; i++)
redisAppendCommand(c,"LRANGE mylist 0 499");
t1 = usec();
for (i = 0; i < num; i++) {
assert(redisGetReply(c, (void*)&replies[i]) == REDIS_OK);
assert(replies[i] != NULL && replies[i]->type == REDIS_REPLY_ARRAY);
assert(replies[i] != NULL && replies[i]->elements == 500);
}
t2 = usec();
for (i = 0; i < num; i++) freeReplyObject(replies[i]);
free(replies);
printf("\t(%dx LRANGE with 500 elements (pipelined): %.3fs)\n", num, (t2-t1)/1000000.0);
disconnect(c);
}
// static long __test_callback_flags = 0;
// static void __test_callback(redisContext *c, void *privdata) {
// ((void)c);
// /* Shift to detect execution order */
// __test_callback_flags <<= 8;
// __test_callback_flags |= (long)privdata;
// }
//
// static void __test_reply_callback(redisContext *c, redisReply *reply, void *privdata) {
// ((void)c);
// /* Shift to detect execution order */
// __test_callback_flags <<= 8;
// __test_callback_flags |= (long)privdata;
// if (reply) freeReplyObject(reply);
// }
//
// static redisContext *__connect_nonblock() {
// /* Reset callback flags */
// __test_callback_flags = 0;
// return redisConnectNonBlock("127.0.0.1", port, NULL);
// }
//
// static void test_nonblocking_connection() {
// redisContext *c;
// int wdone = 0;
//
// test("Calls command callback when command is issued: ");
// c = __connect_nonblock();
// redisSetCommandCallback(c,__test_callback,(void*)1);
// redisCommand(c,"PING");
// test_cond(__test_callback_flags == 1);
// redisFree(c);
//
// test("Calls disconnect callback on redisDisconnect: ");
// c = __connect_nonblock();
// redisSetDisconnectCallback(c,__test_callback,(void*)2);
// redisDisconnect(c);
// test_cond(__test_callback_flags == 2);
// redisFree(c);
//
// test("Calls disconnect callback and free callback on redisFree: ");
// c = __connect_nonblock();
// redisSetDisconnectCallback(c,__test_callback,(void*)2);
// redisSetFreeCallback(c,__test_callback,(void*)4);
// redisFree(c);
// test_cond(__test_callback_flags == ((2 << 8) | 4));
//
// test("redisBufferWrite against empty write buffer: ");
// c = __connect_nonblock();
// test_cond(redisBufferWrite(c,&wdone) == REDIS_OK && wdone == 1);
// redisFree(c);
//
// test("redisBufferWrite against not yet connected fd: ");
// c = __connect_nonblock();
// redisCommand(c,"PING");
// test_cond(redisBufferWrite(c,NULL) == REDIS_ERR &&
// strncmp(c->error,"write:",6) == 0);
// redisFree(c);
//
// test("redisBufferWrite against closed fd: ");
// c = __connect_nonblock();
// redisCommand(c,"PING");
// redisDisconnect(c);
// test_cond(redisBufferWrite(c,NULL) == REDIS_ERR &&
// strncmp(c->error,"write:",6) == 0);
// redisFree(c);
//
// test("Process callbacks in the right sequence: ");
// c = __connect_nonblock();
// redisCommandWithCallback(c,__test_reply_callback,(void*)1,"PING");
// redisCommandWithCallback(c,__test_reply_callback,(void*)2,"PING");
// redisCommandWithCallback(c,__test_reply_callback,(void*)3,"PING");
//
// /* Write output buffer */
// wdone = 0;
// while(!wdone) {
// usleep(500);
// redisBufferWrite(c,&wdone);
// }
//
// /* Read until at least one callback is executed (the 3 replies will
// * arrive in a single packet, causing all callbacks to be executed in
// * a single pass). */
// while(__test_callback_flags == 0) {
// assert(redisBufferRead(c) == REDIS_OK);
// redisProcessCallbacks(c);
// }
// test_cond(__test_callback_flags == 0x010203);
// redisFree(c);
//
// test("redisDisconnect executes pending callbacks with NULL reply: ");
// c = __connect_nonblock();
// redisSetDisconnectCallback(c,__test_callback,(void*)1);
// redisCommandWithCallback(c,__test_reply_callback,(void*)2,"PING");
// redisDisconnect(c);
// test_cond(__test_callback_flags == 0x0201);
// redisFree(c);
// }
int main(int argc, char **argv) {
struct config cfg = {
.tcp = {
.host = "127.0.0.1",
.port = 6379
},
.unix = {
.path = "/tmp/redis.sock"
}
};
int throughput = 1;
/* Ignore broken pipe signal (for I/O error tests). */
signal(SIGPIPE, SIG_IGN);
/* Parse command line options. */
argv++; argc--;
while (argc) {
if (argc >= 2 && !strcmp(argv[0],"-h")) {
argv++; argc--;
cfg.tcp.host = argv[0];
} else if (argc >= 2 && !strcmp(argv[0],"-p")) {
argv++; argc--;
cfg.tcp.port = atoi(argv[0]);
} else if (argc >= 2 && !strcmp(argv[0],"-s")) {
argv++; argc--;
cfg.unix.path = argv[0];
} else if (argc >= 1 && !strcmp(argv[0],"--skip-throughput")) {
throughput = 0;
} else {
fprintf(stderr, "Invalid argument: %s\n", argv[0]);
exit(1);
}
argv++; argc--;
}
test_format_commands();
test_reply_reader();
test_blocking_connection_errors();
printf("\nTesting against TCP connection (%s:%d):\n", cfg.tcp.host, cfg.tcp.port);
cfg.type = CONN_TCP;
test_blocking_connection(cfg);
test_blocking_io_errors(cfg);
if (throughput) test_throughput(cfg);
printf("\nTesting against Unix socket connection (%s):\n", cfg.unix.path);
cfg.type = CONN_UNIX;
test_blocking_connection(cfg);
test_blocking_io_errors(cfg);
if (throughput) test_throughput(cfg);
if (fails) {
printf("*** %d TESTS FAILED ***\n", fails);
return 1;
}
printf("ALL TESTS PASSED\n");
return 0;
}
deps/jemalloc/.gitignore
+25
View File
@@ -0,0 +1,25 @@
/autom4te.cache/
/config.stamp
/config.log
/config.status
/configure
/doc/html.xsl
/doc/manpages.xsl
/doc/jemalloc.xml
/doc/jemalloc.html
/doc/jemalloc.3
/lib/
/Makefile
/include/jemalloc/internal/jemalloc_internal\.h
/include/jemalloc/internal/size_classes\.h
/include/jemalloc/jemalloc\.h
/include/jemalloc/jemalloc_defs\.h
/test/jemalloc_test\.h
/src/*.[od]
/test/*.[od]
/test/*.out
/test/[a-zA-Z_]*
!test/*.c
!test/*.exp
/VERSION
/bin/jemalloc.sh
deps/jemalloc/COPYING
Vendored
+27
View File
@@ -0,0 +1,27 @@
Unless otherwise specified, files in the jemalloc source distribution are
subject to the following license:
--------------------------------------------------------------------------------
Copyright (C) 2002-2012 Jason Evans <jasone@canonware.com>.
All rights reserved.
Copyright (C) 2007-2012 Mozilla Foundation. All rights reserved.
Copyright (C) 2009-2012 Facebook, Inc. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice(s),
this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice(s),
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER(S) ``AS IS'' AND ANY EXPRESS
OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
EVENT SHALL THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE
OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
--------------------------------------------------------------------------------
deps/jemalloc/ChangeLog
Vendored
+380
View File
@@ -0,0 +1,380 @@
Following are change highlights associated with official releases. Important
bug fixes are all mentioned, but internal enhancements are omitted here for
brevity (even though they are more fun to write about). Much more detail can be
found in the git revision history:
http://www.canonware.com/cgi-bin/gitweb.cgi?p=jemalloc.git
git://canonware.com/jemalloc.git
* 3.2.0 (November 9, 2012)
In addition to a couple of bug fixes, this version modifies page run
allocation and dirty page purging algorithms in order to better control
page-level virtual memory fragmentation.
Incompatible changes:
- Change the "opt.lg_dirty_mult" default from 5 to 3 (32:1 to 8:1).
Bug fixes:
- Fix dss/mmap allocation precedence code to use recyclable mmap memory only
after primary dss allocation fails.
- Fix deadlock in the "arenas.purge" mallctl. This regression was introduced
in 3.1.0 by the addition of the "arena.<i>.purge" mallctl.
* 3.1.0 (October 16, 2012)
New features:
- Auto-detect whether running inside Valgrind, thus removing the need to
manually specify MALLOC_CONF=valgrind:true.
- Add the "arenas.extend" mallctl, which allows applications to create
manually managed arenas.
- Add the ALLOCM_ARENA() flag for {,r,d}allocm().
- Add the "opt.dss", "arena.<i>.dss", and "stats.arenas.<i>.dss" mallctls,
which provide control over dss/mmap precedence.
- Add the "arena.<i>.purge" mallctl, which obsoletes "arenas.purge".
- Define LG_QUANTUM for hppa.
Incompatible changes:
- Disable tcache by default if running inside Valgrind, in order to avoid
making unallocated objects appear reachable to Valgrind.
- Drop const from malloc_usable_size() argument on Linux.
Bug fixes:
- Fix heap profiling crash if sampled object is freed via realloc(p, 0).
- Remove const from __*_hook variable declarations, so that glibc can modify
them during process forking.
- Fix mlockall(2)/madvise(2) interaction.
- Fix fork(2)-related deadlocks.
- Fix error return value for "thread.tcache.enabled" mallctl.
* 3.0.0 (May 11, 2012)
Although this version adds some major new features, the primary focus is on
internal code cleanup that facilitates maintainability and portability, most
of which is not reflected in the ChangeLog. This is the first release to
incorporate substantial contributions from numerous other developers, and the
result is a more broadly useful allocator (see the git revision history for
contribution details). Note that the license has been unified, thanks to
Facebook granting a license under the same terms as the other copyright
holders (see COPYING).
New features:
- Implement Valgrind support, redzones, and quarantine.
- Add support for additional platforms:
+ FreeBSD
+ Mac OS X Lion
+ MinGW
+ Windows (no support yet for replacing the system malloc)
- Add support for additional architectures:
+ MIPS
+ SH4
+ Tilera
- Add support for cross compiling.
- Add nallocm(), which rounds a request size up to the nearest size class
without actually allocating.
- Implement aligned_alloc() (blame C11).
- Add the "thread.tcache.enabled" mallctl.
- Add the "opt.prof_final" mallctl.
- Update pprof (from gperftools 2.0).
- Add the --with-mangling option.
- Add the --disable-experimental option.
- Add the --disable-munmap option, and make it the default on Linux.
- Add the --enable-mremap option, which disables use of mremap(2) by default.
Incompatible changes:
- Enable stats by default.
- Enable fill by default.
- Disable lazy locking by default.
- Rename the "tcache.flush" mallctl to "thread.tcache.flush".
- Rename the "arenas.pagesize" mallctl to "arenas.page".
- Change the "opt.lg_prof_sample" default from 0 to 19 (1 B to 512 KiB).
- Change the "opt.prof_accum" default from true to false.
Removed features:
- Remove the swap feature, including the "config.swap", "swap.avail",
"swap.prezeroed", "swap.nfds", and "swap.fds" mallctls.
- Remove highruns statistics, including the
"stats.arenas.<i>.bins.<j>.highruns" and
"stats.arenas.<i>.lruns.<j>.highruns" mallctls.
- As part of small size class refactoring, remove the "opt.lg_[qc]space_max",
"arenas.cacheline", "arenas.subpage", "arenas.[tqcs]space_{min,max}", and
"arenas.[tqcs]bins" mallctls.
- Remove the "arenas.chunksize" mallctl.
- Remove the "opt.lg_prof_tcmax" option.
- Remove the "opt.lg_prof_bt_max" option.
- Remove the "opt.lg_tcache_gc_sweep" option.
- Remove the --disable-tiny option, including the "config.tiny" mallctl.
- Remove the --enable-dynamic-page-shift configure option.
- Remove the --enable-sysv configure option.
Bug fixes:
- Fix a statistics-related bug in the "thread.arena" mallctl that could cause
invalid statistics and crashes.
- Work around TLS deallocation via free() on Linux. This bug could cause
write-after-free memory corruption.
- Fix a potential deadlock that could occur during interval- and
growth-triggered heap profile dumps.
- Fix large calloc() zeroing bugs due to dropping chunk map unzeroed flags.
- Fix chunk_alloc_dss() to stop claiming memory is zeroed. This bug could
cause memory corruption and crashes with --enable-dss specified.
- Fix fork-related bugs that could cause deadlock in children between fork
and exec.
- Fix malloc_stats_print() to honor 'b' and 'l' in the opts parameter.
- Fix realloc(p, 0) to act like free(p).
- Do not enforce minimum alignment in memalign().
- Check for NULL pointer in malloc_usable_size().
- Fix an off-by-one heap profile statistics bug that could be observed in
interval- and growth-triggered heap profiles.
- Fix the "epoch" mallctl to update cached stats even if the passed in epoch
is 0.
- Fix bin->runcur management to fix a layout policy bug. This bug did not
affect correctness.
- Fix a bug in choose_arena_hard() that potentially caused more arenas to be
initialized than necessary.
- Add missing "opt.lg_tcache_max" mallctl implementation.
- Use glibc allocator hooks to make mixed allocator usage less likely.
- Fix build issues for --disable-tcache.
- Don't mangle pthread_create() when --with-private-namespace is specified.
* 2.2.5 (November 14, 2011)
Bug fixes:
- Fix huge_ralloc() race when using mremap(2). This is a serious bug that
could cause memory corruption and/or crashes.
- Fix huge_ralloc() to maintain chunk statistics.
- Fix malloc_stats_print(..., "a") output.
* 2.2.4 (November 5, 2011)
Bug fixes:
- Initialize arenas_tsd before using it. This bug existed for 2.2.[0-3], as
well as for --disable-tls builds in earlier releases.
- Do not assume a 4 KiB page size in test/rallocm.c.
* 2.2.3 (August 31, 2011)
This version fixes numerous bugs related to heap profiling.
Bug fixes:
- Fix a prof-related race condition. This bug could cause memory corruption,
but only occurred in non-default configurations (prof_accum:false).
- Fix off-by-one backtracing issues (make sure that prof_alloc_prep() is
excluded from backtraces).
- Fix a prof-related bug in realloc() (only triggered by OOM errors).
- Fix prof-related bugs in allocm() and rallocm().
- Fix prof_tdata_cleanup() for --disable-tls builds.
- Fix a relative include path, to fix objdir builds.
* 2.2.2 (July 30, 2011)
Bug fixes:
- Fix a build error for --disable-tcache.
- Fix assertions in arena_purge() (for real this time).
- Add the --with-private-namespace option. This is a workaround for symbol
conflicts that can inadvertently arise when using static libraries.
* 2.2.1 (March 30, 2011)
Bug fixes:
- Implement atomic operations for x86/x64. This fixes compilation failures
for versions of gcc that are still in wide use.
- Fix an assertion in arena_purge().
* 2.2.0 (March 22, 2011)
This version incorporates several improvements to algorithms and data
structures that tend to reduce fragmentation and increase speed.
New features:
- Add the "stats.cactive" mallctl.
- Update pprof (from google-perftools 1.7).
- Improve backtracing-related configuration logic, and add the
--disable-prof-libgcc option.
Bug fixes:
- Change default symbol visibility from "internal", to "hidden", which
decreases the overhead of library-internal function calls.
- Fix symbol visibility so that it is also set on OS X.
- Fix a build dependency regression caused by the introduction of the .pic.o
suffix for PIC object files.
- Add missing checks for mutex initialization failures.
- Don't use libgcc-based backtracing except on x64, where it is known to work.
- Fix deadlocks on OS X that were due to memory allocation in
pthread_mutex_lock().
- Heap profiling-specific fixes:
+ Fix memory corruption due to integer overflow in small region index
computation, when using a small enough sample interval that profiling
context pointers are stored in small run headers.
+ Fix a bootstrap ordering bug that only occurred with TLS disabled.
+ Fix a rallocm() rsize bug.
+ Fix error detection bugs for aligned memory allocation.
* 2.1.3 (March 14, 2011)
Bug fixes:
- Fix a cpp logic regression (due to the "thread.{de,}allocatedp" mallctl fix
for OS X in 2.1.2).
- Fix a "thread.arena" mallctl bug.
- Fix a thread cache stats merging bug.
* 2.1.2 (March 2, 2011)
Bug fixes:
- Fix "thread.{de,}allocatedp" mallctl for OS X.
- Add missing jemalloc.a to build system.
* 2.1.1 (January 31, 2011)
Bug fixes:
- Fix aligned huge reallocation (affected allocm()).
- Fix the ALLOCM_LG_ALIGN macro definition.
- Fix a heap dumping deadlock.
- Fix a "thread.arena" mallctl bug.
* 2.1.0 (December 3, 2010)
This version incorporates some optimizations that can't quite be considered
bug fixes.
New features:
- Use Linux's mremap(2) for huge object reallocation when possible.
- Avoid locking in mallctl*() when possible.
- Add the "thread.[de]allocatedp" mallctl's.
- Convert the manual page source from roff to DocBook, and generate both roff
and HTML manuals.
Bug fixes:
- Fix a crash due to incorrect bootstrap ordering. This only impacted
--enable-debug --enable-dss configurations.
- Fix a minor statistics bug for mallctl("swap.avail", ...).
* 2.0.1 (October 29, 2010)
Bug fixes:
- Fix a race condition in heap profiling that could cause undefined behavior
if "opt.prof_accum" were disabled.
- Add missing mutex unlocks for some OOM error paths in the heap profiling
code.
- Fix a compilation error for non-C99 builds.
* 2.0.0 (October 24, 2010)
This version focuses on the experimental *allocm() API, and on improved
run-time configuration/introspection. Nonetheless, numerous performance
improvements are also included.
New features:
- Implement the experimental {,r,s,d}allocm() API, which provides a superset
of the functionality available via malloc(), calloc(), posix_memalign(),
realloc(), malloc_usable_size(), and free(). These functions can be used to
allocate/reallocate aligned zeroed memory, ask for optional extra memory
during reallocation, prevent object movement during reallocation, etc.
- Replace JEMALLOC_OPTIONS/JEMALLOC_PROF_PREFIX with MALLOC_CONF, which is
more human-readable, and more flexible. For example:
JEMALLOC_OPTIONS=AJP
is now:
MALLOC_CONF=abort:true,fill:true,stats_print:true
- Port to Apple OS X. Sponsored by Mozilla.
- Make it possible for the application to control thread-->arena mappings via
the "thread.arena" mallctl.
- Add compile-time support for all TLS-related functionality via pthreads TSD.
This is mainly of interest for OS X, which does not support TLS, but has a
TSD implementation with similar performance.
- Override memalign() and valloc() if they are provided by the system.
- Add the "arenas.purge" mallctl, which can be used to synchronously purge all
dirty unused pages.
- Make cumulative heap profiling data optional, so that it is possible to
limit the amount of memory consumed by heap profiling data structures.
- Add per thread allocation counters that can be accessed via the
"thread.allocated" and "thread.deallocated" mallctls.
Incompatible changes:
- Remove JEMALLOC_OPTIONS and malloc_options (see MALLOC_CONF above).
- Increase default backtrace depth from 4 to 128 for heap profiling.
- Disable interval-based profile dumps by default.
Bug fixes:
- Remove bad assertions in fork handler functions. These assertions could
cause aborts for some combinations of configure settings.
- Fix strerror_r() usage to deal with non-standard semantics in GNU libc.
- Fix leak context reporting. This bug tended to cause the number of contexts
to be underreported (though the reported number of objects and bytes were
correct).
- Fix a realloc() bug for large in-place growing reallocation. This bug could
cause memory corruption, but it was hard to trigger.
- Fix an allocation bug for small allocations that could be triggered if
multiple threads raced to create a new run of backing pages.
- Enhance the heap profiler to trigger samples based on usable size, rather
than request size.
- Fix a heap profiling bug due to sometimes losing track of requested object
size for sampled objects.
* 1.0.3 (August 12, 2010)
Bug fixes:
- Fix the libunwind-based implementation of stack backtracing (used for heap
profiling). This bug could cause zero-length backtraces to be reported.
- Add a missing mutex unlock in library initialization code. If multiple
threads raced to initialize malloc, some of them could end up permanently
blocked.
* 1.0.2 (May 11, 2010)
Bug fixes:
- Fix junk filling of large objects, which could cause memory corruption.
- Add MAP_NORESERVE support for chunk mapping, because otherwise virtual
memory limits could cause swap file configuration to fail. Contributed by
Jordan DeLong.
* 1.0.1 (April 14, 2010)
Bug fixes:
- Fix compilation when --enable-fill is specified.
- Fix threads-related profiling bugs that affected accuracy and caused memory
to be leaked during thread exit.
- Fix dirty page purging race conditions that could cause crashes.
- Fix crash in tcache flushing code during thread destruction.
* 1.0.0 (April 11, 2010)
This release focuses on speed and run-time introspection. Numerous
algorithmic improvements make this release substantially faster than its
predecessors.
New features:
- Implement autoconf-based configuration system.
- Add mallctl*(), for the purposes of introspection and run-time
configuration.
- Make it possible for the application to manually flush a thread's cache, via
the "tcache.flush" mallctl.
- Base maximum dirty page count on proportion of active memory.
- Compute various addtional run-time statistics, including per size class
statistics for large objects.
- Expose malloc_stats_print(), which can be called repeatedly by the
application.
- Simplify the malloc_message() signature to only take one string argument,
and incorporate an opaque data pointer argument for use by the application
in combination with malloc_stats_print().
- Add support for allocation backed by one or more swap files, and allow the
application to disable over-commit if swap files are in use.
- Implement allocation profiling and leak checking.
Removed features:
- Remove the dynamic arena rebalancing code, since thread-specific caching
reduces its utility.
Bug fixes:
- Modify chunk allocation to work when address space layout randomization
(ASLR) is in use.
- Fix thread cleanup bugs related to TLS destruction.
- Handle 0-size allocation requests in posix_memalign().
- Fix a chunk leak. The leaked chunks were never touched, so this impacted
virtual memory usage, but not physical memory usage.
* linux_2008082[78]a (August 27/28, 2008)
These snapshot releases are the simple result of incorporating Linux-specific
support into the FreeBSD malloc sources.
--------------------------------------------------------------------------------
vim:filetype=text:textwidth=80
deps/jemalloc/INSTALL
Vendored
+278
View File
@@ -0,0 +1,278 @@
Building and installing jemalloc can be as simple as typing the following while
in the root directory of the source tree:
./configure
make
make install
=== Advanced configuration =====================================================
The 'configure' script supports numerous options that allow control of which
functionality is enabled, where jemalloc is installed, etc. Optionally, pass
any of the following arguments (not a definitive list) to 'configure':
--help
Print a definitive list of options.
--prefix=<install-root-dir>
Set the base directory in which to install. For example:
./configure --prefix=/usr/local
will cause files to be installed into /usr/local/include, /usr/local/lib,
and /usr/local/man.
--with-rpath=<colon-separated-rpath>
Embed one or more library paths, so that libjemalloc can find the libraries
it is linked to. This works only on ELF-based systems.
--with-mangling=<map>
Mangle public symbols specified in <map> which is a comma-separated list of
name:mangled pairs.
For example, to use ld's --wrap option as an alternative method for
overriding libc's malloc implementation, specify something like:
--with-mangling=malloc:__wrap_malloc,free:__wrap_free[...]
Note that mangling happens prior to application of the prefix specified by
--with-jemalloc-prefix, and mangled symbols are then ignored when applying
the prefix.
--with-jemalloc-prefix=<prefix>
Prefix all public APIs with <prefix>. For example, if <prefix> is
"prefix_", API changes like the following occur:
malloc() --> prefix_malloc()
malloc_conf --> prefix_malloc_conf
/etc/malloc.conf --> /etc/prefix_malloc.conf
MALLOC_CONF --> PREFIX_MALLOC_CONF
This makes it possible to use jemalloc at the same time as the system
allocator, or even to use multiple copies of jemalloc simultaneously.
By default, the prefix is "", except on OS X, where it is "je_". On OS X,
jemalloc overlays the default malloc zone, but makes no attempt to actually
replace the "malloc", "calloc", etc. symbols.
--with-private-namespace=<prefix>
Prefix all library-private APIs with <prefix>. For shared libraries,
symbol visibility mechanisms prevent these symbols from being exported, but
for static libraries, naming collisions are a real possibility. By
default, the prefix is "" (empty string).
--with-install-suffix=<suffix>
Append <suffix> to the base name of all installed files, such that multiple
versions of jemalloc can coexist in the same installation directory. For
example, libjemalloc.so.0 becomes libjemalloc<suffix>.so.0.
--enable-cc-silence
Enable code that silences non-useful compiler warnings. This is helpful
when trying to tell serious warnings from those due to compiler
limitations, but it potentially incurs a performance penalty.
--enable-debug
Enable assertions and validation code. This incurs a substantial
performance hit, but is very useful during application development.
--disable-stats
Disable statistics gathering functionality. See the "opt.stats_print"
option documentation for usage details.
--enable-prof
Enable heap profiling and leak detection functionality. See the "opt.prof"
option documentation for usage details. When enabled, there are several
approaches to backtracing, and the configure script chooses the first one
in the following list that appears to function correctly:
+ libunwind (requires --enable-prof-libunwind)
+ libgcc (unless --disable-prof-libgcc)
+ gcc intrinsics (unless --disable-prof-gcc)
--enable-prof-libunwind
Use the libunwind library (http://www.nongnu.org/libunwind/) for stack
backtracing.
--disable-prof-libgcc
Disable the use of libgcc's backtracing functionality.
--disable-prof-gcc
Disable the use of gcc intrinsics for backtracing.
--with-static-libunwind=<libunwind.a>
Statically link against the specified libunwind.a rather than dynamically
linking with -lunwind.
--disable-tcache
Disable thread-specific caches for small objects. Objects are cached and
released in bulk, thus reducing the total number of mutex operations. See
the "opt.tcache" option for usage details.
--enable-mremap
Enable huge realloc() via mremap(2). mremap() is disabled by default
because the flavor used is specific to Linux, which has a quirk in its
virtual memory allocation algorithm that causes semi-permanent VM map holes
under normal jemalloc operation.
--disable-munmap
Disable virtual memory deallocation via munmap(2); instead keep track of
the virtual memory for later use. munmap() is disabled by default (i.e.
--disable-munmap is implied) on Linux, which has a quirk in its virtual
memory allocation algorithm that causes semi-permanent VM map holes under
normal jemalloc operation.
--enable-dss
Enable support for page allocation/deallocation via sbrk(2), in addition to
mmap(2).
--disable-fill
Disable support for junk/zero filling of memory, quarantine, and redzones.
See the "opt.junk", "opt.zero", "opt.quarantine", and "opt.redzone" option
documentation for usage details.
--disable-valgrind
Disable support for Valgrind.
--disable-experimental
Disable support for the experimental API (*allocm()).
--enable-utrace
Enable utrace(2)-based allocation tracing. This feature is not broadly
portable (FreeBSD has it, but Linux and OS X do not).
--enable-xmalloc
Enable support for optional immediate termination due to out-of-memory
errors, as is commonly implemented by "xmalloc" wrapper function for malloc.
See the "opt.xmalloc" option documentation for usage details.
--enable-lazy-lock
Enable code that wraps pthread_create() to detect when an application
switches from single-threaded to multi-threaded mode, so that it can avoid
mutex locking/unlocking operations while in single-threaded mode. In
practice, this feature usually has little impact on performance unless
thread-specific caching is disabled.
--disable-tls
Disable thread-local storage (TLS), which allows for fast access to
thread-local variables via the __thread keyword. If TLS is available,
jemalloc uses it for several purposes.
--with-xslroot=<path>
Specify where to find DocBook XSL stylesheets when building the
documentation.
The following environment variables (not a definitive list) impact configure's
behavior:
CFLAGS="?"
Pass these flags to the compiler. You probably shouldn't define this unless
you know what you are doing. (Use EXTRA_CFLAGS instead.)
EXTRA_CFLAGS="?"
Append these flags to CFLAGS. This makes it possible to add flags such as
-Werror, while allowing the configure script to determine what other flags
are appropriate for the specified configuration.
The configure script specifically checks whether an optimization flag (-O*)
is specified in EXTRA_CFLAGS, and refrains from specifying an optimization
level if it finds that one has already been specified.
CPPFLAGS="?"
Pass these flags to the C preprocessor. Note that CFLAGS is not passed to
'cpp' when 'configure' is looking for include files, so you must use
CPPFLAGS instead if you need to help 'configure' find header files.
LD_LIBRARY_PATH="?"
'ld' uses this colon-separated list to find libraries.
LDFLAGS="?"
Pass these flags when linking.
PATH="?"
'configure' uses this to find programs.
=== Advanced compilation =======================================================
To build only parts of jemalloc, use the following targets:
build_lib_shared
build_lib_static
build_lib
build_doc_html
build_doc_man
build_doc
To install only parts of jemalloc, use the following targets:
install_bin
install_include
install_lib_shared
install_lib_static
install_lib
install_doc_html
install_doc_man
install_doc
To clean up build results to varying degrees, use the following make targets:
clean
distclean
relclean
=== Advanced installation ======================================================
Optionally, define make variables when invoking make, including (not
exclusively):
INCLUDEDIR="?"
Use this as the installation prefix for header files.
LIBDIR="?"
Use this as the installation prefix for libraries.
MANDIR="?"
Use this as the installation prefix for man pages.
DESTDIR="?"
Prepend DESTDIR to INCLUDEDIR, LIBDIR, DATADIR, and MANDIR. This is useful
when installing to a different path than was specified via --prefix.
CC="?"
Use this to invoke the C compiler.
CFLAGS="?"
Pass these flags to the compiler.
CPPFLAGS="?"
Pass these flags to the C preprocessor.
LDFLAGS="?"
Pass these flags when linking.
PATH="?"
Use this to search for programs used during configuration and building.
=== Development ================================================================
If you intend to make non-trivial changes to jemalloc, use the 'autogen.sh'
script rather than 'configure'. This re-generates 'configure', enables
configuration dependency rules, and enables re-generation of automatically
generated source files.
The build system supports using an object directory separate from the source
tree. For example, you can create an 'obj' directory, and from within that
directory, issue configuration and build commands:
autoconf
mkdir obj
cd obj
../configure --enable-autogen
make
=== Documentation ==============================================================
The manual page is generated in both html and roff formats. Any web browser
can be used to view the html manual. The roff manual page can be formatted
prior to installation via the following command:
nroff -man -t doc/jemalloc.3
deps/jemalloc/Makefile.in
Vendored
+323
View File
@@ -0,0 +1,323 @@
# Clear out all vpaths, then set just one (default vpath) for the main build
# directory.
vpath
vpath % .
# Clear the default suffixes, so that built-in rules are not used.
.SUFFIXES :
SHELL := /bin/sh
CC := @CC@
# Configuration parameters.
DESTDIR =
BINDIR := $(DESTDIR)@BINDIR@
INCLUDEDIR := $(DESTDIR)@INCLUDEDIR@
LIBDIR := $(DESTDIR)@LIBDIR@
DATADIR := $(DESTDIR)@DATADIR@
MANDIR := $(DESTDIR)@MANDIR@
srcroot := @srcroot@
objroot := @objroot@
abs_srcroot := @abs_srcroot@
abs_objroot := @abs_objroot@
# Build parameters.
CPPFLAGS := @CPPFLAGS@ -I$(srcroot)include -I$(objroot)include
CFLAGS := @CFLAGS@
LDFLAGS := @LDFLAGS@
EXTRA_LDFLAGS := @EXTRA_LDFLAGS@
LIBS := @LIBS@
RPATH_EXTRA := @RPATH_EXTRA@
SO := @so@
IMPORTLIB := @importlib@
O := @o@
A := @a@
EXE := @exe@
LIBPREFIX := @libprefix@
REV := @rev@
install_suffix := @install_suffix@
ABI := @abi@
XSLTPROC := @XSLTPROC@
AUTOCONF := @AUTOCONF@
_RPATH = @RPATH@
RPATH = $(if $(1),$(call _RPATH,$(1)))
cfghdrs_in := @cfghdrs_in@
cfghdrs_out := @cfghdrs_out@
cfgoutputs_in := @cfgoutputs_in@
cfgoutputs_out := @cfgoutputs_out@
enable_autogen := @enable_autogen@
enable_experimental := @enable_experimental@
DSO_LDFLAGS = @DSO_LDFLAGS@
SOREV = @SOREV@
PIC_CFLAGS = @PIC_CFLAGS@
CTARGET = @CTARGET@
LDTARGET = @LDTARGET@
MKLIB = @MKLIB@
CC_MM = @CC_MM@
ifeq (macho, $(ABI))
TEST_LIBRARY_PATH := DYLD_FALLBACK_LIBRARY_PATH="$(objroot)lib"
else
ifeq (pecoff, $(ABI))
TEST_LIBRARY_PATH := PATH="$(PATH):$(objroot)lib"
else
TEST_LIBRARY_PATH :=
endif
endif
LIBJEMALLOC := $(LIBPREFIX)jemalloc$(install_suffix)
# Lists of files.
BINS := $(srcroot)bin/pprof $(objroot)bin/jemalloc.sh
CHDRS := $(objroot)include/jemalloc/jemalloc$(install_suffix).h \
$(objroot)include/jemalloc/jemalloc_defs$(install_suffix).h
CSRCS := $(srcroot)src/jemalloc.c $(srcroot)src/arena.c $(srcroot)src/atomic.c \
$(srcroot)src/base.c $(srcroot)src/bitmap.c $(srcroot)src/chunk.c \
$(srcroot)src/chunk_dss.c $(srcroot)src/chunk_mmap.c \
$(srcroot)src/ckh.c $(srcroot)src/ctl.c $(srcroot)src/extent.c \
$(srcroot)src/hash.c $(srcroot)src/huge.c $(srcroot)src/mb.c \
$(srcroot)src/mutex.c $(srcroot)src/prof.c $(srcroot)src/quarantine.c \
$(srcroot)src/rtree.c $(srcroot)src/stats.c $(srcroot)src/tcache.c \
$(srcroot)src/util.c $(srcroot)src/tsd.c
ifeq (macho, $(ABI))
CSRCS += $(srcroot)src/zone.c
endif
ifeq ($(IMPORTLIB),$(SO))
STATIC_LIBS := $(objroot)lib/$(LIBJEMALLOC).$(A)
endif
ifdef PIC_CFLAGS
STATIC_LIBS += $(objroot)lib/$(LIBJEMALLOC)_pic.$(A)
else
STATIC_LIBS += $(objroot)lib/$(LIBJEMALLOC)_s.$(A)
endif
DSOS := $(objroot)lib/$(LIBJEMALLOC).$(SOREV)
ifneq ($(SOREV),$(SO))
DSOS += $(objroot)lib/$(LIBJEMALLOC).$(SO)
endif
MAN3 := $(objroot)doc/jemalloc$(install_suffix).3
DOCS_XML := $(objroot)doc/jemalloc$(install_suffix).xml
DOCS_HTML := $(DOCS_XML:$(objroot)%.xml=$(srcroot)%.html)
DOCS_MAN3 := $(DOCS_XML:$(objroot)%.xml=$(srcroot)%.3)
DOCS := $(DOCS_HTML) $(DOCS_MAN3)
CTESTS := $(srcroot)test/aligned_alloc.c $(srcroot)test/allocated.c \
$(srcroot)test/ALLOCM_ARENA.c $(srcroot)test/bitmap.c \
$(srcroot)test/mremap.c $(srcroot)test/posix_memalign.c \
$(srcroot)test/thread_arena.c $(srcroot)test/thread_tcache_enabled.c
ifeq ($(enable_experimental), 1)
CTESTS += $(srcroot)test/allocm.c $(srcroot)test/rallocm.c
endif
COBJS := $(CSRCS:$(srcroot)%.c=$(objroot)%.$(O))
CPICOBJS := $(CSRCS:$(srcroot)%.c=$(objroot)%.pic.$(O))
CTESTOBJS := $(CTESTS:$(srcroot)%.c=$(objroot)%.$(O))
.PHONY: all dist doc_html doc_man doc
.PHONY: install_bin install_include install_lib
.PHONY: install_html install_man install_doc install
.PHONY: tests check clean distclean relclean
.SECONDARY : $(CTESTOBJS)
# Default target.
all: build
dist: build_doc
$(srcroot)doc/%.html : $(objroot)doc/%.xml $(srcroot)doc/stylesheet.xsl $(objroot)doc/html.xsl
$(XSLTPROC) -o $@ $(objroot)doc/html.xsl $<
$(srcroot)doc/%.3 : $(objroot)doc/%.xml $(srcroot)doc/stylesheet.xsl $(objroot)doc/manpages.xsl
$(XSLTPROC) -o $@ $(objroot)doc/manpages.xsl $<
build_doc_html: $(DOCS_HTML)
build_doc_man: $(DOCS_MAN3)
build_doc: $(DOCS)
#
# Include generated dependency files.
#
ifdef CC_MM
-include $(COBJS:%.$(O)=%.d)
-include $(CPICOBJS:%.$(O)=%.d)
-include $(CTESTOBJS:%.$(O)=%.d)
endif
$(COBJS): $(objroot)src/%.$(O): $(srcroot)src/%.c
$(CPICOBJS): $(objroot)src/%.pic.$(O): $(srcroot)src/%.c
$(CPICOBJS): CFLAGS += $(PIC_CFLAGS)
$(CTESTOBJS): $(objroot)test/%.$(O): $(srcroot)test/%.c
$(CTESTOBJS): CPPFLAGS += -I$(objroot)test
ifneq ($(IMPORTLIB),$(SO))
$(COBJS): CPPFLAGS += -DDLLEXPORT
endif
ifndef CC_MM
# Dependencies
HEADER_DIRS = $(srcroot)include/jemalloc/internal \
$(objroot)include/jemalloc $(objroot)include/jemalloc/internal
HEADERS = $(wildcard $(foreach dir,$(HEADER_DIRS),$(dir)/*.h))
$(COBJS) $(CPICOBJS) $(CTESTOBJS): $(HEADERS)
$(CTESTOBJS): $(objroot)test/jemalloc_test.h
endif
$(COBJS) $(CPICOBJS) $(CTESTOBJS): %.$(O):
@mkdir -p $(@D)
$(CC) $(CFLAGS) -c $(CPPFLAGS) $(CTARGET) $<
ifdef CC_MM
@$(CC) -MM $(CPPFLAGS) -MT $@ -o $(@:%.$(O)=%.d) $<
endif
ifneq ($(SOREV),$(SO))
%.$(SO) : %.$(SOREV)
@mkdir -p $(@D)
ln -sf $(<F) $@
endif
$(objroot)lib/$(LIBJEMALLOC).$(SOREV) : $(if $(PIC_CFLAGS),$(CPICOBJS),$(COBJS))
@mkdir -p $(@D)
$(CC) $(DSO_LDFLAGS) $(call RPATH,$(RPATH_EXTRA)) $(LDTARGET) $+ $(LDFLAGS) $(LIBS) $(EXTRA_LDFLAGS)
$(objroot)lib/$(LIBJEMALLOC)_pic.$(A) : $(CPICOBJS)
$(objroot)lib/$(LIBJEMALLOC).$(A) : $(COBJS)
$(objroot)lib/$(LIBJEMALLOC)_s.$(A) : $(COBJS)
$(STATIC_LIBS):
@mkdir -p $(@D)
$(MKLIB) $+
$(objroot)test/bitmap$(EXE): $(objroot)src/bitmap.$(O)
$(objroot)test/%$(EXE): $(objroot)test/%.$(O) $(objroot)src/util.$(O) $(DSOS)
@mkdir -p $(@D)
$(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(filter -lpthread,$(LIBS)) $(EXTRA_LDFLAGS)
build_lib_shared: $(DSOS)
build_lib_static: $(STATIC_LIBS)
build: build_lib_shared build_lib_static
install_bin:
install -d $(BINDIR)
@for b in $(BINS); do \
echo "install -m 755 $$b $(BINDIR)"; \
install -m 755 $$b $(BINDIR); \
done
install_include:
install -d $(INCLUDEDIR)/jemalloc
@for h in $(CHDRS); do \
echo "install -m 644 $$h $(INCLUDEDIR)/jemalloc"; \
install -m 644 $$h $(INCLUDEDIR)/jemalloc; \
done
install_lib_shared: $(DSOS)
install -d $(LIBDIR)
install -m 755 $(objroot)lib/$(LIBJEMALLOC).$(SOREV) $(LIBDIR)
ifneq ($(SOREV),$(SO))
ln -sf $(LIBJEMALLOC).$(SOREV) $(LIBDIR)/$(LIBJEMALLOC).$(SO)
endif
install_lib_static: $(STATIC_LIBS)
install -d $(LIBDIR)
@for l in $(STATIC_LIBS); do \
echo "install -m 755 $$l $(LIBDIR)"; \
install -m 755 $$l $(LIBDIR); \
done
install_lib: install_lib_shared install_lib_static
install_doc_html:
install -d $(DATADIR)/doc/jemalloc$(install_suffix)
@for d in $(DOCS_HTML); do \
echo "install -m 644 $$d $(DATADIR)/doc/jemalloc$(install_suffix)"; \
install -m 644 $$d $(DATADIR)/doc/jemalloc$(install_suffix); \
done
install_doc_man:
install -d $(MANDIR)/man3
@for d in $(DOCS_MAN3); do \
echo "install -m 644 $$d $(MANDIR)/man3"; \
install -m 644 $$d $(MANDIR)/man3; \
done
install_doc: install_doc_html install_doc_man
install: install_bin install_include install_lib install_doc
tests: $(CTESTS:$(srcroot)%.c=$(objroot)%$(EXE))
check: tests
@mkdir -p $(objroot)test
@$(SHELL) -c 'total=0; \
failures=0; \
echo "========================================="; \
for t in $(CTESTS:$(srcroot)%.c=$(objroot)%); do \
total=`expr $$total + 1`; \
/bin/echo -n "$${t} ... "; \
$(TEST_LIBRARY_PATH) $${t}$(EXE) $(abs_srcroot) \
$(abs_objroot) > $(objroot)$${t}.out 2>&1; \
if test -e "$(srcroot)$${t}.exp"; then \
diff -w -u $(srcroot)$${t}.exp \
$(objroot)$${t}.out >/dev/null 2>&1; \
fail=$$?; \
if test "$${fail}" -eq "1" ; then \
failures=`expr $${failures} + 1`; \
echo "*** FAIL ***"; \
else \
echo "pass"; \
fi; \
else \
echo "*** FAIL *** (.exp file is missing)"; \
failures=`expr $${failures} + 1`; \
fi; \
done; \
echo "========================================="; \
echo "Failures: $${failures}/$${total}"'
clean:
rm -f $(COBJS)
rm -f $(CPICOBJS)
rm -f $(COBJS:%.$(O)=%.d)
rm -f $(CPICOBJS:%.$(O)=%.d)
rm -f $(CTESTOBJS:%.$(O)=%$(EXE))
rm -f $(CTESTOBJS)
rm -f $(CTESTOBJS:%.$(O)=%.d)
rm -f $(CTESTOBJS:%.$(O)=%.out)
rm -f $(DSOS) $(STATIC_LIBS)
distclean: clean
rm -rf $(objroot)autom4te.cache
rm -f $(objroot)config.log
rm -f $(objroot)config.status
rm -f $(objroot)config.stamp
rm -f $(cfghdrs_out)
rm -f $(cfgoutputs_out)
relclean: distclean
rm -f $(objroot)configure
rm -f $(srcroot)VERSION
rm -f $(DOCS_HTML)
rm -f $(DOCS_MAN3)
#===============================================================================
# Re-configuration rules.
ifeq ($(enable_autogen), 1)
$(srcroot)configure : $(srcroot)configure.ac
cd ./$(srcroot) && $(AUTOCONF)
$(objroot)config.status : $(srcroot)configure
./$(objroot)config.status --recheck
$(srcroot)config.stamp.in : $(srcroot)configure.ac
echo stamp > $(srcroot)config.stamp.in
$(objroot)config.stamp : $(cfgoutputs_in) $(cfghdrs_in) $(srcroot)configure
./$(objroot)config.status
@touch $@
# There must be some action in order for make to re-read Makefile when it is
# out of date.
$(cfgoutputs_out) $(cfghdrs_out) : $(objroot)config.stamp
@true
endif
deps/jemalloc/README
Vendored
+16
View File
@@ -0,0 +1,16 @@
jemalloc is a general-purpose scalable concurrent malloc(3) implementation.
This distribution is a "portable" implementation that currently targets
FreeBSD, Linux, Apple OS X, and MinGW. jemalloc is included as the default
allocator in the FreeBSD and NetBSD operating systems, and it is used by the
Mozilla Firefox web browser on Microsoft Windows-related platforms. Depending
on your needs, one of the other divergent versions may suit your needs better
than this distribution.
The COPYING file contains copyright and licensing information.
The INSTALL file contains information on how to configure, build, and install
jemalloc.
The ChangeLog file contains a brief summary of changes for each release.
URL: http://www.canonware.com/jemalloc/
deps/jemalloc/VERSION
Vendored
+1
View File
@@ -0,0 +1 @@
3.2.0-0-g87499f6748ebe4817571e817e9f680ccb5bf54a9
deps/jemalloc/autogen.sh
Vendored Executable
+17
View File
@@ -0,0 +1,17 @@
#!/bin/sh
for i in autoconf; do
echo "$i"
$i
if [ $? -ne 0 ]; then
echo "Error $? in $i"
exit 1
fi
done
echo "./configure --enable-autogen $@"
./configure --enable-autogen $@
if [ $? -ne 0 ]; then
echo "Error $? in ./configure"
exit 1
fi
deps/jemalloc/bin/jemalloc.sh.in
Vendored
+9
View File
@@ -0,0 +1,9 @@
#!/bin/sh
prefix=@prefix@
exec_prefix=@exec_prefix@
libdir=@libdir@
@LD_PRELOAD_VAR@=${libdir}/libjemalloc.@SOREV@
export @LD_PRELOAD_VAR@
exec "$@"
deps/jemalloc/bin/pprof
Vendored Executable
+5348
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/config.guess
Vendored Executable
+1530
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/config.stamp.in
Vendored
View File
deps/jemalloc/config.sub
Vendored Executable
+1773
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/configure
Vendored Executable
+8274
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/configure.ac
Vendored
+1289
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/doc/html.xsl.in
Vendored
+4
View File
@@ -0,0 +1,4 @@
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:import href="@XSLROOT@/html/docbook.xsl"/>
<xsl:import href="@abs_srcroot@doc/stylesheet.xsl"/>
</xsl:stylesheet>
deps/jemalloc/doc/jemalloc.3
Vendored
+1482
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/doc/jemalloc.html
Vendored
+1414
View File
File diff suppressed because one or more lines are too long
deps/jemalloc/doc/jemalloc.xml.in
Vendored
+2173
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/doc/manpages.xsl.in
Vendored
+4
View File
@@ -0,0 +1,4 @@
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:import href="@XSLROOT@/manpages/docbook.xsl"/>
<xsl:import href="@abs_srcroot@doc/stylesheet.xsl"/>
</xsl:stylesheet>
deps/jemalloc/doc/stylesheet.xsl
Vendored
+7
View File
@@ -0,0 +1,7 @@
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:param name="funcsynopsis.style">ansi</xsl:param>
<xsl:param name="function.parens" select="1"/>
<xsl:template match="mallctl">
"<xsl:call-template name="inline.monoseq"/>"
</xsl:template>
</xsl:stylesheet>
deps/jemalloc/include/jemalloc/internal/arena.h
Vendored
+975
View File
@@ -0,0 +1,975 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/*
* RUN_MAX_OVRHD indicates maximum desired run header overhead. Runs are sized
* as small as possible such that this setting is still honored, without
* violating other constraints. The goal is to make runs as small as possible
* without exceeding a per run external fragmentation threshold.
*
* We use binary fixed point math for overhead computations, where the binary
* point is implicitly RUN_BFP bits to the left.
*
* Note that it is possible to set RUN_MAX_OVRHD low enough that it cannot be
* honored for some/all object sizes, since when heap profiling is enabled
* there is one pointer of header overhead per object (plus a constant). This
* constraint is relaxed (ignored) for runs that are so small that the
* per-region overhead is greater than:
*
* (RUN_MAX_OVRHD / (reg_interval << (3+RUN_BFP))
*/
#define RUN_BFP 12
/* \/ Implicit binary fixed point. */
#define RUN_MAX_OVRHD 0x0000003dU
#define RUN_MAX_OVRHD_RELAX 0x00001800U
/* Maximum number of regions in one run. */
#define LG_RUN_MAXREGS 11
#define RUN_MAXREGS (1U << LG_RUN_MAXREGS)
/*
* Minimum redzone size. Redzones may be larger than this if necessary to
* preserve region alignment.
*/
#define REDZONE_MINSIZE 16
/*
* The minimum ratio of active:dirty pages per arena is computed as:
*
* (nactive >> opt_lg_dirty_mult) >= ndirty
*
* So, supposing that opt_lg_dirty_mult is 3, there can be no less than 8 times
* as many active pages as dirty pages.
*/
#define LG_DIRTY_MULT_DEFAULT 3
typedef struct arena_chunk_map_s arena_chunk_map_t;
typedef struct arena_chunk_s arena_chunk_t;
typedef struct arena_run_s arena_run_t;
typedef struct arena_bin_info_s arena_bin_info_t;
typedef struct arena_bin_s arena_bin_t;
typedef struct arena_s arena_t;
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* Each element of the chunk map corresponds to one page within the chunk. */
struct arena_chunk_map_s {
#ifndef JEMALLOC_PROF
/*
* Overlay prof_ctx in order to allow it to be referenced by dead code.
* Such antics aren't warranted for per arena data structures, but
* chunk map overhead accounts for a percentage of memory, rather than
* being just a fixed cost.
*/
union {
#endif
union {
/*
* Linkage for run trees. There are two disjoint uses:
*
* 1) arena_t's runs_avail tree.
* 2) arena_run_t conceptually uses this linkage for in-use
* non-full runs, rather than directly embedding linkage.
*/
rb_node(arena_chunk_map_t) rb_link;
/*
* List of runs currently in purgatory. arena_chunk_purge()
* temporarily allocates runs that contain dirty pages while
* purging, so that other threads cannot use the runs while the
* purging thread is operating without the arena lock held.
*/
ql_elm(arena_chunk_map_t) ql_link;
} u;
/* Profile counters, used for large object runs. */
prof_ctx_t *prof_ctx;
#ifndef JEMALLOC_PROF
}; /* union { ... }; */
#endif
/*
* Run address (or size) and various flags are stored together. The bit
* layout looks like (assuming 32-bit system):
*
* ???????? ???????? ????nnnn nnnndula
*
* ? : Unallocated: Run address for first/last pages, unset for internal
* pages.
* Small: Run page offset.
* Large: Run size for first page, unset for trailing pages.
* n : binind for small size class, BININD_INVALID for large size class.
* d : dirty?
* u : unzeroed?
* l : large?
* a : allocated?
*
* Following are example bit patterns for the three types of runs.
*
* p : run page offset
* s : run size
* n : binind for size class; large objects set these to BININD_INVALID
* except for promoted allocations (see prof_promote)
* x : don't care
* - : 0
* + : 1
* [DULA] : bit set
* [dula] : bit unset
*
* Unallocated (clean):
* ssssssss ssssssss ssss++++ ++++du-a
* xxxxxxxx xxxxxxxx xxxxxxxx xxxx-Uxx
* ssssssss ssssssss ssss++++ ++++dU-a
*
* Unallocated (dirty):
* ssssssss ssssssss ssss++++ ++++D--a
* xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx
* ssssssss ssssssss ssss++++ ++++D--a
*
* Small:
* pppppppp pppppppp ppppnnnn nnnnd--A
* pppppppp pppppppp ppppnnnn nnnn---A
* pppppppp pppppppp ppppnnnn nnnnd--A
*
* Large:
* ssssssss ssssssss ssss++++ ++++D-LA
* xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx
* -------- -------- ----++++ ++++D-LA
*
* Large (sampled, size <= PAGE):
* ssssssss ssssssss ssssnnnn nnnnD-LA
*
* Large (not sampled, size == PAGE):
* ssssssss ssssssss ssss++++ ++++D-LA
*/
size_t bits;
#define CHUNK_MAP_BININD_SHIFT 4
#define BININD_INVALID ((size_t)0xffU)
/* CHUNK_MAP_BININD_MASK == (BININD_INVALID << CHUNK_MAP_BININD_SHIFT) */
#define CHUNK_MAP_BININD_MASK ((size_t)0xff0U)
#define CHUNK_MAP_BININD_INVALID CHUNK_MAP_BININD_MASK
#define CHUNK_MAP_FLAGS_MASK ((size_t)0xcU)
#define CHUNK_MAP_DIRTY ((size_t)0x8U)
#define CHUNK_MAP_UNZEROED ((size_t)0x4U)
#define CHUNK_MAP_LARGE ((size_t)0x2U)
#define CHUNK_MAP_ALLOCATED ((size_t)0x1U)
#define CHUNK_MAP_KEY CHUNK_MAP_ALLOCATED
};
typedef rb_tree(arena_chunk_map_t) arena_avail_tree_t;
typedef rb_tree(arena_chunk_map_t) arena_run_tree_t;
/* Arena chunk header. */
struct arena_chunk_s {
/* Arena that owns the chunk. */
arena_t *arena;
/* Linkage for tree of arena chunks that contain dirty runs. */
rb_node(arena_chunk_t) dirty_link;
/* Number of dirty pages. */
size_t ndirty;
/* Number of available runs. */
size_t nruns_avail;
/*
* Number of available run adjacencies. Clean and dirty available runs
* are not coalesced, which causes virtual memory fragmentation. The
* ratio of (nruns_avail-nruns_adjac):nruns_adjac is used for tracking
* this fragmentation.
* */
size_t nruns_adjac;
/*
* Map of pages within chunk that keeps track of free/large/small. The
* first map_bias entries are omitted, since the chunk header does not
* need to be tracked in the map. This omission saves a header page
* for common chunk sizes (e.g. 4 MiB).
*/
arena_chunk_map_t map[1]; /* Dynamically sized. */
};
typedef rb_tree(arena_chunk_t) arena_chunk_tree_t;
struct arena_run_s {
/* Bin this run is associated with. */
arena_bin_t *bin;
/* Index of next region that has never been allocated, or nregs. */
uint32_t nextind;
/* Number of free regions in run. */
unsigned nfree;
};
/*
* Read-only information associated with each element of arena_t's bins array
* is stored separately, partly to reduce memory usage (only one copy, rather
* than one per arena), but mainly to avoid false cacheline sharing.
*
* Each run has the following layout:
*
* /--------------------\
* | arena_run_t header |
* | ... |
* bitmap_offset | bitmap |
* | ... |
* ctx0_offset | ctx map |
* | ... |
* |--------------------|
* | redzone |
* reg0_offset | region 0 |
* | redzone |
* |--------------------| \
* | redzone | |
* | region 1 | > reg_interval
* | redzone | /
* |--------------------|
* | ... |
* | ... |
* | ... |
* |--------------------|
* | redzone |
* | region nregs-1 |
* | redzone |
* |--------------------|
* | alignment pad? |
* \--------------------/
*
* reg_interval has at least the same minimum alignment as reg_size; this
* preserves the alignment constraint that sa2u() depends on. Alignment pad is
* either 0 or redzone_size; it is present only if needed to align reg0_offset.
*/
struct arena_bin_info_s {
/* Size of regions in a run for this bin's size class. */
size_t reg_size;
/* Redzone size. */
size_t redzone_size;
/* Interval between regions (reg_size + (redzone_size << 1)). */
size_t reg_interval;
/* Total size of a run for this bin's size class. */
size_t run_size;
/* Total number of regions in a run for this bin's size class. */
uint32_t nregs;
/*
* Offset of first bitmap_t element in a run header for this bin's size
* class.
*/
uint32_t bitmap_offset;
/*
* Metadata used to manipulate bitmaps for runs associated with this
* bin.
*/
bitmap_info_t bitmap_info;
/*
* Offset of first (prof_ctx_t *) in a run header for this bin's size
* class, or 0 if (config_prof == false || opt_prof == false).
*/
uint32_t ctx0_offset;
/* Offset of first region in a run for this bin's size class. */
uint32_t reg0_offset;
};
struct arena_bin_s {
/*
* All operations on runcur, runs, and stats require that lock be
* locked. Run allocation/deallocation are protected by the arena lock,
* which may be acquired while holding one or more bin locks, but not
* vise versa.
*/
malloc_mutex_t lock;
/*
* Current run being used to service allocations of this bin's size
* class.
*/
arena_run_t *runcur;
/*
* Tree of non-full runs. This tree is used when looking for an
* existing run when runcur is no longer usable. We choose the
* non-full run that is lowest in memory; this policy tends to keep
* objects packed well, and it can also help reduce the number of
* almost-empty chunks.
*/
arena_run_tree_t runs;
/* Bin statistics. */
malloc_bin_stats_t stats;
};
struct arena_s {
/* This arena's index within the arenas array. */
unsigned ind;
/*
* Number of threads currently assigned to this arena. This field is
* protected by arenas_lock.
*/
unsigned nthreads;
/*
* There are three classes of arena operations from a locking
* perspective:
* 1) Thread asssignment (modifies nthreads) is protected by
* arenas_lock.
* 2) Bin-related operations are protected by bin locks.
* 3) Chunk- and run-related operations are protected by this mutex.
*/
malloc_mutex_t lock;
arena_stats_t stats;
/*
* List of tcaches for extant threads associated with this arena.
* Stats from these are merged incrementally, and at exit.
*/
ql_head(tcache_t) tcache_ql;
uint64_t prof_accumbytes;
dss_prec_t dss_prec;
/* Tree of dirty-page-containing chunks this arena manages. */
arena_chunk_tree_t chunks_dirty;
/*
* In order to avoid rapid chunk allocation/deallocation when an arena
* oscillates right on the cusp of needing a new chunk, cache the most
* recently freed chunk. The spare is left in the arena's chunk trees
* until it is deleted.
*
* There is one spare chunk per arena, rather than one spare total, in
* order to avoid interactions between multiple threads that could make
* a single spare inadequate.
*/
arena_chunk_t *spare;
/* Number of pages in active runs. */
size_t nactive;
/*
* Current count of pages within unused runs that are potentially
* dirty, and for which madvise(... MADV_DONTNEED) has not been called.
* By tracking this, we can institute a limit on how much dirty unused
* memory is mapped for each arena.
*/
size_t ndirty;
/*
* Approximate number of pages being purged. It is possible for
* multiple threads to purge dirty pages concurrently, and they use
* npurgatory to indicate the total number of pages all threads are
* attempting to purge.
*/
size_t npurgatory;
/*
* Size/address-ordered trees of this arena's available runs. The trees
* are used for first-best-fit run allocation.
*/
arena_avail_tree_t runs_avail;
/* bins is used to store trees of free regions. */
arena_bin_t bins[NBINS];
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern ssize_t opt_lg_dirty_mult;
/*
* small_size2bin is a compact lookup table that rounds request sizes up to
* size classes. In order to reduce cache footprint, the table is compressed,
* and all accesses are via the SMALL_SIZE2BIN macro.
*/
extern uint8_t const small_size2bin[];
#define SMALL_SIZE2BIN(s) (small_size2bin[(s-1) >> LG_TINY_MIN])
extern arena_bin_info_t arena_bin_info[NBINS];
/* Number of large size classes. */
#define nlclasses (chunk_npages - map_bias)
void arena_purge_all(arena_t *arena);
void arena_prof_accum(arena_t *arena, uint64_t accumbytes);
void arena_tcache_fill_small(arena_t *arena, tcache_bin_t *tbin,
size_t binind, uint64_t prof_accumbytes);
void arena_alloc_junk_small(void *ptr, arena_bin_info_t *bin_info,
bool zero);
void arena_dalloc_junk_small(void *ptr, arena_bin_info_t *bin_info);
void *arena_malloc_small(arena_t *arena, size_t size, bool zero);
void *arena_malloc_large(arena_t *arena, size_t size, bool zero);
void *arena_palloc(arena_t *arena, size_t size, size_t alignment, bool zero);
void arena_prof_promoted(const void *ptr, size_t size);
void arena_dalloc_bin_locked(arena_t *arena, arena_chunk_t *chunk, void *ptr,
arena_chunk_map_t *mapelm);
void arena_dalloc_bin(arena_t *arena, arena_chunk_t *chunk, void *ptr,
size_t pageind, arena_chunk_map_t *mapelm);
void arena_dalloc_small(arena_t *arena, arena_chunk_t *chunk, void *ptr,
size_t pageind);
void arena_dalloc_large_locked(arena_t *arena, arena_chunk_t *chunk,
void *ptr);
void arena_dalloc_large(arena_t *arena, arena_chunk_t *chunk, void *ptr);
void *arena_ralloc_no_move(void *ptr, size_t oldsize, size_t size,
size_t extra, bool zero);
void *arena_ralloc(arena_t *arena, void *ptr, size_t oldsize, size_t size,
size_t extra, size_t alignment, bool zero, bool try_tcache_alloc,
bool try_tcache_dalloc);
dss_prec_t arena_dss_prec_get(arena_t *arena);
void arena_dss_prec_set(arena_t *arena, dss_prec_t dss_prec);
void arena_stats_merge(arena_t *arena, const char **dss, size_t *nactive,
size_t *ndirty, arena_stats_t *astats, malloc_bin_stats_t *bstats,
malloc_large_stats_t *lstats);
bool arena_new(arena_t *arena, unsigned ind);
void arena_boot(void);
void arena_prefork(arena_t *arena);
void arena_postfork_parent(arena_t *arena);
void arena_postfork_child(arena_t *arena);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
arena_chunk_map_t *arena_mapp_get(arena_chunk_t *chunk, size_t pageind);
size_t *arena_mapbitsp_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_unallocated_size_get(arena_chunk_t *chunk,
size_t pageind);
size_t arena_mapbits_large_size_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_small_runind_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_binind_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_dirty_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_unzeroed_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_large_get(arena_chunk_t *chunk, size_t pageind);
size_t arena_mapbits_allocated_get(arena_chunk_t *chunk, size_t pageind);
void arena_mapbits_unallocated_set(arena_chunk_t *chunk, size_t pageind,
size_t size, size_t flags);
void arena_mapbits_unallocated_size_set(arena_chunk_t *chunk, size_t pageind,
size_t size);
void arena_mapbits_large_set(arena_chunk_t *chunk, size_t pageind,
size_t size, size_t flags);
void arena_mapbits_large_binind_set(arena_chunk_t *chunk, size_t pageind,
size_t binind);
void arena_mapbits_small_set(arena_chunk_t *chunk, size_t pageind,
size_t runind, size_t binind, size_t flags);
void arena_mapbits_unzeroed_set(arena_chunk_t *chunk, size_t pageind,
size_t unzeroed);
size_t arena_ptr_small_binind_get(const void *ptr, size_t mapbits);
size_t arena_bin_index(arena_t *arena, arena_bin_t *bin);
unsigned arena_run_regind(arena_run_t *run, arena_bin_info_t *bin_info,
const void *ptr);
prof_ctx_t *arena_prof_ctx_get(const void *ptr);
void arena_prof_ctx_set(const void *ptr, prof_ctx_t *ctx);
void *arena_malloc(arena_t *arena, size_t size, bool zero, bool try_tcache);
size_t arena_salloc(const void *ptr, bool demote);
void arena_dalloc(arena_t *arena, arena_chunk_t *chunk, void *ptr,
bool try_tcache);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_ARENA_C_))
# ifdef JEMALLOC_ARENA_INLINE_A
JEMALLOC_INLINE arena_chunk_map_t *
arena_mapp_get(arena_chunk_t *chunk, size_t pageind)
{
assert(pageind >= map_bias);
assert(pageind < chunk_npages);
return (&chunk->map[pageind-map_bias]);
}
JEMALLOC_INLINE size_t *
arena_mapbitsp_get(arena_chunk_t *chunk, size_t pageind)
{
return (&arena_mapp_get(chunk, pageind)->bits);
}
JEMALLOC_INLINE size_t
arena_mapbits_get(arena_chunk_t *chunk, size_t pageind)
{
return (*arena_mapbitsp_get(chunk, pageind));
}
JEMALLOC_INLINE size_t
arena_mapbits_unallocated_size_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) == 0);
return (mapbits & ~PAGE_MASK);
}
JEMALLOC_INLINE size_t
arena_mapbits_large_size_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) ==
(CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED));
return (mapbits & ~PAGE_MASK);
}
JEMALLOC_INLINE size_t
arena_mapbits_small_runind_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
assert((mapbits & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) ==
CHUNK_MAP_ALLOCATED);
return (mapbits >> LG_PAGE);
}
JEMALLOC_INLINE size_t
arena_mapbits_binind_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
size_t binind;
mapbits = arena_mapbits_get(chunk, pageind);
binind = (mapbits & CHUNK_MAP_BININD_MASK) >> CHUNK_MAP_BININD_SHIFT;
assert(binind < NBINS || binind == BININD_INVALID);
return (binind);
}
JEMALLOC_INLINE size_t
arena_mapbits_dirty_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
return (mapbits & CHUNK_MAP_DIRTY);
}
JEMALLOC_INLINE size_t
arena_mapbits_unzeroed_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
return (mapbits & CHUNK_MAP_UNZEROED);
}
JEMALLOC_INLINE size_t
arena_mapbits_large_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
return (mapbits & CHUNK_MAP_LARGE);
}
JEMALLOC_INLINE size_t
arena_mapbits_allocated_get(arena_chunk_t *chunk, size_t pageind)
{
size_t mapbits;
mapbits = arena_mapbits_get(chunk, pageind);
return (mapbits & CHUNK_MAP_ALLOCATED);
}
JEMALLOC_INLINE void
arena_mapbits_unallocated_set(arena_chunk_t *chunk, size_t pageind, size_t size,
size_t flags)
{
size_t *mapbitsp;
mapbitsp = arena_mapbitsp_get(chunk, pageind);
assert((size & PAGE_MASK) == 0);
assert((flags & ~CHUNK_MAP_FLAGS_MASK) == 0);
assert((flags & (CHUNK_MAP_DIRTY|CHUNK_MAP_UNZEROED)) == flags);
*mapbitsp = size | CHUNK_MAP_BININD_INVALID | flags;
}
JEMALLOC_INLINE void
arena_mapbits_unallocated_size_set(arena_chunk_t *chunk, size_t pageind,
size_t size)
{
size_t *mapbitsp;
mapbitsp = arena_mapbitsp_get(chunk, pageind);
assert((size & PAGE_MASK) == 0);
assert((*mapbitsp & (CHUNK_MAP_LARGE|CHUNK_MAP_ALLOCATED)) == 0);
*mapbitsp = size | (*mapbitsp & PAGE_MASK);
}
JEMALLOC_INLINE void
arena_mapbits_large_set(arena_chunk_t *chunk, size_t pageind, size_t size,
size_t flags)
{
size_t *mapbitsp;
size_t unzeroed;
mapbitsp = arena_mapbitsp_get(chunk, pageind);
assert((size & PAGE_MASK) == 0);
assert((flags & CHUNK_MAP_DIRTY) == flags);
unzeroed = *mapbitsp & CHUNK_MAP_UNZEROED; /* Preserve unzeroed. */
*mapbitsp = size | CHUNK_MAP_BININD_INVALID | flags | unzeroed |
CHUNK_MAP_LARGE | CHUNK_MAP_ALLOCATED;
}
JEMALLOC_INLINE void
arena_mapbits_large_binind_set(arena_chunk_t *chunk, size_t pageind,
size_t binind)
{
size_t *mapbitsp;
assert(binind <= BININD_INVALID);
mapbitsp = arena_mapbitsp_get(chunk, pageind);
assert(arena_mapbits_large_size_get(chunk, pageind) == PAGE);
*mapbitsp = (*mapbitsp & ~CHUNK_MAP_BININD_MASK) | (binind <<
CHUNK_MAP_BININD_SHIFT);
}
JEMALLOC_INLINE void
arena_mapbits_small_set(arena_chunk_t *chunk, size_t pageind, size_t runind,
size_t binind, size_t flags)
{
size_t *mapbitsp;
size_t unzeroed;
assert(binind < BININD_INVALID);
mapbitsp = arena_mapbitsp_get(chunk, pageind);
assert(pageind - runind >= map_bias);
assert((flags & CHUNK_MAP_DIRTY) == flags);
unzeroed = *mapbitsp & CHUNK_MAP_UNZEROED; /* Preserve unzeroed. */
*mapbitsp = (runind << LG_PAGE) | (binind << CHUNK_MAP_BININD_SHIFT) |
flags | unzeroed | CHUNK_MAP_ALLOCATED;
}
JEMALLOC_INLINE void
arena_mapbits_unzeroed_set(arena_chunk_t *chunk, size_t pageind,
size_t unzeroed)
{
size_t *mapbitsp;
mapbitsp = arena_mapbitsp_get(chunk, pageind);
*mapbitsp = (*mapbitsp & ~CHUNK_MAP_UNZEROED) | unzeroed;
}
JEMALLOC_INLINE size_t
arena_ptr_small_binind_get(const void *ptr, size_t mapbits)
{
size_t binind;
binind = (mapbits & CHUNK_MAP_BININD_MASK) >> CHUNK_MAP_BININD_SHIFT;
if (config_debug) {
arena_chunk_t *chunk;
arena_t *arena;
size_t pageind;
size_t actual_mapbits;
arena_run_t *run;
arena_bin_t *bin;
size_t actual_binind;
arena_bin_info_t *bin_info;
assert(binind != BININD_INVALID);
assert(binind < NBINS);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
arena = chunk->arena;
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE;
actual_mapbits = arena_mapbits_get(chunk, pageind);
assert(mapbits == actual_mapbits);
assert(arena_mapbits_large_get(chunk, pageind) == 0);
assert(arena_mapbits_allocated_get(chunk, pageind) != 0);
run = (arena_run_t *)((uintptr_t)chunk + (uintptr_t)((pageind -
(actual_mapbits >> LG_PAGE)) << LG_PAGE));
bin = run->bin;
actual_binind = bin - arena->bins;
assert(binind == actual_binind);
bin_info = &arena_bin_info[actual_binind];
assert(((uintptr_t)ptr - ((uintptr_t)run +
(uintptr_t)bin_info->reg0_offset)) % bin_info->reg_interval
== 0);
}
return (binind);
}
# endif /* JEMALLOC_ARENA_INLINE_A */
# ifdef JEMALLOC_ARENA_INLINE_B
JEMALLOC_INLINE size_t
arena_bin_index(arena_t *arena, arena_bin_t *bin)
{
size_t binind = bin - arena->bins;
assert(binind < NBINS);
return (binind);
}
JEMALLOC_INLINE unsigned
arena_run_regind(arena_run_t *run, arena_bin_info_t *bin_info, const void *ptr)
{
unsigned shift, diff, regind;
size_t interval;
/*
* Freeing a pointer lower than region zero can cause assertion
* failure.
*/
assert((uintptr_t)ptr >= (uintptr_t)run +
(uintptr_t)bin_info->reg0_offset);
/*
* Avoid doing division with a variable divisor if possible. Using
* actual division here can reduce allocator throughput by over 20%!
*/
diff = (unsigned)((uintptr_t)ptr - (uintptr_t)run -
bin_info->reg0_offset);
/* Rescale (factor powers of 2 out of the numerator and denominator). */
interval = bin_info->reg_interval;
shift = ffs(interval) - 1;
diff >>= shift;
interval >>= shift;
if (interval == 1) {
/* The divisor was a power of 2. */
regind = diff;
} else {
/*
* To divide by a number D that is not a power of two we
* multiply by (2^21 / D) and then right shift by 21 positions.
*
* X / D
*
* becomes
*
* (X * interval_invs[D - 3]) >> SIZE_INV_SHIFT
*
* We can omit the first three elements, because we never
* divide by 0, and 1 and 2 are both powers of two, which are
* handled above.
*/
#define SIZE_INV_SHIFT ((sizeof(unsigned) << 3) - LG_RUN_MAXREGS)
#define SIZE_INV(s) (((1U << SIZE_INV_SHIFT) / (s)) + 1)
static const unsigned interval_invs[] = {
SIZE_INV(3),
SIZE_INV(4), SIZE_INV(5), SIZE_INV(6), SIZE_INV(7),
SIZE_INV(8), SIZE_INV(9), SIZE_INV(10), SIZE_INV(11),
SIZE_INV(12), SIZE_INV(13), SIZE_INV(14), SIZE_INV(15),
SIZE_INV(16), SIZE_INV(17), SIZE_INV(18), SIZE_INV(19),
SIZE_INV(20), SIZE_INV(21), SIZE_INV(22), SIZE_INV(23),
SIZE_INV(24), SIZE_INV(25), SIZE_INV(26), SIZE_INV(27),
SIZE_INV(28), SIZE_INV(29), SIZE_INV(30), SIZE_INV(31)
};
if (interval <= ((sizeof(interval_invs) / sizeof(unsigned)) +
2)) {
regind = (diff * interval_invs[interval - 3]) >>
SIZE_INV_SHIFT;
} else
regind = diff / interval;
#undef SIZE_INV
#undef SIZE_INV_SHIFT
}
assert(diff == regind * interval);
assert(regind < bin_info->nregs);
return (regind);
}
JEMALLOC_INLINE prof_ctx_t *
arena_prof_ctx_get(const void *ptr)
{
prof_ctx_t *ret;
arena_chunk_t *chunk;
size_t pageind, mapbits;
cassert(config_prof);
assert(ptr != NULL);
assert(CHUNK_ADDR2BASE(ptr) != ptr);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE;
mapbits = arena_mapbits_get(chunk, pageind);
assert((mapbits & CHUNK_MAP_ALLOCATED) != 0);
if ((mapbits & CHUNK_MAP_LARGE) == 0) {
if (prof_promote)
ret = (prof_ctx_t *)(uintptr_t)1U;
else {
arena_run_t *run = (arena_run_t *)((uintptr_t)chunk +
(uintptr_t)((pageind - (mapbits >> LG_PAGE)) <<
LG_PAGE));
size_t binind = arena_ptr_small_binind_get(ptr,
mapbits);
arena_bin_info_t *bin_info = &arena_bin_info[binind];
unsigned regind;
regind = arena_run_regind(run, bin_info, ptr);
ret = *(prof_ctx_t **)((uintptr_t)run +
bin_info->ctx0_offset + (regind *
sizeof(prof_ctx_t *)));
}
} else
ret = arena_mapp_get(chunk, pageind)->prof_ctx;
return (ret);
}
JEMALLOC_INLINE void
arena_prof_ctx_set(const void *ptr, prof_ctx_t *ctx)
{
arena_chunk_t *chunk;
size_t pageind, mapbits;
cassert(config_prof);
assert(ptr != NULL);
assert(CHUNK_ADDR2BASE(ptr) != ptr);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE;
mapbits = arena_mapbits_get(chunk, pageind);
assert((mapbits & CHUNK_MAP_ALLOCATED) != 0);
if ((mapbits & CHUNK_MAP_LARGE) == 0) {
if (prof_promote == false) {
arena_run_t *run = (arena_run_t *)((uintptr_t)chunk +
(uintptr_t)((pageind - (mapbits >> LG_PAGE)) <<
LG_PAGE));
size_t binind;
arena_bin_info_t *bin_info;
unsigned regind;
binind = arena_ptr_small_binind_get(ptr, mapbits);
bin_info = &arena_bin_info[binind];
regind = arena_run_regind(run, bin_info, ptr);
*((prof_ctx_t **)((uintptr_t)run + bin_info->ctx0_offset
+ (regind * sizeof(prof_ctx_t *)))) = ctx;
} else
assert((uintptr_t)ctx == (uintptr_t)1U);
} else
arena_mapp_get(chunk, pageind)->prof_ctx = ctx;
}
JEMALLOC_INLINE void *
arena_malloc(arena_t *arena, size_t size, bool zero, bool try_tcache)
{
tcache_t *tcache;
assert(size != 0);
assert(size <= arena_maxclass);
if (size <= SMALL_MAXCLASS) {
if (try_tcache && (tcache = tcache_get(true)) != NULL)
return (tcache_alloc_small(tcache, size, zero));
else {
return (arena_malloc_small(choose_arena(arena), size,
zero));
}
} else {
/*
* Initialize tcache after checking size in order to avoid
* infinite recursion during tcache initialization.
*/
if (try_tcache && size <= tcache_maxclass && (tcache =
tcache_get(true)) != NULL)
return (tcache_alloc_large(tcache, size, zero));
else {
return (arena_malloc_large(choose_arena(arena), size,
zero));
}
}
}
/* Return the size of the allocation pointed to by ptr. */
JEMALLOC_INLINE size_t
arena_salloc(const void *ptr, bool demote)
{
size_t ret;
arena_chunk_t *chunk;
size_t pageind, binind;
assert(ptr != NULL);
assert(CHUNK_ADDR2BASE(ptr) != ptr);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE;
assert(arena_mapbits_allocated_get(chunk, pageind) != 0);
binind = arena_mapbits_binind_get(chunk, pageind);
if (binind == BININD_INVALID || (config_prof && demote == false &&
prof_promote && arena_mapbits_large_get(chunk, pageind) != 0)) {
/*
* Large allocation. In the common case (demote == true), and
* as this is an inline function, most callers will only end up
* looking at binind to determine that ptr is a small
* allocation.
*/
assert(((uintptr_t)ptr & PAGE_MASK) == 0);
ret = arena_mapbits_large_size_get(chunk, pageind);
assert(ret != 0);
assert(pageind + (ret>>LG_PAGE) <= chunk_npages);
assert(ret == PAGE || arena_mapbits_large_size_get(chunk,
pageind+(ret>>LG_PAGE)-1) == 0);
assert(binind == arena_mapbits_binind_get(chunk,
pageind+(ret>>LG_PAGE)-1));
assert(arena_mapbits_dirty_get(chunk, pageind) ==
arena_mapbits_dirty_get(chunk, pageind+(ret>>LG_PAGE)-1));
} else {
/*
* Small allocation (possibly promoted to a large object due to
* prof_promote).
*/
assert(arena_mapbits_large_get(chunk, pageind) != 0 ||
arena_ptr_small_binind_get(ptr, arena_mapbits_get(chunk,
pageind)) == binind);
ret = arena_bin_info[binind].reg_size;
}
return (ret);
}
JEMALLOC_INLINE void
arena_dalloc(arena_t *arena, arena_chunk_t *chunk, void *ptr, bool try_tcache)
{
size_t pageind, mapbits;
tcache_t *tcache;
assert(arena != NULL);
assert(chunk->arena == arena);
assert(ptr != NULL);
assert(CHUNK_ADDR2BASE(ptr) != ptr);
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> LG_PAGE;
mapbits = arena_mapbits_get(chunk, pageind);
assert(arena_mapbits_allocated_get(chunk, pageind) != 0);
if ((mapbits & CHUNK_MAP_LARGE) == 0) {
/* Small allocation. */
if (try_tcache && (tcache = tcache_get(false)) != NULL) {
size_t binind;
binind = arena_ptr_small_binind_get(ptr, mapbits);
tcache_dalloc_small(tcache, ptr, binind);
} else
arena_dalloc_small(arena, chunk, ptr, pageind);
} else {
size_t size = arena_mapbits_large_size_get(chunk, pageind);
assert(((uintptr_t)ptr & PAGE_MASK) == 0);
if (try_tcache && size <= tcache_maxclass && (tcache =
tcache_get(false)) != NULL) {
tcache_dalloc_large(tcache, ptr, size);
} else
arena_dalloc_large(arena, chunk, ptr);
}
}
# endif /* JEMALLOC_ARENA_INLINE_B */
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/atomic.h
Vendored
+304
View File
@@ -0,0 +1,304 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#define atomic_read_uint64(p) atomic_add_uint64(p, 0)
#define atomic_read_uint32(p) atomic_add_uint32(p, 0)
#define atomic_read_z(p) atomic_add_z(p, 0)
#define atomic_read_u(p) atomic_add_u(p, 0)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
uint64_t atomic_add_uint64(uint64_t *p, uint64_t x);
uint64_t atomic_sub_uint64(uint64_t *p, uint64_t x);
uint32_t atomic_add_uint32(uint32_t *p, uint32_t x);
uint32_t atomic_sub_uint32(uint32_t *p, uint32_t x);
size_t atomic_add_z(size_t *p, size_t x);
size_t atomic_sub_z(size_t *p, size_t x);
unsigned atomic_add_u(unsigned *p, unsigned x);
unsigned atomic_sub_u(unsigned *p, unsigned x);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_ATOMIC_C_))
/******************************************************************************/
/* 64-bit operations. */
#if (LG_SIZEOF_PTR == 3 || LG_SIZEOF_INT == 3)
# ifdef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (__sync_sub_and_fetch(p, x));
}
#elif (defined(_MSC_VER))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (InterlockedExchangeAdd64(p, x));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (InterlockedExchangeAdd64(p, -((int64_t)x)));
}
#elif (defined(JEMALLOC_OSATOMIC))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (OSAtomicAdd64((int64_t)x, (int64_t *)p));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (OSAtomicAdd64(-((int64_t)x), (int64_t *)p));
}
# elif (defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
asm volatile (
"lock; xaddq %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
x = (uint64_t)(-(int64_t)x);
asm volatile (
"lock; xaddq %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
# elif (defined(JEMALLOC_ATOMIC9))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
/*
* atomic_fetchadd_64() doesn't exist, but we only ever use this
* function on LP64 systems, so atomic_fetchadd_long() will do.
*/
assert(sizeof(uint64_t) == sizeof(unsigned long));
return (atomic_fetchadd_long(p, (unsigned long)x) + x);
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
assert(sizeof(uint64_t) == sizeof(unsigned long));
return (atomic_fetchadd_long(p, (unsigned long)(-(long)x)) - x);
}
# elif (defined(JE_FORCE_SYNC_COMPARE_AND_SWAP_8))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint64_t
atomic_sub_uint64(uint64_t *p, uint64_t x)
{
return (__sync_sub_and_fetch(p, x));
}
# else
# error "Missing implementation for 64-bit atomic operations"
# endif
#endif
/******************************************************************************/
/* 32-bit operations. */
#ifdef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (__sync_sub_and_fetch(p, x));
}
#elif (defined(_MSC_VER))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (InterlockedExchangeAdd(p, x));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (InterlockedExchangeAdd(p, -((int32_t)x)));
}
#elif (defined(JEMALLOC_OSATOMIC))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (OSAtomicAdd32((int32_t)x, (int32_t *)p));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (OSAtomicAdd32(-((int32_t)x), (int32_t *)p));
}
#elif (defined(__i386__) || defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
asm volatile (
"lock; xaddl %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
x = (uint32_t)(-(int32_t)x);
asm volatile (
"lock; xaddl %0, %1;"
: "+r" (x), "=m" (*p) /* Outputs. */
: "m" (*p) /* Inputs. */
);
return (x);
}
#elif (defined(JEMALLOC_ATOMIC9))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (atomic_fetchadd_32(p, x) + x);
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (atomic_fetchadd_32(p, (uint32_t)(-(int32_t)x)) - x);
}
#elif (defined(JE_FORCE_SYNC_COMPARE_AND_SWAP_4))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
return (__sync_add_and_fetch(p, x));
}
JEMALLOC_INLINE uint32_t
atomic_sub_uint32(uint32_t *p, uint32_t x)
{
return (__sync_sub_and_fetch(p, x));
}
#else
# error "Missing implementation for 32-bit atomic operations"
#endif
/******************************************************************************/
/* size_t operations. */
JEMALLOC_INLINE size_t
atomic_add_z(size_t *p, size_t x)
{
#if (LG_SIZEOF_PTR == 3)
return ((size_t)atomic_add_uint64((uint64_t *)p, (uint64_t)x));
#elif (LG_SIZEOF_PTR == 2)
return ((size_t)atomic_add_uint32((uint32_t *)p, (uint32_t)x));
#endif
}
JEMALLOC_INLINE size_t
atomic_sub_z(size_t *p, size_t x)
{
#if (LG_SIZEOF_PTR == 3)
return ((size_t)atomic_add_uint64((uint64_t *)p,
(uint64_t)-((int64_t)x)));
#elif (LG_SIZEOF_PTR == 2)
return ((size_t)atomic_add_uint32((uint32_t *)p,
(uint32_t)-((int32_t)x)));
#endif
}
/******************************************************************************/
/* unsigned operations. */
JEMALLOC_INLINE unsigned
atomic_add_u(unsigned *p, unsigned x)
{
#if (LG_SIZEOF_INT == 3)
return ((unsigned)atomic_add_uint64((uint64_t *)p, (uint64_t)x));
#elif (LG_SIZEOF_INT == 2)
return ((unsigned)atomic_add_uint32((uint32_t *)p, (uint32_t)x));
#endif
}
JEMALLOC_INLINE unsigned
atomic_sub_u(unsigned *p, unsigned x)
{
#if (LG_SIZEOF_INT == 3)
return ((unsigned)atomic_add_uint64((uint64_t *)p,
(uint64_t)-((int64_t)x)));
#elif (LG_SIZEOF_INT == 2)
return ((unsigned)atomic_add_uint32((uint32_t *)p,
(uint32_t)-((int32_t)x)));
#endif
}
/******************************************************************************/
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/base.h
Vendored
+26
View File
@@ -0,0 +1,26 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
void *base_alloc(size_t size);
void *base_calloc(size_t number, size_t size);
extent_node_t *base_node_alloc(void);
void base_node_dealloc(extent_node_t *node);
bool base_boot(void);
void base_prefork(void);
void base_postfork_parent(void);
void base_postfork_child(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/bitmap.h
Vendored
+184
View File
@@ -0,0 +1,184 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/* Maximum bitmap bit count is 2^LG_BITMAP_MAXBITS. */
#define LG_BITMAP_MAXBITS LG_RUN_MAXREGS
typedef struct bitmap_level_s bitmap_level_t;
typedef struct bitmap_info_s bitmap_info_t;
typedef unsigned long bitmap_t;
#define LG_SIZEOF_BITMAP LG_SIZEOF_LONG
/* Number of bits per group. */
#define LG_BITMAP_GROUP_NBITS (LG_SIZEOF_BITMAP + 3)
#define BITMAP_GROUP_NBITS (ZU(1) << LG_BITMAP_GROUP_NBITS)
#define BITMAP_GROUP_NBITS_MASK (BITMAP_GROUP_NBITS-1)
/* Maximum number of levels possible. */
#define BITMAP_MAX_LEVELS \
(LG_BITMAP_MAXBITS / LG_SIZEOF_BITMAP) \
+ !!(LG_BITMAP_MAXBITS % LG_SIZEOF_BITMAP)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct bitmap_level_s {
/* Offset of this level's groups within the array of groups. */
size_t group_offset;
};
struct bitmap_info_s {
/* Logical number of bits in bitmap (stored at bottom level). */
size_t nbits;
/* Number of levels necessary for nbits. */
unsigned nlevels;
/*
* Only the first (nlevels+1) elements are used, and levels are ordered
* bottom to top (e.g. the bottom level is stored in levels[0]).
*/
bitmap_level_t levels[BITMAP_MAX_LEVELS+1];
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
void bitmap_info_init(bitmap_info_t *binfo, size_t nbits);
size_t bitmap_info_ngroups(const bitmap_info_t *binfo);
size_t bitmap_size(size_t nbits);
void bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
bool bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo);
bool bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit);
void bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit);
size_t bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo);
void bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_BITMAP_C_))
JEMALLOC_INLINE bool
bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo)
{
unsigned rgoff = binfo->levels[binfo->nlevels].group_offset - 1;
bitmap_t rg = bitmap[rgoff];
/* The bitmap is full iff the root group is 0. */
return (rg == 0);
}
JEMALLOC_INLINE bool
bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
{
size_t goff;
bitmap_t g;
assert(bit < binfo->nbits);
goff = bit >> LG_BITMAP_GROUP_NBITS;
g = bitmap[goff];
return (!(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))));
}
JEMALLOC_INLINE void
bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
{
size_t goff;
bitmap_t *gp;
bitmap_t g;
assert(bit < binfo->nbits);
assert(bitmap_get(bitmap, binfo, bit) == false);
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[goff];
g = *gp;
assert(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
assert(bitmap_get(bitmap, binfo, bit));
/* Propagate group state transitions up the tree. */
if (g == 0) {
unsigned i;
for (i = 1; i < binfo->nlevels; i++) {
bit = goff;
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[binfo->levels[i].group_offset + goff];
g = *gp;
assert(g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
if (g != 0)
break;
}
}
}
/* sfu: set first unset. */
JEMALLOC_INLINE size_t
bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo)
{
size_t bit;
bitmap_t g;
unsigned i;
assert(bitmap_full(bitmap, binfo) == false);
i = binfo->nlevels - 1;
g = bitmap[binfo->levels[i].group_offset];
bit = ffsl(g) - 1;
while (i > 0) {
i--;
g = bitmap[binfo->levels[i].group_offset + bit];
bit = (bit << LG_BITMAP_GROUP_NBITS) + (ffsl(g) - 1);
}
bitmap_set(bitmap, binfo, bit);
return (bit);
}
JEMALLOC_INLINE void
bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit)
{
size_t goff;
bitmap_t *gp;
bitmap_t g;
bool propagate;
assert(bit < binfo->nbits);
assert(bitmap_get(bitmap, binfo, bit));
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[goff];
g = *gp;
propagate = (g == 0);
assert((g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK))) == 0);
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
assert(bitmap_get(bitmap, binfo, bit) == false);
/* Propagate group state transitions up the tree. */
if (propagate) {
unsigned i;
for (i = 1; i < binfo->nlevels; i++) {
bit = goff;
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[binfo->levels[i].group_offset + goff];
g = *gp;
propagate = (g == 0);
assert((g & (1LU << (bit & BITMAP_GROUP_NBITS_MASK)))
== 0);
g ^= 1LU << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
if (propagate == false)
break;
}
}
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/chunk.h
Vendored
+63
View File
@@ -0,0 +1,63 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/*
* Size and alignment of memory chunks that are allocated by the OS's virtual
* memory system.
*/
#define LG_CHUNK_DEFAULT 22
/* Return the chunk address for allocation address a. */
#define CHUNK_ADDR2BASE(a) \
((void *)((uintptr_t)(a) & ~chunksize_mask))
/* Return the chunk offset of address a. */
#define CHUNK_ADDR2OFFSET(a) \
((size_t)((uintptr_t)(a) & chunksize_mask))
/* Return the smallest chunk multiple that is >= s. */
#define CHUNK_CEILING(s) \
(((s) + chunksize_mask) & ~chunksize_mask)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern size_t opt_lg_chunk;
extern const char *opt_dss;
/* Protects stats_chunks; currently not used for any other purpose. */
extern malloc_mutex_t chunks_mtx;
/* Chunk statistics. */
extern chunk_stats_t stats_chunks;
extern rtree_t *chunks_rtree;
extern size_t chunksize;
extern size_t chunksize_mask; /* (chunksize - 1). */
extern size_t chunk_npages;
extern size_t map_bias; /* Number of arena chunk header pages. */
extern size_t arena_maxclass; /* Max size class for arenas. */
void *chunk_alloc(size_t size, size_t alignment, bool base, bool *zero,
dss_prec_t dss_prec);
void chunk_unmap(void *chunk, size_t size);
void chunk_dealloc(void *chunk, size_t size, bool unmap);
bool chunk_boot(void);
void chunk_prefork(void);
void chunk_postfork_parent(void);
void chunk_postfork_child(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#include "jemalloc/internal/chunk_dss.h"
#include "jemalloc/internal/chunk_mmap.h"
deps/jemalloc/include/jemalloc/internal/chunk_dss.h
Vendored
+38
View File
@@ -0,0 +1,38 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef enum {
dss_prec_disabled = 0,
dss_prec_primary = 1,
dss_prec_secondary = 2,
dss_prec_limit = 3
} dss_prec_t ;
#define DSS_PREC_DEFAULT dss_prec_secondary
#define DSS_DEFAULT "secondary"
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
extern const char *dss_prec_names[];
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
dss_prec_t chunk_dss_prec_get(void);
bool chunk_dss_prec_set(dss_prec_t dss_prec);
void *chunk_alloc_dss(size_t size, size_t alignment, bool *zero);
bool chunk_in_dss(void *chunk);
bool chunk_dss_boot(void);
void chunk_dss_prefork(void);
void chunk_dss_postfork_parent(void);
void chunk_dss_postfork_child(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/chunk_mmap.h
Vendored
+22
View File
@@ -0,0 +1,22 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
bool pages_purge(void *addr, size_t length);
void *chunk_alloc_mmap(size_t size, size_t alignment, bool *zero);
bool chunk_dealloc_mmap(void *chunk, size_t size);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/ckh.h
Vendored
+90
View File
@@ -0,0 +1,90 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct ckh_s ckh_t;
typedef struct ckhc_s ckhc_t;
/* Typedefs to allow easy function pointer passing. */
typedef void ckh_hash_t (const void *, unsigned, size_t *, size_t *);
typedef bool ckh_keycomp_t (const void *, const void *);
/* Maintain counters used to get an idea of performance. */
/* #define CKH_COUNT */
/* Print counter values in ckh_delete() (requires CKH_COUNT). */
/* #define CKH_VERBOSE */
/*
* There are 2^LG_CKH_BUCKET_CELLS cells in each hash table bucket. Try to fit
* one bucket per L1 cache line.
*/
#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* Hash table cell. */
struct ckhc_s {
const void *key;
const void *data;
};
struct ckh_s {
#ifdef CKH_COUNT
/* Counters used to get an idea of performance. */
uint64_t ngrows;
uint64_t nshrinks;
uint64_t nshrinkfails;
uint64_t ninserts;
uint64_t nrelocs;
#endif
/* Used for pseudo-random number generation. */
#define CKH_A 1103515241
#define CKH_C 12347
uint32_t prng_state;
/* Total number of items. */
size_t count;
/*
* Minimum and current number of hash table buckets. There are
* 2^LG_CKH_BUCKET_CELLS cells per bucket.
*/
unsigned lg_minbuckets;
unsigned lg_curbuckets;
/* Hash and comparison functions. */
ckh_hash_t *hash;
ckh_keycomp_t *keycomp;
/* Hash table with 2^lg_curbuckets buckets. */
ckhc_t *tab;
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
bool ckh_new(ckh_t *ckh, size_t minitems, ckh_hash_t *hash,
ckh_keycomp_t *keycomp);
void ckh_delete(ckh_t *ckh);
size_t ckh_count(ckh_t *ckh);
bool ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data);
bool ckh_insert(ckh_t *ckh, const void *key, const void *data);
bool ckh_remove(ckh_t *ckh, const void *searchkey, void **key,
void **data);
bool ckh_search(ckh_t *ckh, const void *seachkey, void **key, void **data);
void ckh_string_hash(const void *key, unsigned minbits, size_t *hash1,
size_t *hash2);
bool ckh_string_keycomp(const void *k1, const void *k2);
void ckh_pointer_hash(const void *key, unsigned minbits, size_t *hash1,
size_t *hash2);
bool ckh_pointer_keycomp(const void *k1, const void *k2);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/ctl.h
Vendored
+117
View File
@@ -0,0 +1,117 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct ctl_node_s ctl_node_t;
typedef struct ctl_named_node_s ctl_named_node_t;
typedef struct ctl_indexed_node_s ctl_indexed_node_t;
typedef struct ctl_arena_stats_s ctl_arena_stats_t;
typedef struct ctl_stats_s ctl_stats_t;
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct ctl_node_s {
bool named;
};
struct ctl_named_node_s {
struct ctl_node_s node;
const char *name;
/* If (nchildren == 0), this is a terminal node. */
unsigned nchildren;
const ctl_node_t *children;
int (*ctl)(const size_t *, size_t, void *, size_t *,
void *, size_t);
};
struct ctl_indexed_node_s {
struct ctl_node_s node;
const ctl_named_node_t *(*index)(const size_t *, size_t, size_t);
};
struct ctl_arena_stats_s {
bool initialized;
unsigned nthreads;
const char *dss;
size_t pactive;
size_t pdirty;
arena_stats_t astats;
/* Aggregate stats for small size classes, based on bin stats. */
size_t allocated_small;
uint64_t nmalloc_small;
uint64_t ndalloc_small;
uint64_t nrequests_small;
malloc_bin_stats_t bstats[NBINS];
malloc_large_stats_t *lstats; /* nlclasses elements. */
};
struct ctl_stats_s {
size_t allocated;
size_t active;
size_t mapped;
struct {
size_t current; /* stats_chunks.curchunks */
uint64_t total; /* stats_chunks.nchunks */
size_t high; /* stats_chunks.highchunks */
} chunks;
struct {
size_t allocated; /* huge_allocated */
uint64_t nmalloc; /* huge_nmalloc */
uint64_t ndalloc; /* huge_ndalloc */
} huge;
unsigned narenas;
ctl_arena_stats_t *arenas; /* (narenas + 1) elements. */
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
int ctl_byname(const char *name, void *oldp, size_t *oldlenp, void *newp,
size_t newlen);
int ctl_nametomib(const char *name, size_t *mibp, size_t *miblenp);
int ctl_bymib(const size_t *mib, size_t miblen, void *oldp, size_t *oldlenp,
void *newp, size_t newlen);
bool ctl_boot(void);
void ctl_prefork(void);
void ctl_postfork_parent(void);
void ctl_postfork_child(void);
#define xmallctl(name, oldp, oldlenp, newp, newlen) do { \
if (je_mallctl(name, oldp, oldlenp, newp, newlen) \
!= 0) { \
malloc_printf( \
"<jemalloc>: Failure in xmallctl(\"%s\", ...)\n", \
name); \
abort(); \
} \
} while (0)
#define xmallctlnametomib(name, mibp, miblenp) do { \
if (je_mallctlnametomib(name, mibp, miblenp) != 0) { \
malloc_printf("<jemalloc>: Failure in " \
"xmallctlnametomib(\"%s\", ...)\n", name); \
abort(); \
} \
} while (0)
#define xmallctlbymib(mib, miblen, oldp, oldlenp, newp, newlen) do { \
if (je_mallctlbymib(mib, miblen, oldp, oldlenp, newp, \
newlen) != 0) { \
malloc_write( \
"<jemalloc>: Failure in xmallctlbymib()\n"); \
abort(); \
} \
} while (0)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/extent.h
Vendored
+46
View File
@@ -0,0 +1,46 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct extent_node_s extent_node_t;
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
/* Tree of extents. */
struct extent_node_s {
/* Linkage for the size/address-ordered tree. */
rb_node(extent_node_t) link_szad;
/* Linkage for the address-ordered tree. */
rb_node(extent_node_t) link_ad;
/* Profile counters, used for huge objects. */
prof_ctx_t *prof_ctx;
/* Pointer to the extent that this tree node is responsible for. */
void *addr;
/* Total region size. */
size_t size;
/* True if zero-filled; used by chunk recycling code. */
bool zeroed;
};
typedef rb_tree(extent_node_t) extent_tree_t;
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
rb_proto(, extent_tree_szad_, extent_tree_t, extent_node_t)
rb_proto(, extent_tree_ad_, extent_tree_t, extent_node_t)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/hash.h
Vendored
+70
View File
@@ -0,0 +1,70 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
uint64_t hash(const void *key, size_t len, uint64_t seed);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_HASH_C_))
/*
* The following hash function is based on MurmurHash64A(), placed into the
* public domain by Austin Appleby. See http://murmurhash.googlepages.com/ for
* details.
*/
JEMALLOC_INLINE uint64_t
hash(const void *key, size_t len, uint64_t seed)
{
const uint64_t m = UINT64_C(0xc6a4a7935bd1e995);
const int r = 47;
uint64_t h = seed ^ (len * m);
const uint64_t *data = (const uint64_t *)key;
const uint64_t *end = data + (len/8);
const unsigned char *data2;
assert(((uintptr_t)key & 0x7) == 0);
while(data != end) {
uint64_t k = *data++;
k *= m;
k ^= k >> r;
k *= m;
h ^= k;
h *= m;
}
data2 = (const unsigned char *)data;
switch(len & 7) {
case 7: h ^= ((uint64_t)(data2[6])) << 48;
case 6: h ^= ((uint64_t)(data2[5])) << 40;
case 5: h ^= ((uint64_t)(data2[4])) << 32;
case 4: h ^= ((uint64_t)(data2[3])) << 24;
case 3: h ^= ((uint64_t)(data2[2])) << 16;
case 2: h ^= ((uint64_t)(data2[1])) << 8;
case 1: h ^= ((uint64_t)(data2[0]));
h *= m;
}
h ^= h >> r;
h *= m;
h ^= h >> r;
return (h);
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/huge.h
Vendored
+40
View File
@@ -0,0 +1,40 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
/* Huge allocation statistics. */
extern uint64_t huge_nmalloc;
extern uint64_t huge_ndalloc;
extern size_t huge_allocated;
/* Protects chunk-related data structures. */
extern malloc_mutex_t huge_mtx;
void *huge_malloc(size_t size, bool zero);
void *huge_palloc(size_t size, size_t alignment, bool zero);
void *huge_ralloc_no_move(void *ptr, size_t oldsize, size_t size,
size_t extra);
void *huge_ralloc(void *ptr, size_t oldsize, size_t size, size_t extra,
size_t alignment, bool zero, bool try_tcache_dalloc);
void huge_dalloc(void *ptr, bool unmap);
size_t huge_salloc(const void *ptr);
prof_ctx_t *huge_prof_ctx_get(const void *ptr);
void huge_prof_ctx_set(const void *ptr, prof_ctx_t *ctx);
bool huge_boot(void);
void huge_prefork(void);
void huge_postfork_parent(void);
void huge_postfork_child(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/jemalloc_internal.h.in
Vendored
+1010
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/include/jemalloc/internal/mb.h
Vendored
+115
View File
@@ -0,0 +1,115 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
void mb_write(void);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_MB_C_))
#ifdef __i386__
/*
* According to the Intel Architecture Software Developer's Manual, current
* processors execute instructions in order from the perspective of other
* processors in a multiprocessor system, but 1) Intel reserves the right to
* change that, and 2) the compiler's optimizer could re-order instructions if
* there weren't some form of barrier. Therefore, even if running on an
* architecture that does not need memory barriers (everything through at least
* i686), an "optimizer barrier" is necessary.
*/
JEMALLOC_INLINE void
mb_write(void)
{
# if 0
/* This is a true memory barrier. */
asm volatile ("pusha;"
"xor %%eax,%%eax;"
"cpuid;"
"popa;"
: /* Outputs. */
: /* Inputs. */
: "memory" /* Clobbers. */
);
#else
/*
* This is hopefully enough to keep the compiler from reordering
* instructions around this one.
*/
asm volatile ("nop;"
: /* Outputs. */
: /* Inputs. */
: "memory" /* Clobbers. */
);
#endif
}
#elif (defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE void
mb_write(void)
{
asm volatile ("sfence"
: /* Outputs. */
: /* Inputs. */
: "memory" /* Clobbers. */
);
}
#elif defined(__powerpc__)
JEMALLOC_INLINE void
mb_write(void)
{
asm volatile ("eieio"
: /* Outputs. */
: /* Inputs. */
: "memory" /* Clobbers. */
);
}
#elif defined(__sparc64__)
JEMALLOC_INLINE void
mb_write(void)
{
asm volatile ("membar #StoreStore"
: /* Outputs. */
: /* Inputs. */
: "memory" /* Clobbers. */
);
}
#elif defined(__tile__)
JEMALLOC_INLINE void
mb_write(void)
{
__sync_synchronize();
}
#else
/*
* This is much slower than a simple memory barrier, but the semantics of mutex
* unlock make this work.
*/
JEMALLOC_INLINE void
mb_write(void)
{
malloc_mutex_t mtx;
malloc_mutex_init(&mtx);
malloc_mutex_lock(&mtx);
malloc_mutex_unlock(&mtx);
}
#endif
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/mutex.h
Vendored
+99
View File
@@ -0,0 +1,99 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct malloc_mutex_s malloc_mutex_t;
#ifdef _WIN32
# define MALLOC_MUTEX_INITIALIZER
#elif (defined(JEMALLOC_OSSPIN))
# define MALLOC_MUTEX_INITIALIZER {0}
#elif (defined(JEMALLOC_MUTEX_INIT_CB))
# define MALLOC_MUTEX_INITIALIZER {PTHREAD_MUTEX_INITIALIZER, NULL}
#else
# if (defined(PTHREAD_MUTEX_ADAPTIVE_NP) && \
defined(PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP))
# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_ADAPTIVE_NP
# define MALLOC_MUTEX_INITIALIZER {PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP}
# else
# define MALLOC_MUTEX_TYPE PTHREAD_MUTEX_DEFAULT
# define MALLOC_MUTEX_INITIALIZER {PTHREAD_MUTEX_INITIALIZER}
# endif
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct malloc_mutex_s {
#ifdef _WIN32
CRITICAL_SECTION lock;
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLock lock;
#elif (defined(JEMALLOC_MUTEX_INIT_CB))
pthread_mutex_t lock;
malloc_mutex_t *postponed_next;
#else
pthread_mutex_t lock;
#endif
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#ifdef JEMALLOC_LAZY_LOCK
extern bool isthreaded;
#else
# undef isthreaded /* Undo private_namespace.h definition. */
# define isthreaded true
#endif
bool malloc_mutex_init(malloc_mutex_t *mutex);
void malloc_mutex_prefork(malloc_mutex_t *mutex);
void malloc_mutex_postfork_parent(malloc_mutex_t *mutex);
void malloc_mutex_postfork_child(malloc_mutex_t *mutex);
bool mutex_boot(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
void malloc_mutex_lock(malloc_mutex_t *mutex);
void malloc_mutex_unlock(malloc_mutex_t *mutex);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_MUTEX_C_))
JEMALLOC_INLINE void
malloc_mutex_lock(malloc_mutex_t *mutex)
{
if (isthreaded) {
#ifdef _WIN32
EnterCriticalSection(&mutex->lock);
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLockLock(&mutex->lock);
#else
pthread_mutex_lock(&mutex->lock);
#endif
}
}
JEMALLOC_INLINE void
malloc_mutex_unlock(malloc_mutex_t *mutex)
{
if (isthreaded) {
#ifdef _WIN32
LeaveCriticalSection(&mutex->lock);
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLockUnlock(&mutex->lock);
#else
pthread_mutex_unlock(&mutex->lock);
#endif
}
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/private_namespace.h
Vendored
+367
View File
@@ -0,0 +1,367 @@
#define a0calloc JEMALLOC_N(a0calloc)
#define a0free JEMALLOC_N(a0free)
#define a0malloc JEMALLOC_N(a0malloc)
#define arena_alloc_junk_small JEMALLOC_N(arena_alloc_junk_small)
#define arena_bin_index JEMALLOC_N(arena_bin_index)
#define arena_bin_info JEMALLOC_N(arena_bin_info)
#define arena_boot JEMALLOC_N(arena_boot)
#define arena_dalloc JEMALLOC_N(arena_dalloc)
#define arena_dalloc_bin JEMALLOC_N(arena_dalloc_bin)
#define arena_dalloc_bin_locked JEMALLOC_N(arena_dalloc_bin_locked)
#define arena_dalloc_junk_small JEMALLOC_N(arena_dalloc_junk_small)
#define arena_dalloc_large JEMALLOC_N(arena_dalloc_large)
#define arena_dalloc_large_locked JEMALLOC_N(arena_dalloc_large_locked)
#define arena_dalloc_small JEMALLOC_N(arena_dalloc_small)
#define arena_dss_prec_get JEMALLOC_N(arena_dss_prec_get)
#define arena_dss_prec_set JEMALLOC_N(arena_dss_prec_set)
#define arena_malloc JEMALLOC_N(arena_malloc)
#define arena_malloc_large JEMALLOC_N(arena_malloc_large)
#define arena_malloc_small JEMALLOC_N(arena_malloc_small)
#define arena_mapbits_allocated_get JEMALLOC_N(arena_mapbits_allocated_get)
#define arena_mapbits_binind_get JEMALLOC_N(arena_mapbits_binind_get)
#define arena_mapbits_dirty_get JEMALLOC_N(arena_mapbits_dirty_get)
#define arena_mapbits_get JEMALLOC_N(arena_mapbits_get)
#define arena_mapbits_large_binind_set JEMALLOC_N(arena_mapbits_large_binind_set)
#define arena_mapbits_large_get JEMALLOC_N(arena_mapbits_large_get)
#define arena_mapbits_large_set JEMALLOC_N(arena_mapbits_large_set)
#define arena_mapbits_large_size_get JEMALLOC_N(arena_mapbits_large_size_get)
#define arena_mapbits_small_runind_get JEMALLOC_N(arena_mapbits_small_runind_get)
#define arena_mapbits_small_set JEMALLOC_N(arena_mapbits_small_set)
#define arena_mapbits_unallocated_set JEMALLOC_N(arena_mapbits_unallocated_set)
#define arena_mapbits_unallocated_size_get JEMALLOC_N(arena_mapbits_unallocated_size_get)
#define arena_mapbits_unallocated_size_set JEMALLOC_N(arena_mapbits_unallocated_size_set)
#define arena_mapbits_unzeroed_get JEMALLOC_N(arena_mapbits_unzeroed_get)
#define arena_mapbits_unzeroed_set JEMALLOC_N(arena_mapbits_unzeroed_set)
#define arena_mapbitsp_get JEMALLOC_N(arena_mapbitsp_get)
#define arena_mapp_get JEMALLOC_N(arena_mapp_get)
#define arena_maxclass JEMALLOC_N(arena_maxclass)
#define arena_new JEMALLOC_N(arena_new)
#define arena_palloc JEMALLOC_N(arena_palloc)
#define arena_postfork_child JEMALLOC_N(arena_postfork_child)
#define arena_postfork_parent JEMALLOC_N(arena_postfork_parent)
#define arena_prefork JEMALLOC_N(arena_prefork)
#define arena_prof_accum JEMALLOC_N(arena_prof_accum)
#define arena_prof_ctx_get JEMALLOC_N(arena_prof_ctx_get)
#define arena_prof_ctx_set JEMALLOC_N(arena_prof_ctx_set)
#define arena_prof_promoted JEMALLOC_N(arena_prof_promoted)
#define arena_ptr_small_binind_get JEMALLOC_N(arena_ptr_small_binind_get)
#define arena_purge_all JEMALLOC_N(arena_purge_all)
#define arena_ralloc JEMALLOC_N(arena_ralloc)
#define arena_ralloc_no_move JEMALLOC_N(arena_ralloc_no_move)
#define arena_run_regind JEMALLOC_N(arena_run_regind)
#define arena_salloc JEMALLOC_N(arena_salloc)
#define arena_stats_merge JEMALLOC_N(arena_stats_merge)
#define arena_tcache_fill_small JEMALLOC_N(arena_tcache_fill_small)
#define arenas JEMALLOC_N(arenas)
#define arenas_booted JEMALLOC_N(arenas_booted)
#define arenas_cleanup JEMALLOC_N(arenas_cleanup)
#define arenas_extend JEMALLOC_N(arenas_extend)
#define arenas_initialized JEMALLOC_N(arenas_initialized)
#define arenas_lock JEMALLOC_N(arenas_lock)
#define arenas_tls JEMALLOC_N(arenas_tls)
#define arenas_tsd JEMALLOC_N(arenas_tsd)
#define arenas_tsd_boot JEMALLOC_N(arenas_tsd_boot)
#define arenas_tsd_cleanup_wrapper JEMALLOC_N(arenas_tsd_cleanup_wrapper)
#define arenas_tsd_get JEMALLOC_N(arenas_tsd_get)
#define arenas_tsd_set JEMALLOC_N(arenas_tsd_set)
#define atomic_add_u JEMALLOC_N(atomic_add_u)
#define atomic_add_uint32 JEMALLOC_N(atomic_add_uint32)
#define atomic_add_uint64 JEMALLOC_N(atomic_add_uint64)
#define atomic_add_z JEMALLOC_N(atomic_add_z)
#define atomic_sub_u JEMALLOC_N(atomic_sub_u)
#define atomic_sub_uint32 JEMALLOC_N(atomic_sub_uint32)
#define atomic_sub_uint64 JEMALLOC_N(atomic_sub_uint64)
#define atomic_sub_z JEMALLOC_N(atomic_sub_z)
#define base_alloc JEMALLOC_N(base_alloc)
#define base_boot JEMALLOC_N(base_boot)
#define base_calloc JEMALLOC_N(base_calloc)
#define base_node_alloc JEMALLOC_N(base_node_alloc)
#define base_node_dealloc JEMALLOC_N(base_node_dealloc)
#define base_postfork_child JEMALLOC_N(base_postfork_child)
#define base_postfork_parent JEMALLOC_N(base_postfork_parent)
#define base_prefork JEMALLOC_N(base_prefork)
#define bitmap_full JEMALLOC_N(bitmap_full)
#define bitmap_get JEMALLOC_N(bitmap_get)
#define bitmap_info_init JEMALLOC_N(bitmap_info_init)
#define bitmap_info_ngroups JEMALLOC_N(bitmap_info_ngroups)
#define bitmap_init JEMALLOC_N(bitmap_init)
#define bitmap_set JEMALLOC_N(bitmap_set)
#define bitmap_sfu JEMALLOC_N(bitmap_sfu)
#define bitmap_size JEMALLOC_N(bitmap_size)
#define bitmap_unset JEMALLOC_N(bitmap_unset)
#define bt_init JEMALLOC_N(bt_init)
#define buferror JEMALLOC_N(buferror)
#define choose_arena JEMALLOC_N(choose_arena)
#define choose_arena_hard JEMALLOC_N(choose_arena_hard)
#define chunk_alloc JEMALLOC_N(chunk_alloc)
#define chunk_alloc_dss JEMALLOC_N(chunk_alloc_dss)
#define chunk_alloc_mmap JEMALLOC_N(chunk_alloc_mmap)
#define chunk_boot JEMALLOC_N(chunk_boot)
#define chunk_dealloc JEMALLOC_N(chunk_dealloc)
#define chunk_dealloc_mmap JEMALLOC_N(chunk_dealloc_mmap)
#define chunk_dss_boot JEMALLOC_N(chunk_dss_boot)
#define chunk_dss_postfork_child JEMALLOC_N(chunk_dss_postfork_child)
#define chunk_dss_postfork_parent JEMALLOC_N(chunk_dss_postfork_parent)
#define chunk_dss_prec_get JEMALLOC_N(chunk_dss_prec_get)
#define chunk_dss_prec_set JEMALLOC_N(chunk_dss_prec_set)
#define chunk_dss_prefork JEMALLOC_N(chunk_dss_prefork)
#define chunk_in_dss JEMALLOC_N(chunk_in_dss)
#define chunk_npages JEMALLOC_N(chunk_npages)
#define chunk_postfork_child JEMALLOC_N(chunk_postfork_child)
#define chunk_postfork_parent JEMALLOC_N(chunk_postfork_parent)
#define chunk_prefork JEMALLOC_N(chunk_prefork)
#define chunk_unmap JEMALLOC_N(chunk_unmap)
#define chunks_mtx JEMALLOC_N(chunks_mtx)
#define chunks_rtree JEMALLOC_N(chunks_rtree)
#define chunksize JEMALLOC_N(chunksize)
#define chunksize_mask JEMALLOC_N(chunksize_mask)
#define ckh_bucket_search JEMALLOC_N(ckh_bucket_search)
#define ckh_count JEMALLOC_N(ckh_count)
#define ckh_delete JEMALLOC_N(ckh_delete)
#define ckh_evict_reloc_insert JEMALLOC_N(ckh_evict_reloc_insert)
#define ckh_insert JEMALLOC_N(ckh_insert)
#define ckh_isearch JEMALLOC_N(ckh_isearch)
#define ckh_iter JEMALLOC_N(ckh_iter)
#define ckh_new JEMALLOC_N(ckh_new)
#define ckh_pointer_hash JEMALLOC_N(ckh_pointer_hash)
#define ckh_pointer_keycomp JEMALLOC_N(ckh_pointer_keycomp)
#define ckh_rebuild JEMALLOC_N(ckh_rebuild)
#define ckh_remove JEMALLOC_N(ckh_remove)
#define ckh_search JEMALLOC_N(ckh_search)
#define ckh_string_hash JEMALLOC_N(ckh_string_hash)
#define ckh_string_keycomp JEMALLOC_N(ckh_string_keycomp)
#define ckh_try_bucket_insert JEMALLOC_N(ckh_try_bucket_insert)
#define ckh_try_insert JEMALLOC_N(ckh_try_insert)
#define ctl_boot JEMALLOC_N(ctl_boot)
#define ctl_bymib JEMALLOC_N(ctl_bymib)
#define ctl_byname JEMALLOC_N(ctl_byname)
#define ctl_nametomib JEMALLOC_N(ctl_nametomib)
#define ctl_postfork_child JEMALLOC_N(ctl_postfork_child)
#define ctl_postfork_parent JEMALLOC_N(ctl_postfork_parent)
#define ctl_prefork JEMALLOC_N(ctl_prefork)
#define dss_prec_names JEMALLOC_N(dss_prec_names)
#define extent_tree_ad_first JEMALLOC_N(extent_tree_ad_first)
#define extent_tree_ad_insert JEMALLOC_N(extent_tree_ad_insert)
#define extent_tree_ad_iter JEMALLOC_N(extent_tree_ad_iter)
#define extent_tree_ad_iter_recurse JEMALLOC_N(extent_tree_ad_iter_recurse)
#define extent_tree_ad_iter_start JEMALLOC_N(extent_tree_ad_iter_start)
#define extent_tree_ad_last JEMALLOC_N(extent_tree_ad_last)
#define extent_tree_ad_new JEMALLOC_N(extent_tree_ad_new)
#define extent_tree_ad_next JEMALLOC_N(extent_tree_ad_next)
#define extent_tree_ad_nsearch JEMALLOC_N(extent_tree_ad_nsearch)
#define extent_tree_ad_prev JEMALLOC_N(extent_tree_ad_prev)
#define extent_tree_ad_psearch JEMALLOC_N(extent_tree_ad_psearch)
#define extent_tree_ad_remove JEMALLOC_N(extent_tree_ad_remove)
#define extent_tree_ad_reverse_iter JEMALLOC_N(extent_tree_ad_reverse_iter)
#define extent_tree_ad_reverse_iter_recurse JEMALLOC_N(extent_tree_ad_reverse_iter_recurse)
#define extent_tree_ad_reverse_iter_start JEMALLOC_N(extent_tree_ad_reverse_iter_start)
#define extent_tree_ad_search JEMALLOC_N(extent_tree_ad_search)
#define extent_tree_szad_first JEMALLOC_N(extent_tree_szad_first)
#define extent_tree_szad_insert JEMALLOC_N(extent_tree_szad_insert)
#define extent_tree_szad_iter JEMALLOC_N(extent_tree_szad_iter)
#define extent_tree_szad_iter_recurse JEMALLOC_N(extent_tree_szad_iter_recurse)
#define extent_tree_szad_iter_start JEMALLOC_N(extent_tree_szad_iter_start)
#define extent_tree_szad_last JEMALLOC_N(extent_tree_szad_last)
#define extent_tree_szad_new JEMALLOC_N(extent_tree_szad_new)
#define extent_tree_szad_next JEMALLOC_N(extent_tree_szad_next)
#define extent_tree_szad_nsearch JEMALLOC_N(extent_tree_szad_nsearch)
#define extent_tree_szad_prev JEMALLOC_N(extent_tree_szad_prev)
#define extent_tree_szad_psearch JEMALLOC_N(extent_tree_szad_psearch)
#define extent_tree_szad_remove JEMALLOC_N(extent_tree_szad_remove)
#define extent_tree_szad_reverse_iter JEMALLOC_N(extent_tree_szad_reverse_iter)
#define extent_tree_szad_reverse_iter_recurse JEMALLOC_N(extent_tree_szad_reverse_iter_recurse)
#define extent_tree_szad_reverse_iter_start JEMALLOC_N(extent_tree_szad_reverse_iter_start)
#define extent_tree_szad_search JEMALLOC_N(extent_tree_szad_search)
#define get_errno JEMALLOC_N(get_errno)
#define hash JEMALLOC_N(hash)
#define huge_allocated JEMALLOC_N(huge_allocated)
#define huge_boot JEMALLOC_N(huge_boot)
#define huge_dalloc JEMALLOC_N(huge_dalloc)
#define huge_malloc JEMALLOC_N(huge_malloc)
#define huge_mtx JEMALLOC_N(huge_mtx)
#define huge_ndalloc JEMALLOC_N(huge_ndalloc)
#define huge_nmalloc JEMALLOC_N(huge_nmalloc)
#define huge_palloc JEMALLOC_N(huge_palloc)
#define huge_postfork_child JEMALLOC_N(huge_postfork_child)
#define huge_postfork_parent JEMALLOC_N(huge_postfork_parent)
#define huge_prefork JEMALLOC_N(huge_prefork)
#define huge_prof_ctx_get JEMALLOC_N(huge_prof_ctx_get)
#define huge_prof_ctx_set JEMALLOC_N(huge_prof_ctx_set)
#define huge_ralloc JEMALLOC_N(huge_ralloc)
#define huge_ralloc_no_move JEMALLOC_N(huge_ralloc_no_move)
#define huge_salloc JEMALLOC_N(huge_salloc)
#define iallocm JEMALLOC_N(iallocm)
#define icalloc JEMALLOC_N(icalloc)
#define icallocx JEMALLOC_N(icallocx)
#define idalloc JEMALLOC_N(idalloc)
#define idallocx JEMALLOC_N(idallocx)
#define imalloc JEMALLOC_N(imalloc)
#define imallocx JEMALLOC_N(imallocx)
#define ipalloc JEMALLOC_N(ipalloc)
#define ipallocx JEMALLOC_N(ipallocx)
#define iqalloc JEMALLOC_N(iqalloc)
#define iqallocx JEMALLOC_N(iqallocx)
#define iralloc JEMALLOC_N(iralloc)
#define irallocx JEMALLOC_N(irallocx)
#define isalloc JEMALLOC_N(isalloc)
#define isthreaded JEMALLOC_N(isthreaded)
#define ivsalloc JEMALLOC_N(ivsalloc)
#define jemalloc_postfork_child JEMALLOC_N(jemalloc_postfork_child)
#define jemalloc_postfork_parent JEMALLOC_N(jemalloc_postfork_parent)
#define jemalloc_prefork JEMALLOC_N(jemalloc_prefork)
#define malloc_cprintf JEMALLOC_N(malloc_cprintf)
#define malloc_mutex_init JEMALLOC_N(malloc_mutex_init)
#define malloc_mutex_lock JEMALLOC_N(malloc_mutex_lock)
#define malloc_mutex_postfork_child JEMALLOC_N(malloc_mutex_postfork_child)
#define malloc_mutex_postfork_parent JEMALLOC_N(malloc_mutex_postfork_parent)
#define malloc_mutex_prefork JEMALLOC_N(malloc_mutex_prefork)
#define malloc_mutex_unlock JEMALLOC_N(malloc_mutex_unlock)
#define malloc_printf JEMALLOC_N(malloc_printf)
#define malloc_snprintf JEMALLOC_N(malloc_snprintf)
#define malloc_strtoumax JEMALLOC_N(malloc_strtoumax)
#define malloc_tsd_boot JEMALLOC_N(malloc_tsd_boot)
#define malloc_tsd_cleanup_register JEMALLOC_N(malloc_tsd_cleanup_register)
#define malloc_tsd_dalloc JEMALLOC_N(malloc_tsd_dalloc)
#define malloc_tsd_malloc JEMALLOC_N(malloc_tsd_malloc)
#define malloc_tsd_no_cleanup JEMALLOC_N(malloc_tsd_no_cleanup)
#define malloc_vcprintf JEMALLOC_N(malloc_vcprintf)
#define malloc_vsnprintf JEMALLOC_N(malloc_vsnprintf)
#define malloc_write JEMALLOC_N(malloc_write)
#define map_bias JEMALLOC_N(map_bias)
#define mb_write JEMALLOC_N(mb_write)
#define mutex_boot JEMALLOC_N(mutex_boot)
#define narenas_auto JEMALLOC_N(narenas_auto)
#define narenas_total JEMALLOC_N(narenas_total)
#define narenas_total_get JEMALLOC_N(narenas_total_get)
#define ncpus JEMALLOC_N(ncpus)
#define nhbins JEMALLOC_N(nhbins)
#define opt_abort JEMALLOC_N(opt_abort)
#define opt_junk JEMALLOC_N(opt_junk)
#define opt_lg_chunk JEMALLOC_N(opt_lg_chunk)
#define opt_lg_dirty_mult JEMALLOC_N(opt_lg_dirty_mult)
#define opt_lg_prof_interval JEMALLOC_N(opt_lg_prof_interval)
#define opt_lg_prof_sample JEMALLOC_N(opt_lg_prof_sample)
#define opt_lg_tcache_max JEMALLOC_N(opt_lg_tcache_max)
#define opt_narenas JEMALLOC_N(opt_narenas)
#define opt_prof JEMALLOC_N(opt_prof)
#define opt_prof_accum JEMALLOC_N(opt_prof_accum)
#define opt_prof_active JEMALLOC_N(opt_prof_active)
#define opt_prof_final JEMALLOC_N(opt_prof_final)
#define opt_prof_gdump JEMALLOC_N(opt_prof_gdump)
#define opt_prof_leak JEMALLOC_N(opt_prof_leak)
#define opt_prof_prefix JEMALLOC_N(opt_prof_prefix)
#define opt_quarantine JEMALLOC_N(opt_quarantine)
#define opt_redzone JEMALLOC_N(opt_redzone)
#define opt_stats_print JEMALLOC_N(opt_stats_print)
#define opt_tcache JEMALLOC_N(opt_tcache)
#define opt_utrace JEMALLOC_N(opt_utrace)
#define opt_valgrind JEMALLOC_N(opt_valgrind)
#define opt_xmalloc JEMALLOC_N(opt_xmalloc)
#define opt_zero JEMALLOC_N(opt_zero)
#define p2rz JEMALLOC_N(p2rz)
#define pages_purge JEMALLOC_N(pages_purge)
#define pow2_ceil JEMALLOC_N(pow2_ceil)
#define prof_backtrace JEMALLOC_N(prof_backtrace)
#define prof_boot0 JEMALLOC_N(prof_boot0)
#define prof_boot1 JEMALLOC_N(prof_boot1)
#define prof_boot2 JEMALLOC_N(prof_boot2)
#define prof_ctx_get JEMALLOC_N(prof_ctx_get)
#define prof_ctx_set JEMALLOC_N(prof_ctx_set)
#define prof_free JEMALLOC_N(prof_free)
#define prof_gdump JEMALLOC_N(prof_gdump)
#define prof_idump JEMALLOC_N(prof_idump)
#define prof_interval JEMALLOC_N(prof_interval)
#define prof_lookup JEMALLOC_N(prof_lookup)
#define prof_malloc JEMALLOC_N(prof_malloc)
#define prof_mdump JEMALLOC_N(prof_mdump)
#define prof_postfork_child JEMALLOC_N(prof_postfork_child)
#define prof_postfork_parent JEMALLOC_N(prof_postfork_parent)
#define prof_prefork JEMALLOC_N(prof_prefork)
#define prof_promote JEMALLOC_N(prof_promote)
#define prof_realloc JEMALLOC_N(prof_realloc)
#define prof_sample_accum_update JEMALLOC_N(prof_sample_accum_update)
#define prof_sample_threshold_update JEMALLOC_N(prof_sample_threshold_update)
#define prof_tdata_booted JEMALLOC_N(prof_tdata_booted)
#define prof_tdata_cleanup JEMALLOC_N(prof_tdata_cleanup)
#define prof_tdata_get JEMALLOC_N(prof_tdata_get)
#define prof_tdata_init JEMALLOC_N(prof_tdata_init)
#define prof_tdata_initialized JEMALLOC_N(prof_tdata_initialized)
#define prof_tdata_tls JEMALLOC_N(prof_tdata_tls)
#define prof_tdata_tsd JEMALLOC_N(prof_tdata_tsd)
#define prof_tdata_tsd_boot JEMALLOC_N(prof_tdata_tsd_boot)
#define prof_tdata_tsd_cleanup_wrapper JEMALLOC_N(prof_tdata_tsd_cleanup_wrapper)
#define prof_tdata_tsd_get JEMALLOC_N(prof_tdata_tsd_get)
#define prof_tdata_tsd_set JEMALLOC_N(prof_tdata_tsd_set)
#define quarantine JEMALLOC_N(quarantine)
#define quarantine_boot JEMALLOC_N(quarantine_boot)
#define quarantine_tsd_boot JEMALLOC_N(quarantine_tsd_boot)
#define quarantine_tsd_cleanup_wrapper JEMALLOC_N(quarantine_tsd_cleanup_wrapper)
#define quarantine_tsd_get JEMALLOC_N(quarantine_tsd_get)
#define quarantine_tsd_set JEMALLOC_N(quarantine_tsd_set)
#define register_zone JEMALLOC_N(register_zone)
#define rtree_get JEMALLOC_N(rtree_get)
#define rtree_get_locked JEMALLOC_N(rtree_get_locked)
#define rtree_new JEMALLOC_N(rtree_new)
#define rtree_postfork_child JEMALLOC_N(rtree_postfork_child)
#define rtree_postfork_parent JEMALLOC_N(rtree_postfork_parent)
#define rtree_prefork JEMALLOC_N(rtree_prefork)
#define rtree_set JEMALLOC_N(rtree_set)
#define s2u JEMALLOC_N(s2u)
#define sa2u JEMALLOC_N(sa2u)
#define set_errno JEMALLOC_N(set_errno)
#define stats_cactive JEMALLOC_N(stats_cactive)
#define stats_cactive_add JEMALLOC_N(stats_cactive_add)
#define stats_cactive_get JEMALLOC_N(stats_cactive_get)
#define stats_cactive_sub JEMALLOC_N(stats_cactive_sub)
#define stats_chunks JEMALLOC_N(stats_chunks)
#define stats_print JEMALLOC_N(stats_print)
#define tcache_alloc_easy JEMALLOC_N(tcache_alloc_easy)
#define tcache_alloc_large JEMALLOC_N(tcache_alloc_large)
#define tcache_alloc_small JEMALLOC_N(tcache_alloc_small)
#define tcache_alloc_small_hard JEMALLOC_N(tcache_alloc_small_hard)
#define tcache_arena_associate JEMALLOC_N(tcache_arena_associate)
#define tcache_arena_dissociate JEMALLOC_N(tcache_arena_dissociate)
#define tcache_bin_flush_large JEMALLOC_N(tcache_bin_flush_large)
#define tcache_bin_flush_small JEMALLOC_N(tcache_bin_flush_small)
#define tcache_bin_info JEMALLOC_N(tcache_bin_info)
#define tcache_boot0 JEMALLOC_N(tcache_boot0)
#define tcache_boot1 JEMALLOC_N(tcache_boot1)
#define tcache_booted JEMALLOC_N(tcache_booted)
#define tcache_create JEMALLOC_N(tcache_create)
#define tcache_dalloc_large JEMALLOC_N(tcache_dalloc_large)
#define tcache_dalloc_small JEMALLOC_N(tcache_dalloc_small)
#define tcache_destroy JEMALLOC_N(tcache_destroy)
#define tcache_enabled_booted JEMALLOC_N(tcache_enabled_booted)
#define tcache_enabled_get JEMALLOC_N(tcache_enabled_get)
#define tcache_enabled_initialized JEMALLOC_N(tcache_enabled_initialized)
#define tcache_enabled_set JEMALLOC_N(tcache_enabled_set)
#define tcache_enabled_tls JEMALLOC_N(tcache_enabled_tls)
#define tcache_enabled_tsd JEMALLOC_N(tcache_enabled_tsd)
#define tcache_enabled_tsd_boot JEMALLOC_N(tcache_enabled_tsd_boot)
#define tcache_enabled_tsd_cleanup_wrapper JEMALLOC_N(tcache_enabled_tsd_cleanup_wrapper)
#define tcache_enabled_tsd_get JEMALLOC_N(tcache_enabled_tsd_get)
#define tcache_enabled_tsd_set JEMALLOC_N(tcache_enabled_tsd_set)
#define tcache_event JEMALLOC_N(tcache_event)
#define tcache_event_hard JEMALLOC_N(tcache_event_hard)
#define tcache_flush JEMALLOC_N(tcache_flush)
#define tcache_get JEMALLOC_N(tcache_get)
#define tcache_initialized JEMALLOC_N(tcache_initialized)
#define tcache_maxclass JEMALLOC_N(tcache_maxclass)
#define tcache_salloc JEMALLOC_N(tcache_salloc)
#define tcache_stats_merge JEMALLOC_N(tcache_stats_merge)
#define tcache_thread_cleanup JEMALLOC_N(tcache_thread_cleanup)
#define tcache_tls JEMALLOC_N(tcache_tls)
#define tcache_tsd JEMALLOC_N(tcache_tsd)
#define tcache_tsd_boot JEMALLOC_N(tcache_tsd_boot)
#define tcache_tsd_cleanup_wrapper JEMALLOC_N(tcache_tsd_cleanup_wrapper)
#define tcache_tsd_get JEMALLOC_N(tcache_tsd_get)
#define tcache_tsd_set JEMALLOC_N(tcache_tsd_set)
#define thread_allocated_booted JEMALLOC_N(thread_allocated_booted)
#define thread_allocated_initialized JEMALLOC_N(thread_allocated_initialized)
#define thread_allocated_tls JEMALLOC_N(thread_allocated_tls)
#define thread_allocated_tsd JEMALLOC_N(thread_allocated_tsd)
#define thread_allocated_tsd_boot JEMALLOC_N(thread_allocated_tsd_boot)
#define thread_allocated_tsd_cleanup_wrapper JEMALLOC_N(thread_allocated_tsd_cleanup_wrapper)
#define thread_allocated_tsd_get JEMALLOC_N(thread_allocated_tsd_get)
#define thread_allocated_tsd_set JEMALLOC_N(thread_allocated_tsd_set)
#define u2rz JEMALLOC_N(u2rz)
deps/jemalloc/include/jemalloc/internal/prng.h
Vendored
+60
View File
@@ -0,0 +1,60 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/*
* Simple linear congruential pseudo-random number generator:
*
* prng(y) = (a*x + c) % m
*
* where the following constants ensure maximal period:
*
* a == Odd number (relatively prime to 2^n), and (a-1) is a multiple of 4.
* c == Odd number (relatively prime to 2^n).
* m == 2^32
*
* See Knuth's TAOCP 3rd Ed., Vol. 2, pg. 17 for details on these constraints.
*
* This choice of m has the disadvantage that the quality of the bits is
* proportional to bit position. For example. the lowest bit has a cycle of 2,
* the next has a cycle of 4, etc. For this reason, we prefer to use the upper
* bits.
*
* Macro parameters:
* uint32_t r : Result.
* unsigned lg_range : (0..32], number of least significant bits to return.
* uint32_t state : Seed value.
* const uint32_t a, c : See above discussion.
*/
#define prng32(r, lg_range, state, a, c) do { \
assert(lg_range > 0); \
assert(lg_range <= 32); \
\
r = (state * (a)) + (c); \
state = r; \
r >>= (32 - lg_range); \
} while (false)
/* Same as prng32(), but 64 bits of pseudo-randomness, using uint64_t. */
#define prng64(r, lg_range, state, a, c) do { \
assert(lg_range > 0); \
assert(lg_range <= 64); \
\
r = (state * (a)) + (c); \
state = r; \
r >>= (64 - lg_range); \
} while (false)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/prof.h
Vendored
+581
View File
@@ -0,0 +1,581 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct prof_bt_s prof_bt_t;
typedef struct prof_cnt_s prof_cnt_t;
typedef struct prof_thr_cnt_s prof_thr_cnt_t;
typedef struct prof_ctx_s prof_ctx_t;
typedef struct prof_tdata_s prof_tdata_t;
/* Option defaults. */
#define PROF_PREFIX_DEFAULT "jeprof"
#define LG_PROF_SAMPLE_DEFAULT 19
#define LG_PROF_INTERVAL_DEFAULT -1
/*
* Hard limit on stack backtrace depth. The version of prof_backtrace() that
* is based on __builtin_return_address() necessarily has a hard-coded number
* of backtrace frame handlers, and should be kept in sync with this setting.
*/
#define PROF_BT_MAX 128
/* Maximum number of backtraces to store in each per thread LRU cache. */
#define PROF_TCMAX 1024
/* Initial hash table size. */
#define PROF_CKH_MINITEMS 64
/* Size of memory buffer to use when writing dump files. */
#define PROF_DUMP_BUFSIZE 65536
/* Size of stack-allocated buffer used by prof_printf(). */
#define PROF_PRINTF_BUFSIZE 128
/*
* Number of mutexes shared among all ctx's. No space is allocated for these
* unless profiling is enabled, so it's okay to over-provision.
*/
#define PROF_NCTX_LOCKS 1024
/*
* prof_tdata pointers close to NULL are used to encode state information that
* is used for cleaning up during thread shutdown.
*/
#define PROF_TDATA_STATE_REINCARNATED ((prof_tdata_t *)(uintptr_t)1)
#define PROF_TDATA_STATE_PURGATORY ((prof_tdata_t *)(uintptr_t)2)
#define PROF_TDATA_STATE_MAX PROF_TDATA_STATE_PURGATORY
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct prof_bt_s {
/* Backtrace, stored as len program counters. */
void **vec;
unsigned len;
};
#ifdef JEMALLOC_PROF_LIBGCC
/* Data structure passed to libgcc _Unwind_Backtrace() callback functions. */
typedef struct {
prof_bt_t *bt;
unsigned nignore;
unsigned max;
} prof_unwind_data_t;
#endif
struct prof_cnt_s {
/*
* Profiling counters. An allocation/deallocation pair can operate on
* different prof_thr_cnt_t objects that are linked into the same
* prof_ctx_t cnts_ql, so it is possible for the cur* counters to go
* negative. In principle it is possible for the *bytes counters to
* overflow/underflow, but a general solution would require something
* like 128-bit counters; this implementation doesn't bother to solve
* that problem.
*/
int64_t curobjs;
int64_t curbytes;
uint64_t accumobjs;
uint64_t accumbytes;
};
struct prof_thr_cnt_s {
/* Linkage into prof_ctx_t's cnts_ql. */
ql_elm(prof_thr_cnt_t) cnts_link;
/* Linkage into thread's LRU. */
ql_elm(prof_thr_cnt_t) lru_link;
/*
* Associated context. If a thread frees an object that it did not
* allocate, it is possible that the context is not cached in the
* thread's hash table, in which case it must be able to look up the
* context, insert a new prof_thr_cnt_t into the thread's hash table,
* and link it into the prof_ctx_t's cnts_ql.
*/
prof_ctx_t *ctx;
/*
* Threads use memory barriers to update the counters. Since there is
* only ever one writer, the only challenge is for the reader to get a
* consistent read of the counters.
*
* The writer uses this series of operations:
*
* 1) Increment epoch to an odd number.
* 2) Update counters.
* 3) Increment epoch to an even number.
*
* The reader must assure 1) that the epoch is even while it reads the
* counters, and 2) that the epoch doesn't change between the time it
* starts and finishes reading the counters.
*/
unsigned epoch;
/* Profiling counters. */
prof_cnt_t cnts;
};
struct prof_ctx_s {
/* Associated backtrace. */
prof_bt_t *bt;
/* Protects nlimbo, cnt_merged, and cnts_ql. */
malloc_mutex_t *lock;
/*
* Number of threads that currently cause this ctx to be in a state of
* limbo due to one of:
* - Initializing per thread counters associated with this ctx.
* - Preparing to destroy this ctx.
* nlimbo must be 1 (single destroyer) in order to safely destroy the
* ctx.
*/
unsigned nlimbo;
/* Temporary storage for summation during dump. */
prof_cnt_t cnt_summed;
/* When threads exit, they merge their stats into cnt_merged. */
prof_cnt_t cnt_merged;
/*
* List of profile counters, one for each thread that has allocated in
* this context.
*/
ql_head(prof_thr_cnt_t) cnts_ql;
};
struct prof_tdata_s {
/*
* Hash of (prof_bt_t *)-->(prof_thr_cnt_t *). Each thread keeps a
* cache of backtraces, with associated thread-specific prof_thr_cnt_t
* objects. Other threads may read the prof_thr_cnt_t contents, but no
* others will ever write them.
*
* Upon thread exit, the thread must merge all the prof_thr_cnt_t
* counter data into the associated prof_ctx_t objects, and unlink/free
* the prof_thr_cnt_t objects.
*/
ckh_t bt2cnt;
/* LRU for contents of bt2cnt. */
ql_head(prof_thr_cnt_t) lru_ql;
/* Backtrace vector, used for calls to prof_backtrace(). */
void **vec;
/* Sampling state. */
uint64_t prng_state;
uint64_t threshold;
uint64_t accum;
/* State used to avoid dumping while operating on prof internals. */
bool enq;
bool enq_idump;
bool enq_gdump;
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern bool opt_prof;
/*
* Even if opt_prof is true, sampling can be temporarily disabled by setting
* opt_prof_active to false. No locking is used when updating opt_prof_active,
* so there are no guarantees regarding how long it will take for all threads
* to notice state changes.
*/
extern bool opt_prof_active;
extern size_t opt_lg_prof_sample; /* Mean bytes between samples. */
extern ssize_t opt_lg_prof_interval; /* lg(prof_interval). */
extern bool opt_prof_gdump; /* High-water memory dumping. */
extern bool opt_prof_final; /* Final profile dumping. */
extern bool opt_prof_leak; /* Dump leak summary at exit. */
extern bool opt_prof_accum; /* Report cumulative bytes. */
extern char opt_prof_prefix[PATH_MAX + 1];
/*
* Profile dump interval, measured in bytes allocated. Each arena triggers a
* profile dump when it reaches this threshold. The effect is that the
* interval between profile dumps averages prof_interval, though the actual
* interval between dumps will tend to be sporadic, and the interval will be a
* maximum of approximately (prof_interval * narenas).
*/
extern uint64_t prof_interval;
/*
* If true, promote small sampled objects to large objects, since small run
* headers do not have embedded profile context pointers.
*/
extern bool prof_promote;
void bt_init(prof_bt_t *bt, void **vec);
void prof_backtrace(prof_bt_t *bt, unsigned nignore);
prof_thr_cnt_t *prof_lookup(prof_bt_t *bt);
void prof_idump(void);
bool prof_mdump(const char *filename);
void prof_gdump(void);
prof_tdata_t *prof_tdata_init(void);
void prof_tdata_cleanup(void *arg);
void prof_boot0(void);
void prof_boot1(void);
bool prof_boot2(void);
void prof_prefork(void);
void prof_postfork_parent(void);
void prof_postfork_child(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#define PROF_ALLOC_PREP(nignore, size, ret) do { \
prof_tdata_t *prof_tdata; \
prof_bt_t bt; \
\
assert(size == s2u(size)); \
\
prof_tdata = prof_tdata_get(); \
if ((uintptr_t)prof_tdata <= (uintptr_t)PROF_TDATA_STATE_MAX) { \
if (prof_tdata != NULL) \
ret = (prof_thr_cnt_t *)(uintptr_t)1U; \
else \
ret = NULL; \
break; \
} \
\
if (opt_prof_active == false) { \
/* Sampling is currently inactive, so avoid sampling. */\
ret = (prof_thr_cnt_t *)(uintptr_t)1U; \
} else if (opt_lg_prof_sample == 0) { \
/* Don't bother with sampling logic, since sampling */\
/* interval is 1. */\
bt_init(&bt, prof_tdata->vec); \
prof_backtrace(&bt, nignore); \
ret = prof_lookup(&bt); \
} else { \
if (prof_tdata->threshold == 0) { \
/* Initialize. Seed the prng differently for */\
/* each thread. */\
prof_tdata->prng_state = \
(uint64_t)(uintptr_t)&size; \
prof_sample_threshold_update(prof_tdata); \
} \
\
/* Determine whether to capture a backtrace based on */\
/* whether size is enough for prof_accum to reach */\
/* prof_tdata->threshold. However, delay updating */\
/* these variables until prof_{m,re}alloc(), because */\
/* we don't know for sure that the allocation will */\
/* succeed. */\
/* */\
/* Use subtraction rather than addition to avoid */\
/* potential integer overflow. */\
if (size >= prof_tdata->threshold - \
prof_tdata->accum) { \
bt_init(&bt, prof_tdata->vec); \
prof_backtrace(&bt, nignore); \
ret = prof_lookup(&bt); \
} else \
ret = (prof_thr_cnt_t *)(uintptr_t)1U; \
} \
} while (0)
#ifndef JEMALLOC_ENABLE_INLINE
malloc_tsd_protos(JEMALLOC_ATTR(unused), prof_tdata, prof_tdata_t *)
prof_tdata_t *prof_tdata_get(void);
void prof_sample_threshold_update(prof_tdata_t *prof_tdata);
prof_ctx_t *prof_ctx_get(const void *ptr);
void prof_ctx_set(const void *ptr, prof_ctx_t *ctx);
bool prof_sample_accum_update(size_t size);
void prof_malloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt);
void prof_realloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt,
size_t old_size, prof_ctx_t *old_ctx);
void prof_free(const void *ptr, size_t size);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_PROF_C_))
/* Thread-specific backtrace cache, used to reduce bt2ctx contention. */
malloc_tsd_externs(prof_tdata, prof_tdata_t *)
malloc_tsd_funcs(JEMALLOC_INLINE, prof_tdata, prof_tdata_t *, NULL,
prof_tdata_cleanup)
JEMALLOC_INLINE prof_tdata_t *
prof_tdata_get(void)
{
prof_tdata_t *prof_tdata;
cassert(config_prof);
prof_tdata = *prof_tdata_tsd_get();
if ((uintptr_t)prof_tdata <= (uintptr_t)PROF_TDATA_STATE_MAX) {
if (prof_tdata == NULL)
prof_tdata = prof_tdata_init();
}
return (prof_tdata);
}
JEMALLOC_INLINE void
prof_sample_threshold_update(prof_tdata_t *prof_tdata)
{
uint64_t r;
double u;
cassert(config_prof);
/*
* Compute sample threshold as a geometrically distributed random
* variable with mean (2^opt_lg_prof_sample).
*
* __ __
* | log(u) | 1
* prof_tdata->threshold = | -------- |, where p = -------------------
* | log(1-p) | opt_lg_prof_sample
* 2
*
* For more information on the math, see:
*
* Non-Uniform Random Variate Generation
* Luc Devroye
* Springer-Verlag, New York, 1986
* pp 500
* (http://cg.scs.carleton.ca/~luc/rnbookindex.html)
*/
prng64(r, 53, prof_tdata->prng_state,
UINT64_C(6364136223846793005), UINT64_C(1442695040888963407));
u = (double)r * (1.0/9007199254740992.0L);
prof_tdata->threshold = (uint64_t)(log(u) /
log(1.0 - (1.0 / (double)((uint64_t)1U << opt_lg_prof_sample))))
+ (uint64_t)1U;
}
JEMALLOC_INLINE prof_ctx_t *
prof_ctx_get(const void *ptr)
{
prof_ctx_t *ret;
arena_chunk_t *chunk;
cassert(config_prof);
assert(ptr != NULL);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
if (chunk != ptr) {
/* Region. */
ret = arena_prof_ctx_get(ptr);
} else
ret = huge_prof_ctx_get(ptr);
return (ret);
}
JEMALLOC_INLINE void
prof_ctx_set(const void *ptr, prof_ctx_t *ctx)
{
arena_chunk_t *chunk;
cassert(config_prof);
assert(ptr != NULL);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
if (chunk != ptr) {
/* Region. */
arena_prof_ctx_set(ptr, ctx);
} else
huge_prof_ctx_set(ptr, ctx);
}
JEMALLOC_INLINE bool
prof_sample_accum_update(size_t size)
{
prof_tdata_t *prof_tdata;
cassert(config_prof);
/* Sampling logic is unnecessary if the interval is 1. */
assert(opt_lg_prof_sample != 0);
prof_tdata = *prof_tdata_tsd_get();
if ((uintptr_t)prof_tdata <= (uintptr_t)PROF_TDATA_STATE_MAX)
return (true);
/* Take care to avoid integer overflow. */
if (size >= prof_tdata->threshold - prof_tdata->accum) {
prof_tdata->accum -= (prof_tdata->threshold - size);
/* Compute new sample threshold. */
prof_sample_threshold_update(prof_tdata);
while (prof_tdata->accum >= prof_tdata->threshold) {
prof_tdata->accum -= prof_tdata->threshold;
prof_sample_threshold_update(prof_tdata);
}
return (false);
} else {
prof_tdata->accum += size;
return (true);
}
}
JEMALLOC_INLINE void
prof_malloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt)
{
cassert(config_prof);
assert(ptr != NULL);
assert(size == isalloc(ptr, true));
if (opt_lg_prof_sample != 0) {
if (prof_sample_accum_update(size)) {
/*
* Don't sample. For malloc()-like allocation, it is
* always possible to tell in advance how large an
* object's usable size will be, so there should never
* be a difference between the size passed to
* PROF_ALLOC_PREP() and prof_malloc().
*/
assert((uintptr_t)cnt == (uintptr_t)1U);
}
}
if ((uintptr_t)cnt > (uintptr_t)1U) {
prof_ctx_set(ptr, cnt->ctx);
cnt->epoch++;
/*********/
mb_write();
/*********/
cnt->cnts.curobjs++;
cnt->cnts.curbytes += size;
if (opt_prof_accum) {
cnt->cnts.accumobjs++;
cnt->cnts.accumbytes += size;
}
/*********/
mb_write();
/*********/
cnt->epoch++;
/*********/
mb_write();
/*********/
} else
prof_ctx_set(ptr, (prof_ctx_t *)(uintptr_t)1U);
}
JEMALLOC_INLINE void
prof_realloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt,
size_t old_size, prof_ctx_t *old_ctx)
{
prof_thr_cnt_t *told_cnt;
cassert(config_prof);
assert(ptr != NULL || (uintptr_t)cnt <= (uintptr_t)1U);
if (ptr != NULL) {
assert(size == isalloc(ptr, true));
if (opt_lg_prof_sample != 0) {
if (prof_sample_accum_update(size)) {
/*
* Don't sample. The size passed to
* PROF_ALLOC_PREP() was larger than what
* actually got allocated, so a backtrace was
* captured for this allocation, even though
* its actual size was insufficient to cross
* the sample threshold.
*/
cnt = (prof_thr_cnt_t *)(uintptr_t)1U;
}
}
}
if ((uintptr_t)old_ctx > (uintptr_t)1U) {
told_cnt = prof_lookup(old_ctx->bt);
if (told_cnt == NULL) {
/*
* It's too late to propagate OOM for this realloc(),
* so operate directly on old_cnt->ctx->cnt_merged.
*/
malloc_mutex_lock(old_ctx->lock);
old_ctx->cnt_merged.curobjs--;
old_ctx->cnt_merged.curbytes -= old_size;
malloc_mutex_unlock(old_ctx->lock);
told_cnt = (prof_thr_cnt_t *)(uintptr_t)1U;
}
} else
told_cnt = (prof_thr_cnt_t *)(uintptr_t)1U;
if ((uintptr_t)told_cnt > (uintptr_t)1U)
told_cnt->epoch++;
if ((uintptr_t)cnt > (uintptr_t)1U) {
prof_ctx_set(ptr, cnt->ctx);
cnt->epoch++;
} else if (ptr != NULL)
prof_ctx_set(ptr, (prof_ctx_t *)(uintptr_t)1U);
/*********/
mb_write();
/*********/
if ((uintptr_t)told_cnt > (uintptr_t)1U) {
told_cnt->cnts.curobjs--;
told_cnt->cnts.curbytes -= old_size;
}
if ((uintptr_t)cnt > (uintptr_t)1U) {
cnt->cnts.curobjs++;
cnt->cnts.curbytes += size;
if (opt_prof_accum) {
cnt->cnts.accumobjs++;
cnt->cnts.accumbytes += size;
}
}
/*********/
mb_write();
/*********/
if ((uintptr_t)told_cnt > (uintptr_t)1U)
told_cnt->epoch++;
if ((uintptr_t)cnt > (uintptr_t)1U)
cnt->epoch++;
/*********/
mb_write(); /* Not strictly necessary. */
}
JEMALLOC_INLINE void
prof_free(const void *ptr, size_t size)
{
prof_ctx_t *ctx = prof_ctx_get(ptr);
cassert(config_prof);
if ((uintptr_t)ctx > (uintptr_t)1) {
prof_thr_cnt_t *tcnt;
assert(size == isalloc(ptr, true));
tcnt = prof_lookup(ctx->bt);
if (tcnt != NULL) {
tcnt->epoch++;
/*********/
mb_write();
/*********/
tcnt->cnts.curobjs--;
tcnt->cnts.curbytes -= size;
/*********/
mb_write();
/*********/
tcnt->epoch++;
/*********/
mb_write();
/*********/
} else {
/*
* OOM during free() cannot be propagated, so operate
* directly on cnt->ctx->cnt_merged.
*/
malloc_mutex_lock(ctx->lock);
ctx->cnt_merged.curobjs--;
ctx->cnt_merged.curbytes -= size;
malloc_mutex_unlock(ctx->lock);
}
}
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/ql.h
Vendored
+83
View File
@@ -0,0 +1,83 @@
/*
* List definitions.
*/
#define ql_head(a_type) \
struct { \
a_type *qlh_first; \
}
#define ql_head_initializer(a_head) {NULL}
#define ql_elm(a_type) qr(a_type)
/* List functions. */
#define ql_new(a_head) do { \
(a_head)->qlh_first = NULL; \
} while (0)
#define ql_elm_new(a_elm, a_field) qr_new((a_elm), a_field)
#define ql_first(a_head) ((a_head)->qlh_first)
#define ql_last(a_head, a_field) \
((ql_first(a_head) != NULL) \
? qr_prev(ql_first(a_head), a_field) : NULL)
#define ql_next(a_head, a_elm, a_field) \
((ql_last(a_head, a_field) != (a_elm)) \
? qr_next((a_elm), a_field) : NULL)
#define ql_prev(a_head, a_elm, a_field) \
((ql_first(a_head) != (a_elm)) ? qr_prev((a_elm), a_field) \
: NULL)
#define ql_before_insert(a_head, a_qlelm, a_elm, a_field) do { \
qr_before_insert((a_qlelm), (a_elm), a_field); \
if (ql_first(a_head) == (a_qlelm)) { \
ql_first(a_head) = (a_elm); \
} \
} while (0)
#define ql_after_insert(a_qlelm, a_elm, a_field) \
qr_after_insert((a_qlelm), (a_elm), a_field)
#define ql_head_insert(a_head, a_elm, a_field) do { \
if (ql_first(a_head) != NULL) { \
qr_before_insert(ql_first(a_head), (a_elm), a_field); \
} \
ql_first(a_head) = (a_elm); \
} while (0)
#define ql_tail_insert(a_head, a_elm, a_field) do { \
if (ql_first(a_head) != NULL) { \
qr_before_insert(ql_first(a_head), (a_elm), a_field); \
} \
ql_first(a_head) = qr_next((a_elm), a_field); \
} while (0)
#define ql_remove(a_head, a_elm, a_field) do { \
if (ql_first(a_head) == (a_elm)) { \
ql_first(a_head) = qr_next(ql_first(a_head), a_field); \
} \
if (ql_first(a_head) != (a_elm)) { \
qr_remove((a_elm), a_field); \
} else { \
ql_first(a_head) = NULL; \
} \
} while (0)
#define ql_head_remove(a_head, a_type, a_field) do { \
a_type *t = ql_first(a_head); \
ql_remove((a_head), t, a_field); \
} while (0)
#define ql_tail_remove(a_head, a_type, a_field) do { \
a_type *t = ql_last(a_head, a_field); \
ql_remove((a_head), t, a_field); \
} while (0)
#define ql_foreach(a_var, a_head, a_field) \
qr_foreach((a_var), ql_first(a_head), a_field)
#define ql_reverse_foreach(a_var, a_head, a_field) \
qr_reverse_foreach((a_var), ql_first(a_head), a_field)
deps/jemalloc/include/jemalloc/internal/qr.h
Vendored
+67
View File
@@ -0,0 +1,67 @@
/* Ring definitions. */
#define qr(a_type) \
struct { \
a_type *qre_next; \
a_type *qre_prev; \
}
/* Ring functions. */
#define qr_new(a_qr, a_field) do { \
(a_qr)->a_field.qre_next = (a_qr); \
(a_qr)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_next(a_qr, a_field) ((a_qr)->a_field.qre_next)
#define qr_prev(a_qr, a_field) ((a_qr)->a_field.qre_prev)
#define qr_before_insert(a_qrelm, a_qr, a_field) do { \
(a_qr)->a_field.qre_prev = (a_qrelm)->a_field.qre_prev; \
(a_qr)->a_field.qre_next = (a_qrelm); \
(a_qr)->a_field.qre_prev->a_field.qre_next = (a_qr); \
(a_qrelm)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_after_insert(a_qrelm, a_qr, a_field) \
do \
{ \
(a_qr)->a_field.qre_next = (a_qrelm)->a_field.qre_next; \
(a_qr)->a_field.qre_prev = (a_qrelm); \
(a_qr)->a_field.qre_next->a_field.qre_prev = (a_qr); \
(a_qrelm)->a_field.qre_next = (a_qr); \
} while (0)
#define qr_meld(a_qr_a, a_qr_b, a_field) do { \
void *t; \
(a_qr_a)->a_field.qre_prev->a_field.qre_next = (a_qr_b); \
(a_qr_b)->a_field.qre_prev->a_field.qre_next = (a_qr_a); \
t = (a_qr_a)->a_field.qre_prev; \
(a_qr_a)->a_field.qre_prev = (a_qr_b)->a_field.qre_prev; \
(a_qr_b)->a_field.qre_prev = t; \
} while (0)
/* qr_meld() and qr_split() are functionally equivalent, so there's no need to
* have two copies of the code. */
#define qr_split(a_qr_a, a_qr_b, a_field) \
qr_meld((a_qr_a), (a_qr_b), a_field)
#define qr_remove(a_qr, a_field) do { \
(a_qr)->a_field.qre_prev->a_field.qre_next \
= (a_qr)->a_field.qre_next; \
(a_qr)->a_field.qre_next->a_field.qre_prev \
= (a_qr)->a_field.qre_prev; \
(a_qr)->a_field.qre_next = (a_qr); \
(a_qr)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_foreach(var, a_qr, a_field) \
for ((var) = (a_qr); \
(var) != NULL; \
(var) = (((var)->a_field.qre_next != (a_qr)) \
? (var)->a_field.qre_next : NULL))
#define qr_reverse_foreach(var, a_qr, a_field) \
for ((var) = ((a_qr) != NULL) ? qr_prev(a_qr, a_field) : NULL; \
(var) != NULL; \
(var) = (((var) != (a_qr)) \
? (var)->a_field.qre_prev : NULL))
deps/jemalloc/include/jemalloc/internal/quarantine.h
Vendored
+24
View File
@@ -0,0 +1,24 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/* Default per thread quarantine size if valgrind is enabled. */
#define JEMALLOC_VALGRIND_QUARANTINE_DEFAULT (ZU(1) << 24)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
void quarantine(void *ptr);
bool quarantine_boot(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/rb.h
Vendored
+973
View File
@@ -0,0 +1,973 @@
/*-
*******************************************************************************
*
* cpp macro implementation of left-leaning 2-3 red-black trees. Parent
* pointers are not used, and color bits are stored in the least significant
* bit of right-child pointers (if RB_COMPACT is defined), thus making node
* linkage as compact as is possible for red-black trees.
*
* Usage:
*
* #include <stdint.h>
* #include <stdbool.h>
* #define NDEBUG // (Optional, see assert(3).)
* #include <assert.h>
* #define RB_COMPACT // (Optional, embed color bits in right-child pointers.)
* #include <rb.h>
* ...
*
*******************************************************************************
*/
#ifndef RB_H_
#define RB_H_
#if 0
__FBSDID("$FreeBSD: head/lib/libc/stdlib/rb.h 204493 2010-02-28 22:57:13Z jasone $");
#endif
#ifdef RB_COMPACT
/* Node structure. */
#define rb_node(a_type) \
struct { \
a_type *rbn_left; \
a_type *rbn_right_red; \
}
#else
#define rb_node(a_type) \
struct { \
a_type *rbn_left; \
a_type *rbn_right; \
bool rbn_red; \
}
#endif
/* Root structure. */
#define rb_tree(a_type) \
struct { \
a_type *rbt_root; \
a_type rbt_nil; \
}
/* Left accessors. */
#define rbtn_left_get(a_type, a_field, a_node) \
((a_node)->a_field.rbn_left)
#define rbtn_left_set(a_type, a_field, a_node, a_left) do { \
(a_node)->a_field.rbn_left = a_left; \
} while (0)
#ifdef RB_COMPACT
/* Right accessors. */
#define rbtn_right_get(a_type, a_field, a_node) \
((a_type *) (((intptr_t) (a_node)->a_field.rbn_right_red) \
& ((ssize_t)-2)))
#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \
(a_node)->a_field.rbn_right_red = (a_type *) (((uintptr_t) a_right) \
| (((uintptr_t) (a_node)->a_field.rbn_right_red) & ((size_t)1))); \
} while (0)
/* Color accessors. */
#define rbtn_red_get(a_type, a_field, a_node) \
((bool) (((uintptr_t) (a_node)->a_field.rbn_right_red) \
& ((size_t)1)))
#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \
(a_node)->a_field.rbn_right_red = (a_type *) ((((intptr_t) \
(a_node)->a_field.rbn_right_red) & ((ssize_t)-2)) \
| ((ssize_t)a_red)); \
} while (0)
#define rbtn_red_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_right_red = (a_type *) (((uintptr_t) \
(a_node)->a_field.rbn_right_red) | ((size_t)1)); \
} while (0)
#define rbtn_black_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_right_red = (a_type *) (((intptr_t) \
(a_node)->a_field.rbn_right_red) & ((ssize_t)-2)); \
} while (0)
#else
/* Right accessors. */
#define rbtn_right_get(a_type, a_field, a_node) \
((a_node)->a_field.rbn_right)
#define rbtn_right_set(a_type, a_field, a_node, a_right) do { \
(a_node)->a_field.rbn_right = a_right; \
} while (0)
/* Color accessors. */
#define rbtn_red_get(a_type, a_field, a_node) \
((a_node)->a_field.rbn_red)
#define rbtn_color_set(a_type, a_field, a_node, a_red) do { \
(a_node)->a_field.rbn_red = (a_red); \
} while (0)
#define rbtn_red_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_red = true; \
} while (0)
#define rbtn_black_set(a_type, a_field, a_node) do { \
(a_node)->a_field.rbn_red = false; \
} while (0)
#endif
/* Node initializer. */
#define rbt_node_new(a_type, a_field, a_rbt, a_node) do { \
rbtn_left_set(a_type, a_field, (a_node), &(a_rbt)->rbt_nil); \
rbtn_right_set(a_type, a_field, (a_node), &(a_rbt)->rbt_nil); \
rbtn_red_set(a_type, a_field, (a_node)); \
} while (0)
/* Tree initializer. */
#define rb_new(a_type, a_field, a_rbt) do { \
(a_rbt)->rbt_root = &(a_rbt)->rbt_nil; \
rbt_node_new(a_type, a_field, a_rbt, &(a_rbt)->rbt_nil); \
rbtn_black_set(a_type, a_field, &(a_rbt)->rbt_nil); \
} while (0)
/* Internal utility macros. */
#define rbtn_first(a_type, a_field, a_rbt, a_root, r_node) do { \
(r_node) = (a_root); \
if ((r_node) != &(a_rbt)->rbt_nil) { \
for (; \
rbtn_left_get(a_type, a_field, (r_node)) != &(a_rbt)->rbt_nil;\
(r_node) = rbtn_left_get(a_type, a_field, (r_node))) { \
} \
} \
} while (0)
#define rbtn_last(a_type, a_field, a_rbt, a_root, r_node) do { \
(r_node) = (a_root); \
if ((r_node) != &(a_rbt)->rbt_nil) { \
for (; rbtn_right_get(a_type, a_field, (r_node)) != \
&(a_rbt)->rbt_nil; (r_node) = rbtn_right_get(a_type, a_field, \
(r_node))) { \
} \
} \
} while (0)
#define rbtn_rotate_left(a_type, a_field, a_node, r_node) do { \
(r_node) = rbtn_right_get(a_type, a_field, (a_node)); \
rbtn_right_set(a_type, a_field, (a_node), \
rbtn_left_get(a_type, a_field, (r_node))); \
rbtn_left_set(a_type, a_field, (r_node), (a_node)); \
} while (0)
#define rbtn_rotate_right(a_type, a_field, a_node, r_node) do { \
(r_node) = rbtn_left_get(a_type, a_field, (a_node)); \
rbtn_left_set(a_type, a_field, (a_node), \
rbtn_right_get(a_type, a_field, (r_node))); \
rbtn_right_set(a_type, a_field, (r_node), (a_node)); \
} while (0)
/*
* The rb_proto() macro generates function prototypes that correspond to the
* functions generated by an equivalently parameterized call to rb_gen().
*/
#define rb_proto(a_attr, a_prefix, a_rbt_type, a_type) \
a_attr void \
a_prefix##new(a_rbt_type *rbtree); \
a_attr a_type * \
a_prefix##first(a_rbt_type *rbtree); \
a_attr a_type * \
a_prefix##last(a_rbt_type *rbtree); \
a_attr a_type * \
a_prefix##next(a_rbt_type *rbtree, a_type *node); \
a_attr a_type * \
a_prefix##prev(a_rbt_type *rbtree, a_type *node); \
a_attr a_type * \
a_prefix##search(a_rbt_type *rbtree, a_type *key); \
a_attr a_type * \
a_prefix##nsearch(a_rbt_type *rbtree, a_type *key); \
a_attr a_type * \
a_prefix##psearch(a_rbt_type *rbtree, a_type *key); \
a_attr void \
a_prefix##insert(a_rbt_type *rbtree, a_type *node); \
a_attr void \
a_prefix##remove(a_rbt_type *rbtree, a_type *node); \
a_attr a_type * \
a_prefix##iter(a_rbt_type *rbtree, a_type *start, a_type *(*cb)( \
a_rbt_type *, a_type *, void *), void *arg); \
a_attr a_type * \
a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg);
/*
* The rb_gen() macro generates a type-specific red-black tree implementation,
* based on the above cpp macros.
*
* Arguments:
*
* a_attr : Function attribute for generated functions (ex: static).
* a_prefix : Prefix for generated functions (ex: ex_).
* a_rb_type : Type for red-black tree data structure (ex: ex_t).
* a_type : Type for red-black tree node data structure (ex: ex_node_t).
* a_field : Name of red-black tree node linkage (ex: ex_link).
* a_cmp : Node comparison function name, with the following prototype:
* int (a_cmp *)(a_type *a_node, a_type *a_other);
* ^^^^^^
* or a_key
* Interpretation of comparision function return values:
* -1 : a_node < a_other
* 0 : a_node == a_other
* 1 : a_node > a_other
* In all cases, the a_node or a_key macro argument is the first
* argument to the comparison function, which makes it possible
* to write comparison functions that treat the first argument
* specially.
*
* Assuming the following setup:
*
* typedef struct ex_node_s ex_node_t;
* struct ex_node_s {
* rb_node(ex_node_t) ex_link;
* };
* typedef rb_tree(ex_node_t) ex_t;
* rb_gen(static, ex_, ex_t, ex_node_t, ex_link, ex_cmp)
*
* The following API is generated:
*
* static void
* ex_new(ex_t *tree);
* Description: Initialize a red-black tree structure.
* Args:
* tree: Pointer to an uninitialized red-black tree object.
*
* static ex_node_t *
* ex_first(ex_t *tree);
* static ex_node_t *
* ex_last(ex_t *tree);
* Description: Get the first/last node in tree.
* Args:
* tree: Pointer to an initialized red-black tree object.
* Ret: First/last node in tree, or NULL if tree is empty.
*
* static ex_node_t *
* ex_next(ex_t *tree, ex_node_t *node);
* static ex_node_t *
* ex_prev(ex_t *tree, ex_node_t *node);
* Description: Get node's successor/predecessor.
* Args:
* tree: Pointer to an initialized red-black tree object.
* node: A node in tree.
* Ret: node's successor/predecessor in tree, or NULL if node is
* last/first.
*
* static ex_node_t *
* ex_search(ex_t *tree, ex_node_t *key);
* Description: Search for node that matches key.
* Args:
* tree: Pointer to an initialized red-black tree object.
* key : Search key.
* Ret: Node in tree that matches key, or NULL if no match.
*
* static ex_node_t *
* ex_nsearch(ex_t *tree, ex_node_t *key);
* static ex_node_t *
* ex_psearch(ex_t *tree, ex_node_t *key);
* Description: Search for node that matches key. If no match is found,
* return what would be key's successor/predecessor, were
* key in tree.
* Args:
* tree: Pointer to an initialized red-black tree object.
* key : Search key.
* Ret: Node in tree that matches key, or if no match, hypothetical node's
* successor/predecessor (NULL if no successor/predecessor).
*
* static void
* ex_insert(ex_t *tree, ex_node_t *node);
* Description: Insert node into tree.
* Args:
* tree: Pointer to an initialized red-black tree object.
* node: Node to be inserted into tree.
*
* static void
* ex_remove(ex_t *tree, ex_node_t *node);
* Description: Remove node from tree.
* Args:
* tree: Pointer to an initialized red-black tree object.
* node: Node in tree to be removed.
*
* static ex_node_t *
* ex_iter(ex_t *tree, ex_node_t *start, ex_node_t *(*cb)(ex_t *,
* ex_node_t *, void *), void *arg);
* static ex_node_t *
* ex_reverse_iter(ex_t *tree, ex_node_t *start, ex_node *(*cb)(ex_t *,
* ex_node_t *, void *), void *arg);
* Description: Iterate forward/backward over tree, starting at node. If
* tree is modified, iteration must be immediately
* terminated by the callback function that causes the
* modification.
* Args:
* tree : Pointer to an initialized red-black tree object.
* start: Node at which to start iteration, or NULL to start at
* first/last node.
* cb : Callback function, which is called for each node during
* iteration. Under normal circumstances the callback function
* should return NULL, which causes iteration to continue. If a
* callback function returns non-NULL, iteration is immediately
* terminated and the non-NULL return value is returned by the
* iterator. This is useful for re-starting iteration after
* modifying tree.
* arg : Opaque pointer passed to cb().
* Ret: NULL if iteration completed, or the non-NULL callback return value
* that caused termination of the iteration.
*/
#define rb_gen(a_attr, a_prefix, a_rbt_type, a_type, a_field, a_cmp) \
a_attr void \
a_prefix##new(a_rbt_type *rbtree) { \
rb_new(a_type, a_field, rbtree); \
} \
a_attr a_type * \
a_prefix##first(a_rbt_type *rbtree) { \
a_type *ret; \
rbtn_first(a_type, a_field, rbtree, rbtree->rbt_root, ret); \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##last(a_rbt_type *rbtree) { \
a_type *ret; \
rbtn_last(a_type, a_field, rbtree, rbtree->rbt_root, ret); \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##next(a_rbt_type *rbtree, a_type *node) { \
a_type *ret; \
if (rbtn_right_get(a_type, a_field, node) != &rbtree->rbt_nil) { \
rbtn_first(a_type, a_field, rbtree, rbtn_right_get(a_type, \
a_field, node), ret); \
} else { \
a_type *tnode = rbtree->rbt_root; \
assert(tnode != &rbtree->rbt_nil); \
ret = &rbtree->rbt_nil; \
while (true) { \
int cmp = (a_cmp)(node, tnode); \
if (cmp < 0) { \
ret = tnode; \
tnode = rbtn_left_get(a_type, a_field, tnode); \
} else if (cmp > 0) { \
tnode = rbtn_right_get(a_type, a_field, tnode); \
} else { \
break; \
} \
assert(tnode != &rbtree->rbt_nil); \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##prev(a_rbt_type *rbtree, a_type *node) { \
a_type *ret; \
if (rbtn_left_get(a_type, a_field, node) != &rbtree->rbt_nil) { \
rbtn_last(a_type, a_field, rbtree, rbtn_left_get(a_type, \
a_field, node), ret); \
} else { \
a_type *tnode = rbtree->rbt_root; \
assert(tnode != &rbtree->rbt_nil); \
ret = &rbtree->rbt_nil; \
while (true) { \
int cmp = (a_cmp)(node, tnode); \
if (cmp < 0) { \
tnode = rbtn_left_get(a_type, a_field, tnode); \
} else if (cmp > 0) { \
ret = tnode; \
tnode = rbtn_right_get(a_type, a_field, tnode); \
} else { \
break; \
} \
assert(tnode != &rbtree->rbt_nil); \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##search(a_rbt_type *rbtree, a_type *key) { \
a_type *ret; \
int cmp; \
ret = rbtree->rbt_root; \
while (ret != &rbtree->rbt_nil \
&& (cmp = (a_cmp)(key, ret)) != 0) { \
if (cmp < 0) { \
ret = rbtn_left_get(a_type, a_field, ret); \
} else { \
ret = rbtn_right_get(a_type, a_field, ret); \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##nsearch(a_rbt_type *rbtree, a_type *key) { \
a_type *ret; \
a_type *tnode = rbtree->rbt_root; \
ret = &rbtree->rbt_nil; \
while (tnode != &rbtree->rbt_nil) { \
int cmp = (a_cmp)(key, tnode); \
if (cmp < 0) { \
ret = tnode; \
tnode = rbtn_left_get(a_type, a_field, tnode); \
} else if (cmp > 0) { \
tnode = rbtn_right_get(a_type, a_field, tnode); \
} else { \
ret = tnode; \
break; \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##psearch(a_rbt_type *rbtree, a_type *key) { \
a_type *ret; \
a_type *tnode = rbtree->rbt_root; \
ret = &rbtree->rbt_nil; \
while (tnode != &rbtree->rbt_nil) { \
int cmp = (a_cmp)(key, tnode); \
if (cmp < 0) { \
tnode = rbtn_left_get(a_type, a_field, tnode); \
} else if (cmp > 0) { \
ret = tnode; \
tnode = rbtn_right_get(a_type, a_field, tnode); \
} else { \
ret = tnode; \
break; \
} \
} \
if (ret == &rbtree->rbt_nil) { \
ret = (NULL); \
} \
return (ret); \
} \
a_attr void \
a_prefix##insert(a_rbt_type *rbtree, a_type *node) { \
struct { \
a_type *node; \
int cmp; \
} path[sizeof(void *) << 4], *pathp; \
rbt_node_new(a_type, a_field, rbtree, node); \
/* Wind. */ \
path->node = rbtree->rbt_root; \
for (pathp = path; pathp->node != &rbtree->rbt_nil; pathp++) { \
int cmp = pathp->cmp = a_cmp(node, pathp->node); \
assert(cmp != 0); \
if (cmp < 0) { \
pathp[1].node = rbtn_left_get(a_type, a_field, \
pathp->node); \
} else { \
pathp[1].node = rbtn_right_get(a_type, a_field, \
pathp->node); \
} \
} \
pathp->node = node; \
/* Unwind. */ \
for (pathp--; (uintptr_t)pathp >= (uintptr_t)path; pathp--) { \
a_type *cnode = pathp->node; \
if (pathp->cmp < 0) { \
a_type *left = pathp[1].node; \
rbtn_left_set(a_type, a_field, cnode, left); \
if (rbtn_red_get(a_type, a_field, left)) { \
a_type *leftleft = rbtn_left_get(a_type, a_field, left);\
if (rbtn_red_get(a_type, a_field, leftleft)) { \
/* Fix up 4-node. */ \
a_type *tnode; \
rbtn_black_set(a_type, a_field, leftleft); \
rbtn_rotate_right(a_type, a_field, cnode, tnode); \
cnode = tnode; \
} \
} else { \
return; \
} \
} else { \
a_type *right = pathp[1].node; \
rbtn_right_set(a_type, a_field, cnode, right); \
if (rbtn_red_get(a_type, a_field, right)) { \
a_type *left = rbtn_left_get(a_type, a_field, cnode); \
if (rbtn_red_get(a_type, a_field, left)) { \
/* Split 4-node. */ \
rbtn_black_set(a_type, a_field, left); \
rbtn_black_set(a_type, a_field, right); \
rbtn_red_set(a_type, a_field, cnode); \
} else { \
/* Lean left. */ \
a_type *tnode; \
bool tred = rbtn_red_get(a_type, a_field, cnode); \
rbtn_rotate_left(a_type, a_field, cnode, tnode); \
rbtn_color_set(a_type, a_field, tnode, tred); \
rbtn_red_set(a_type, a_field, cnode); \
cnode = tnode; \
} \
} else { \
return; \
} \
} \
pathp->node = cnode; \
} \
/* Set root, and make it black. */ \
rbtree->rbt_root = path->node; \
rbtn_black_set(a_type, a_field, rbtree->rbt_root); \
} \
a_attr void \
a_prefix##remove(a_rbt_type *rbtree, a_type *node) { \
struct { \
a_type *node; \
int cmp; \
} *pathp, *nodep, path[sizeof(void *) << 4]; \
/* Wind. */ \
nodep = NULL; /* Silence compiler warning. */ \
path->node = rbtree->rbt_root; \
for (pathp = path; pathp->node != &rbtree->rbt_nil; pathp++) { \
int cmp = pathp->cmp = a_cmp(node, pathp->node); \
if (cmp < 0) { \
pathp[1].node = rbtn_left_get(a_type, a_field, \
pathp->node); \
} else { \
pathp[1].node = rbtn_right_get(a_type, a_field, \
pathp->node); \
if (cmp == 0) { \
/* Find node's successor, in preparation for swap. */ \
pathp->cmp = 1; \
nodep = pathp; \
for (pathp++; pathp->node != &rbtree->rbt_nil; \
pathp++) { \
pathp->cmp = -1; \
pathp[1].node = rbtn_left_get(a_type, a_field, \
pathp->node); \
} \
break; \
} \
} \
} \
assert(nodep->node == node); \
pathp--; \
if (pathp->node != node) { \
/* Swap node with its successor. */ \
bool tred = rbtn_red_get(a_type, a_field, pathp->node); \
rbtn_color_set(a_type, a_field, pathp->node, \
rbtn_red_get(a_type, a_field, node)); \
rbtn_left_set(a_type, a_field, pathp->node, \
rbtn_left_get(a_type, a_field, node)); \
/* If node's successor is its right child, the following code */\
/* will do the wrong thing for the right child pointer. */\
/* However, it doesn't matter, because the pointer will be */\
/* properly set when the successor is pruned. */\
rbtn_right_set(a_type, a_field, pathp->node, \
rbtn_right_get(a_type, a_field, node)); \
rbtn_color_set(a_type, a_field, node, tred); \
/* The pruned leaf node's child pointers are never accessed */\
/* again, so don't bother setting them to nil. */\
nodep->node = pathp->node; \
pathp->node = node; \
if (nodep == path) { \
rbtree->rbt_root = nodep->node; \
} else { \
if (nodep[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, nodep[-1].node, \
nodep->node); \
} else { \
rbtn_right_set(a_type, a_field, nodep[-1].node, \
nodep->node); \
} \
} \
} else { \
a_type *left = rbtn_left_get(a_type, a_field, node); \
if (left != &rbtree->rbt_nil) { \
/* node has no successor, but it has a left child. */\
/* Splice node out, without losing the left child. */\
assert(rbtn_red_get(a_type, a_field, node) == false); \
assert(rbtn_red_get(a_type, a_field, left)); \
rbtn_black_set(a_type, a_field, left); \
if (pathp == path) { \
rbtree->rbt_root = left; \
} else { \
if (pathp[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, pathp[-1].node, \
left); \
} else { \
rbtn_right_set(a_type, a_field, pathp[-1].node, \
left); \
} \
} \
return; \
} else if (pathp == path) { \
/* The tree only contained one node. */ \
rbtree->rbt_root = &rbtree->rbt_nil; \
return; \
} \
} \
if (rbtn_red_get(a_type, a_field, pathp->node)) { \
/* Prune red node, which requires no fixup. */ \
assert(pathp[-1].cmp < 0); \
rbtn_left_set(a_type, a_field, pathp[-1].node, \
&rbtree->rbt_nil); \
return; \
} \
/* The node to be pruned is black, so unwind until balance is */\
/* restored. */\
pathp->node = &rbtree->rbt_nil; \
for (pathp--; (uintptr_t)pathp >= (uintptr_t)path; pathp--) { \
assert(pathp->cmp != 0); \
if (pathp->cmp < 0) { \
rbtn_left_set(a_type, a_field, pathp->node, \
pathp[1].node); \
assert(rbtn_red_get(a_type, a_field, pathp[1].node) \
== false); \
if (rbtn_red_get(a_type, a_field, pathp->node)) { \
a_type *right = rbtn_right_get(a_type, a_field, \
pathp->node); \
a_type *rightleft = rbtn_left_get(a_type, a_field, \
right); \
a_type *tnode; \
if (rbtn_red_get(a_type, a_field, rightleft)) { \
/* In the following diagrams, ||, //, and \\ */\
/* indicate the path to the removed node. */\
/* */\
/* || */\
/* pathp(r) */\
/* // \ */\
/* (b) (b) */\
/* / */\
/* (r) */\
/* */\
rbtn_black_set(a_type, a_field, pathp->node); \
rbtn_rotate_right(a_type, a_field, right, tnode); \
rbtn_right_set(a_type, a_field, pathp->node, tnode);\
rbtn_rotate_left(a_type, a_field, pathp->node, \
tnode); \
} else { \
/* || */\
/* pathp(r) */\
/* // \ */\
/* (b) (b) */\
/* / */\
/* (b) */\
/* */\
rbtn_rotate_left(a_type, a_field, pathp->node, \
tnode); \
} \
/* Balance restored, but rotation modified subtree */\
/* root. */\
assert((uintptr_t)pathp > (uintptr_t)path); \
if (pathp[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, pathp[-1].node, \
tnode); \
} else { \
rbtn_right_set(a_type, a_field, pathp[-1].node, \
tnode); \
} \
return; \
} else { \
a_type *right = rbtn_right_get(a_type, a_field, \
pathp->node); \
a_type *rightleft = rbtn_left_get(a_type, a_field, \
right); \
if (rbtn_red_get(a_type, a_field, rightleft)) { \
/* || */\
/* pathp(b) */\
/* // \ */\
/* (b) (b) */\
/* / */\
/* (r) */\
a_type *tnode; \
rbtn_black_set(a_type, a_field, rightleft); \
rbtn_rotate_right(a_type, a_field, right, tnode); \
rbtn_right_set(a_type, a_field, pathp->node, tnode);\
rbtn_rotate_left(a_type, a_field, pathp->node, \
tnode); \
/* Balance restored, but rotation modified */\
/* subree root, which may actually be the tree */\
/* root. */\
if (pathp == path) { \
/* Set root. */ \
rbtree->rbt_root = tnode; \
} else { \
if (pathp[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, \
pathp[-1].node, tnode); \
} else { \
rbtn_right_set(a_type, a_field, \
pathp[-1].node, tnode); \
} \
} \
return; \
} else { \
/* || */\
/* pathp(b) */\
/* // \ */\
/* (b) (b) */\
/* / */\
/* (b) */\
a_type *tnode; \
rbtn_red_set(a_type, a_field, pathp->node); \
rbtn_rotate_left(a_type, a_field, pathp->node, \
tnode); \
pathp->node = tnode; \
} \
} \
} else { \
a_type *left; \
rbtn_right_set(a_type, a_field, pathp->node, \
pathp[1].node); \
left = rbtn_left_get(a_type, a_field, pathp->node); \
if (rbtn_red_get(a_type, a_field, left)) { \
a_type *tnode; \
a_type *leftright = rbtn_right_get(a_type, a_field, \
left); \
a_type *leftrightleft = rbtn_left_get(a_type, a_field, \
leftright); \
if (rbtn_red_get(a_type, a_field, leftrightleft)) { \
/* || */\
/* pathp(b) */\
/* / \\ */\
/* (r) (b) */\
/* \ */\
/* (b) */\
/* / */\
/* (r) */\
a_type *unode; \
rbtn_black_set(a_type, a_field, leftrightleft); \
rbtn_rotate_right(a_type, a_field, pathp->node, \
unode); \
rbtn_rotate_right(a_type, a_field, pathp->node, \
tnode); \
rbtn_right_set(a_type, a_field, unode, tnode); \
rbtn_rotate_left(a_type, a_field, unode, tnode); \
} else { \
/* || */\
/* pathp(b) */\
/* / \\ */\
/* (r) (b) */\
/* \ */\
/* (b) */\
/* / */\
/* (b) */\
assert(leftright != &rbtree->rbt_nil); \
rbtn_red_set(a_type, a_field, leftright); \
rbtn_rotate_right(a_type, a_field, pathp->node, \
tnode); \
rbtn_black_set(a_type, a_field, tnode); \
} \
/* Balance restored, but rotation modified subtree */\
/* root, which may actually be the tree root. */\
if (pathp == path) { \
/* Set root. */ \
rbtree->rbt_root = tnode; \
} else { \
if (pathp[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, pathp[-1].node, \
tnode); \
} else { \
rbtn_right_set(a_type, a_field, pathp[-1].node, \
tnode); \
} \
} \
return; \
} else if (rbtn_red_get(a_type, a_field, pathp->node)) { \
a_type *leftleft = rbtn_left_get(a_type, a_field, left);\
if (rbtn_red_get(a_type, a_field, leftleft)) { \
/* || */\
/* pathp(r) */\
/* / \\ */\
/* (b) (b) */\
/* / */\
/* (r) */\
a_type *tnode; \
rbtn_black_set(a_type, a_field, pathp->node); \
rbtn_red_set(a_type, a_field, left); \
rbtn_black_set(a_type, a_field, leftleft); \
rbtn_rotate_right(a_type, a_field, pathp->node, \
tnode); \
/* Balance restored, but rotation modified */\
/* subtree root. */\
assert((uintptr_t)pathp > (uintptr_t)path); \
if (pathp[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, pathp[-1].node, \
tnode); \
} else { \
rbtn_right_set(a_type, a_field, pathp[-1].node, \
tnode); \
} \
return; \
} else { \
/* || */\
/* pathp(r) */\
/* / \\ */\
/* (b) (b) */\
/* / */\
/* (b) */\
rbtn_red_set(a_type, a_field, left); \
rbtn_black_set(a_type, a_field, pathp->node); \
/* Balance restored. */ \
return; \
} \
} else { \
a_type *leftleft = rbtn_left_get(a_type, a_field, left);\
if (rbtn_red_get(a_type, a_field, leftleft)) { \
/* || */\
/* pathp(b) */\
/* / \\ */\
/* (b) (b) */\
/* / */\
/* (r) */\
a_type *tnode; \
rbtn_black_set(a_type, a_field, leftleft); \
rbtn_rotate_right(a_type, a_field, pathp->node, \
tnode); \
/* Balance restored, but rotation modified */\
/* subtree root, which may actually be the tree */\
/* root. */\
if (pathp == path) { \
/* Set root. */ \
rbtree->rbt_root = tnode; \
} else { \
if (pathp[-1].cmp < 0) { \
rbtn_left_set(a_type, a_field, \
pathp[-1].node, tnode); \
} else { \
rbtn_right_set(a_type, a_field, \
pathp[-1].node, tnode); \
} \
} \
return; \
} else { \
/* || */\
/* pathp(b) */\
/* / \\ */\
/* (b) (b) */\
/* / */\
/* (b) */\
rbtn_red_set(a_type, a_field, left); \
} \
} \
} \
} \
/* Set root. */ \
rbtree->rbt_root = path->node; \
assert(rbtn_red_get(a_type, a_field, rbtree->rbt_root) == false); \
} \
a_attr a_type * \
a_prefix##iter_recurse(a_rbt_type *rbtree, a_type *node, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \
if (node == &rbtree->rbt_nil) { \
return (&rbtree->rbt_nil); \
} else { \
a_type *ret; \
if ((ret = a_prefix##iter_recurse(rbtree, rbtn_left_get(a_type, \
a_field, node), cb, arg)) != &rbtree->rbt_nil \
|| (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
} \
return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg)); \
} \
} \
a_attr a_type * \
a_prefix##iter_start(a_rbt_type *rbtree, a_type *start, a_type *node, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \
int cmp = a_cmp(start, node); \
if (cmp < 0) { \
a_type *ret; \
if ((ret = a_prefix##iter_start(rbtree, start, \
rbtn_left_get(a_type, a_field, node), cb, arg)) != \
&rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
} \
return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg)); \
} else if (cmp > 0) { \
return (a_prefix##iter_start(rbtree, start, \
rbtn_right_get(a_type, a_field, node), cb, arg)); \
} else { \
a_type *ret; \
if ((ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
} \
return (a_prefix##iter_recurse(rbtree, rbtn_right_get(a_type, \
a_field, node), cb, arg)); \
} \
} \
a_attr a_type * \
a_prefix##iter(a_rbt_type *rbtree, a_type *start, a_type *(*cb)( \
a_rbt_type *, a_type *, void *), void *arg) { \
a_type *ret; \
if (start != NULL) { \
ret = a_prefix##iter_start(rbtree, start, rbtree->rbt_root, \
cb, arg); \
} else { \
ret = a_prefix##iter_recurse(rbtree, rbtree->rbt_root, cb, arg);\
} \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
} \
a_attr a_type * \
a_prefix##reverse_iter_recurse(a_rbt_type *rbtree, a_type *node, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \
if (node == &rbtree->rbt_nil) { \
return (&rbtree->rbt_nil); \
} else { \
a_type *ret; \
if ((ret = a_prefix##reverse_iter_recurse(rbtree, \
rbtn_right_get(a_type, a_field, node), cb, arg)) != \
&rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
} \
return (a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
} \
} \
a_attr a_type * \
a_prefix##reverse_iter_start(a_rbt_type *rbtree, a_type *start, \
a_type *node, a_type *(*cb)(a_rbt_type *, a_type *, void *), \
void *arg) { \
int cmp = a_cmp(start, node); \
if (cmp > 0) { \
a_type *ret; \
if ((ret = a_prefix##reverse_iter_start(rbtree, start, \
rbtn_right_get(a_type, a_field, node), cb, arg)) != \
&rbtree->rbt_nil || (ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
} \
return (a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
} else if (cmp < 0) { \
return (a_prefix##reverse_iter_start(rbtree, start, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
} else { \
a_type *ret; \
if ((ret = cb(rbtree, node, arg)) != NULL) { \
return (ret); \
} \
return (a_prefix##reverse_iter_recurse(rbtree, \
rbtn_left_get(a_type, a_field, node), cb, arg)); \
} \
} \
a_attr a_type * \
a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
a_type *(*cb)(a_rbt_type *, a_type *, void *), void *arg) { \
a_type *ret; \
if (start != NULL) { \
ret = a_prefix##reverse_iter_start(rbtree, start, \
rbtree->rbt_root, cb, arg); \
} else { \
ret = a_prefix##reverse_iter_recurse(rbtree, rbtree->rbt_root, \
cb, arg); \
} \
if (ret == &rbtree->rbt_nil) { \
ret = NULL; \
} \
return (ret); \
}
#endif /* RB_H_ */
deps/jemalloc/include/jemalloc/internal/rtree.h
Vendored
+164
View File
@@ -0,0 +1,164 @@
/*
* This radix tree implementation is tailored to the singular purpose of
* tracking which chunks are currently owned by jemalloc. This functionality
* is mandatory for OS X, where jemalloc must be able to respond to object
* ownership queries.
*
*******************************************************************************
*/
#ifdef JEMALLOC_H_TYPES
typedef struct rtree_s rtree_t;
/*
* Size of each radix tree node (must be a power of 2). This impacts tree
* depth.
*/
#if (LG_SIZEOF_PTR == 2)
# define RTREE_NODESIZE (1U << 14)
#else
# define RTREE_NODESIZE CACHELINE
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct rtree_s {
malloc_mutex_t mutex;
void **root;
unsigned height;
unsigned level2bits[1]; /* Dynamically sized. */
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
rtree_t *rtree_new(unsigned bits);
void rtree_prefork(rtree_t *rtree);
void rtree_postfork_parent(rtree_t *rtree);
void rtree_postfork_child(rtree_t *rtree);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
#ifndef JEMALLOC_DEBUG
void *rtree_get_locked(rtree_t *rtree, uintptr_t key);
#endif
void *rtree_get(rtree_t *rtree, uintptr_t key);
bool rtree_set(rtree_t *rtree, uintptr_t key, void *val);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_RTREE_C_))
#define RTREE_GET_GENERATE(f) \
/* The least significant bits of the key are ignored. */ \
JEMALLOC_INLINE void * \
f(rtree_t *rtree, uintptr_t key) \
{ \
void *ret; \
uintptr_t subkey; \
unsigned i, lshift, height, bits; \
void **node, **child; \
\
RTREE_LOCK(&rtree->mutex); \
for (i = lshift = 0, height = rtree->height, node = rtree->root;\
i < height - 1; \
i++, lshift += bits, node = child) { \
bits = rtree->level2bits[i]; \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR + \
3)) - bits); \
child = (void**)node[subkey]; \
if (child == NULL) { \
RTREE_UNLOCK(&rtree->mutex); \
return (NULL); \
} \
} \
\
/* \
* node is a leaf, so it contains values rather than node \
* pointers. \
*/ \
bits = rtree->level2bits[i]; \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) - \
bits); \
ret = node[subkey]; \
RTREE_UNLOCK(&rtree->mutex); \
\
RTREE_GET_VALIDATE \
return (ret); \
}
#ifdef JEMALLOC_DEBUG
# define RTREE_LOCK(l) malloc_mutex_lock(l)
# define RTREE_UNLOCK(l) malloc_mutex_unlock(l)
# define RTREE_GET_VALIDATE
RTREE_GET_GENERATE(rtree_get_locked)
# undef RTREE_LOCK
# undef RTREE_UNLOCK
# undef RTREE_GET_VALIDATE
#endif
#define RTREE_LOCK(l)
#define RTREE_UNLOCK(l)
#ifdef JEMALLOC_DEBUG
/*
* Suppose that it were possible for a jemalloc-allocated chunk to be
* munmap()ped, followed by a different allocator in another thread re-using
* overlapping virtual memory, all without invalidating the cached rtree
* value. The result would be a false positive (the rtree would claim that
* jemalloc owns memory that it had actually discarded). This scenario
* seems impossible, but the following assertion is a prudent sanity check.
*/
# define RTREE_GET_VALIDATE \
assert(rtree_get_locked(rtree, key) == ret);
#else
# define RTREE_GET_VALIDATE
#endif
RTREE_GET_GENERATE(rtree_get)
#undef RTREE_LOCK
#undef RTREE_UNLOCK
#undef RTREE_GET_VALIDATE
JEMALLOC_INLINE bool
rtree_set(rtree_t *rtree, uintptr_t key, void *val)
{
uintptr_t subkey;
unsigned i, lshift, height, bits;
void **node, **child;
malloc_mutex_lock(&rtree->mutex);
for (i = lshift = 0, height = rtree->height, node = rtree->root;
i < height - 1;
i++, lshift += bits, node = child) {
bits = rtree->level2bits[i];
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) -
bits);
child = (void**)node[subkey];
if (child == NULL) {
child = (void**)base_alloc(sizeof(void *) <<
rtree->level2bits[i+1]);
if (child == NULL) {
malloc_mutex_unlock(&rtree->mutex);
return (true);
}
memset(child, 0, sizeof(void *) <<
rtree->level2bits[i+1]);
node[subkey] = child;
}
}
/* node is a leaf, so it contains values rather than node pointers. */
bits = rtree->level2bits[i];
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) - bits);
node[subkey] = val;
malloc_mutex_unlock(&rtree->mutex);
return (false);
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/size_classes.sh
Vendored Executable
+122
View File
@@ -0,0 +1,122 @@
#!/bin/sh
# The following limits are chosen such that they cover all supported platforms.
# Range of quanta.
lg_qmin=3
lg_qmax=4
# The range of tiny size classes is [2^lg_tmin..2^(lg_q-1)].
lg_tmin=3
# Range of page sizes.
lg_pmin=12
lg_pmax=16
pow2() {
e=$1
pow2_result=1
while [ ${e} -gt 0 ] ; do
pow2_result=$((${pow2_result} + ${pow2_result}))
e=$((${e} - 1))
done
}
cat <<EOF
/* This file was automatically generated by size_classes.sh. */
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
EOF
lg_q=${lg_qmin}
while [ ${lg_q} -le ${lg_qmax} ] ; do
lg_t=${lg_tmin}
while [ ${lg_t} -le ${lg_q} ] ; do
lg_p=${lg_pmin}
while [ ${lg_p} -le ${lg_pmax} ] ; do
echo "#if (LG_TINY_MIN == ${lg_t} && LG_QUANTUM == ${lg_q} && LG_PAGE == ${lg_p})"
echo "#define SIZE_CLASSES_DEFINED"
pow2 ${lg_q}; q=${pow2_result}
pow2 ${lg_t}; t=${pow2_result}
pow2 ${lg_p}; p=${pow2_result}
bin=0
psz=0
sz=${t}
delta=$((${sz} - ${psz}))
echo "/* SIZE_CLASS(bin, delta, sz) */"
echo "#define SIZE_CLASSES \\"
# Tiny size classes.
while [ ${sz} -lt ${q} ] ; do
echo " SIZE_CLASS(${bin}, ${delta}, ${sz}) \\"
bin=$((${bin} + 1))
psz=${sz}
sz=$((${sz} + ${sz}))
delta=$((${sz} - ${psz}))
done
# Quantum-multiple size classes. For each doubling of sz, as many as 4
# size classes exist. Their spacing is the greater of:
# - q
# - sz/4, where sz is a power of 2
while [ ${sz} -lt ${p} ] ; do
if [ ${sz} -ge $((${q} * 4)) ] ; then
i=$((${sz} / 4))
else
i=${q}
fi
next_2pow=$((${sz} * 2))
while [ ${sz} -lt $next_2pow ] ; do
echo " SIZE_CLASS(${bin}, ${delta}, ${sz}) \\"
bin=$((${bin} + 1))
psz=${sz}
sz=$((${sz} + ${i}))
delta=$((${sz} - ${psz}))
done
done
echo
echo "#define NBINS ${bin}"
echo "#define SMALL_MAXCLASS ${psz}"
echo "#endif"
echo
lg_p=$((${lg_p} + 1))
done
lg_t=$((${lg_t} + 1))
done
lg_q=$((${lg_q} + 1))
done
cat <<EOF
#ifndef SIZE_CLASSES_DEFINED
# error "No size class definitions match configuration"
#endif
#undef SIZE_CLASSES_DEFINED
/*
* The small_size2bin lookup table uses uint8_t to encode each bin index, so we
* cannot support more than 256 small size classes. Further constrain NBINS to
* 255 to support prof_promote, since all small size classes, plus a "not
* small" size class must be stored in 8 bits of arena_chunk_map_t's bits
* field.
*/
#if (NBINS > 255)
# error "Too many small size classes"
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
EOF
deps/jemalloc/include/jemalloc/internal/stats.h
Vendored
+173
View File
@@ -0,0 +1,173 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct tcache_bin_stats_s tcache_bin_stats_t;
typedef struct malloc_bin_stats_s malloc_bin_stats_t;
typedef struct malloc_large_stats_s malloc_large_stats_t;
typedef struct arena_stats_s arena_stats_t;
typedef struct chunk_stats_s chunk_stats_t;
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct tcache_bin_stats_s {
/*
* Number of allocation requests that corresponded to the size of this
* bin.
*/
uint64_t nrequests;
};
struct malloc_bin_stats_s {
/*
* Current number of bytes allocated, including objects currently
* cached by tcache.
*/
size_t allocated;
/*
* Total number of allocation/deallocation requests served directly by
* the bin. Note that tcache may allocate an object, then recycle it
* many times, resulting many increments to nrequests, but only one
* each to nmalloc and ndalloc.
*/
uint64_t nmalloc;
uint64_t ndalloc;
/*
* Number of allocation requests that correspond to the size of this
* bin. This includes requests served by tcache, though tcache only
* periodically merges into this counter.
*/
uint64_t nrequests;
/* Number of tcache fills from this bin. */
uint64_t nfills;
/* Number of tcache flushes to this bin. */
uint64_t nflushes;
/* Total number of runs created for this bin's size class. */
uint64_t nruns;
/*
* Total number of runs reused by extracting them from the runs tree for
* this bin's size class.
*/
uint64_t reruns;
/* Current number of runs in this bin. */
size_t curruns;
};
struct malloc_large_stats_s {
/*
* Total number of allocation/deallocation requests served directly by
* the arena. Note that tcache may allocate an object, then recycle it
* many times, resulting many increments to nrequests, but only one
* each to nmalloc and ndalloc.
*/
uint64_t nmalloc;
uint64_t ndalloc;
/*
* Number of allocation requests that correspond to this size class.
* This includes requests served by tcache, though tcache only
* periodically merges into this counter.
*/
uint64_t nrequests;
/* Current number of runs of this size class. */
size_t curruns;
};
struct arena_stats_s {
/* Number of bytes currently mapped. */
size_t mapped;
/*
* Total number of purge sweeps, total number of madvise calls made,
* and total pages purged in order to keep dirty unused memory under
* control.
*/
uint64_t npurge;
uint64_t nmadvise;
uint64_t purged;
/* Per-size-category statistics. */
size_t allocated_large;
uint64_t nmalloc_large;
uint64_t ndalloc_large;
uint64_t nrequests_large;
/*
* One element for each possible size class, including sizes that
* overlap with bin size classes. This is necessary because ipalloc()
* sometimes has to use such large objects in order to assure proper
* alignment.
*/
malloc_large_stats_t *lstats;
};
struct chunk_stats_s {
/* Number of chunks that were allocated. */
uint64_t nchunks;
/* High-water mark for number of chunks allocated. */
size_t highchunks;
/*
* Current number of chunks allocated. This value isn't maintained for
* any other purpose, so keep track of it in order to be able to set
* highchunks.
*/
size_t curchunks;
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern bool opt_stats_print;
extern size_t stats_cactive;
void stats_print(void (*write)(void *, const char *), void *cbopaque,
const char *opts);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
size_t stats_cactive_get(void);
void stats_cactive_add(size_t size);
void stats_cactive_sub(size_t size);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_STATS_C_))
JEMALLOC_INLINE size_t
stats_cactive_get(void)
{
return (atomic_read_z(&stats_cactive));
}
JEMALLOC_INLINE void
stats_cactive_add(size_t size)
{
atomic_add_z(&stats_cactive, size);
}
JEMALLOC_INLINE void
stats_cactive_sub(size_t size)
{
atomic_sub_z(&stats_cactive, size);
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/tcache.h
Vendored
+440
View File
@@ -0,0 +1,440 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct tcache_bin_info_s tcache_bin_info_t;
typedef struct tcache_bin_s tcache_bin_t;
typedef struct tcache_s tcache_t;
/*
* tcache pointers close to NULL are used to encode state information that is
* used for two purposes: preventing thread caching on a per thread basis and
* cleaning up during thread shutdown.
*/
#define TCACHE_STATE_DISABLED ((tcache_t *)(uintptr_t)1)
#define TCACHE_STATE_REINCARNATED ((tcache_t *)(uintptr_t)2)
#define TCACHE_STATE_PURGATORY ((tcache_t *)(uintptr_t)3)
#define TCACHE_STATE_MAX TCACHE_STATE_PURGATORY
/*
* Absolute maximum number of cache slots for each small bin in the thread
* cache. This is an additional constraint beyond that imposed as: twice the
* number of regions per run for this size class.
*
* This constant must be an even number.
*/
#define TCACHE_NSLOTS_SMALL_MAX 200
/* Number of cache slots for large size classes. */
#define TCACHE_NSLOTS_LARGE 20
/* (1U << opt_lg_tcache_max) is used to compute tcache_maxclass. */
#define LG_TCACHE_MAXCLASS_DEFAULT 15
/*
* TCACHE_GC_SWEEP is the approximate number of allocation events between
* full GC sweeps. Integer rounding may cause the actual number to be
* slightly higher, since GC is performed incrementally.
*/
#define TCACHE_GC_SWEEP 8192
/* Number of tcache allocation/deallocation events between incremental GCs. */
#define TCACHE_GC_INCR \
((TCACHE_GC_SWEEP / NBINS) + ((TCACHE_GC_SWEEP / NBINS == 0) ? 0 : 1))
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
typedef enum {
tcache_enabled_false = 0, /* Enable cast to/from bool. */
tcache_enabled_true = 1,
tcache_enabled_default = 2
} tcache_enabled_t;
/*
* Read-only information associated with each element of tcache_t's tbins array
* is stored separately, mainly to reduce memory usage.
*/
struct tcache_bin_info_s {
unsigned ncached_max; /* Upper limit on ncached. */
};
struct tcache_bin_s {
tcache_bin_stats_t tstats;
int low_water; /* Min # cached since last GC. */
unsigned lg_fill_div; /* Fill (ncached_max >> lg_fill_div). */
unsigned ncached; /* # of cached objects. */
void **avail; /* Stack of available objects. */
};
struct tcache_s {
ql_elm(tcache_t) link; /* Used for aggregating stats. */
uint64_t prof_accumbytes;/* Cleared after arena_prof_accum() */
arena_t *arena; /* This thread's arena. */
unsigned ev_cnt; /* Event count since incremental GC. */
unsigned next_gc_bin; /* Next bin to GC. */
tcache_bin_t tbins[1]; /* Dynamically sized. */
/*
* The pointer stacks associated with tbins follow as a contiguous
* array. During tcache initialization, the avail pointer in each
* element of tbins is initialized to point to the proper offset within
* this array.
*/
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern bool opt_tcache;
extern ssize_t opt_lg_tcache_max;
extern tcache_bin_info_t *tcache_bin_info;
/*
* Number of tcache bins. There are NBINS small-object bins, plus 0 or more
* large-object bins.
*/
extern size_t nhbins;
/* Maximum cached size class. */
extern size_t tcache_maxclass;
size_t tcache_salloc(const void *ptr);
void tcache_event_hard(tcache_t *tcache);
void *tcache_alloc_small_hard(tcache_t *tcache, tcache_bin_t *tbin,
size_t binind);
void tcache_bin_flush_small(tcache_bin_t *tbin, size_t binind, unsigned rem,
tcache_t *tcache);
void tcache_bin_flush_large(tcache_bin_t *tbin, size_t binind, unsigned rem,
tcache_t *tcache);
void tcache_arena_associate(tcache_t *tcache, arena_t *arena);
void tcache_arena_dissociate(tcache_t *tcache);
tcache_t *tcache_create(arena_t *arena);
void tcache_destroy(tcache_t *tcache);
void tcache_thread_cleanup(void *arg);
void tcache_stats_merge(tcache_t *tcache, arena_t *arena);
bool tcache_boot0(void);
bool tcache_boot1(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
malloc_tsd_protos(JEMALLOC_ATTR(unused), tcache, tcache_t *)
malloc_tsd_protos(JEMALLOC_ATTR(unused), tcache_enabled, tcache_enabled_t)
void tcache_event(tcache_t *tcache);
void tcache_flush(void);
bool tcache_enabled_get(void);
tcache_t *tcache_get(bool create);
void tcache_enabled_set(bool enabled);
void *tcache_alloc_easy(tcache_bin_t *tbin);
void *tcache_alloc_small(tcache_t *tcache, size_t size, bool zero);
void *tcache_alloc_large(tcache_t *tcache, size_t size, bool zero);
void tcache_dalloc_small(tcache_t *tcache, void *ptr, size_t binind);
void tcache_dalloc_large(tcache_t *tcache, void *ptr, size_t size);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_TCACHE_C_))
/* Map of thread-specific caches. */
malloc_tsd_externs(tcache, tcache_t *)
malloc_tsd_funcs(JEMALLOC_INLINE, tcache, tcache_t *, NULL,
tcache_thread_cleanup)
/* Per thread flag that allows thread caches to be disabled. */
malloc_tsd_externs(tcache_enabled, tcache_enabled_t)
malloc_tsd_funcs(JEMALLOC_INLINE, tcache_enabled, tcache_enabled_t,
tcache_enabled_default, malloc_tsd_no_cleanup)
JEMALLOC_INLINE void
tcache_flush(void)
{
tcache_t *tcache;
cassert(config_tcache);
tcache = *tcache_tsd_get();
if ((uintptr_t)tcache <= (uintptr_t)TCACHE_STATE_MAX)
return;
tcache_destroy(tcache);
tcache = NULL;
tcache_tsd_set(&tcache);
}
JEMALLOC_INLINE bool
tcache_enabled_get(void)
{
tcache_enabled_t tcache_enabled;
cassert(config_tcache);
tcache_enabled = *tcache_enabled_tsd_get();
if (tcache_enabled == tcache_enabled_default) {
tcache_enabled = (tcache_enabled_t)opt_tcache;
tcache_enabled_tsd_set(&tcache_enabled);
}
return ((bool)tcache_enabled);
}
JEMALLOC_INLINE void
tcache_enabled_set(bool enabled)
{
tcache_enabled_t tcache_enabled;
tcache_t *tcache;
cassert(config_tcache);
tcache_enabled = (tcache_enabled_t)enabled;
tcache_enabled_tsd_set(&tcache_enabled);
tcache = *tcache_tsd_get();
if (enabled) {
if (tcache == TCACHE_STATE_DISABLED) {
tcache = NULL;
tcache_tsd_set(&tcache);
}
} else /* disabled */ {
if (tcache > TCACHE_STATE_MAX) {
tcache_destroy(tcache);
tcache = NULL;
}
if (tcache == NULL) {
tcache = TCACHE_STATE_DISABLED;
tcache_tsd_set(&tcache);
}
}
}
JEMALLOC_INLINE tcache_t *
tcache_get(bool create)
{
tcache_t *tcache;
if (config_tcache == false)
return (NULL);
if (config_lazy_lock && isthreaded == false)
return (NULL);
tcache = *tcache_tsd_get();
if ((uintptr_t)tcache <= (uintptr_t)TCACHE_STATE_MAX) {
if (tcache == TCACHE_STATE_DISABLED)
return (NULL);
if (tcache == NULL) {
if (create == false) {
/*
* Creating a tcache here would cause
* allocation as a side effect of free().
* Ordinarily that would be okay since
* tcache_create() failure is a soft failure
* that doesn't propagate. However, if TLS
* data are freed via free() as in glibc,
* subtle corruption could result from setting
* a TLS variable after its backing memory is
* freed.
*/
return (NULL);
}
if (tcache_enabled_get() == false) {
tcache_enabled_set(false); /* Memoize. */
return (NULL);
}
return (tcache_create(choose_arena(NULL)));
}
if (tcache == TCACHE_STATE_PURGATORY) {
/*
* Make a note that an allocator function was called
* after tcache_thread_cleanup() was called.
*/
tcache = TCACHE_STATE_REINCARNATED;
tcache_tsd_set(&tcache);
return (NULL);
}
if (tcache == TCACHE_STATE_REINCARNATED)
return (NULL);
not_reached();
}
return (tcache);
}
JEMALLOC_INLINE void
tcache_event(tcache_t *tcache)
{
if (TCACHE_GC_INCR == 0)
return;
tcache->ev_cnt++;
assert(tcache->ev_cnt <= TCACHE_GC_INCR);
if (tcache->ev_cnt == TCACHE_GC_INCR)
tcache_event_hard(tcache);
}
JEMALLOC_INLINE void *
tcache_alloc_easy(tcache_bin_t *tbin)
{
void *ret;
if (tbin->ncached == 0) {
tbin->low_water = -1;
return (NULL);
}
tbin->ncached--;
if ((int)tbin->ncached < tbin->low_water)
tbin->low_water = tbin->ncached;
ret = tbin->avail[tbin->ncached];
return (ret);
}
JEMALLOC_INLINE void *
tcache_alloc_small(tcache_t *tcache, size_t size, bool zero)
{
void *ret;
size_t binind;
tcache_bin_t *tbin;
binind = SMALL_SIZE2BIN(size);
assert(binind < NBINS);
tbin = &tcache->tbins[binind];
ret = tcache_alloc_easy(tbin);
if (ret == NULL) {
ret = tcache_alloc_small_hard(tcache, tbin, binind);
if (ret == NULL)
return (NULL);
}
assert(tcache_salloc(ret) == arena_bin_info[binind].reg_size);
if (zero == false) {
if (config_fill) {
if (opt_junk) {
arena_alloc_junk_small(ret,
&arena_bin_info[binind], false);
} else if (opt_zero)
memset(ret, 0, size);
}
} else {
if (config_fill && opt_junk) {
arena_alloc_junk_small(ret, &arena_bin_info[binind],
true);
}
VALGRIND_MAKE_MEM_UNDEFINED(ret, size);
memset(ret, 0, size);
}
if (config_stats)
tbin->tstats.nrequests++;
if (config_prof)
tcache->prof_accumbytes += arena_bin_info[binind].reg_size;
tcache_event(tcache);
return (ret);
}
JEMALLOC_INLINE void *
tcache_alloc_large(tcache_t *tcache, size_t size, bool zero)
{
void *ret;
size_t binind;
tcache_bin_t *tbin;
size = PAGE_CEILING(size);
assert(size <= tcache_maxclass);
binind = NBINS + (size >> LG_PAGE) - 1;
assert(binind < nhbins);
tbin = &tcache->tbins[binind];
ret = tcache_alloc_easy(tbin);
if (ret == NULL) {
/*
* Only allocate one large object at a time, because it's quite
* expensive to create one and not use it.
*/
ret = arena_malloc_large(tcache->arena, size, zero);
if (ret == NULL)
return (NULL);
} else {
if (config_prof && prof_promote && size == PAGE) {
arena_chunk_t *chunk =
(arena_chunk_t *)CHUNK_ADDR2BASE(ret);
size_t pageind = (((uintptr_t)ret - (uintptr_t)chunk) >>
LG_PAGE);
arena_mapbits_large_binind_set(chunk, pageind,
BININD_INVALID);
}
if (zero == false) {
if (config_fill) {
if (opt_junk)
memset(ret, 0xa5, size);
else if (opt_zero)
memset(ret, 0, size);
}
} else {
VALGRIND_MAKE_MEM_UNDEFINED(ret, size);
memset(ret, 0, size);
}
if (config_stats)
tbin->tstats.nrequests++;
if (config_prof)
tcache->prof_accumbytes += size;
}
tcache_event(tcache);
return (ret);
}
JEMALLOC_INLINE void
tcache_dalloc_small(tcache_t *tcache, void *ptr, size_t binind)
{
tcache_bin_t *tbin;
tcache_bin_info_t *tbin_info;
assert(tcache_salloc(ptr) <= SMALL_MAXCLASS);
if (config_fill && opt_junk)
arena_dalloc_junk_small(ptr, &arena_bin_info[binind]);
tbin = &tcache->tbins[binind];
tbin_info = &tcache_bin_info[binind];
if (tbin->ncached == tbin_info->ncached_max) {
tcache_bin_flush_small(tbin, binind, (tbin_info->ncached_max >>
1), tcache);
}
assert(tbin->ncached < tbin_info->ncached_max);
tbin->avail[tbin->ncached] = ptr;
tbin->ncached++;
tcache_event(tcache);
}
JEMALLOC_INLINE void
tcache_dalloc_large(tcache_t *tcache, void *ptr, size_t size)
{
size_t binind;
tcache_bin_t *tbin;
tcache_bin_info_t *tbin_info;
assert((size & PAGE_MASK) == 0);
assert(tcache_salloc(ptr) > SMALL_MAXCLASS);
assert(tcache_salloc(ptr) <= tcache_maxclass);
binind = NBINS + (size >> LG_PAGE) - 1;
if (config_fill && opt_junk)
memset(ptr, 0x5a, size);
tbin = &tcache->tbins[binind];
tbin_info = &tcache_bin_info[binind];
if (tbin->ncached == tbin_info->ncached_max) {
tcache_bin_flush_large(tbin, binind, (tbin_info->ncached_max >>
1), tcache);
}
assert(tbin->ncached < tbin_info->ncached_max);
tbin->avail[tbin->ncached] = ptr;
tbin->ncached++;
tcache_event(tcache);
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/tsd.h
Vendored
+397
View File
@@ -0,0 +1,397 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/* Maximum number of malloc_tsd users with cleanup functions. */
#define MALLOC_TSD_CLEANUPS_MAX 8
typedef bool (*malloc_tsd_cleanup_t)(void);
/*
* TLS/TSD-agnostic macro-based implementation of thread-specific data. There
* are four macros that support (at least) three use cases: file-private,
* library-private, and library-private inlined. Following is an example
* library-private tsd variable:
*
* In example.h:
* typedef struct {
* int x;
* int y;
* } example_t;
* #define EX_INITIALIZER JEMALLOC_CONCAT({0, 0})
* malloc_tsd_protos(, example, example_t *)
* malloc_tsd_externs(example, example_t *)
* In example.c:
* malloc_tsd_data(, example, example_t *, EX_INITIALIZER)
* malloc_tsd_funcs(, example, example_t *, EX_INITIALIZER,
* example_tsd_cleanup)
*
* The result is a set of generated functions, e.g.:
*
* bool example_tsd_boot(void) {...}
* example_t **example_tsd_get() {...}
* void example_tsd_set(example_t **val) {...}
*
* Note that all of the functions deal in terms of (a_type *) rather than
* (a_type) so that it is possible to support non-pointer types (unlike
* pthreads TSD). example_tsd_cleanup() is passed an (a_type *) pointer that is
* cast to (void *). This means that the cleanup function needs to cast *and*
* dereference the function argument, e.g.:
*
* void
* example_tsd_cleanup(void *arg)
* {
* example_t *example = *(example_t **)arg;
*
* [...]
* if ([want the cleanup function to be called again]) {
* example_tsd_set(&example);
* }
* }
*
* If example_tsd_set() is called within example_tsd_cleanup(), it will be
* called again. This is similar to how pthreads TSD destruction works, except
* that pthreads only calls the cleanup function again if the value was set to
* non-NULL.
*/
/* malloc_tsd_protos(). */
#define malloc_tsd_protos(a_attr, a_name, a_type) \
a_attr bool \
a_name##_tsd_boot(void); \
a_attr a_type * \
a_name##_tsd_get(void); \
a_attr void \
a_name##_tsd_set(a_type *val);
/* malloc_tsd_externs(). */
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#define malloc_tsd_externs(a_name, a_type) \
extern __thread a_type a_name##_tls; \
extern __thread bool a_name##_initialized; \
extern bool a_name##_booted;
#elif (defined(JEMALLOC_TLS))
#define malloc_tsd_externs(a_name, a_type) \
extern __thread a_type a_name##_tls; \
extern pthread_key_t a_name##_tsd; \
extern bool a_name##_booted;
#elif (defined(_WIN32))
#define malloc_tsd_externs(a_name, a_type) \
extern DWORD a_name##_tsd; \
extern bool a_name##_booted;
#else
#define malloc_tsd_externs(a_name, a_type) \
extern pthread_key_t a_name##_tsd; \
extern bool a_name##_booted;
#endif
/* malloc_tsd_data(). */
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr __thread a_type JEMALLOC_TLS_MODEL \
a_name##_tls = a_initializer; \
a_attr __thread bool JEMALLOC_TLS_MODEL \
a_name##_initialized = false; \
a_attr bool a_name##_booted = false;
#elif (defined(JEMALLOC_TLS))
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr __thread a_type JEMALLOC_TLS_MODEL \
a_name##_tls = a_initializer; \
a_attr pthread_key_t a_name##_tsd; \
a_attr bool a_name##_booted = false;
#elif (defined(_WIN32))
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr DWORD a_name##_tsd; \
a_attr bool a_name##_booted = false;
#else
#define malloc_tsd_data(a_attr, a_name, a_type, a_initializer) \
a_attr pthread_key_t a_name##_tsd; \
a_attr bool a_name##_booted = false;
#endif
/* malloc_tsd_funcs(). */
#ifdef JEMALLOC_MALLOC_THREAD_CLEANUP
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Initialization/cleanup. */ \
a_attr bool \
a_name##_tsd_cleanup_wrapper(void) \
{ \
\
if (a_name##_initialized) { \
a_name##_initialized = false; \
a_cleanup(&a_name##_tls); \
} \
return (a_name##_initialized); \
} \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
if (a_cleanup != malloc_tsd_no_cleanup) { \
malloc_tsd_cleanup_register( \
&a_name##_tsd_cleanup_wrapper); \
} \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
\
assert(a_name##_booted); \
return (&a_name##_tls); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
\
assert(a_name##_booted); \
a_name##_tls = (*val); \
if (a_cleanup != malloc_tsd_no_cleanup) \
a_name##_initialized = true; \
}
#elif (defined(JEMALLOC_TLS))
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Initialization/cleanup. */ \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
if (a_cleanup != malloc_tsd_no_cleanup) { \
if (pthread_key_create(&a_name##_tsd, a_cleanup) != 0) \
return (true); \
} \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
\
assert(a_name##_booted); \
return (&a_name##_tls); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
\
assert(a_name##_booted); \
a_name##_tls = (*val); \
if (a_cleanup != malloc_tsd_no_cleanup) { \
if (pthread_setspecific(a_name##_tsd, \
(void *)(&a_name##_tls))) { \
malloc_write("<jemalloc>: Error" \
" setting TSD for "#a_name"\n"); \
if (opt_abort) \
abort(); \
} \
} \
}
#elif (defined(_WIN32))
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Data structure. */ \
typedef struct { \
bool initialized; \
a_type val; \
} a_name##_tsd_wrapper_t; \
/* Initialization/cleanup. */ \
a_attr bool \
a_name##_tsd_cleanup_wrapper(void) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
wrapper = (a_name##_tsd_wrapper_t *) TlsGetValue(a_name##_tsd); \
if (wrapper == NULL) \
return (false); \
if (a_cleanup != malloc_tsd_no_cleanup && \
wrapper->initialized) { \
a_type val = wrapper->val; \
a_type tsd_static_data = a_initializer; \
wrapper->initialized = false; \
wrapper->val = tsd_static_data; \
a_cleanup(&val); \
if (wrapper->initialized) { \
/* Trigger another cleanup round. */ \
return (true); \
} \
} \
malloc_tsd_dalloc(wrapper); \
return (false); \
} \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
a_name##_tsd = TlsAlloc(); \
if (a_name##_tsd == TLS_OUT_OF_INDEXES) \
return (true); \
if (a_cleanup != malloc_tsd_no_cleanup) { \
malloc_tsd_cleanup_register( \
&a_name##_tsd_cleanup_wrapper); \
} \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_name##_tsd_wrapper_t * \
a_name##_tsd_get_wrapper(void) \
{ \
a_name##_tsd_wrapper_t *wrapper = (a_name##_tsd_wrapper_t *) \
TlsGetValue(a_name##_tsd); \
\
if (wrapper == NULL) { \
wrapper = (a_name##_tsd_wrapper_t *) \
malloc_tsd_malloc(sizeof(a_name##_tsd_wrapper_t)); \
if (wrapper == NULL) { \
malloc_write("<jemalloc>: Error allocating" \
" TSD for "#a_name"\n"); \
abort(); \
} else { \
static a_type tsd_static_data = a_initializer; \
wrapper->initialized = false; \
wrapper->val = tsd_static_data; \
} \
if (!TlsSetValue(a_name##_tsd, (void *)wrapper)) { \
malloc_write("<jemalloc>: Error setting" \
" TSD for "#a_name"\n"); \
abort(); \
} \
} \
return (wrapper); \
} \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
return (&wrapper->val); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
wrapper->val = *(val); \
if (a_cleanup != malloc_tsd_no_cleanup) \
wrapper->initialized = true; \
}
#else
#define malloc_tsd_funcs(a_attr, a_name, a_type, a_initializer, \
a_cleanup) \
/* Data structure. */ \
typedef struct { \
bool initialized; \
a_type val; \
} a_name##_tsd_wrapper_t; \
/* Initialization/cleanup. */ \
a_attr void \
a_name##_tsd_cleanup_wrapper(void *arg) \
{ \
a_name##_tsd_wrapper_t *wrapper = (a_name##_tsd_wrapper_t *)arg;\
\
if (a_cleanup != malloc_tsd_no_cleanup && \
wrapper->initialized) { \
wrapper->initialized = false; \
a_cleanup(&wrapper->val); \
if (wrapper->initialized) { \
/* Trigger another cleanup round. */ \
if (pthread_setspecific(a_name##_tsd, \
(void *)wrapper)) { \
malloc_write("<jemalloc>: Error" \
" setting TSD for "#a_name"\n"); \
if (opt_abort) \
abort(); \
} \
return; \
} \
} \
malloc_tsd_dalloc(wrapper); \
} \
a_attr bool \
a_name##_tsd_boot(void) \
{ \
\
if (pthread_key_create(&a_name##_tsd, \
a_name##_tsd_cleanup_wrapper) != 0) \
return (true); \
a_name##_booted = true; \
return (false); \
} \
/* Get/set. */ \
a_attr a_name##_tsd_wrapper_t * \
a_name##_tsd_get_wrapper(void) \
{ \
a_name##_tsd_wrapper_t *wrapper = (a_name##_tsd_wrapper_t *) \
pthread_getspecific(a_name##_tsd); \
\
if (wrapper == NULL) { \
wrapper = (a_name##_tsd_wrapper_t *) \
malloc_tsd_malloc(sizeof(a_name##_tsd_wrapper_t)); \
if (wrapper == NULL) { \
malloc_write("<jemalloc>: Error allocating" \
" TSD for "#a_name"\n"); \
abort(); \
} else { \
static a_type tsd_static_data = a_initializer; \
wrapper->initialized = false; \
wrapper->val = tsd_static_data; \
} \
if (pthread_setspecific(a_name##_tsd, \
(void *)wrapper)) { \
malloc_write("<jemalloc>: Error setting" \
" TSD for "#a_name"\n"); \
abort(); \
} \
} \
return (wrapper); \
} \
a_attr a_type * \
a_name##_tsd_get(void) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
return (&wrapper->val); \
} \
a_attr void \
a_name##_tsd_set(a_type *val) \
{ \
a_name##_tsd_wrapper_t *wrapper; \
\
assert(a_name##_booted); \
wrapper = a_name##_tsd_get_wrapper(); \
wrapper->val = *(val); \
if (a_cleanup != malloc_tsd_no_cleanup) \
wrapper->initialized = true; \
}
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
void *malloc_tsd_malloc(size_t size);
void malloc_tsd_dalloc(void *wrapper);
void malloc_tsd_no_cleanup(void *);
void malloc_tsd_cleanup_register(bool (*f)(void));
void malloc_tsd_boot(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/util.h
Vendored
+160
View File
@@ -0,0 +1,160 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
/* Size of stack-allocated buffer passed to buferror(). */
#define BUFERROR_BUF 64
/*
* Size of stack-allocated buffer used by malloc_{,v,vc}printf(). This must be
* large enough for all possible uses within jemalloc.
*/
#define MALLOC_PRINTF_BUFSIZE 4096
/*
* Wrap a cpp argument that contains commas such that it isn't broken up into
* multiple arguments.
*/
#define JEMALLOC_CONCAT(...) __VA_ARGS__
/*
* Silence compiler warnings due to uninitialized values. This is used
* wherever the compiler fails to recognize that the variable is never used
* uninitialized.
*/
#ifdef JEMALLOC_CC_SILENCE
# define JEMALLOC_CC_SILENCE_INIT(v) = v
#else
# define JEMALLOC_CC_SILENCE_INIT(v)
#endif
/*
* Define a custom assert() in order to reduce the chances of deadlock during
* assertion failure.
*/
#ifndef assert
#define assert(e) do { \
if (config_debug && !(e)) { \
malloc_printf( \
"<jemalloc>: %s:%d: Failed assertion: \"%s\"\n", \
__FILE__, __LINE__, #e); \
abort(); \
} \
} while (0)
#endif
/* Use to assert a particular configuration, e.g., cassert(config_debug). */
#define cassert(c) do { \
if ((c) == false) \
assert(false); \
} while (0)
#ifndef not_reached
#define not_reached() do { \
if (config_debug) { \
malloc_printf( \
"<jemalloc>: %s:%d: Unreachable code reached\n", \
__FILE__, __LINE__); \
abort(); \
} \
} while (0)
#endif
#ifndef not_implemented
#define not_implemented() do { \
if (config_debug) { \
malloc_printf("<jemalloc>: %s:%d: Not implemented\n", \
__FILE__, __LINE__); \
abort(); \
} \
} while (0)
#endif
#define assert_not_implemented(e) do { \
if (config_debug && !(e)) \
not_implemented(); \
} while (0)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
int buferror(char *buf, size_t buflen);
uintmax_t malloc_strtoumax(const char *nptr, char **endptr, int base);
void malloc_write(const char *s);
/*
* malloc_vsnprintf() supports a subset of snprintf(3) that avoids floating
* point math.
*/
int malloc_vsnprintf(char *str, size_t size, const char *format,
va_list ap);
int malloc_snprintf(char *str, size_t size, const char *format, ...)
JEMALLOC_ATTR(format(printf, 3, 4));
void malloc_vcprintf(void (*write_cb)(void *, const char *), void *cbopaque,
const char *format, va_list ap);
void malloc_cprintf(void (*write)(void *, const char *), void *cbopaque,
const char *format, ...) JEMALLOC_ATTR(format(printf, 3, 4));
void malloc_printf(const char *format, ...)
JEMALLOC_ATTR(format(printf, 1, 2));
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
size_t pow2_ceil(size_t x);
void malloc_write(const char *s);
void set_errno(int errnum);
int get_errno(void);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_UTIL_C_))
/* Compute the smallest power of 2 that is >= x. */
JEMALLOC_INLINE size_t
pow2_ceil(size_t x)
{
x--;
x |= x >> 1;
x |= x >> 2;
x |= x >> 4;
x |= x >> 8;
x |= x >> 16;
#if (LG_SIZEOF_PTR == 3)
x |= x >> 32;
#endif
x++;
return (x);
}
/* Sets error code */
JEMALLOC_INLINE void
set_errno(int errnum)
{
#ifdef _WIN32
SetLastError(errnum);
#else
errno = errnum;
#endif
}
/* Get last error code */
JEMALLOC_INLINE int
get_errno(void)
{
#ifdef _WIN32
return (GetLastError());
#else
return (errno);
#endif
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/jemalloc.h.in
Vendored
+157
View File
@@ -0,0 +1,157 @@
#ifndef JEMALLOC_H_
#define JEMALLOC_H_
#ifdef __cplusplus
extern "C" {
#endif
#include <limits.h>
#include <strings.h>
#define JEMALLOC_VERSION "@jemalloc_version@"
#define JEMALLOC_VERSION_MAJOR @jemalloc_version_major@
#define JEMALLOC_VERSION_MINOR @jemalloc_version_minor@
#define JEMALLOC_VERSION_BUGFIX @jemalloc_version_bugfix@
#define JEMALLOC_VERSION_NREV @jemalloc_version_nrev@
#define JEMALLOC_VERSION_GID "@jemalloc_version_gid@"
#include "jemalloc_defs@install_suffix@.h"
#ifdef JEMALLOC_EXPERIMENTAL
#define ALLOCM_LG_ALIGN(la) (la)
#if LG_SIZEOF_PTR == 2
#define ALLOCM_ALIGN(a) (ffs(a)-1)
#else
#define ALLOCM_ALIGN(a) ((a < (size_t)INT_MAX) ? ffs(a)-1 : ffs(a>>32)+31)
#endif
#define ALLOCM_ZERO ((int)0x40)
#define ALLOCM_NO_MOVE ((int)0x80)
/* Bias arena index bits so that 0 encodes "ALLOCM_ARENA() unspecified". */
#define ALLOCM_ARENA(a) ((int)(((a)+1) << 8))
#define ALLOCM_SUCCESS 0
#define ALLOCM_ERR_OOM 1
#define ALLOCM_ERR_NOT_MOVED 2
#endif
/*
* The je_ prefix on the following public symbol declarations is an artifact of
* namespace management, and should be omitted in application code unless
* JEMALLOC_NO_DEMANGLE is defined (see below).
*/
extern JEMALLOC_EXPORT const char *je_malloc_conf;
extern JEMALLOC_EXPORT void (*je_malloc_message)(void *cbopaque,
const char *s);
JEMALLOC_EXPORT void *je_malloc(size_t size) JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT void *je_calloc(size_t num, size_t size)
JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT int je_posix_memalign(void **memptr, size_t alignment,
size_t size) JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT void *je_aligned_alloc(size_t alignment, size_t size)
JEMALLOC_ATTR(malloc);
JEMALLOC_EXPORT void *je_realloc(void *ptr, size_t size);
JEMALLOC_EXPORT void je_free(void *ptr);
#ifdef JEMALLOC_OVERRIDE_MEMALIGN
JEMALLOC_EXPORT void * je_memalign(size_t alignment, size_t size)
JEMALLOC_ATTR(malloc);
#endif
#ifdef JEMALLOC_OVERRIDE_VALLOC
JEMALLOC_EXPORT void * je_valloc(size_t size) JEMALLOC_ATTR(malloc);
#endif
JEMALLOC_EXPORT size_t je_malloc_usable_size(
JEMALLOC_USABLE_SIZE_CONST void *ptr);
JEMALLOC_EXPORT void je_malloc_stats_print(void (*write_cb)(void *,
const char *), void *je_cbopaque, const char *opts);
JEMALLOC_EXPORT int je_mallctl(const char *name, void *oldp,
size_t *oldlenp, void *newp, size_t newlen);
JEMALLOC_EXPORT int je_mallctlnametomib(const char *name, size_t *mibp,
size_t *miblenp);
JEMALLOC_EXPORT int je_mallctlbymib(const size_t *mib, size_t miblen,
void *oldp, size_t *oldlenp, void *newp, size_t newlen);
#ifdef JEMALLOC_EXPERIMENTAL
JEMALLOC_EXPORT int je_allocm(void **ptr, size_t *rsize, size_t size,
int flags) JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_rallocm(void **ptr, size_t *rsize, size_t size,
size_t extra, int flags) JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_sallocm(const void *ptr, size_t *rsize, int flags)
JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_dallocm(void *ptr, int flags)
JEMALLOC_ATTR(nonnull(1));
JEMALLOC_EXPORT int je_nallocm(size_t *rsize, size_t size, int flags);
#endif
/*
* By default application code must explicitly refer to mangled symbol names,
* so that it is possible to use jemalloc in conjunction with another allocator
* in the same application. Define JEMALLOC_MANGLE in order to cause automatic
* name mangling that matches the API prefixing that happened as a result of
* --with-mangling and/or --with-jemalloc-prefix configuration settings.
*/
#ifdef JEMALLOC_MANGLE
#ifndef JEMALLOC_NO_DEMANGLE
#define JEMALLOC_NO_DEMANGLE
#endif
#define malloc_conf je_malloc_conf
#define malloc_message je_malloc_message
#define malloc je_malloc
#define calloc je_calloc
#define posix_memalign je_posix_memalign
#define aligned_alloc je_aligned_alloc
#define realloc je_realloc
#define free je_free
#define malloc_usable_size je_malloc_usable_size
#define malloc_stats_print je_malloc_stats_print
#define mallctl je_mallctl
#define mallctlnametomib je_mallctlnametomib
#define mallctlbymib je_mallctlbymib
#define memalign je_memalign
#define valloc je_valloc
#ifdef JEMALLOC_EXPERIMENTAL
#define allocm je_allocm
#define rallocm je_rallocm
#define sallocm je_sallocm
#define dallocm je_dallocm
#define nallocm je_nallocm
#endif
#endif
/*
* The je_* macros can be used as stable alternative names for the public
* jemalloc API if JEMALLOC_NO_DEMANGLE is defined. This is primarily meant
* for use in jemalloc itself, but it can be used by application code to
* provide isolation from the name mangling specified via --with-mangling
* and/or --with-jemalloc-prefix.
*/
#ifndef JEMALLOC_NO_DEMANGLE
#undef je_malloc_conf
#undef je_malloc_message
#undef je_malloc
#undef je_calloc
#undef je_posix_memalign
#undef je_aligned_alloc
#undef je_realloc
#undef je_free
#undef je_malloc_usable_size
#undef je_malloc_stats_print
#undef je_mallctl
#undef je_mallctlnametomib
#undef je_mallctlbymib
#undef je_memalign
#undef je_valloc
#ifdef JEMALLOC_EXPERIMENTAL
#undef je_allocm
#undef je_rallocm
#undef je_sallocm
#undef je_dallocm
#undef je_nallocm
#endif
#endif
#ifdef __cplusplus
};
#endif
#endif /* JEMALLOC_H_ */
deps/jemalloc/include/jemalloc/jemalloc_defs.h.in
Vendored
+262
View File
@@ -0,0 +1,262 @@
/*
* If JEMALLOC_PREFIX is defined via --with-jemalloc-prefix, it will cause all
* public APIs to be prefixed. This makes it possible, with some care, to use
* multiple allocators simultaneously.
*/
#undef JEMALLOC_PREFIX
#undef JEMALLOC_CPREFIX
/*
* Name mangling for public symbols is controlled by --with-mangling and
* --with-jemalloc-prefix. With default settings the je_ prefix is stripped by
* these macro definitions.
*/
#undef je_malloc_conf
#undef je_malloc_message
#undef je_malloc
#undef je_calloc
#undef je_posix_memalign
#undef je_aligned_alloc
#undef je_realloc
#undef je_free
#undef je_malloc_usable_size
#undef je_malloc_stats_print
#undef je_mallctl
#undef je_mallctlnametomib
#undef je_mallctlbymib
#undef je_memalign
#undef je_valloc
#undef je_allocm
#undef je_rallocm
#undef je_sallocm
#undef je_dallocm
#undef je_nallocm
/*
* JEMALLOC_PRIVATE_NAMESPACE is used as a prefix for all library-private APIs.
* For shared libraries, symbol visibility mechanisms prevent these symbols
* from being exported, but for static libraries, naming collisions are a real
* possibility.
*/
#undef JEMALLOC_PRIVATE_NAMESPACE
#undef JEMALLOC_N
/*
* Hyper-threaded CPUs may need a special instruction inside spin loops in
* order to yield to another virtual CPU.
*/
#undef CPU_SPINWAIT
/* Defined if the equivalent of FreeBSD's atomic(9) functions are available. */
#undef JEMALLOC_ATOMIC9
/*
* Defined if OSAtomic*() functions are available, as provided by Darwin, and
* documented in the atomic(3) manual page.
*/
#undef JEMALLOC_OSATOMIC
/*
* Defined if __sync_add_and_fetch(uint32_t *, uint32_t) and
* __sync_sub_and_fetch(uint32_t *, uint32_t) are available, despite
* __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 not being defined (which means the
* functions are defined in libgcc instead of being inlines)
*/
#undef JE_FORCE_SYNC_COMPARE_AND_SWAP_4
/*
* Defined if __sync_add_and_fetch(uint64_t *, uint64_t) and
* __sync_sub_and_fetch(uint64_t *, uint64_t) are available, despite
* __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8 not being defined (which means the
* functions are defined in libgcc instead of being inlines)
*/
#undef JE_FORCE_SYNC_COMPARE_AND_SWAP_8
/*
* Defined if OSSpin*() functions are available, as provided by Darwin, and
* documented in the spinlock(3) manual page.
*/
#undef JEMALLOC_OSSPIN
/*
* Defined if _malloc_thread_cleanup() exists. At least in the case of
* FreeBSD, pthread_key_create() allocates, which if used during malloc
* bootstrapping will cause recursion into the pthreads library. Therefore, if
* _malloc_thread_cleanup() exists, use it as the basis for thread cleanup in
* malloc_tsd.
*/
#undef JEMALLOC_MALLOC_THREAD_CLEANUP
/*
* Defined if threaded initialization is known to be safe on this platform.
* Among other things, it must be possible to initialize a mutex without
* triggering allocation in order for threaded allocation to be safe.
*/
#undef JEMALLOC_THREADED_INIT
/*
* Defined if the pthreads implementation defines
* _pthread_mutex_init_calloc_cb(), in which case the function is used in order
* to avoid recursive allocation during mutex initialization.
*/
#undef JEMALLOC_MUTEX_INIT_CB
/* Defined if __attribute__((...)) syntax is supported. */
#undef JEMALLOC_HAVE_ATTR
#ifdef JEMALLOC_HAVE_ATTR
# define JEMALLOC_ATTR(s) __attribute__((s))
# define JEMALLOC_EXPORT JEMALLOC_ATTR(visibility("default"))
# define JEMALLOC_ALIGNED(s) JEMALLOC_ATTR(aligned(s))
# define JEMALLOC_SECTION(s) JEMALLOC_ATTR(section(s))
# define JEMALLOC_NOINLINE JEMALLOC_ATTR(noinline)
#elif _MSC_VER
# define JEMALLOC_ATTR(s)
# ifdef DLLEXPORT
# define JEMALLOC_EXPORT __declspec(dllexport)
# else
# define JEMALLOC_EXPORT __declspec(dllimport)
# endif
# define JEMALLOC_ALIGNED(s) __declspec(align(s))
# define JEMALLOC_SECTION(s) __declspec(allocate(s))
# define JEMALLOC_NOINLINE __declspec(noinline)
#else
# define JEMALLOC_ATTR(s)
# define JEMALLOC_EXPORT
# define JEMALLOC_ALIGNED(s)
# define JEMALLOC_SECTION(s)
# define JEMALLOC_NOINLINE
#endif
/* Defined if sbrk() is supported. */
#undef JEMALLOC_HAVE_SBRK
/* Non-empty if the tls_model attribute is supported. */
#undef JEMALLOC_TLS_MODEL
/* JEMALLOC_CC_SILENCE enables code that silences unuseful compiler warnings. */
#undef JEMALLOC_CC_SILENCE
/*
* JEMALLOC_DEBUG enables assertions and other sanity checks, and disables
* inline functions.
*/
#undef JEMALLOC_DEBUG
/* JEMALLOC_STATS enables statistics calculation. */
#undef JEMALLOC_STATS
/* JEMALLOC_PROF enables allocation profiling. */
#undef JEMALLOC_PROF
/* Use libunwind for profile backtracing if defined. */
#undef JEMALLOC_PROF_LIBUNWIND
/* Use libgcc for profile backtracing if defined. */
#undef JEMALLOC_PROF_LIBGCC
/* Use gcc intrinsics for profile backtracing if defined. */
#undef JEMALLOC_PROF_GCC
/*
* JEMALLOC_TCACHE enables a thread-specific caching layer for small objects.
* This makes it possible to allocate/deallocate objects without any locking
* when the cache is in the steady state.
*/
#undef JEMALLOC_TCACHE
/*
* JEMALLOC_DSS enables use of sbrk(2) to allocate chunks from the data storage
* segment (DSS).
*/
#undef JEMALLOC_DSS
/* Support memory filling (junk/zero/quarantine/redzone). */
#undef JEMALLOC_FILL
/* Support the experimental API. */
#undef JEMALLOC_EXPERIMENTAL
/* Support utrace(2)-based tracing. */
#undef JEMALLOC_UTRACE
/* Support Valgrind. */
#undef JEMALLOC_VALGRIND
/* Support optional abort() on OOM. */
#undef JEMALLOC_XMALLOC
/* Support lazy locking (avoid locking unless a second thread is launched). */
#undef JEMALLOC_LAZY_LOCK
/* One page is 2^STATIC_PAGE_SHIFT bytes. */
#undef STATIC_PAGE_SHIFT
/*
* If defined, use munmap() to unmap freed chunks, rather than storing them for
* later reuse. This is disabled by default on Linux because common sequences
* of mmap()/munmap() calls will cause virtual memory map holes.
*/
#undef JEMALLOC_MUNMAP
/*
* If defined, use mremap(...MREMAP_FIXED...) for huge realloc(). This is
* disabled by default because it is Linux-specific and it will cause virtual
* memory map holes, much like munmap(2) does.
*/
#undef JEMALLOC_MREMAP
/* TLS is used to map arenas and magazine caches to threads. */
#undef JEMALLOC_TLS
/*
* JEMALLOC_IVSALLOC enables ivsalloc(), which verifies that pointers reside
* within jemalloc-owned chunks before dereferencing them.
*/
#undef JEMALLOC_IVSALLOC
/*
* Define overrides for non-standard allocator-related functions if they
* are present on the system.
*/
#undef JEMALLOC_OVERRIDE_MEMALIGN
#undef JEMALLOC_OVERRIDE_VALLOC
/*
* At least Linux omits the "const" in:
*
* size_t malloc_usable_size(const void *ptr);
*
* Match the operating system's prototype.
*/
#undef JEMALLOC_USABLE_SIZE_CONST
/*
* Darwin (OS X) uses zones to work around Mach-O symbol override shortcomings.
*/
#undef JEMALLOC_ZONE
#undef JEMALLOC_ZONE_VERSION
/*
* Methods for purging unused pages differ between operating systems.
*
* madvise(..., MADV_DONTNEED) : On Linux, this immediately discards pages,
* such that new pages will be demand-zeroed if
* the address region is later touched.
* madvise(..., MADV_FREE) : On FreeBSD and Darwin, this marks pages as being
* unused, such that they will be discarded rather
* than swapped out.
*/
#undef JEMALLOC_PURGE_MADVISE_DONTNEED
#undef JEMALLOC_PURGE_MADVISE_FREE
/* sizeof(void *) == 2^LG_SIZEOF_PTR. */
#undef LG_SIZEOF_PTR
/* sizeof(int) == 2^LG_SIZEOF_INT. */
#undef LG_SIZEOF_INT
/* sizeof(long) == 2^LG_SIZEOF_LONG. */
#undef LG_SIZEOF_LONG
/* sizeof(intmax_t) == 2^LG_SIZEOF_INTMAX_T. */
#undef LG_SIZEOF_INTMAX_T
deps/jemalloc/include/msvc_compat/inttypes.h
Vendored
+313
View File
@@ -0,0 +1,313 @@
// ISO C9x compliant inttypes.h for Microsoft Visual Studio
// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124
//
// Copyright (c) 2006 Alexander Chemeris
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
//
// 3. The name of the author may be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
///////////////////////////////////////////////////////////////////////////////
#ifndef _MSC_VER // [
#error "Use this header only with Microsoft Visual C++ compilers!"
#endif // _MSC_VER ]
#ifndef _MSC_INTTYPES_H_ // [
#define _MSC_INTTYPES_H_
#if _MSC_VER > 1000
#pragma once
#endif
#include "stdint.h"
// 7.8 Format conversion of integer types
typedef struct {
intmax_t quot;
intmax_t rem;
} imaxdiv_t;
// 7.8.1 Macros for format specifiers
#if !defined(__cplusplus) || defined(__STDC_FORMAT_MACROS) // [ See footnote 185 at page 198
#ifdef _WIN64
# define __PRI64_PREFIX "l"
# define __PRIPTR_PREFIX "l"
#else
# define __PRI64_PREFIX "ll"
# define __PRIPTR_PREFIX
#endif
// The fprintf macros for signed integers are:
#define PRId8 "d"
#define PRIi8 "i"
#define PRIdLEAST8 "d"
#define PRIiLEAST8 "i"
#define PRIdFAST8 "d"
#define PRIiFAST8 "i"
#define PRId16 "hd"
#define PRIi16 "hi"
#define PRIdLEAST16 "hd"
#define PRIiLEAST16 "hi"
#define PRIdFAST16 "hd"
#define PRIiFAST16 "hi"
#define PRId32 "d"
#define PRIi32 "i"
#define PRIdLEAST32 "d"
#define PRIiLEAST32 "i"
#define PRIdFAST32 "d"
#define PRIiFAST32 "i"
#define PRId64 __PRI64_PREFIX "d"
#define PRIi64 __PRI64_PREFIX "i"
#define PRIdLEAST64 __PRI64_PREFIX "d"
#define PRIiLEAST64 __PRI64_PREFIX "i"
#define PRIdFAST64 __PRI64_PREFIX "d"
#define PRIiFAST64 __PRI64_PREFIX "i"
#define PRIdMAX __PRI64_PREFIX "d"
#define PRIiMAX __PRI64_PREFIX "i"
#define PRIdPTR __PRIPTR_PREFIX "d"
#define PRIiPTR __PRIPTR_PREFIX "i"
// The fprintf macros for unsigned integers are:
#define PRIo8 "o"
#define PRIu8 "u"
#define PRIx8 "x"
#define PRIX8 "X"
#define PRIoLEAST8 "o"
#define PRIuLEAST8 "u"
#define PRIxLEAST8 "x"
#define PRIXLEAST8 "X"
#define PRIoFAST8 "o"
#define PRIuFAST8 "u"
#define PRIxFAST8 "x"
#define PRIXFAST8 "X"
#define PRIo16 "ho"
#define PRIu16 "hu"
#define PRIx16 "hx"
#define PRIX16 "hX"
#define PRIoLEAST16 "ho"
#define PRIuLEAST16 "hu"
#define PRIxLEAST16 "hx"
#define PRIXLEAST16 "hX"
#define PRIoFAST16 "ho"
#define PRIuFAST16 "hu"
#define PRIxFAST16 "hx"
#define PRIXFAST16 "hX"
#define PRIo32 "o"
#define PRIu32 "u"
#define PRIx32 "x"
#define PRIX32 "X"
#define PRIoLEAST32 "o"
#define PRIuLEAST32 "u"
#define PRIxLEAST32 "x"
#define PRIXLEAST32 "X"
#define PRIoFAST32 "o"
#define PRIuFAST32 "u"
#define PRIxFAST32 "x"
#define PRIXFAST32 "X"
#define PRIo64 __PRI64_PREFIX "o"
#define PRIu64 __PRI64_PREFIX "u"
#define PRIx64 __PRI64_PREFIX "x"
#define PRIX64 __PRI64_PREFIX "X"
#define PRIoLEAST64 __PRI64_PREFIX "o"
#define PRIuLEAST64 __PRI64_PREFIX "u"
#define PRIxLEAST64 __PRI64_PREFIX "x"
#define PRIXLEAST64 __PRI64_PREFIX "X"
#define PRIoFAST64 __PRI64_PREFIX "o"
#define PRIuFAST64 __PRI64_PREFIX "u"
#define PRIxFAST64 __PRI64_PREFIX "x"
#define PRIXFAST64 __PRI64_PREFIX "X"
#define PRIoMAX __PRI64_PREFIX "o"
#define PRIuMAX __PRI64_PREFIX "u"
#define PRIxMAX __PRI64_PREFIX "x"
#define PRIXMAX __PRI64_PREFIX "X"
#define PRIoPTR __PRIPTR_PREFIX "o"
#define PRIuPTR __PRIPTR_PREFIX "u"
#define PRIxPTR __PRIPTR_PREFIX "x"
#define PRIXPTR __PRIPTR_PREFIX "X"
// The fscanf macros for signed integers are:
#define SCNd8 "d"
#define SCNi8 "i"
#define SCNdLEAST8 "d"
#define SCNiLEAST8 "i"
#define SCNdFAST8 "d"
#define SCNiFAST8 "i"
#define SCNd16 "hd"
#define SCNi16 "hi"
#define SCNdLEAST16 "hd"
#define SCNiLEAST16 "hi"
#define SCNdFAST16 "hd"
#define SCNiFAST16 "hi"
#define SCNd32 "ld"
#define SCNi32 "li"
#define SCNdLEAST32 "ld"
#define SCNiLEAST32 "li"
#define SCNdFAST32 "ld"
#define SCNiFAST32 "li"
#define SCNd64 "I64d"
#define SCNi64 "I64i"
#define SCNdLEAST64 "I64d"
#define SCNiLEAST64 "I64i"
#define SCNdFAST64 "I64d"
#define SCNiFAST64 "I64i"
#define SCNdMAX "I64d"
#define SCNiMAX "I64i"
#ifdef _WIN64 // [
# define SCNdPTR "I64d"
# define SCNiPTR "I64i"
#else // _WIN64 ][
# define SCNdPTR "ld"
# define SCNiPTR "li"
#endif // _WIN64 ]
// The fscanf macros for unsigned integers are:
#define SCNo8 "o"
#define SCNu8 "u"
#define SCNx8 "x"
#define SCNX8 "X"
#define SCNoLEAST8 "o"
#define SCNuLEAST8 "u"
#define SCNxLEAST8 "x"
#define SCNXLEAST8 "X"
#define SCNoFAST8 "o"
#define SCNuFAST8 "u"
#define SCNxFAST8 "x"
#define SCNXFAST8 "X"
#define SCNo16 "ho"
#define SCNu16 "hu"
#define SCNx16 "hx"
#define SCNX16 "hX"
#define SCNoLEAST16 "ho"
#define SCNuLEAST16 "hu"
#define SCNxLEAST16 "hx"
#define SCNXLEAST16 "hX"
#define SCNoFAST16 "ho"
#define SCNuFAST16 "hu"
#define SCNxFAST16 "hx"
#define SCNXFAST16 "hX"
#define SCNo32 "lo"
#define SCNu32 "lu"
#define SCNx32 "lx"
#define SCNX32 "lX"
#define SCNoLEAST32 "lo"
#define SCNuLEAST32 "lu"
#define SCNxLEAST32 "lx"
#define SCNXLEAST32 "lX"
#define SCNoFAST32 "lo"
#define SCNuFAST32 "lu"
#define SCNxFAST32 "lx"
#define SCNXFAST32 "lX"
#define SCNo64 "I64o"
#define SCNu64 "I64u"
#define SCNx64 "I64x"
#define SCNX64 "I64X"
#define SCNoLEAST64 "I64o"
#define SCNuLEAST64 "I64u"
#define SCNxLEAST64 "I64x"
#define SCNXLEAST64 "I64X"
#define SCNoFAST64 "I64o"
#define SCNuFAST64 "I64u"
#define SCNxFAST64 "I64x"
#define SCNXFAST64 "I64X"
#define SCNoMAX "I64o"
#define SCNuMAX "I64u"
#define SCNxMAX "I64x"
#define SCNXMAX "I64X"
#ifdef _WIN64 // [
# define SCNoPTR "I64o"
# define SCNuPTR "I64u"
# define SCNxPTR "I64x"
# define SCNXPTR "I64X"
#else // _WIN64 ][
# define SCNoPTR "lo"
# define SCNuPTR "lu"
# define SCNxPTR "lx"
# define SCNXPTR "lX"
#endif // _WIN64 ]
#endif // __STDC_FORMAT_MACROS ]
// 7.8.2 Functions for greatest-width integer types
// 7.8.2.1 The imaxabs function
#define imaxabs _abs64
// 7.8.2.2 The imaxdiv function
// This is modified version of div() function from Microsoft's div.c found
// in %MSVC.NET%\crt\src\div.c
#ifdef STATIC_IMAXDIV // [
static
#else // STATIC_IMAXDIV ][
_inline
#endif // STATIC_IMAXDIV ]
imaxdiv_t __cdecl imaxdiv(intmax_t numer, intmax_t denom)
{
imaxdiv_t result;
result.quot = numer / denom;
result.rem = numer % denom;
if (numer < 0 && result.rem > 0) {
// did division wrong; must fix up
++result.quot;
result.rem -= denom;
}
return result;
}
// 7.8.2.3 The strtoimax and strtoumax functions
#define strtoimax _strtoi64
#define strtoumax _strtoui64
// 7.8.2.4 The wcstoimax and wcstoumax functions
#define wcstoimax _wcstoi64
#define wcstoumax _wcstoui64
#endif // _MSC_INTTYPES_H_ ]
deps/jemalloc/include/msvc_compat/stdbool.h
Vendored
+16
View File
@@ -0,0 +1,16 @@
#ifndef stdbool_h
#define stdbool_h
#include <wtypes.h>
/* MSVC doesn't define _Bool or bool in C, but does have BOOL */
/* Note this doesn't pass autoconf's test because (bool) 0.5 != true */
typedef BOOL _Bool;
#define bool _Bool
#define true 1
#define false 0
#define __bool_true_false_are_defined 1
#endif /* stdbool_h */
deps/jemalloc/include/msvc_compat/stdint.h
Vendored
+247
View File
@@ -0,0 +1,247 @@
// ISO C9x compliant stdint.h for Microsoft Visual Studio
// Based on ISO/IEC 9899:TC2 Committee draft (May 6, 2005) WG14/N1124
//
// Copyright (c) 2006-2008 Alexander Chemeris
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
//
// 3. The name of the author may be used to endorse or promote products
// derived from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
// WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
// MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
// EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
// OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
// WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
// OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
// ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
//
///////////////////////////////////////////////////////////////////////////////
#ifndef _MSC_VER // [
#error "Use this header only with Microsoft Visual C++ compilers!"
#endif // _MSC_VER ]
#ifndef _MSC_STDINT_H_ // [
#define _MSC_STDINT_H_
#if _MSC_VER > 1000
#pragma once
#endif
#include <limits.h>
// For Visual Studio 6 in C++ mode and for many Visual Studio versions when
// compiling for ARM we should wrap <wchar.h> include with 'extern "C++" {}'
// or compiler give many errors like this:
// error C2733: second C linkage of overloaded function 'wmemchr' not allowed
#ifdef __cplusplus
extern "C" {
#endif
# include <wchar.h>
#ifdef __cplusplus
}
#endif
// Define _W64 macros to mark types changing their size, like intptr_t.
#ifndef _W64
# if !defined(__midl) && (defined(_X86_) || defined(_M_IX86)) && _MSC_VER >= 1300
# define _W64 __w64
# else
# define _W64
# endif
#endif
// 7.18.1 Integer types
// 7.18.1.1 Exact-width integer types
// Visual Studio 6 and Embedded Visual C++ 4 doesn't
// realize that, e.g. char has the same size as __int8
// so we give up on __intX for them.
#if (_MSC_VER < 1300)
typedef signed char int8_t;
typedef signed short int16_t;
typedef signed int int32_t;
typedef unsigned char uint8_t;
typedef unsigned short uint16_t;
typedef unsigned int uint32_t;
#else
typedef signed __int8 int8_t;
typedef signed __int16 int16_t;
typedef signed __int32 int32_t;
typedef unsigned __int8 uint8_t;
typedef unsigned __int16 uint16_t;
typedef unsigned __int32 uint32_t;
#endif
typedef signed __int64 int64_t;
typedef unsigned __int64 uint64_t;
// 7.18.1.2 Minimum-width integer types
typedef int8_t int_least8_t;
typedef int16_t int_least16_t;
typedef int32_t int_least32_t;
typedef int64_t int_least64_t;
typedef uint8_t uint_least8_t;
typedef uint16_t uint_least16_t;
typedef uint32_t uint_least32_t;
typedef uint64_t uint_least64_t;
// 7.18.1.3 Fastest minimum-width integer types
typedef int8_t int_fast8_t;
typedef int16_t int_fast16_t;
typedef int32_t int_fast32_t;
typedef int64_t int_fast64_t;
typedef uint8_t uint_fast8_t;
typedef uint16_t uint_fast16_t;
typedef uint32_t uint_fast32_t;
typedef uint64_t uint_fast64_t;
// 7.18.1.4 Integer types capable of holding object pointers
#ifdef _WIN64 // [
typedef signed __int64 intptr_t;
typedef unsigned __int64 uintptr_t;
#else // _WIN64 ][
typedef _W64 signed int intptr_t;
typedef _W64 unsigned int uintptr_t;
#endif // _WIN64 ]
// 7.18.1.5 Greatest-width integer types
typedef int64_t intmax_t;
typedef uint64_t uintmax_t;
// 7.18.2 Limits of specified-width integer types
#if !defined(__cplusplus) || defined(__STDC_LIMIT_MACROS) // [ See footnote 220 at page 257 and footnote 221 at page 259
// 7.18.2.1 Limits of exact-width integer types
#define INT8_MIN ((int8_t)_I8_MIN)
#define INT8_MAX _I8_MAX
#define INT16_MIN ((int16_t)_I16_MIN)
#define INT16_MAX _I16_MAX
#define INT32_MIN ((int32_t)_I32_MIN)
#define INT32_MAX _I32_MAX
#define INT64_MIN ((int64_t)_I64_MIN)
#define INT64_MAX _I64_MAX
#define UINT8_MAX _UI8_MAX
#define UINT16_MAX _UI16_MAX
#define UINT32_MAX _UI32_MAX
#define UINT64_MAX _UI64_MAX
// 7.18.2.2 Limits of minimum-width integer types
#define INT_LEAST8_MIN INT8_MIN
#define INT_LEAST8_MAX INT8_MAX
#define INT_LEAST16_MIN INT16_MIN
#define INT_LEAST16_MAX INT16_MAX
#define INT_LEAST32_MIN INT32_MIN
#define INT_LEAST32_MAX INT32_MAX
#define INT_LEAST64_MIN INT64_MIN
#define INT_LEAST64_MAX INT64_MAX
#define UINT_LEAST8_MAX UINT8_MAX
#define UINT_LEAST16_MAX UINT16_MAX
#define UINT_LEAST32_MAX UINT32_MAX
#define UINT_LEAST64_MAX UINT64_MAX
// 7.18.2.3 Limits of fastest minimum-width integer types
#define INT_FAST8_MIN INT8_MIN
#define INT_FAST8_MAX INT8_MAX
#define INT_FAST16_MIN INT16_MIN
#define INT_FAST16_MAX INT16_MAX
#define INT_FAST32_MIN INT32_MIN
#define INT_FAST32_MAX INT32_MAX
#define INT_FAST64_MIN INT64_MIN
#define INT_FAST64_MAX INT64_MAX
#define UINT_FAST8_MAX UINT8_MAX
#define UINT_FAST16_MAX UINT16_MAX
#define UINT_FAST32_MAX UINT32_MAX
#define UINT_FAST64_MAX UINT64_MAX
// 7.18.2.4 Limits of integer types capable of holding object pointers
#ifdef _WIN64 // [
# define INTPTR_MIN INT64_MIN
# define INTPTR_MAX INT64_MAX
# define UINTPTR_MAX UINT64_MAX
#else // _WIN64 ][
# define INTPTR_MIN INT32_MIN
# define INTPTR_MAX INT32_MAX
# define UINTPTR_MAX UINT32_MAX
#endif // _WIN64 ]
// 7.18.2.5 Limits of greatest-width integer types
#define INTMAX_MIN INT64_MIN
#define INTMAX_MAX INT64_MAX
#define UINTMAX_MAX UINT64_MAX
// 7.18.3 Limits of other integer types
#ifdef _WIN64 // [
# define PTRDIFF_MIN _I64_MIN
# define PTRDIFF_MAX _I64_MAX
#else // _WIN64 ][
# define PTRDIFF_MIN _I32_MIN
# define PTRDIFF_MAX _I32_MAX
#endif // _WIN64 ]
#define SIG_ATOMIC_MIN INT_MIN
#define SIG_ATOMIC_MAX INT_MAX
#ifndef SIZE_MAX // [
# ifdef _WIN64 // [
# define SIZE_MAX _UI64_MAX
# else // _WIN64 ][
# define SIZE_MAX _UI32_MAX
# endif // _WIN64 ]
#endif // SIZE_MAX ]
// WCHAR_MIN and WCHAR_MAX are also defined in <wchar.h>
#ifndef WCHAR_MIN // [
# define WCHAR_MIN 0
#endif // WCHAR_MIN ]
#ifndef WCHAR_MAX // [
# define WCHAR_MAX _UI16_MAX
#endif // WCHAR_MAX ]
#define WINT_MIN 0
#define WINT_MAX _UI16_MAX
#endif // __STDC_LIMIT_MACROS ]
// 7.18.4 Limits of other integer types
#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) // [ See footnote 224 at page 260
// 7.18.4.1 Macros for minimum-width integer constants
#define INT8_C(val) val##i8
#define INT16_C(val) val##i16
#define INT32_C(val) val##i32
#define INT64_C(val) val##i64
#define UINT8_C(val) val##ui8
#define UINT16_C(val) val##ui16
#define UINT32_C(val) val##ui32
#define UINT64_C(val) val##ui64
// 7.18.4.2 Macros for greatest-width integer constants
#define INTMAX_C INT64_C
#define UINTMAX_C UINT64_C
#endif // __STDC_CONSTANT_MACROS ]
#endif // _MSC_STDINT_H_ ]
deps/jemalloc/include/msvc_compat/strings.h
Vendored
+23
View File
@@ -0,0 +1,23 @@
#ifndef strings_h
#define strings_h
/* MSVC doesn't define ffs/ffsl. This dummy strings.h header is provided
* for both */
#include <intrin.h>
#pragma intrinsic(_BitScanForward)
static __forceinline int ffsl(long x)
{
unsigned long i;
if (_BitScanForward(&i, x))
return (i + 1);
return (0);
}
static __forceinline int ffs(int x)
{
return (ffsl(x));
}
#endif
deps/jemalloc/install-sh
Vendored Executable
+250
View File
@@ -0,0 +1,250 @@
#! /bin/sh
#
# install - install a program, script, or datafile
# This comes from X11R5 (mit/util/scripts/install.sh).
#
# Copyright 1991 by the Massachusetts Institute of Technology
#
# Permission to use, copy, modify, distribute, and sell this software and its
# documentation for any purpose is hereby granted without fee, provided that
# the above copyright notice appear in all copies and that both that
# copyright notice and this permission notice appear in supporting
# documentation, and that the name of M.I.T. not be used in advertising or
# publicity pertaining to distribution of the software without specific,
# written prior permission. M.I.T. makes no representations about the
# suitability of this software for any purpose. It is provided "as is"
# without express or implied warranty.
#
# Calling this script install-sh is preferred over install.sh, to prevent
# `make' implicit rules from creating a file called install from it
# when there is no Makefile.
#
# This script is compatible with the BSD install script, but was written
# from scratch. It can only install one file at a time, a restriction
# shared with many OS's install programs.
# set DOITPROG to echo to test this script
# Don't use :- since 4.3BSD and earlier shells don't like it.
doit="${DOITPROG-}"
# put in absolute paths if you don't have them in your path; or use env. vars.
mvprog="${MVPROG-mv}"
cpprog="${CPPROG-cp}"
chmodprog="${CHMODPROG-chmod}"
chownprog="${CHOWNPROG-chown}"
chgrpprog="${CHGRPPROG-chgrp}"
stripprog="${STRIPPROG-strip}"
rmprog="${RMPROG-rm}"
mkdirprog="${MKDIRPROG-mkdir}"
transformbasename=""
transform_arg=""
instcmd="$mvprog"
chmodcmd="$chmodprog 0755"
chowncmd=""
chgrpcmd=""
stripcmd=""
rmcmd="$rmprog -f"
mvcmd="$mvprog"
src=""
dst=""
dir_arg=""
while [ x"$1" != x ]; do
case $1 in
-c) instcmd="$cpprog"
shift
continue;;
-d) dir_arg=true
shift
continue;;
-m) chmodcmd="$chmodprog $2"
shift
shift
continue;;
-o) chowncmd="$chownprog $2"
shift
shift
continue;;
-g) chgrpcmd="$chgrpprog $2"
shift
shift
continue;;
-s) stripcmd="$stripprog"
shift
continue;;
-t=*) transformarg=`echo $1 | sed 's/-t=//'`
shift
continue;;
-b=*) transformbasename=`echo $1 | sed 's/-b=//'`
shift
continue;;
*) if [ x"$src" = x ]
then
src=$1
else
# this colon is to work around a 386BSD /bin/sh bug
:
dst=$1
fi
shift
continue;;
esac
done
if [ x"$src" = x ]
then
echo "install: no input file specified"
exit 1
else
true
fi
if [ x"$dir_arg" != x ]; then
dst=$src
src=""
if [ -d $dst ]; then
instcmd=:
else
instcmd=mkdir
fi
else
# Waiting for this to be detected by the "$instcmd $src $dsttmp" command
# might cause directories to be created, which would be especially bad
# if $src (and thus $dsttmp) contains '*'.
if [ -f $src -o -d $src ]
then
true
else
echo "install: $src does not exist"
exit 1
fi
if [ x"$dst" = x ]
then
echo "install: no destination specified"
exit 1
else
true
fi
# If destination is a directory, append the input filename; if your system
# does not like double slashes in filenames, you may need to add some logic
if [ -d $dst ]
then
dst="$dst"/`basename $src`
else
true
fi
fi
## this sed command emulates the dirname command
dstdir=`echo $dst | sed -e 's,[^/]*$,,;s,/$,,;s,^$,.,'`
# Make sure that the destination directory exists.
# this part is taken from Noah Friedman's mkinstalldirs script
# Skip lots of stat calls in the usual case.
if [ ! -d "$dstdir" ]; then
defaultIFS='
'
IFS="${IFS-${defaultIFS}}"
oIFS="${IFS}"
# Some sh's can't handle IFS=/ for some reason.
IFS='%'
set - `echo ${dstdir} | sed -e 's@/@%@g' -e 's@^%@/@'`
IFS="${oIFS}"
pathcomp=''
while [ $# -ne 0 ] ; do
pathcomp="${pathcomp}${1}"
shift
if [ ! -d "${pathcomp}" ] ;
then
$mkdirprog "${pathcomp}"
else
true
fi
pathcomp="${pathcomp}/"
done
fi
if [ x"$dir_arg" != x ]
then
$doit $instcmd $dst &&
if [ x"$chowncmd" != x ]; then $doit $chowncmd $dst; else true ; fi &&
if [ x"$chgrpcmd" != x ]; then $doit $chgrpcmd $dst; else true ; fi &&
if [ x"$stripcmd" != x ]; then $doit $stripcmd $dst; else true ; fi &&
if [ x"$chmodcmd" != x ]; then $doit $chmodcmd $dst; else true ; fi
else
# If we're going to rename the final executable, determine the name now.
if [ x"$transformarg" = x ]
then
dstfile=`basename $dst`
else
dstfile=`basename $dst $transformbasename |
sed $transformarg`$transformbasename
fi
# don't allow the sed command to completely eliminate the filename
if [ x"$dstfile" = x ]
then
dstfile=`basename $dst`
else
true
fi
# Make a temp file name in the proper directory.
dsttmp=$dstdir/#inst.$$#
# Move or copy the file name to the temp name
$doit $instcmd $src $dsttmp &&
trap "rm -f ${dsttmp}" 0 &&
# and set any options; do chmod last to preserve setuid bits
# If any of these fail, we abort the whole thing. If we want to
# ignore errors from any of these, just make sure not to ignore
# errors from the above "$doit $instcmd $src $dsttmp" command.
if [ x"$chowncmd" != x ]; then $doit $chowncmd $dsttmp; else true;fi &&
if [ x"$chgrpcmd" != x ]; then $doit $chgrpcmd $dsttmp; else true;fi &&
if [ x"$stripcmd" != x ]; then $doit $stripcmd $dsttmp; else true;fi &&
if [ x"$chmodcmd" != x ]; then $doit $chmodcmd $dsttmp; else true;fi &&
# Now rename the file to the real destination.
$doit $rmcmd -f $dstdir/$dstfile &&
$doit $mvcmd $dsttmp $dstdir/$dstfile
fi &&
exit 0
deps/jemalloc/src/arena.c
Vendored
+2376
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/src/atomic.c
Vendored
+2
View File
@@ -0,0 +1,2 @@
#define JEMALLOC_ATOMIC_C_
#include "jemalloc/internal/jemalloc_internal.h"
deps/jemalloc/src/base.c
Vendored
+139
View File
@@ -0,0 +1,139 @@
#define JEMALLOC_BASE_C_
#include "jemalloc/internal/jemalloc_internal.h"
/******************************************************************************/
/* Data. */
static malloc_mutex_t base_mtx;
/*
* Current pages that are being used for internal memory allocations. These
* pages are carved up in cacheline-size quanta, so that there is no chance of
* false cache line sharing.
*/
static void *base_pages;
static void *base_next_addr;
static void *base_past_addr; /* Addr immediately past base_pages. */
static extent_node_t *base_nodes;
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static bool base_pages_alloc(size_t minsize);
/******************************************************************************/
static bool
base_pages_alloc(size_t minsize)
{
size_t csize;
bool zero;
assert(minsize != 0);
csize = CHUNK_CEILING(minsize);
zero = false;
base_pages = chunk_alloc(csize, chunksize, true, &zero,
chunk_dss_prec_get());
if (base_pages == NULL)
return (true);
base_next_addr = base_pages;
base_past_addr = (void *)((uintptr_t)base_pages + csize);
return (false);
}
void *
base_alloc(size_t size)
{
void *ret;
size_t csize;
/* Round size up to nearest multiple of the cacheline size. */
csize = CACHELINE_CEILING(size);
malloc_mutex_lock(&base_mtx);
/* Make sure there's enough space for the allocation. */
if ((uintptr_t)base_next_addr + csize > (uintptr_t)base_past_addr) {
if (base_pages_alloc(csize)) {
malloc_mutex_unlock(&base_mtx);
return (NULL);
}
}
/* Allocate. */
ret = base_next_addr;
base_next_addr = (void *)((uintptr_t)base_next_addr + csize);
malloc_mutex_unlock(&base_mtx);
return (ret);
}
void *
base_calloc(size_t number, size_t size)
{
void *ret = base_alloc(number * size);
if (ret != NULL)
memset(ret, 0, number * size);
return (ret);
}
extent_node_t *
base_node_alloc(void)
{
extent_node_t *ret;
malloc_mutex_lock(&base_mtx);
if (base_nodes != NULL) {
ret = base_nodes;
base_nodes = *(extent_node_t **)ret;
malloc_mutex_unlock(&base_mtx);
} else {
malloc_mutex_unlock(&base_mtx);
ret = (extent_node_t *)base_alloc(sizeof(extent_node_t));
}
return (ret);
}
void
base_node_dealloc(extent_node_t *node)
{
malloc_mutex_lock(&base_mtx);
*(extent_node_t **)node = base_nodes;
base_nodes = node;
malloc_mutex_unlock(&base_mtx);
}
bool
base_boot(void)
{
base_nodes = NULL;
if (malloc_mutex_init(&base_mtx))
return (true);
return (false);
}
void
base_prefork(void)
{
malloc_mutex_prefork(&base_mtx);
}
void
base_postfork_parent(void)
{
malloc_mutex_postfork_parent(&base_mtx);
}
void
base_postfork_child(void)
{
malloc_mutex_postfork_child(&base_mtx);
}
deps/jemalloc/src/bitmap.c
Vendored
+90
View File
@@ -0,0 +1,90 @@
#define JEMALLOC_BITMAP_C_
#include "jemalloc/internal/jemalloc_internal.h"
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static size_t bits2groups(size_t nbits);
/******************************************************************************/
static size_t
bits2groups(size_t nbits)
{
return ((nbits >> LG_BITMAP_GROUP_NBITS) +
!!(nbits & BITMAP_GROUP_NBITS_MASK));
}
void
bitmap_info_init(bitmap_info_t *binfo, size_t nbits)
{
unsigned i;
size_t group_count;
assert(nbits > 0);
assert(nbits <= (ZU(1) << LG_BITMAP_MAXBITS));
/*
* Compute the number of groups necessary to store nbits bits, and
* progressively work upward through the levels until reaching a level
* that requires only one group.
*/
binfo->levels[0].group_offset = 0;
group_count = bits2groups(nbits);
for (i = 1; group_count > 1; i++) {
assert(i < BITMAP_MAX_LEVELS);
binfo->levels[i].group_offset = binfo->levels[i-1].group_offset
+ group_count;
group_count = bits2groups(group_count);
}
binfo->levels[i].group_offset = binfo->levels[i-1].group_offset
+ group_count;
binfo->nlevels = i;
binfo->nbits = nbits;
}
size_t
bitmap_info_ngroups(const bitmap_info_t *binfo)
{
return (binfo->levels[binfo->nlevels].group_offset << LG_SIZEOF_BITMAP);
}
size_t
bitmap_size(size_t nbits)
{
bitmap_info_t binfo;
bitmap_info_init(&binfo, nbits);
return (bitmap_info_ngroups(&binfo));
}
void
bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo)
{
size_t extra;
unsigned i;
/*
* Bits are actually inverted with regard to the external bitmap
* interface, so the bitmap starts out with all 1 bits, except for
* trailing unused bits (if any). Note that each group uses bit 0 to
* correspond to the first logical bit in the group, so extra bits
* are the most significant bits of the last group.
*/
memset(bitmap, 0xffU, binfo->levels[binfo->nlevels].group_offset <<
LG_SIZEOF_BITMAP);
extra = (BITMAP_GROUP_NBITS - (binfo->nbits & BITMAP_GROUP_NBITS_MASK))
& BITMAP_GROUP_NBITS_MASK;
if (extra != 0)
bitmap[binfo->levels[1].group_offset - 1] >>= extra;
for (i = 1; i < binfo->nlevels; i++) {
size_t group_count = binfo->levels[i].group_offset -
binfo->levels[i-1].group_offset;
extra = (BITMAP_GROUP_NBITS - (group_count &
BITMAP_GROUP_NBITS_MASK)) & BITMAP_GROUP_NBITS_MASK;
if (extra != 0)
bitmap[binfo->levels[i+1].group_offset - 1] >>= extra;
}
}
deps/jemalloc/src/chunk.c
Vendored
+381
View File
@@ -0,0 +1,381 @@
#define JEMALLOC_CHUNK_C_
#include "jemalloc/internal/jemalloc_internal.h"
/******************************************************************************/
/* Data. */
const char *opt_dss = DSS_DEFAULT;
size_t opt_lg_chunk = LG_CHUNK_DEFAULT;
malloc_mutex_t chunks_mtx;
chunk_stats_t stats_chunks;
/*
* Trees of chunks that were previously allocated (trees differ only in node
* ordering). These are used when allocating chunks, in an attempt to re-use
* address space. Depending on function, different tree orderings are needed,
* which is why there are two trees with the same contents.
*/
static extent_tree_t chunks_szad_mmap;
static extent_tree_t chunks_ad_mmap;
static extent_tree_t chunks_szad_dss;
static extent_tree_t chunks_ad_dss;
rtree_t *chunks_rtree;
/* Various chunk-related settings. */
size_t chunksize;
size_t chunksize_mask; /* (chunksize - 1). */
size_t chunk_npages;
size_t map_bias;
size_t arena_maxclass; /* Max size class for arenas. */
/******************************************************************************/
/* Function prototypes for non-inline static functions. */
static void *chunk_recycle(extent_tree_t *chunks_szad,
extent_tree_t *chunks_ad, size_t size, size_t alignment, bool base,
bool *zero);
static void chunk_record(extent_tree_t *chunks_szad,
extent_tree_t *chunks_ad, void *chunk, size_t size);
/******************************************************************************/
static void *
chunk_recycle(extent_tree_t *chunks_szad, extent_tree_t *chunks_ad, size_t size,
size_t alignment, bool base, bool *zero)
{
void *ret;
extent_node_t *node;
extent_node_t key;
size_t alloc_size, leadsize, trailsize;
bool zeroed;
if (base) {
/*
* This function may need to call base_node_{,de}alloc(), but
* the current chunk allocation request is on behalf of the
* base allocator. Avoid deadlock (and if that weren't an
* issue, potential for infinite recursion) by returning NULL.
*/
return (NULL);
}
alloc_size = size + alignment - chunksize;
/* Beware size_t wrap-around. */
if (alloc_size < size)
return (NULL);
key.addr = NULL;
key.size = alloc_size;
malloc_mutex_lock(&chunks_mtx);
node = extent_tree_szad_nsearch(chunks_szad, &key);
if (node == NULL) {
malloc_mutex_unlock(&chunks_mtx);
return (NULL);
}
leadsize = ALIGNMENT_CEILING((uintptr_t)node->addr, alignment) -
(uintptr_t)node->addr;
assert(node->size >= leadsize + size);
trailsize = node->size - leadsize - size;
ret = (void *)((uintptr_t)node->addr + leadsize);
/* Remove node from the tree. */
extent_tree_szad_remove(chunks_szad, node);
extent_tree_ad_remove(chunks_ad, node);
if (leadsize != 0) {
/* Insert the leading space as a smaller chunk. */
node->size = leadsize;
extent_tree_szad_insert(chunks_szad, node);
extent_tree_ad_insert(chunks_ad, node);
node = NULL;
}
if (trailsize != 0) {
/* Insert the trailing space as a smaller chunk. */
if (node == NULL) {
/*
* An additional node is required, but
* base_node_alloc() can cause a new base chunk to be
* allocated. Drop chunks_mtx in order to avoid
* deadlock, and if node allocation fails, deallocate
* the result before returning an error.
*/
malloc_mutex_unlock(&chunks_mtx);
node = base_node_alloc();
if (node == NULL) {
chunk_dealloc(ret, size, true);
return (NULL);
}
malloc_mutex_lock(&chunks_mtx);
}
node->addr = (void *)((uintptr_t)(ret) + size);
node->size = trailsize;
extent_tree_szad_insert(chunks_szad, node);
extent_tree_ad_insert(chunks_ad, node);
node = NULL;
}
malloc_mutex_unlock(&chunks_mtx);
zeroed = false;
if (node != NULL) {
if (node->zeroed) {
zeroed = true;
*zero = true;
}
base_node_dealloc(node);
}
if (zeroed == false && *zero) {
VALGRIND_MAKE_MEM_UNDEFINED(ret, size);
memset(ret, 0, size);
}
return (ret);
}
/*
* If the caller specifies (*zero == false), it is still possible to receive
* zeroed memory, in which case *zero is toggled to true. arena_chunk_alloc()
* takes advantage of this to avoid demanding zeroed chunks, but taking
* advantage of them if they are returned.
*/
void *
chunk_alloc(size_t size, size_t alignment, bool base, bool *zero,
dss_prec_t dss_prec)
{
void *ret;
assert(size != 0);
assert((size & chunksize_mask) == 0);
assert(alignment != 0);
assert((alignment & chunksize_mask) == 0);
/* "primary" dss. */
if (config_dss && dss_prec == dss_prec_primary) {
if ((ret = chunk_recycle(&chunks_szad_dss, &chunks_ad_dss, size,
alignment, base, zero)) != NULL)
goto label_return;
if ((ret = chunk_alloc_dss(size, alignment, zero)) != NULL)
goto label_return;
}
/* mmap. */
if ((ret = chunk_recycle(&chunks_szad_mmap, &chunks_ad_mmap, size,
alignment, base, zero)) != NULL)
goto label_return;
if ((ret = chunk_alloc_mmap(size, alignment, zero)) != NULL)
goto label_return;
/* "secondary" dss. */
if (config_dss && dss_prec == dss_prec_secondary) {
if ((ret = chunk_recycle(&chunks_szad_dss, &chunks_ad_dss, size,
alignment, base, zero)) != NULL)
goto label_return;
if ((ret = chunk_alloc_dss(size, alignment, zero)) != NULL)
goto label_return;
}
/* All strategies for allocation failed. */
ret = NULL;
label_return:
if (config_ivsalloc && base == false && ret != NULL) {
if (rtree_set(chunks_rtree, (uintptr_t)ret, ret)) {
chunk_dealloc(ret, size, true);
return (NULL);
}
}
if ((config_stats || config_prof) && ret != NULL) {
bool gdump;
malloc_mutex_lock(&chunks_mtx);
if (config_stats)
stats_chunks.nchunks += (size / chunksize);
stats_chunks.curchunks += (size / chunksize);
if (stats_chunks.curchunks > stats_chunks.highchunks) {
stats_chunks.highchunks = stats_chunks.curchunks;
if (config_prof)
gdump = true;
} else if (config_prof)
gdump = false;
malloc_mutex_unlock(&chunks_mtx);
if (config_prof && opt_prof && opt_prof_gdump && gdump)
prof_gdump();
}
if (config_debug && *zero && ret != NULL) {
size_t i;
size_t *p = (size_t *)(uintptr_t)ret;
VALGRIND_MAKE_MEM_DEFINED(ret, size);
for (i = 0; i < size / sizeof(size_t); i++)
assert(p[i] == 0);
}
assert(CHUNK_ADDR2BASE(ret) == ret);
return (ret);
}
static void
chunk_record(extent_tree_t *chunks_szad, extent_tree_t *chunks_ad, void *chunk,
size_t size)
{
bool unzeroed;
extent_node_t *xnode, *node, *prev, key;
unzeroed = pages_purge(chunk, size);
/*
* Allocate a node before acquiring chunks_mtx even though it might not
* be needed, because base_node_alloc() may cause a new base chunk to
* be allocated, which could cause deadlock if chunks_mtx were already
* held.
*/
xnode = base_node_alloc();
malloc_mutex_lock(&chunks_mtx);
key.addr = (void *)((uintptr_t)chunk + size);
node = extent_tree_ad_nsearch(chunks_ad, &key);
/* Try to coalesce forward. */
if (node != NULL && node->addr == key.addr) {
/*
* Coalesce chunk with the following address range. This does
* not change the position within chunks_ad, so only
* remove/insert from/into chunks_szad.
*/
extent_tree_szad_remove(chunks_szad, node);
node->addr = chunk;
node->size += size;
node->zeroed = (node->zeroed && (unzeroed == false));
extent_tree_szad_insert(chunks_szad, node);
if (xnode != NULL)
base_node_dealloc(xnode);
} else {
/* Coalescing forward failed, so insert a new node. */
if (xnode == NULL) {
/*
* base_node_alloc() failed, which is an exceedingly
* unlikely failure. Leak chunk; its pages have
* already been purged, so this is only a virtual
* memory leak.
*/
malloc_mutex_unlock(&chunks_mtx);
return;
}
node = xnode;
node->addr = chunk;
node->size = size;
node->zeroed = (unzeroed == false);
extent_tree_ad_insert(chunks_ad, node);
extent_tree_szad_insert(chunks_szad, node);
}
/* Try to coalesce backward. */
prev = extent_tree_ad_prev(chunks_ad, node);
if (prev != NULL && (void *)((uintptr_t)prev->addr + prev->size) ==
chunk) {
/*
* Coalesce chunk with the previous address range. This does
* not change the position within chunks_ad, so only
* remove/insert node from/into chunks_szad.
*/
extent_tree_szad_remove(chunks_szad, prev);
extent_tree_ad_remove(chunks_ad, prev);
extent_tree_szad_remove(chunks_szad, node);
node->addr = prev->addr;
node->size += prev->size;
node->zeroed = (node->zeroed && prev->zeroed);
extent_tree_szad_insert(chunks_szad, node);
base_node_dealloc(prev);
}
malloc_mutex_unlock(&chunks_mtx);
}
void
chunk_unmap(void *chunk, size_t size)
{
assert(chunk != NULL);
assert(CHUNK_ADDR2BASE(chunk) == chunk);
assert(size != 0);
assert((size & chunksize_mask) == 0);
if (config_dss && chunk_in_dss(chunk))
chunk_record(&chunks_szad_dss, &chunks_ad_dss, chunk, size);
else if (chunk_dealloc_mmap(chunk, size))
chunk_record(&chunks_szad_mmap, &chunks_ad_mmap, chunk, size);
}
void
chunk_dealloc(void *chunk, size_t size, bool unmap)
{
assert(chunk != NULL);
assert(CHUNK_ADDR2BASE(chunk) == chunk);
assert(size != 0);
assert((size & chunksize_mask) == 0);
if (config_ivsalloc)
rtree_set(chunks_rtree, (uintptr_t)chunk, NULL);
if (config_stats || config_prof) {
malloc_mutex_lock(&chunks_mtx);
assert(stats_chunks.curchunks >= (size / chunksize));
stats_chunks.curchunks -= (size / chunksize);
malloc_mutex_unlock(&chunks_mtx);
}
if (unmap)
chunk_unmap(chunk, size);
}
bool
chunk_boot(void)
{
/* Set variables according to the value of opt_lg_chunk. */
chunksize = (ZU(1) << opt_lg_chunk);
assert(chunksize >= PAGE);
chunksize_mask = chunksize - 1;
chunk_npages = (chunksize >> LG_PAGE);
if (config_stats || config_prof) {
if (malloc_mutex_init(&chunks_mtx))
return (true);
memset(&stats_chunks, 0, sizeof(chunk_stats_t));
}
if (config_dss && chunk_dss_boot())
return (true);
extent_tree_szad_new(&chunks_szad_mmap);
extent_tree_ad_new(&chunks_ad_mmap);
extent_tree_szad_new(&chunks_szad_dss);
extent_tree_ad_new(&chunks_ad_dss);
if (config_ivsalloc) {
chunks_rtree = rtree_new((ZU(1) << (LG_SIZEOF_PTR+3)) -
opt_lg_chunk);
if (chunks_rtree == NULL)
return (true);
}
return (false);
}
void
chunk_prefork(void)
{
malloc_mutex_lock(&chunks_mtx);
if (config_ivsalloc)
rtree_prefork(chunks_rtree);
chunk_dss_prefork();
}
void
chunk_postfork_parent(void)
{
chunk_dss_postfork_parent();
if (config_ivsalloc)
rtree_postfork_parent(chunks_rtree);
malloc_mutex_postfork_parent(&chunks_mtx);
}
void
chunk_postfork_child(void)
{
chunk_dss_postfork_child();
if (config_ivsalloc)
rtree_postfork_child(chunks_rtree);
malloc_mutex_postfork_child(&chunks_mtx);
}

Some files were not shown because too many files have changed in this diff Show More