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:2.8.5
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:3.0.0-beta7
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
1909 Commits
2.8.5 ... 3.0.0-beta7
Author SHA1 Message Date
antirez f884be6854 Redis 2.9.56 (Redis 3.0.0 beta-7). 2014-06-30 12:11:22 +02:00
antirez 05b6b147ec Test: find_available_port: check that cluster port is free as well. 
The function will only return ports that have also port+10000 free, so
that Redis Cluster instances can be executed at the returned port.
 2014-06-30 12:11:22 +02:00
antirez a5aa3b6491 Test: fix instances.tcl restart_instance abort error. 2014-06-30 12:11:22 +02:00
antirez b18c92b9e7 DEBUG CMDKEYS moved to COMMAND GETKEYS. 2014-06-27 18:39:33 +02:00
antirez d3f43db20f COMMAND COUNT subcommand added. 2014-06-27 18:39:33 +02:00
antirez b205119810 COMMAND: fix argument parsing. 
This fixes detection of wrong subcommand (that resulted in the default
all-commands output instead) and allows COMMAND INFO to be called
without arguments (resulting into an empty array) which is useful in
programmtically generated calls like the following (in Ruby):

    redis.commands("command","info",*mycommands)

Note: mycommands may be empty.
 2014-06-27 18:39:33 +02:00
antirez 3364b2d12b COMMANDS command renamed COMMAND. 2014-06-27 18:39:33 +02:00
antirez 0aab8eb1e3 COMMANDS command: remove static + aesthetic changes. 
Static was removed since it is needed in order to get symbols in stack
traces. Minor changes in the source code were operated to make it more
similar to the existing Redis code base.
 2014-06-27 18:39:33 +02:00
Matt Stancliff 60d62a99d2 Cluster: Add COMMANDS command 
COMMANDS returns a nested multibulk reply for each
command in the command table.  The reply for each
command contains:
  - command name
  - arity
  - array of command flags
  - start key position
  - end key position
  - key offset step
  - optional: if the keys are not deterministic and
    Redis uses an internal key evaluation function,
    the 6th field appears and is defined as a status
    reply of: REQUIRES ARGUMENT PARSING

Cluster clients need to know where the keys are in each
command to implement proper routing to cluster nodes.

Redis commands can have multiple keys, keys at offset steps, or other
issues where you can't always assume the first element after
the command name is the cluster routing key.

Using the information exposed by COMMANDS, client implementations
can have live, accurate key extraction details for all commands.

Also implements COMMANDS INFO [commands...] to return only a
specific set of commands instead of all 160+ commands live in Redis.
 2014-06-27 18:39:33 +02:00
antirez e530c2ccb2 Fixed assert conditional in ROLE command test. 2014-06-26 22:13:54 +02:00
antirez b411678679 No more trailing spaces in Redis source code. 2014-06-26 18:48:47 +02:00
antirez b4d378f137 CLIENT KILL: don't kill the master as a normal client. 
Technically the problem is due to the client type API that does not
return a special value for the master, however fixing it locally in the
CLIENT KILL command is better currently because otherwise we would
introduce a new output buffer limit class as a side effect.
 2014-06-26 18:43:28 +02:00
antirez 76e7673259 Remove infinite loop from PSYNC test. 
Added for debugging and forgot there.
 2014-06-26 18:43:28 +02:00
antirez 2e01ac52c7 Test: hopefully more robust PSYNC test. 
This is supposed to fix issue #1417, but we'll know if this is enough
only after a couple of runs of the CI test without false positives.
 2014-06-26 18:43:28 +02:00
Matt Stancliff 77fb9fd4fd Allow __powerpc__ to define HAVE_ATOMIC too 
From mailing list post https://groups.google.com/forum/#!topic/redis-db/D3k7KmJmYgM

In the file “config.h”, the definition HAVE_ATOMIC is used to indicate
if an architecture on which redis is implemented supports atomic
synchronization primitives.  Powerpc  supports atomic synchronization
primitives, however, it is not listed as one of the architectures
supported in config.h. This patch  adds the __powerpc__ to the list of
architectures supporting these primitives. The improvement of redis
due to the atomic synchronization on powerpc is significant,
around 30% to 40%, over the default implementation using pthreads.

This proposal adds __powerpc__ to the list of architectures designated
to support atomic builtins.
 2014-06-26 15:16:23 +02:00
Matt Stancliff 27e30e2f8b Allow atomic memory count update with C11 builtins 
From mailing list post https://groups.google.com/forum/#!topic/redis-db/QLjiQe4D7LA

In zmalloc.c the following primitives are currently used
to synchronize access to single global variable:
__sync_add_and_fetch
__sync_sub_and_fetch

In some architectures such as powerpc these primitives are overhead
intensive. More efficient C11 __atomic builtins are available with
newer GCC versions, see
http://gcc.gnu.org/onlinedocs/gcc-4.8.2/gcc/_005f_005fatomic-Builtins.html#_005f_005fatomic-Builtins

By substituting the following  __atomic… builtins:
__atomic_add_fetch
__atomic_sub_fetch

the performance improvement on certain architectures such as powerpc can be significant,
around 10% to 15%, over the implementation using __sync builtins while there is only slight uptick on
Intel architectures because it was already enforcing Intel Strongly ordered memory semantics.

The selection of __atomic built-ins can be predicated on the definition of ATOMIC_RELAXED
which Is available on in gcc 4.8.2 and later versions.
 2014-06-26 15:16:23 +02:00
Matt Stancliff 4d14f2744a Use predefined macro for used_memory() update 2014-06-26 15:16:23 +02:00
antirez a7bef1bb7c CLUSTER SLOTS: don't output failing slaves. 
While we have to output failing masters in order to provide an accurate
map (that may be the one of a Redis Cluster in down state because not
all slots are served by a working master), to provide slaves in FAIL
state is not a good idea since those are not necesarely needed, and the
client will likely incur into a latency penalty trying to connect with a
slave which is down.

Note that this means that CLUSTER SLOTS does not provide a *complete*
map of slaves, however this would not be of any help since slaves may be
added later, and a client that needs to scale reads and requires to
stay updated with the list of slaves, need to do a refresh of the map
from time to time, anyway.
 2014-06-26 15:16:23 +02:00
antirez 94bbcf400d CLUSTER SLOTS: main loop should skip only slaves and zero slot masters. 2014-06-26 15:16:23 +02:00
Matt Stancliff 237188bf1b Cluster: Add CLUSTER SLOTS command 
CLUSTER SLOTS returns a Redis-formatted mapping from
slot ranges to IP/Port pairs serving that slot range.

The outer return elements group return values by slot ranges.

The first two entires in each result are the min and max slots for the range.

The third entry in each result is guaranteed to be either
an IP/Port of the master for that slot range - OR - null
if that slot range, for some reason, has no master

The 4th and higher entries in each result are replica instances
for the slot range.

Output comparison:
127.0.0.1:7001> cluster nodes
f853501ec8ae1618df0e0f0e86fd7abcfca36207 127.0.0.1:7001 myself,master - 0 0 2 connected 4096-8191
5a2caa782042187277647661ffc5da739b3e0805 127.0.0.1:7005 slave f853501ec8ae1618df0e0f0e86fd7abcfca36207 0 1402622415859 6 connected
6c70b49813e2ffc9dd4b8ec1e108276566fcf59f 127.0.0.1:7007 slave 26f4729ca0a5a992822667fc16b5220b13368f32 0 1402622415357 8 connected
2bd5a0e3bb7afb2b56a2120d3fef2f2e4333de1d 127.0.0.1:7006 slave 32adf4b8474fdc938189dba00dc8ed60ce635b0f 0 1402622419373 7 connected
5a9450e8279df36ff8e6bb1c139ce4d5268d1390 127.0.0.1:7000 master - 0 1402622418872 1 connected 0-4095
32adf4b8474fdc938189dba00dc8ed60ce635b0f 127.0.0.1:7002 master - 0 1402622419874 3 connected 8192-12287
5db7d05c245267afdfe48c83e7de899348d2bdb6 127.0.0.1:7004 slave 5a9450e8279df36ff8e6bb1c139ce4d5268d1390 0 1402622417867 5 connected
26f4729ca0a5a992822667fc16b5220b13368f32 127.0.0.1:7003 master - 0 1402622420877 4 connected 12288-16383

127.0.0.1:7001> cluster slots
1) 1) (integer) 0
   2) (integer) 4095
   3) 1) "127.0.0.1"
      2) (integer) 7000
   4) 1) "127.0.0.1"
      2) (integer) 7004
2) 1) (integer) 12288
   2) (integer) 16383
   3) 1) "127.0.0.1"
      2) (integer) 7003
   4) 1) "127.0.0.1"
      2) (integer) 7007
3) 1) (integer) 4096
   2) (integer) 8191
   3) 1) "127.0.0.1"
      2) (integer) 7001
   4) 1) "127.0.0.1"
      2) (integer) 7005
4) 1) (integer) 8192
   2) (integer) 12287
   3) 1) "127.0.0.1"
      2) (integer) 7002
   4) 1) "127.0.0.1"
      2) (integer) 7006
 2014-06-26 15:16:23 +02:00
antirez 8fdc857a9f Cluster: myself->ip autodiscovery. 
Instead of having an hardcoded IP address in the node configuration, we
autodiscover it via MEET messages for automatic update when the node is
restarted with a different IP address.

This mechanism was discussed in the context of PR #1782.
 2014-06-26 15:16:23 +02:00
antirez b56849cd82 Old form of CLIENT KILL should still allow suicide. 2014-06-24 12:49:26 +02:00
antirez 6adf0e092c Sentinel test: more correct sentinels config reset. 
In the initialization test for each instance we used to unregister the
old master and register it again to clear the config.
However there is a race condition doing this: as soon as we unregister
and re-register "mymaster", another Sentinel can update the new
configuration with the old state because of gossip "hello" messages.

So the correct procedure is instead, unregister "mymaster" from all the
sentinel instances, and re-register it everywhere again.
 2014-06-23 14:07:45 +02:00
antirez 2f778b1f6b Sentinel implementation of ROLE. 2014-06-23 12:07:52 +02:00
antirez 8322ede059 Silence different signs comparison warning in sds.c. 2014-06-23 11:54:46 +02:00
antirez 26291dc7f3 Use Redis updated sds.c for deps/hiredis. 2014-06-23 11:54:46 +02:00
Matt Stancliff 02b94801ad Sentinel: bind source address 
Some deployments need traffic sent from a specific address.  This
change uses the same policy as Cluster where the first listed bindaddr
becomes the source address for outgoing Sentinel communication.

Fixes #1667
 2014-06-23 11:54:45 +02:00
Matt Stancliff a2632f2692 Add REDIS_BIND_ADDR access macro 
We need to access (bindaddr[0] || NULL) in a few places, so centralize
access with a nice macro.
 2014-06-23 11:54:45 +02:00
antirez ba472111d3 Dummy zmalloc.h restored into deps/hiredis. 2014-06-23 11:54:45 +02:00
Matt Stancliff b95332a6a6 hiredis: Update to latest version 
This is hiredis f225c276be7fd0646019b51023e3f41566633dfe

This update includes all changes that diverged inside of Redis since
the last update.  This version also allows optional source address
binding for connections which we need for some Sentinel deployments.
 2014-06-23 11:54:45 +02:00
Matt Stancliff ddad61bfcc Cancel SHUTDOWN if initial AOF is being written 
Fixes #1826 (and many other reports of the same problem)
 2014-06-23 10:09:22 +02:00
antirez 5eb5a06078 Basic tests for the ROLE command. 2014-06-23 09:09:07 +02:00
antirez 82caeee12e Allow to call ROLE in LOADING state. 2014-06-21 15:18:33 +02:00
antirez 96f556b27e ROLE command: array len fixed for slave output. 2014-06-21 15:18:33 +02:00
antirez 5c988b334e Jemalloc updated to 3.6.0. 
Not a single bug in about 3 months, and our previous version was
too old (3.2.0).
 2014-06-21 15:18:33 +02:00
antirez f94d46ba9a Jemalloc update script added. 2014-06-21 15:18:33 +02:00
antirez 9eb02c6a1e Cluster: clear NOADDR flag when updating node address. 2014-06-21 15:18:33 +02:00
antirez 52456bf8c3 Sentinel test: tolerate larger delays in init tests. 2014-06-21 15:18:33 +02:00
antirez 716a77d347 Sentinel test: unit 02, avoid some time related false positives. 2014-06-21 15:18:33 +02:00
antirez 54756adf44 Sentinel: send hello messages ASAP after config change. 
Eventual configuration convergence is guaranteed by our periodic hello
messages to all the instances, however when there are important notices
to share, better make a phone call. With this commit we force an hello
message to other Sentinal and Redis instances within the next 100
milliseconds of a config update, which is practically better than
waiting a few seconds.
 2014-06-21 15:18:33 +02:00
antirez 32adcda857 Sentinel test: add manual failover test. 2014-06-21 15:18:33 +02:00
antirez 6588ef2944 Sentinel: handle SRI_PROMOTED flag correctly. 
Lack of check of the SRI_PROMOTED flag caused Sentienl to act with the
promoted slave turned into a master during failover like if it was a
normal instance.

Normally this problem was not apparent because during real failovers the
old master is down so the bugged code path was not entered, however with
manual failovers via the SENTINEL FAILOVER command, the problem was
easily triggered.

This commit prevents promoted slaves from getting reconfigured, moreover
we now explicitly check that during a failover the slave turning into a
master is the one we selected for promotion and not a different one.
 2014-06-21 15:18:33 +02:00
antirez 64e618d10c Test: use higher level redis.tcl proc to read replies. 2014-06-21 15:18:33 +02:00
antirez 28c86fe463 Sentienl/Redis test: enable link reconnection in instances.tcl. 2014-06-21 15:18:32 +02:00
antirez 3cba0c3ebb Test: Tcl client initial support for automatic reconnection. 2014-06-21 15:18:32 +02:00
antirez 256046fdb4 Test: tcl client, unset deferred var on close. 2014-06-21 15:18:32 +02:00
Matt Stancliff 4dfc18596e Add correct exit value to failed tests 2014-06-21 15:18:32 +02:00
Alex Suraci 2b1047ffd3 add missing signal.h include 2014-06-21 15:18:32 +02:00
Matt Stancliff 65bab61159 Add SIGINT handler to cli --intrinsic-latency 
If we run a long latency session and want to Ctrl-C out of it,
it's nice to still get the summary output at the end.
 2014-06-21 15:18:32 +02:00
antirez 8d75a20ab1 Sentinel: send SLAVEOF with MULTI, CLIENT KILL, CONFIG REWRITE. 
This implements the new Sentinel-Client protocol for the Sentinel part:
now instances are reconfigured using a transaction that ensures that the
config is rewritten in the target instance, and that clients lose the
connection with the instance, in order to be forced to: ask Sentinel,
reconnect to the instance, and verify the instance role with the new
ROLE command.
 2014-06-21 15:18:32 +02:00
antirez 52f1edb2ab CLIENT KILL API modified. 
Added a new SKIPME option that is true by default, that prevents the
client sending the command to be killed, unless SKIPME NO is sent.
 2014-06-21 15:18:32 +02:00
antirez 7e9863637c CLIENT KILL: fix closing link of the current client. 2014-06-21 15:18:32 +02:00
antirez ca262a58c6 New features for CLIENT KILL. 2014-06-21 15:18:32 +02:00
antirez b7dd451892 Assign an unique non-repeating ID to each new client. 
This will be used by CLIENT KILL and is also a good way to ensure a
given client is still the same across CLIENT LIST calls.

The output of CLIENT LIST was modified to include the new ID, but this
change is considered to be backward compatible as the API does not imply
you can do positional parsing, since each filed as a different name.
 2014-06-21 15:18:32 +02:00
antirez d97d1a6430 Client types generalized. 
Because of output buffer limits Redis internals had this idea of type of
clients: normal, pubsub, slave. It is possible to set different output
buffer limits for the three kinds of clients.

However all the macros and API were named after output buffer limit
classes, while the idea of a client type is a generic one that can be
reused.

This commit does two things:

1) Rename the API and defines with more general names.
2) Change the class of clients executing the MONITOR command from "slave"
   to "normal".

"2" is a good idea because you want to have very special settings for
slaves, that are not a good idea for MONITOR clients that are instead
normal clients even if they are conceptually slave-alike (since it is a
push protocol).

The backward-compatibility breakage resulting from "2" is considered to
be minimal to care, since MONITOR is a debugging command, and because
anyway this change is not going to break the format or the behavior, but
just when a connection is closed on big output buffer issues.
 2014-06-21 15:18:32 +02:00
antirez 59cfa49c21 Scripting: regression test for issue #1811. 2014-06-21 15:18:32 +02:00
antirez 2b57d96ea0 Fix semantics of Lua calls to SELECT. 
Lua scripts are executed in the context of the currently selected
database (as selected by the caller of the script).

However Lua scripts are also free to use the SELECT command in order to
affect other DBs. When SELECT is called frm Lua, the old behavior, before
this commit, was to automatically set the Lua caller selected DB to the
last DB selected by Lua. See for example the following sequence of
commands:

    SELECT 0
    SET x 10
    EVAL "redis.call('select','1')" 0
    SET x 20

Before this commit after the execution of this sequence of commands,
we'll have x=10 in DB 0, and x=20 in DB 1.

Because of the problem above, there was a bug affecting replication of
Lua scripts, because of the actual implementation of replication. It was
possible to fix the implementation of Lua scripts in order to fix the
issue, but looking closely, the bug is the consequence of the behavior
of Lua ability to set the caller's DB.

Under the old semantics, a script selecting a different DB, has no simple
ways to restore the state and select back the previously selected DB.
Moreover the script auhtor must remember that the restore is needed,
otherwise the new commands executed by the caller, will be executed in
the context of a different DB.

So this commit fixes both the replication issue, and this hard-to-use
semantics, by removing the ability of Lua, after the script execution,
to force the caller to switch to the DB selected by the Lua script.

The new behavior of the previous sequence of commadns is to just set
X=20 in DB 0. However Lua scripts are still capable of writing / reading
from different DBs if needed.

WARNING: This is a semantical change that will break programs that are
conceived to select the client selected DB via Lua scripts.

This fixes issue #1811.
 2014-06-21 15:18:32 +02:00
antirez 62bddfe3af Cluser test: write multiple keys in unit 03. 
The write-failover-read test is better if we write 100 keys so that we
are sure all the nodes are affected by the write.
 2014-06-21 15:18:32 +02:00
antirez 1381a6bf3d Cluster test: rewrite config when resetting nodes. 2014-06-21 15:18:32 +02:00
antirez 0f4b925de5 Scripting: Fix for a #1118 regression simplified. 
It is more straightforward to just test for a numerical type avoiding
Lua's automatic conversion. The code is technically more correct now,
however Lua should automatically convert to number only if the original
type is a string that "looks like a number", and not from other types,
so practically speaking the fix is identical AFAIK.
 2014-06-21 15:18:32 +02:00
Matt Stancliff 9ca83c82bf Scripting: Fix regression from #1118 
The new check-for-number behavior of Lua arguments broke
users who use large strings of just integers.

The Lua number check would convert the string to a number, but
that breaks user data because
Lua numbers have limited precision compared to an arbitrarily
precise number wrapped in a string.

Regression fixed and new test added.

Fixes #1118 again.
 2014-06-21 15:18:32 +02:00
antirez 6b566ab78a Cluster: fix an error message when logging failover auth denied. 2014-06-21 15:18:32 +02:00
antirez 326145b9ef Cluster: better comment for clusterSendFailoverAuthIfNeeded() epoch test. 2014-06-21 15:18:32 +02:00
antirez 053bcf034e Cluster test: set config epochs when creating clusters. 2014-06-21 15:18:32 +02:00
antirez 78a687ec17 Cluster: log granted failover authorizations. 2014-06-21 15:18:32 +02:00
antirez 61910a6708 Cluster: log configEpoch updates to myself. 2014-06-21 15:18:32 +02:00
antirez 111a5556aa Cluster: log when a master denies a failover auth. 2014-06-21 15:18:32 +02:00
antirez 6cd068d421 Cluster test: unit 03, check epochs correctness. 2014-06-21 15:18:32 +02:00
antirez 90b4f91e1a Cluster test: better failover detection in unit 03. 2014-06-21 15:18:32 +02:00
antirez 9912a60d01 Cluster test: added show-redis-logs command in the interactive console. 2014-06-21 15:18:32 +02:00
antirez 7079ef0341 Cluster test: new unit 03, failover loop stress testing. 2014-06-21 15:18:32 +02:00
antirez 1d896303af Cluster test: 02 unit more reliable waiting for slave sync. 2014-06-21 15:18:32 +02:00
antirez d7a4025828 Cluster test: unit 02 should wait for failover. 2014-06-21 15:18:31 +02:00
antirez 33f09f0ff3 Cluster test: check master -> slave role switch. 2014-06-21 15:18:31 +02:00
antirez 5cdefb7c99 Cluster: cluster_my_epoch added to CLUSTER INFO output. 2014-06-21 15:18:31 +02:00
antirez 2213f283c5 ROLE output improved for slaves. 
Info about the replication state with the master added.
 2014-06-21 15:18:31 +02:00
antirez 9b460e803a ROLE command added. 
The new ROLE command is designed in order to provide a client with
informations about the replication in a fast and easy to use way
compared to the INFO command where the same information is also
available.
 2014-06-21 15:18:31 +02:00
antirez ea5335fb9b Redis 2.9.55 (Redis 3.0.0 beta-6). 2014-06-09 16:01:12 +02:00
Matt Stancliff 726d343ad7 Fix lack of strtold under Cygwin 
Renaming strtold to strtod then casting
the result is the standard way of dealing with
no strtold in Cygwin.
 2014-06-09 11:43:17 +02:00
Matt Stancliff 28fef5c57b Fix lack of SA_ONSTACK under Cygwin 
Fixes #232
 2014-06-09 11:43:17 +02:00
Matt Stancliff 7fc1fc8ca7 Fix blocking operations from missing new lists 
Behrad Zari discovered [1] and Josiah reported [2]: if you block
and wait for a list to exist, but the list creates from
a non-push command, the blocked client never gets notified.

This commit adds notification of blocked clients into
the DB layer and away from individual commands.

Lists can be created by [LR]PUSH, SORT..STORE, RENAME, MOVE,
and RESTORE.  Previously, blocked client notifications were
only triggered by [LR]PUSH.  Your client would never get
notified if a list were created by SORT..STORE or RENAME or
a RESTORE, etc.

Blocked client notification now happens in one unified place:
  - dbAdd() triggers notification when adding a list to the DB

Two new tests are added that fail prior to this commit.

All test pass.

Fixes #1668

[1]: https://groups.google.com/forum/#!topic/redis-db/k4oWfMkN1NU
[2]: #1668
 2014-06-09 11:38:55 +02:00
antirez 8b059f06d7 Cluster: check that configEpoch never goes back. 
Since there are ways to alter the configEpoch outside of the failover
procedure (for exampel CLUSTER SET-CONFIG-EPOCH and via the configEpoch
collision resolution algorithm), make always sure, before replacing our
configEpoch with a new one, that it is greater than the current one.
 2014-06-07 14:45:53 +02:00
antirez 67029323a3 Cluster: SET-CONFIG-EPOCH should update currentEpoch. 
SET-CONFIG-EPOCH, used by redis-trib at cluster creation time, failed to
update the currentEpoch, making it possible after a failover for a
server to set its configEpoch to a value smaller than the current one
(since configEpochs are obtained using currentEpoch).

The bug totally break the Redis Cluster algorithms and protocols
allowing for permanent split brain conditions about the slots
configuration as shown in issue #1799.
 2014-06-07 14:28:21 +02:00
Andy Grunwald 531e577e2d Fixed typo in word avarege in result message of --intrinsic-latency analyzer 2014-06-06 11:20:00 +02:00
Jan-Erik Rediger eca41f48d3 Small typo fixed 2014-06-06 10:45:44 +02:00
yoav a20d5c50d0 Fix eval usage in tests to conform with eval semantics 2014-06-06 10:44:04 +02:00
zionwu a970f5ca70 fix issue 1787 2014-06-06 10:38:01 +02:00
antirez 9f5ab8699d Don't process min-slaves-to-write for slaves. 
Replication is totally broken when a slave has this option, since it
stops accepting updates from masters.

This fixes issue #1434.
 2014-06-05 10:50:33 +02:00
antirez cfe2d563bc Tests for min-slaves-* feature. 2014-06-05 10:50:32 +02:00
antirez 751c8698d4 Fixed dbuf variable scope in luaRedisGenericCommand(). 
I'm not sure if while the visibility is the inner block, the fact we
point to 'dbuf' is a problem or not, probably the stack var isx
guaranteed to live until the function returns. However obvious code is
better anyway.
 2014-06-04 18:58:56 +02:00
antirez b71720772e Regression test for issue #1118. 2014-06-04 18:51:33 +02:00
antirez c3967f42d7 Scripting: better Lua number -> string conversion in luaRedisGenericCommand(). 
The lua_to*string() family of functions use a non optimal format
specifier when converting integers to strings. This has both the problem
of the number being converted in exponential notation, which we don't
use as a Redis return value when floating point numbers are involed,
and, moreover, there is a loss of precision since the default format
specifier is not able to represent numbers that must be represented
exactly in the IEEE 754 number mantissa.

The new code handles it as a special case using a saner conversion.

This fixes issue #1118.
 2014-06-04 18:45:45 +02:00
antirez 3614176349 More trailing spaces in sentinel.c removed. 2014-05-28 15:46:11 +02:00
antirez 53bc8a77d1 Cluster test: add tmp dir to Git repo. 2014-05-26 18:06:48 +02:00
Matt Stancliff 033d717295 Fix test framework to detect proper server PID 
Previously the PID format was:
[PID] Timestamp

But it recently changed to:
PID:X Timestamp

The tcl testing framework was grabbing the PID from \[\d+\], but
that's not valid anymore.

Now we grab the pid from "PID: <PID>" in the part of Redis startup
output to the right of the ASCII logo.
 2014-05-26 17:58:49 +02:00
Matt Stancliff 186adefd56 Disable recursive watchdog signal handler 
If we are in the signal handler, we don't want to handle
the signal again.  In extreme cases, this can cause a stack overflow
and segfault Redis.

Fixes #1771
 2014-05-26 17:53:37 +02:00
antirez 4fe4c4b34d Redis 2.9.54 (Redis 3.0.0 beta-5). 2014-05-26 16:38:11 +02:00
antirez eea0d41fb9 Cluster: always allow ok -> fail switch in clusterUpdateState(). 
There is a time defined by REDIS_CLUSTER_WRITABLE_DELAY where fail -> ok
switch is not possible after startup as a master for some time, however
the contrary (ok -> fail) should always be possible.
 2014-05-26 16:36:14 +02:00
antirez 672e2672aa Cluster test: catch FLUSHALL errors on node reset. 
FLUSHALL will fail on read-only slaves, but there the command is not
needed in order to reset the instance with CLUSTER RESET so errors can
be ignored.
 2014-05-26 16:36:11 +02:00
antirez df0c8c28ad Sentinel example config: explain you don't need to specify slaves. 2014-05-26 10:18:04 +02:00
antirez b5f0d0d6f9 Redis 2.9.53 (Redis 3.0.0 beta-4). 2014-05-23 16:08:38 +02:00
antirez 7f8ea7a50f Cluster test: basic failover unit added. 2014-05-23 15:54:53 +02:00
antirez 29aed9c48f Cluster test: move basic read/write test into a procedure. 2014-05-23 15:54:53 +02:00
antirez 81af783fec Cluster test: more reliable 01-faildet unit. 
Do things in a sequence that prevents failover during failure detection.
 2014-05-23 15:54:53 +02:00
antirez e62e3c4cb2 3.0 Changelog typo fixed. lost -> list. 2014-05-23 09:49:45 +02:00
antirez 5697f4f85d Document log format change in 3.0 Changelog. 2014-05-23 09:28:32 +02:00
antirez 90bbf1159b redisLogFromHandler() format changed to match new logs format. 2014-05-23 09:26:04 +02:00
antirez 9996c37e74 Tag every log line with role. 
Every log contains, just after the pid, a single character that provides
information about the role of an instance:

S - Slave
M - Master
C - Writing child
X - Sentinel
 2014-05-23 09:26:04 +02:00
antirez 33f63ff988 Cluster: slave validity factor is now user configurable. 
Check the commit changes in the example redis.conf for more information.
 2014-05-23 09:26:04 +02:00
antirez 308c570e04 Test: AOF test false positive when running in slow hosts. 
The bug was triggered by running the test with Valgrind (which is a lot
slower and more sensible to timing issues) after the recent changes
that made Redis more promptly able to reply with the -LOADING error.
 2014-05-23 09:26:04 +02:00
antirez 1d70312028 Test: dump.tcl fixed for RESTORE new error msg. 2014-05-23 09:26:04 +02:00
antirez 0e1b5ff63b Fix an error in redis-trib where we always talk with same node. 
While iterating the list of nodes we want to set the slot as stable in
the current node, not always in the first node of the list.
 2014-05-23 09:26:04 +02:00
antirez c268d1ebd5 redis-trib fix improved: move keys from N nodes to owner. 2014-05-23 09:26:04 +02:00
antirez 32ca949c76 redis-trib fix: use MIGRATE REPLACE when fixing slots. 
This fixes issue #1765.
 2014-05-23 09:26:04 +02:00
antirez 122157f220 Regression test for issue #1764. 2014-05-20 17:45:50 +02:00
antirez d357e3ee8f Remove trailing spaces from scripting.c 2014-05-20 17:45:50 +02:00
antirez d0c84acc97 Remove trailing spaces from sentinel.c. 2014-05-20 17:45:50 +02:00
michael-grunder 6f49395148 Fix LUA_OBJCACHE segfault. 
When scanning the argument list inside of a redis.call() invocation
for pre-cached values, there was no check being done that the
argument we were on was in fact within the bounds of the cache size.

So if a redis.call() command was ever executed with more than 32
arguments (current cache size #define setting) redis-server could
segfault.
 2014-05-20 17:45:50 +02:00
antirez 440ee603b7 Cluster test: better failure detection test and framework improvements. 2014-05-20 17:45:50 +02:00
antirez a7fd2eb594 Cluster test: failure detection initial tests. 2014-05-20 17:45:50 +02:00
antirez a82c000bdb Cluster test: proper initialization at unit startup. 2014-05-20 17:45:50 +02:00
antirez 8c6e86805c Cluster: use clusterSetNodeAsMaster() during slave failover. 
clusterHandleSlaveFailover() was reimplementing what
clusterSetNodeAsMaster() without any good reason.
 2014-05-20 17:45:50 +02:00
antirez 41a7241603 Cluster: clear todo_before_sleep flags when executing actions. 
Thanks to this change, when there is some code like:

    clusterDoBeforeSleep(CLUSTER_TODO_UPDATE_STATE|...);
    ... and later before returning to the event loop ...
    clusterUpdateState();

The clusterUpdateState() function will clar the flag and will not be
repeated in the clusterBeforeSleep() function. This especially important
for config save/fsync flags which are slow to execute and not a good
idea to repeat without a good reason.

This is implemented for all the CLUSTER_TODO flags.
 2014-05-20 17:45:50 +02:00
antirez 5685d15af1 Fixed typo in CLUSTER RESET implementation. 2014-05-20 17:45:50 +02:00
antirez 5efa5501e8 CLUSTER RESET implemented. 
The new command is able to reset a cluster node so that it starts again
as a fresh node. By default the command performs a soft reset (the same
as calling it as CLUSTER RESET SOFT), and the following steps are
performed:

1) All slots are set as unassigned.
2) The list of known nodes is flushed.
3) Node is set as master if it is a slave.

When an hard reset is performed with CLUSTER RESET HARD the following
additional operations are performed:

4) A new Node ID is created at random.
5) Epochs are set to 0.

CLUSTER RESET is useful both when the sysadmin wants to reconfigure a
node with a different role (for example turning a slave into a master)
and for testing purposes.

It also may play a role in automatically provisioned Redis Clusters,
since it allows to reset a node back to the initial state in order to be
reconfigured.
 2014-05-20 17:45:50 +02:00
antirez 687f84a377 Remove trailing spaces from cluster.c file. 2014-05-20 17:45:50 +02:00
antirez d279024081 Cluster test: added function assert_cluster_state. 2014-05-20 17:45:50 +02:00
antirez be1594905f Cluster: don't accept cluster bus connections during startup. 2014-05-20 17:45:50 +02:00
antirez b8a71e5a77 Cluster: better handling of stolen slots. 
The previous code handling a lost slot (by another master with an higher
configuration for the slot) was defensive, considering it an error and
putting the cluster in an odd state requiring redis-cli fix.

This was changed, because actually this only happens either in a
legitimate way, with failovers, or when the admin messed with the config
in order to reconfigure the cluster. So the new code instead will try to
make sure that the keys stored match the new slots map, by removing all
the keys in the slots we lost ownership from.

The function that deletes the keys from the lost slots is called only
if the node does not lose all its slots (resulting in a reconfiguration
as a slave of the node that got ownership). This is an optimization
since the replication code will anyway flush all the instance data in
a faster way.
 2014-05-20 17:45:49 +02:00
antirez f727000f0c cluster.tcl: fix redis links leak in refresh_nodes_map. 2014-05-20 17:45:49 +02:00
antirez 4db4a5d5d6 cluster.tcl: saner error handling. 
Better handling of connection errors in order to update the table and
recovery, populate the startup nodes table after fetching the list of
nodes.

More work to do about it, it is still not as reliable as
redis-rb-cluster implementation which is the minimal reference
implementation for Redis Cluster clients.
 2014-05-20 17:45:49 +02:00
antirez d9227938c8 redis.tcl: return I/O error message when peer closes connection. 2014-05-20 17:45:49 +02:00
antirez e10ee0728a Cluster: fixed data_age computation / check integer overflow. 2014-05-20 17:45:49 +02:00
antirez e84dcabf83 Cluster: forced failover implemented. 
Using CLUSTER FAILOVER FORCE it is now possible to failover a master in
a forced way, which means:

1) No check to understand if the master is up is performed.
2) No data age of the slave is checked. Evan a slave with very old data
   can manually failover a master in this way.
3) No chat with the master is attempted to reach its replication offset:
   the master can just be down.
 2014-05-20 17:45:49 +02:00
antirez b5cdd42bf3 Cluster: bypass data_age check for manual failovers. 
Automatic failovers only happen in Redis Cluster if the slave trying to
be elected was disconnected from its master for no more than 10 times
the node-timeout value. However there should be no such a check for
manual failovers, since these are initiated by the sysadmin that, in
theory, knows what she is doing when a slave is selected to be promoted.
 2014-05-20 17:45:49 +02:00
antirez 26eb7ec731 HyperLogLog regression test for issue #1762. 2014-05-19 15:45:22 +02:00
Mike Trinkala 73737c4b85 Correct the HyperLogLog stale cache flag to prevent unnecessary computations. 
Set the MSB as documented.
 2014-05-19 15:45:22 +02:00
Akos Vandra 15a1c261ea Fixed possible buffer overflow bug if RDB file is corrupted. 
(Note: commit message modified by @antirez for clarity).
 2014-05-12 15:56:17 +02:00
Akos Vandra ef92a30767 fixed possible buffer overflow error 2014-05-12 15:56:17 +02:00
antirez d3ddeff2d2 redis-trib create: use CONFIG SET-CONFIG-EPOCH before joining the cluster. 
This way there is no need for the conflict resolution algo to be used in
order to start with a cluster where each node has a different
configEpoch.
 2014-05-12 15:56:17 +02:00
antirez ce5e625a5e Sentinel: Add "dir /tmp" directive in example sentinel.conf. 2014-05-12 15:56:17 +02:00
antirez 34b68b7e89 redis-trib import: trap MIGRATE errors. 2014-05-12 15:56:17 +02:00
antirez bdfa26fb51 redis-trib.rb: MIGRATE hardcoded timeout set to 15 sec. 
Will be configurable / adaptive at some point but let's start with a
saner value compared to 1 sec which is not a good idea for big data
structures stored into a single key.
 2014-05-12 15:56:17 +02:00
antirez 0a707babb1 RESTORE: reply with -BUSYKEY special error code. 
The error when the target key is busy was a generic one, while it makes
sense to be able to distinguish between the target key busy error and
the others easily.
 2014-05-12 15:56:17 +02:00
antirez 85d2a8091c Cluster: initial ability to import data from standalone instance. 2014-05-12 15:56:17 +02:00
antirez fafe29cd4c CLUSTER MEET: better error messages when address is invalid. 
Fixes issue #1734.
 2014-05-12 15:56:17 +02:00
antirez 5fef9b63d2 redis-trib: allow support for mandatory options. 2014-05-12 15:56:16 +02:00
antirez 8522f295fa DEBUG POPULATE: call dictExpand() to avoid useless rehashing. 2014-05-09 15:05:26 +02:00
antirez 83c92d50a6 Cluster: bulk-accept new nodes connections. 
The same change was operated for normal client connections. This is
important for Cluster as well, since when a node rejoins the cluster,
when a partition heals or after a restart, it gets flooded with new
connection attempts by all the other nodes trying to form a full
mesh again.
 2014-05-09 15:05:26 +02:00
antirez ed18587f52 Cluster: clusterAcceptHandler() comments updated to match the code. 2014-05-09 15:05:26 +02:00
antirez 6eadeb7236 Sentinel: log when a failover will be attempted again. 
When a Sentinel performs a failover (successful or not), or when a
Sentinel votes for a different Sentinel trying to start a failover, it
sets a min delay before it will try to get elected for a failover.

While not strictly needed, because if multiple Sentinels will try
to failover the same master at the same time, only one configuration
will eventually win, this serialization is practically very useful.
Normal failovers are cleaner: one Sentinel starts to failover, the
others update their config when the Sentinel performing the failover
is able to get the selected slave to move from the role of slave to the
one of master.

However currently this timeout was implicit, so users could see
Sentinels not reacting, after a failed failover, for some time, without
giving any feedback in the logs to the poor sysadmin waiting for clues.

This commit makes Sentinels more verbose about the delay: when a master
is down and a failover attempt is not performed because the delay has
still not elaped, something like that will be logged:

    Next failover delay: I will not start a failover
    before Thu May  8 16:48:59 2014
 2014-05-08 16:49:41 +02:00
antirez c738f0f4fc Sentinel: generate +config-update-from event when a new config is received. 
This event makes clear, before the switch-master event is generated,
that a Sentinel received a configuration update from another Sentinel.
 2014-05-08 16:49:41 +02:00
antirez d8ffd2b2ba REDIS_ENCODING_EMBSTR_SIZE_LIMIT set to 39. 
The new value is the limit for the robj + SDS header + string +
null-term to stay inside the 64 bytes Jemalloc arena in 64 bits
systems.
 2014-05-08 15:40:43 +02:00
antirez 35b94422f6 Scripting test: check that Lua can call commands rewirting argv. 
SPOP, tested in the new test, is among the commands rewritng the
client->argv argument vector (it gets rewritten as SREM) for command
replication purposes.

Because of recent optimizations to client->argv caching in the context
of the Lua internal Redis client, it is important to test for SPOP to be
callable from Lua without bad effects to the other commands.
 2014-05-07 16:12:38 +02:00
antirez 1255460c96 Test: handle new osx 'leaks' error. 
Sometimes the process is still there but no longer in a state that can
be checked (after being killed). This used to happen after a call to
SHUTDOWN NOSAVE in the scripting unit, causing a false positive.
 2014-05-07 16:12:38 +02:00
antirez baf930176b Scripting: objects caching for Lua c->argv creation. 
Reusing small objects when possible is a major speedup under certain
conditions, since it is able to avoid the malloc/free pattern that
otherwise is performed for every argument in the client command vector.
 2014-05-07 16:12:38 +02:00
antirez 77503edf10 Scripting: Use faster API for Lua client c->argv creation. 
Replace the three calls to Lua API lua_tostring, lua_lua_strlen,
and lua_isstring, with a single call to lua_tolstring.

~ 5% consistent speed gain measured.
 2014-05-07 16:12:38 +02:00
antirez d757d400c1 Scripting: don't call lua_gc() after Lua script run. 
Calling lua_gc() after every script execution is too expensive, and
apparently does not make the execution smoother: the same peak latency
was measured before and after the commit.

This change accounts for scripts execution speedup in the order of 10%.
 2014-05-07 16:12:38 +02:00
antirez ef164836b1 Scripting: cache argv in luaRedisGenericCommand(). 
~ 4% consistently measured speed improvement.
 2014-05-07 16:12:38 +02:00
antirez 5c291ba9f6 Fixed missing c->bufpos reset in luaRedisGenericCommand(). 
Bug introduced when adding a fast path to avoid copying the reply buffer
for small replies that fit into the client static buffer.
 2014-05-07 16:12:38 +02:00
antirez 4b1cb0c44f Scripting: replace tolower() with faster code in evalGenericCommand(). 
The function showed up consuming a non trivial amount of time in the
profiler output. After this change benchmarking gives a 6% speed
improvement that can be consistently measured.
 2014-05-07 16:12:38 +02:00
antirez b1f3bf8858 Scripting: luaRedisGenericCommand() fast path for buffer-only replies. 
When the reply is only contained in the client static output buffer, use
a fast path avoiding the dynamic allocation of an SDS string to
concatenate the client reply objects.
 2014-05-07 16:12:38 +02:00
antirez 473a8bd7d8 Define HAVE_ATOMIC for clang. 2014-05-07 16:12:38 +02:00
antirez 650b48ea7c Scripting: simpler reply buffer creation in luaRedisGenericCommand(). 
It if faster to just create the string with a single sdsnewlen() call.
If c->bufpos is zero, the call will simply be like sdsemtpy().
 2014-05-07 16:12:38 +02:00
antirez b7a6a6c486 Redis 2.9.52 (Redis 3.0.0 beta-3). 2014-05-05 10:29:00 +02:00
antirez 402b1170bc Test: cluster/base, check that we can write/read from cluster. 2014-05-05 09:37:39 +02:00
antirez f188fa4a15 Cluster: Tcl cluster client: handle MOVED/ASK. 2014-05-05 09:37:39 +02:00
antirez 5069e0b9cd Cluster: Tcl cluster client: slots-nodes map and close method. 
Now the client is able to actually run commands in a Redis Cluster
assuming the slots->nodes map is stable.
 2014-05-05 09:37:39 +02:00
antirez 9fe28612b9 Cluster: Tcl cluster client: build nodes representation. 2014-05-05 09:37:39 +02:00
antirez a5b7065bdf Cluster: Tcl cluster client: get nodes description. 2014-05-05 09:37:39 +02:00
antirez 1826167519 Cluster: Tcl cluster client key -> hashslot. 2014-05-05 09:37:39 +02:00
antirez 43570e9319 Cluster test: Tcl cluster library initial skeleton. 2014-05-05 09:37:39 +02:00
antirez 4ca7c5bcde Cluster test: check for state=ok after slot allocation. 2014-05-05 09:37:39 +02:00
antirez c4c7389fea CLUSTER SET-CONFIG-EPOCH implemented. 
Initially Redis Cluster accepted that after cluster creation all the
nodes were at configEpoch 0, evolving from zero as failovers happen.

However later the semantic was made more strict in order to make sure a
cluster has always all the master nodes with a different configEpoch,
which is more robust in some corner case (especially resulting from
errors by the system administrator).

To assign different configEpochs to different nodes at startup was a
task performed naturally by the config conflicts resolution algorithm
(see the Cluster specification). However this works well only for small
clusters or when there are actually just a few collisions, since it is
designed for exceptional cases.

When a large cluster is created hundred of nodes can be at epoch 0, so
the conflict resolution code is slow to provide an unique config to each
node. For this reason this new command was introduced. It can be called
only when a node is totally fresh: no other nodes known, and configEpoch
set to zero, so it is safe even against misuses.

redis-trib will use the new command in order to start the cluster
already setting an incremental unique config to every node.
 2014-05-05 09:37:39 +02:00
antirez 6fc17adcd8 Cluster test: slots allocation. 2014-05-05 09:37:39 +02:00
antirez 0906ca8a3e Cluster test: use 20 instances. 
This makes tests a bit slower, but it is better to test things at a
decent scale instead of using just a few nodes, and for a few tests we
actually need so many nodes.
 2014-05-05 09:37:39 +02:00
antirez 402815cc3f Cluster / Sentinel test: instances count moved to run.tcl. 2014-05-05 09:37:39 +02:00
antirez 6e60e2467b Cluster test: config epoch conflict resolution. 2014-05-05 09:37:39 +02:00
antirez b1f6f2ace0 Cluster test: auto-discovery to form full mesh. 2014-05-05 09:37:39 +02:00
antirez e244bee60c Cluster test: check that every node has a different ID. 2014-05-05 09:37:39 +02:00
antirez 6b4e9558bd Cluster test: basic cluster nodes info access functions. 2014-05-05 09:37:39 +02:00
antirez a11ae385e9 CLIENT LIST speedup via peerid caching + smart allocation. 
This commit adds peer ID caching in the client structure plus an API
change and the use of sdsMakeRoomFor() in order to improve the
reallocation pattern to generate the CLIENT LIST output.

Both the changes account for a very significant speedup.
 2014-04-28 18:17:02 +02:00
antirez bb00154f0e Use sdscatfmt() in getClientInfoString() to make it faster. 2014-04-28 18:17:02 +02:00
antirez e285a57073 Added new sdscatfmt() %u and %U format specifiers. 
This commit also fixes a bug in the implementation of sdscatfmt()
resulting from stale references to the SDS string header after
sdsMakeRoomFor() calls.
 2014-04-28 18:17:02 +02:00
antirez 0ab06bb500 sdscatfmt() added to SDS library. 
sdscatprintf() relies on printf() family libc functions and is sometimes
too slow in critical code paths. sdscatfmt() is an alternative which is:

1) Far less capable.
2) Format specifier uncompatible.
3) Faster.

It is suitable to be used in those speed critical code paths such as
CLIENT LIST output generation.
 2014-04-28 18:17:02 +02:00
antirez eeab60d1d5 Test: fixed scripting.tcl test false positive. 2014-04-28 18:17:02 +02:00
antirez cdd2bd5632 Process events with processEventsWhileBlocked() when blocked. 
When we are blocked and a few events a processed from time to time, it
is smarter to call the event handler a few times in order to handle the
accept, read, write, close cycle of a client in a single pass, otherwise
there is too much latency added for clients to receive a reply while the
server is busy in some way (for example during the DB loading).
 2014-04-28 18:17:02 +02:00
antirez 357b039fdb Accept multiple clients per iteration. 
When the listening sockets readable event is fired, we have the chance
to accept multiple clients instead of accepting a single one. This makes
Redis more responsive when there is a mass-connect event (for example
after the server startup), and in workloads where a connect-disconnect
pattern is used often, so that multiple clients are waiting to be
accepted continuously.

As a side effect, this commit makes the LOADING, BUSY, and similar
errors much faster to deliver to the client, making Redis more
responsive when there is to return errors to inform the clients that the
server is blocked in an not interruptible operation.
 2014-04-28 18:17:02 +02:00
antirez ba86119414 AE_ERR -> ANET_ERR in acceptUnixHandler(). 
No actual changes since the value is the same.
 2014-04-28 18:17:02 +02:00
antirez a8464828b6 Redis Cluster test framework skeleton. 2014-04-28 18:17:02 +02:00
antirez 3acae3dee3 While ANET_ERR is -1, check syscall retval for -1 itself. 2014-04-28 18:17:02 +02:00
antirez b008863edb clusterLoadConfig() REDIS_ERR retval semantics refined. 
We should return REDIS_ERR to signal we can't read the configuration
because there is no config file only after checking errno, othewise
we risk to rewrite an existing file that was not accessible for some
other reason.
 2014-04-28 18:17:02 +02:00
antirez 71d71814a9 Lock nodes.conf to avoid multiple processes using the same file. 
This was a common source of problems among users.
The solution adopted is not bullet-proof as if the user deletes the
nodes.conf file manually, and starts a new instance with the same
nodes.conf file path, two instances will use the same file. However
following this reasoning the user may drop a nuclear bomb into the
datacenter as well.
 2014-04-28 18:17:02 +02:00
antirez a0050e5030 Sentinel test files / directories layout improved. 
The test now runs in a self-contained directory.
The general abstractions to run the tests in an environment where
mutliple instances are executed at the same time was extrapolated into
instances.tcl, that will be reused to test Redis Cluster.
 2014-04-28 18:17:02 +02:00
antirez 3bff0291d8 Test: spawn_instance now supports additional config. 2014-04-28 18:17:02 +02:00
Matt Stancliff 08174e730f Add test for deleting an expired key 
Verify proper expire-before-delete behavior.

This test passes with the expire-before-delete commit and fails
without it.
 2014-04-23 16:14:34 +02:00
Matt Stancliff e687621e15 Check key expiration before deleting 
Deleting an expired key should return 0, not success.

Fixes #1648
 2014-04-23 16:14:34 +02:00
Glauber Costa 0c8fdb349c fix null pointer access with no file pointer 
I happen to be working on a system that lacks urandom. While the code does try
to handle this case and artificially create some bytes if the file pointer is
empty, it does try to close it unconditionally, leading to a segfault.
 2014-04-23 15:35:07 +02:00
kingsumos 19ac9af09f fix cluster node description showing wrong slot allocation 2014-04-23 15:35:07 +02:00
Kevin Menard 7b25ff4bf7 Fixed typos. 2014-04-23 15:34:47 +02:00
antirez 7b670805cf Missing return REDIS_ERR added to processMultibulkBuffer(). 
When we set a protocol error we should return with REDIS_ERR to let the
caller know it should stop processing the client.

Bug found in a code auditing related to issue #1699.
 2014-04-23 10:21:30 +02:00
antirez e99e45be38 redis-cli help.h updated. 2014-04-22 16:14:59 +02:00
antirez 25f72e9e0d generate-command-help.rb updated with new hyperloglog group. 2014-04-22 16:14:59 +02:00
antirez aff94a706e Fuzzy test for ZREMRANGEBYLEX added. 2014-04-18 16:14:34 +02:00
antirez 5b5dd73093 ZREMRANGEBYLEX memory leak removed calling zslFreeLexRange(). 2014-04-18 16:14:34 +02:00
antirez ca6c793844 PFCOUNT multi-key test added. 2014-04-18 16:14:34 +02:00
antirez 82556c1687 Speedup hllRawSum() processing 8 bytes per iteration. 
The internal HLL raw encoding used by PFCOUNT when merging multiple keys
is aligned to 8 bits (1 byte per register) so we can exploit this to
improve performances by processing multiple bytes per iteration.

In benchmarks the new code was several times faster with HLLs with many
registers set to zero, while no slowdown was observed with populated
HLLs.
 2014-04-18 16:14:34 +02:00
antirez 5e928eff64 Speedup SUM(2^-reg[m]) in HyperLogLog computation. 
When the register is set to zero, we need to add 2^-0 to E, which is 1,
but it is faster to just add 'ez' at the end, which is the number of
registers set to zero, a value we need to compute anyway.
 2014-04-18 16:14:34 +02:00
antirez 00726b2e0e PFCOUNT support for multi-key union. 2014-04-18 16:14:34 +02:00
antirez 57b34a73c2 HyperLogLog low level merge extracted from PFMERGE. 2014-04-18 16:14:34 +02:00
antirez 816e22bc6f ZREMRANGEBYLEX implemented. 2014-04-18 16:14:34 +02:00
antirez d28b239fe9 Always pass sorted set range objects by reference. 2014-04-18 16:14:34 +02:00
antirez 69d0585d61 ZREMRANGE* commands refactored into a single generic function. 2014-04-18 16:14:34 +02:00
antirez 53ca563ff7 Sorted set lex ranges stress tester. 2014-04-18 16:14:34 +02:00
antirez 5e1224dc1d Basic ZRANGEBYLEX / ZLEXCOUNT tests. 2014-04-17 00:08:35 +02:00
antirez 964faba650 Pass by pointer and release of lex ranges. 
Given that the code was written with a 2 years pause... something
strange happened in the middle. So there was no function to free a
lex range min/max objects, and in some places the range was passed by
value.
 2014-04-17 00:08:35 +02:00
antirez 5e21acec54 ZLEXCOUNT implemented. 
Like ZCOUNT for lexicographical ranges.
 2014-04-16 15:09:47 +02:00
antirez 00aa96f34b More HyperLogLog tests. 2014-04-16 15:09:47 +02:00
antirez a3848a7429 HyperLogLog invalid representation error code set to INVALIDOBJ. 2014-04-16 15:09:47 +02:00
antirez 030336d4ef PFDEBUG TODENSE added. 
Converts HyperLogLogs from sparse to dense. Used for testing.
 2014-04-16 15:09:47 +02:00
antirez e4c2afb9c0 User-defined switch point between sparse-dense HLL encodings. 2014-04-16 15:09:47 +02:00
antirez e19e078c77 PFSELFTEST improved with sparse encoding checks. 2014-04-16 15:09:47 +02:00
antirez 66f67208d3 PFDEBUG ENCODING added. 2014-04-16 15:09:47 +02:00
antirez db929041b5 Set HLL_SPARSE_MAX to 3000. 
After running a few benchmarks, 3000 looks like a reasonable value to
keep HLLs with a few thousand elements small while the CPU cost is
still not huge.

This covers all the cases where the dense representation would use N
orders of magnitude more space, like in the case of many HLLs with
carinality of a few tens or hundreds.

It is not impossible that in the future this gets user configurable,
however it is easy to pick an unreasoable value just looking at savings
in the space dimension without checking what happens in the time
dimension.
 2014-04-16 15:09:47 +02:00
antirez 90b67e416a Error message for invalid HLL objects unified. 2014-04-16 15:09:47 +02:00
antirez fdace9bc50 PFMERGE fixed to work with sparse encoding. 2014-04-16 15:09:47 +02:00
antirez 64fbd15706 Mark PFDEBUG as write command in the commands table. 
It is safer since it is able to have side effects.
 2014-04-16 15:09:47 +02:00
antirez 159926a170 Correctly replicate PFDEBUG GETREG. 
Even if it is a debugging command, make sure that when it forces a
change in encoding, the command is propagated.
 2014-04-16 15:09:47 +02:00
antirez fc33779283 Added assertion in hllSparseAdd() when promotion to dense occurs. 
If we converted to dense, a register must be updated in the dense
representation.
 2014-04-16 15:09:47 +02:00
antirez 67e15a0ff7 hllSparseAdd(): speed optimization. 
Mostly by reordering opcodes check conditional by frequency of opcodes
in larger sparse-encoded HLLs.
 2014-04-16 15:09:47 +02:00
antirez ed52bbe134 Detect corrupted sparse HLLs in hllSparseSum(). 2014-04-16 15:09:46 +02:00
antirez fd137b0781 hllSparseAdd(): faster code removing conditional. 
Bottleneck found profiling. Big run time improvement found when testing
after the change.
 2014-04-16 15:09:46 +02:00
antirez 8b3d5a8eb0 Comment typo in hllSparseAdd(). first -> fits. 2014-04-16 15:09:46 +02:00
antirez 46f9a78d9c Merge adjacent VAL opcodes in hllSparseAdd(). 
As more values are added splitting ZERO or XZERO opcodes, try to merge
adjacent VAL opcodes if they have the same value.
 2014-04-16 15:09:46 +02:00
antirez 9f12467c9c More robust HLL_SPARSE macros protecting 'p' with parens. 
Now the macros will work with arguments such as "ptr+1".
 2014-04-16 15:09:46 +02:00
antirez 10bc0aea8d hllSparseAdd() opcode seek stop condition fixed. 2014-04-16 15:09:46 +02:00
antirez 8be9394ec7 Fixed error message generation in PFDEBUG GETREG. 
Bulk length for registers was emitted too early, so if there was a bug
the reply looked like a long array with just one element, blocking the
client as result.
 2014-04-16 15:09:46 +02:00
antirez 8cc09325c8 Fixed memmove() count in hllSparseAdd(). 2014-04-16 15:09:46 +02:00
antirez da084bdf84 hllSparseAdd(): more correct dense conversion conditional. 
We want to promote if the total string size exceeds the resulting size
after the upgrade.
 2014-04-16 15:09:46 +02:00
antirez a7dbd28cff hllSparseToDense(): sanity check added. 
The function checks if all the HLL_REGISTERS were processed during the
convertion from sparse to dense encoding, returning REDIS_OK or
REDIS_ERR to signal a corruption problem.

A bug in PFDEBUG GETREG was fixed: when the object is converted to the
dense representation we need to reassign the new pointer to the header
structure pointer.
 2014-04-16 15:09:46 +02:00
antirez cf500998e8 PFDEBUG DECODE added. 
Provides a human readable description of the opcodes composing a
run-length encoded HLL (sparse encoding).
The command is only useful for debugging / development tasks.
 2014-04-16 15:09:46 +02:00
antirez 2c4a1eccda PFDEBUG added, PFGETREG removed. 
PFDEBUG will be the interface to do debugging tasks with a key
containing an HLL object.
 2014-04-16 15:09:46 +02:00
antirez 5786846122 hllSparseToDense API changed to take ref to object. 
The new API takes directly the object doing everything needed to
turn it into a dense representation, including setting the new
representation as object->ptr.
 2014-04-16 15:09:46 +02:00
antirez 68fb3019a0 hllSparseAdd() sanity check for span != 0 added. 2014-04-16 15:09:46 +02:00
antirez e7e6aa49d0 Fix hllSparseAdd() new sequence replacement when next is NULL. 
sdsIncrLen() must be called anyway even if we are replacing the last
oppcode of the sparse representation.
 2014-04-16 15:09:46 +02:00
antirez b964380817 Fix seqlen computation in hllSparseAdd(). 2014-04-16 15:09:46 +02:00
antirez 1c6671ab90 Abstract hllSparseAdd() / hllDenseAdd() via hllAdd(). 2014-04-16 15:09:46 +02:00
antirez e3f5a38695 hllSparseSum(): multiply 1 * runlen for zero entries. 2014-04-16 15:09:46 +02:00
antirez 5b69b984b3 Macro HLL_SPARSE_XZERO_LEN fixed. 2014-04-16 15:09:46 +02:00
antirez 2c1fa11124 Fix HLL sparse object creation #2. 
Two vars initialized to wrong values in createHLLObject().
 2014-04-16 15:09:46 +02:00
antirez 0b9f09c4e0 Increment pointer while iterating sparse HLL object. 2014-04-16 15:09:46 +02:00
antirez e526d32635 Fix HLL sparse object creation. 
The function didn't considered the fact that each XZERO opcode is
two bytes.
 2014-04-16 15:09:46 +02:00
antirez ccbabf01e1 Create HyperLogLog objects with sparse encoding. 2014-04-16 15:09:46 +02:00
antirez df5782f4d0 HyperLogLog sparse to dense conversion function. 2014-04-16 15:09:46 +02:00
antirez e872394176 HyperLogLog sparse representation initial implementation. 
Code never tested, but the basic layout is shaped in this commit.
Also missing:

1) Sparse -> Dense conversion function.
2) New HLL object creation using the sparse representation.
3) Implementation of PFMERGE for the sparse representation.
 2014-04-16 15:09:46 +02:00
antirez 4e12ecdba5 hllCount() refactored to support multiple representations. 2014-04-16 15:09:46 +02:00
antirez 05b5032302 hllAdd() refactored into two functions. 
Also dense representation access macro renamed accordingly.
 2014-04-16 15:09:46 +02:00
antirez 4407ca6fc8 HyperLogLog refactoring to support different encodings. 
Metadata are now placed at the start of the representation as an header.
There is a proper structure to access the representation.
Still work to do in order to truly abstract the implementation from the
representation, commands still work assuming dense representation.
 2014-04-16 15:09:46 +02:00
antirez 099ae637ab HyperLogLog sparse representation slightly modified. 
After running a few simulations with different alternative encodings,
it was found that the VAL opcode performs better using 5 bits for the
value and 2 bits for the run length, at least for cardinalities in the
range of interest.
 2014-04-16 15:09:46 +02:00
antirez 2b531f729d HyperLogLog sparse representation description and macros. 2014-04-16 15:09:46 +02:00
antirez 437cddee0d Add casting to match printf format. 
adjustOpenFilesLimit() and clusterUpdateSlotsWithConfig() that were
assuming uint64_t is the same as unsigned long long, which is true
probably for all the systems out there that we target, but still GCC
emitted a warning since technically they are two different types.
 2014-04-16 15:09:46 +02:00
antirez 109ec9c6a5 ZRANGEBYLEX and ZREVRANGEBYLEX implementation. 2014-04-16 15:09:46 +02:00
antirez f5bdaf366e PFCOUNT: always unshare/decode the object. 
This will be a non-op most of the times since the object will be
unshared / decoded, however it is more technically correct to start this
way since the object may be decoded even in the read-only code path.
 2014-04-16 15:09:46 +02:00
antirez ae6103ebf6 tryObjectEncoding() refactoring. 
We also avoid to re-create an object that is already in EMBSTR encoding.
 2014-04-16 15:09:46 +02:00
antirez 63ba5f5a8c Changed HyperLogLog hash seed to a non-zero value. 
Using a seed of zero has the side effect of having the empty string
hashing to what is a very special case in the context of HyperLogLog: a
very long run of zeroes.

This did not influenced the correctness of the result with 16k registers
because of the harmonic mean, but still it is inconvenient that a so
obvious value maps to a so special hash.

The seed 0xadc83b19 is used instead, which is the first 64 bits of the
SHA1 of the empty string.

Reference: issue #1657.
 2014-04-16 15:09:46 +02:00
antirez e0389f031e Initial HyperLogLog tests. 2014-04-16 15:09:45 +02:00
antirez 5056450b02 Return "WRONGTYPE" error on PF* type mismatch. 2014-04-16 15:09:45 +02:00
antirez f553fc442f Fix PFADD infinite loop. 
We need to guarantee that the last bit is 1, otherwise an element may
hash to just zeroes with probability 1/(2^64) and trigger an infinite
loop.

See issue #1657.
 2014-04-16 15:09:45 +02:00
antirez 19938b9103 Make hll-gnuplot-graph.rb callable from cli. 2014-04-16 15:09:45 +02:00
antirez 628b9fa535 Remove HyperLogLog type checking duplicated code. 2014-04-16 15:09:45 +02:00
antirez aecb59b0b2 PFGETREG added for testing purposes. 
The new command allows to get a dump of the registers stored
into an HyperLogLog data structure for testing / debugging purposes.
 2014-04-16 15:09:45 +02:00
antirez e6fda1c9ed PFCOUNT: unshare the object when cached cardinality is modified. 2014-04-16 15:09:45 +02:00
antirez 9c7bb351ab PFSELFTEST improved to test the approximation error. 2014-04-16 15:09:45 +02:00
antirez 80d7f7081c hll-gnuplot-graph.rb improved with new filter. 
The function to generate graphs is also more flexible as now includes
step and max value. The step of the samples generation function is no
longer limited to min step of 1000.
 2014-04-16 15:09:45 +02:00
antirez 6415fe5623 HyperLogLog: added magic / version. 
This will allow future changes like compressed representations.
Currently the magic is not checked for performance reasons but this may
change in the future, for example if we add new types encoded in strings
that may have the same size of HyperLogLogs.
 2014-04-16 15:09:45 +02:00
Raymond Myers f403fca654 Fixed pfadd/pfcount commands emitting hll* events instead of pf* events 2014-04-16 15:09:45 +02:00
Raymond Myers 21c9372454 Change HLL* to PF* in error messages 2014-04-16 15:09:45 +02:00
antirez f307a51049 Include redis.h before other stuff in hyperloglog.c. 
Otherwise fmacros.h is included later and this may break compilation on
different systems.
 2014-04-16 15:09:45 +02:00
antirez 08516c1ab6 HyperLogLog API prefix modified from "P" to "PF". 
Using both the initials of Philippe Flajolet instead of just "P".
 2014-04-16 15:09:45 +02:00
antirez 0c01e293ef Makefile.dep updated with hyperloglog.o deps. 2014-04-16 15:09:45 +02:00
antirez e262442fd4 HyperLogLog: make API use the P prefix in honor of Philippe Flajolet. 2014-04-16 15:09:45 +02:00
antirez 8ad9123a14 HLLMERGE fixed by adding a... missing loop! 2014-04-16 15:09:45 +02:00
antirez 9cb2b5003f hll-gnuplot-graph.rb: added new filter "all". 2014-04-16 15:09:45 +02:00
antirez 07daaa6ec4 HyperLogLog apply bias correction using a polynomial. 
Better results can be achieved by compensating for the bias of the raw
approximation just after 2.5m (when LINEARCOUNTING is no longer used) by
using a polynomial that approximates the bias at a given cardinality.

The curve used was found using this web page:

    http://www.xuru.org/rt/PR.asp

That performs polynomial regression given a set of values.
 2014-04-16 15:09:45 +02:00
antirez 691846e38d HLLMERGE implemented. 
Merge N HLL data structures by selecting the max value for every
M[i] register among the set of HLLs.
 2014-04-16 15:09:45 +02:00
antirez 7f18689a00 HLLCOUNT is technically a write command 
When we update the cached value, we need to propagate the command and
signal the key as modified for WATCH.
 2014-04-16 15:09:45 +02:00
antirez 6672e101e8 HLLADD: propagate write when only variable name is given. 
The following form is given:

    HLLADD myhll

No element is provided in the above case so if 'myhll' var does not
exist the result is to just create an empty HLL structure, and no update
will be performed on the registers.

In this case, the DB should still be set dirty and the command
propagated.
 2014-04-16 15:09:45 +02:00
antirez df2c02785a hll-gnuplot-graph.rb: Use |error| when filter is :max 2014-04-16 15:09:45 +02:00
antirez dc774e1bdc Ignore txt files inside utils/hyperloglog. 
Those are generated to trace graphs using gnuplot.
 2014-04-16 15:09:45 +02:00
antirez 00cb4c5dee HyperLogLog: use LINEARCOUNTING up to 3m. 
The HyperLogLog original paper suggests using LINEARCOUNTING for
cardinalities < 2.5m, however for P=14 the median / max error
curves show that a value of '3' is the best pick for m = 16384.
 2014-04-16 15:09:45 +02:00
antirez 2c1b4dcd40 hll-gnuplot-graph.rb added to plot HyperLogLog error graphs. 2014-04-16 15:09:45 +02:00
antirez 7acccc7ef9 HyperLogLog approximated cardinality caching. 
The more we add elements to an HyperLogLog counter, the smaller is
the probability that we actually update some register.

From this observation it is easy to see how it is possible to use
caching of a previously computed cardinality and reuse it to serve
HLLCOUNT queries as long as no register was updated in the data
structure.

This commit does exactly this by using just additional 8 bytes for the
data structure to store a 64 bit unsigned integer value cached
cardinality. When the most significant bit of the 64 bit integer is set,
it means that the value computed is no longer usable since at least a
single register was modified and we need to recompute it at the next
call of HLLCOUNT.

The value is always stored in little endian format regardless of the
actual CPU endianess.
 2014-04-16 15:09:45 +02:00
antirez 8be30eca17 String value unsharing refactored into proper function. 
All the Redis functions that need to modify the string value of a key in
a destructive way (APPEND, SETBIT, SETRANGE, ...) require to make the
object unshared (if refcount > 1) and encoded in raw format (if encoding
is not already REDIS_ENCODING_RAW).

This was cut & pasted many times in multiple places of the code. This
commit puts the small logic needed into a function called
dbUnshareStringValue().
 2014-04-16 15:09:45 +02:00
antirez f3aa086b6d Use endian neutral hash function for HyperLogLog. 
We need to be sure that you can save a dataset in a Redis instance,
reload it in a different architecture, and continue to count in the same
HyperLogLog structure.

So 32 and 64 bit, little or bit endian, must all guarantee to output the
same hash for the same element.
 2014-04-16 15:09:45 +02:00
antirez f704678d07 HyperLogLog internal representation modified. 
The new representation is more obvious, starting from the LSB of the
first byte and using bits going to MSB, and passing to next byte as
needed.

There was also a subtle error: first two bits were unused, everything
was carried over on the right of two bits, even if it worked because of
the code requirement of always having a byte more at the end.

During the rewrite the code was made safer trying to avoid undefined
behavior due to shifting an uint8_t for more than 8 bits.
 2014-04-16 15:09:45 +02:00
antirez c1f97815ca Remove a few useless operations from hllCount() fast path. 2014-04-16 15:09:45 +02:00
antirez c16f638dc5 HLLCOUNT 3x faster taking fast path for default params. 2014-04-16 15:09:45 +02:00
antirez e6d57660ed Use processor base types in HLL_(GET|SET)_REGISTER. 
This speedups the macros by a noticeable factor.
 2014-04-16 15:09:45 +02:00
antirez 650b21fb2c HyperLogLog: use precomputed table for 2^(-M[i]). 2014-04-16 15:09:45 +02:00
antirez 6aa1b23269 hll-err.rb: speedup using pipelining. 2014-04-16 15:09:44 +02:00
antirez a0b27c329a hll-err.rb added to test error rate of Redis HyperLogLog. 2014-04-16 15:09:44 +02:00
antirez 7d04f823e1 HyperLogLog algorithm fixed in two ways. 
There was an error in the computation of 2^register, and the sequence of
zeroes computed after the hashing did not included the "1".
 2014-04-16 15:09:44 +02:00
antirez 93e5876a72 HLLCOUNT implemented. 2014-04-16 15:09:44 +02:00
antirez bb57749a86 HLLADD implemented. 2014-04-16 15:09:44 +02:00
antirez 0faae58091 hllAdd() low level HyperLogLog "add" implemented. 2014-04-16 15:09:44 +02:00
antirez c03e7ceaaa HyperLogLog: redefine constants using "P". 2014-04-16 15:09:44 +02:00
antirez f4839dfb27 HLL_SET_REGISTER fixed. 
There was an error in the first version of the macro.
Now the HLLSELFTEST test reports success.
 2014-04-16 15:09:44 +02:00
antirez 87dd31427c Use REDIS_HLL_REGISTER_MAX when possible. 2014-04-16 15:09:44 +02:00
antirez 3610af42b2 HLL_(SET|GET)_REGISTER types fixed. 2014-04-16 15:09:44 +02:00
antirez c7344144f4 HLLSELFTEST command implemented. 
To test the bitfield array of counters set/get macros from the Redis Tcl
suite is hard, so a specialized command that is able to test the
internals was developed.
 2014-04-16 15:09:44 +02:00
antirez be20651435 HyperLogLog: initial sketch of registers access. 2014-04-16 15:09:44 +02:00
antirez 8c6ce3dd6a Cluster: last_vote_epoch -> lastVoteEpoch. 
Use cammel case for epochs that are persisted on disk.
 2014-04-16 15:09:44 +02:00
antirez 1080e272de Cluster: save/restore vars that must persist after recovery. 
This fixes issue #1479.
 2014-04-16 15:09:44 +02:00
antirez 47fbbd9fbc Cluster: handshake "already known" error logged to VERBOSE. 
This is not really an error but something that always happens for
example when creating a new cluster, or if the sysadmin rejoins manually
a node that is already known.

Since useless logs don't help, moved to VERBOSE level.
 2014-04-16 15:09:44 +02:00
antirez 6875a15858 Cluster: clusterHandleConfigEpochCollision() fixed. 
New config epochs must always be obtained incrementing the currentEpoch,
that is itself guaranteed to be >= the max configEpoch currently known
to the node.
 2014-04-16 15:09:44 +02:00
antirez 74fd89b321 Cluster: better logging for clusterUpdateSlotsConfigWith(). 2014-04-16 15:09:44 +02:00
antirez cf8f72c184 Cluster: CLUSTER SETSLOT implementation comment updated. 
Update the comment since the implementation details changed.
 2014-04-16 15:09:44 +02:00
antirez 6837863902 Cluster: redis-trib cluster allocation more even across nodes. 
redis-trib used to allocate slots not considering fractions of nodes
when computing the slots_per_node amount. So the fractional part was
carried over till the end of the allocation, where the last node
received a few more slots than any other (or a lot more if the cluster
was composed of many nodes).

The computation was changed to allocate slots more evenly when they are
not exactly divisible for the number of masters we have.
 2014-04-16 15:09:44 +02:00
antirez 431573ae98 Cluster: configEpoch collisions resolution. 
The slave election in Redis Cluster guarantees that slaves promoted to
masters always end with unique config epochs, however failures during
manual reshardings, software bugs and operational errors may in theory
cause two nodes to have the same configEpoch.

This commit introduces a mechanism to eventually always end with different
configEpochs if a collision ever happens.

As a (wanted) side effect, this also ensures that after a new cluster
is created, all nodes will end with a different configEpoch automatically.
 2014-04-16 15:09:44 +02:00
antirez 8a3dde208d Cluster: stay within 80 cols. 2014-04-16 15:09:44 +02:00
antirez 046e691a6f Fix off by one bug in freeMemoryIfNeeded() eviction pool. 
Bug found by the continuous integration test running the Redis
with valgrind:

==6245== Invalid read of size 8
==6245==    at 0x4C2DEEF: memcpy@GLIBC_2.2.5 (mc_replace_strmem.c:876)
==6245==    by 0x41F9E6: freeMemoryIfNeeded (redis.c:3010)
==6245==    by 0x41D2CC: processCommand (redis.c:2069)

memmove() size argument was accounting for an extra element, going
outside the bounds of the array.
 2014-03-25 10:32:50 +01:00
antirez 74d86dc13b Test: do not complain when "leaks" can't run because process died. 2014-03-25 09:34:07 +01:00
antirez c69c40a4ae adjustOpenFilesLimit() refactoring. 
In this commit:
* Decrement steps are semantically differentiated from the reserved FDs.
  Previously both values were 32 but the meaning was different.
* Make it clear that we save setrlimit errno.
* Don't explicitly handle wrapping of 'f', but prevent it from
  happening.
* Add comments to make the function flow more readable.

This integrates PR #1630
 2014-03-25 09:07:17 +01:00
Matt Stancliff b2a02cc92d Fix potentially incorrect errno usage 
errno may be reset by the previous call to redisLog, so capture
the original value for proper error reporting.
 2014-03-25 09:07:17 +01:00
Matt Stancliff 01fe750c6a Add REDIS_MIN_RESERVED_FDS define for open fds 
Also update the original REDIS_EVENTLOOP_FDSET_INCR to
include REDIS_MIN_RESERVED_FDS. REDIS_EVENTLOOP_FDSET_INCR
exists to make sure more than (maxclients+RESERVED) entries
are allocated, but we can only guarantee that if we include
the current value of REDIS_MIN_RESERVED_FDS as a minimum
for the INCR size.
 2014-03-25 09:07:17 +01:00
Matt Stancliff 1e7b99809b Fix infinite loop on startup if ulimit too low 
Fun fact: rlim_t is an unsigned long long on all platforms.

Continually subtracting from a rlim_t makes it get smaller
and smaller until it wraps, then you're up to 2^64-1.

This was causing an infinite loop on Redis startup if
your ulimit was extremely (almost comically) low.

The case of (f > oldlimit) would never be met in a case like:

    f = 150
    while (f > 20) f -= 128

Since f is unsigned, it can't go negative and would
take on values of:

    Iteration 1: 150 - 128 => 22
    Iteration 2:  22 - 128 => 18446744073709551510
    Iterations 3-∞: ...

To catch the wraparound, we use the previous value of f
stored in limit.rlimit_cur.  If we subtract from f and
get a larger number than the value it had previously,
we print an error and exit since we don't have enough
file descriptors to help the user at this point.

Thanks to @bs3g for the inspiration to fix this problem.
Patches existed from @bs3g at antirez#1227, but I needed to repair a few other
parts of Redis simultaneously, so I didn't get a chance to use them.
 2014-03-25 09:07:17 +01:00
Matt Stancliff f701a347b8 Improve error handling around setting ulimits 
The log messages about open file limits have always
been slightly opaque and confusing.  Here's an attempt to
fix their wording, detail, and meaning.  Users will have a
better understanding of how to fix very common problems
with these reworded messages.

Also, we handle a new error case when maxclients becomes less
than one, essentially rendering the server unusable.  We
now exit on startup instead of leaving the user with a server
unable to handle any connections.

This fixes antirez#356 as well.
 2014-03-25 09:07:17 +01:00
Matt Stancliff 6f4be45997 Replace magic 32 with REDIS_EVENTLOOP_FDSET_INCR 
32 was the additional number of file descriptors Redis
would reserve when managing a too-low ulimit.  The
number 32 was in too many places statically, so now
we use a macro instead that looks more appropriate.

When Redis sets up the server event loop, it uses:
    server.maxclients+REDIS_EVENTLOOP_FDSET_INCR

So, when reserving file descriptors, it makes sense to
reserve at least REDIS_EVENTLOOP_FDSET_INCR FDs instead
of only 32.  Currently, REDIS_EVENTLOOP_FDSET_INCR is
set to 128 in redis.h.

Also, I replaced the static 128 in the while f < old loop
with REDIS_EVENTLOOP_FDSET_INCR as well, which results
in no change since it was already 128.

Impact: Users now need at least maxclients+128 as
their open file limit instead of maxclients+32 to obtain
actual "maxclients" number of clients.  Redis will carve
the extra REDIS_EVENTLOOP_FDSET_INCR file descriptors it
needs out of the "maxclients" range instead of failing
to start (unless the local ulimit -n is too low to accomidate
the request).
 2014-03-25 09:07:17 +01:00
Matt Stancliff 49b576cb6c Fix maxclients error handling 
Everywhere in the Redis code base, maxclients is treated
as an int with (int)maxclients or `maxclients = atoi(source)`,
so let's make maxclients an int.

This fixes a bug where someone could specify a negative maxclients
on startup and it would work (as well as set maxclients very high)
because:

    unsigned int maxclients;
    char *update = "-300";
    maxclients = atoi(update);
    if (maxclients < 1) goto fail;

But, (maxclients < 1) can only catch the case when maxclients
is exactly 0.  maxclients happily sets itself to -300, which isn't
-300, but rather 4294966996, which isn't < 1, so... everything
"worked."

maxclients config parsing checks for the case of < 1, but maxclients
CONFIG SET parsing was checking for case of < 0 (allowing
maxclients to be set to 0).  CONFIG SET parsing is now updated to
match config parsing of < 1.

It's tempting to add a MINIMUM_CLIENTS define, but... I didn't.

These changes were inspired by antirez#356, but this doesn't
fix that issue.
 2014-03-25 09:07:17 +01:00
antirez 317ec18214 Fixed undefined variable value with certain code paths. 
In sentinelFlushConfig() fd could be undefined when the following if
statement was true:

        if (rewrite_status == -1) goto werr;

This could cause random file descriptors to get closed.
 2014-03-24 21:10:00 +01:00
Matt Stancliff 19ecf7cdba Use LRU_CLOCK() instead of function getLRUClock() 
lookupKey() uses LRU_CLOCK(), so it seems object
creation should use LRU_CLOCK() too.
 2014-03-24 21:10:00 +01:00
Matt Stancliff 73c2bcca85 Sentinel: Notify user when config can't be saved 2014-03-24 21:10:00 +01:00
Matt Stancliff 65e0452886 Fix data loss when save AOF/RDB with no free space 
Previously, the (!fp) would only catch lack of free space
under OS X.  Linux waits to discover it can't write until
it actually writes contents to disk.

(fwrite() returns success even if the underlying file
has no free space to write into.  All the errors
only show up at flush/sync/close time.)

Fixes antirez/redis#1604
 2014-03-24 21:10:00 +01:00
Matt Stancliff 4af72d326e Cluster: Restore proper trib master iteration 
This got removed in 2e5c394 during a new feature addition.

The prior commit had "break if masters.length == masters_count"
but we are guaranteed to aready have that condition met since
otherwise we would haven't gotten this far.

Without this break statement, it's possible some masters may
be forgotten and have zero replicas while other masters have
more than their requested number of replicas.

Thanks to carlos for pointing out this regression at:
https://groups.google.com/forum/#!topic/redis-db/_WVVqDw5B7c
 2014-03-24 21:10:00 +01:00
Matt Stancliff ee18135a19 Cluster: Fix trib create when masters==replicas 
This bug was introduced in 2e5c394f during a refactor.

It took me a while to understand what was going on with
the code, so I've refactored it further by:
  - Replacing boolean values with meaningful symbols
  - Replacing 'i' with a meaningful variable name
  - Adding the proper abort check
  - Factoring out now duplicated conditionals
  - Adding optional verbose logging (we're inside *four*
    different looping constructs, so it takes a while to
    figure out where all the moving parts are)
  - Updating comment for the section

This fixes a problem when the number of master instances
equaled the number of replica instances.  Before, when
there were equal numbers of both, nodes_count would go to
zero, but the while loop would spin in i < @replicas because
i would never be updated (because the nodes_list of each ip
was length == 0, which triggered an endless loop of
next -> i = 0 -> 0 < 1? -> true -> next -> i = 0 ...)

Thanks to carlo who found this problem at:
https://groups.google.com/forum/#!topic/redis-db/_WVVqDw5B7c
 2014-03-24 21:10:00 +01:00
Jan-Erik Rediger 9fa96697a9 Fixed a few typos. 2014-03-24 21:10:00 +01:00
Matt Stancliff a7478079a1 Cluster: remove variable causing warning 
GCC-4.9 warned about this, but clang didn't.

This commit fixes warning:
sentinel.c: In function 'sentinelReceiveHelloMessages':
sentinel.c:2156:43: warning: variable 'master' set but not used [-Wunused-but-set-variable]
     sentinelRedisInstance *ri = c->data, *master;
 2014-03-24 21:10:00 +01:00
Jan-Erik Rediger 46904ae22b Finally fix the install_server.sh script. 
Includes changes from a dozen bug reports and pull requests.
Was tested on Ubuntu, Debian and CentOS.
 2014-03-24 18:12:55 +01:00
antirez 2dd8c46249 Sample and cache RSS in serverCron(). 
Obtaining the RSS (Resident Set Size) info is slow in Linux and OSX.
This slowed down the generation of the INFO 'memory' section.

Since the RSS does not require to be a real-time measurement, we
now sample it with server.hz frequency (10 times per second by default)
and use this value both to show the INFO rss field and to compute the
fragmentation ratio.

Practically this does not make any difference for memory profiling of
Redis but speeds up the INFO call significantly.
 2014-03-24 12:03:41 +01:00
antirez ae211be98b sdscatvprintf(): Try to use a static buffer. 
For small content the function now tries to use a static buffer to avoid
a malloc/free cycle that is too costly when the function is used in the
context of performance critical code path such as INFO output generation.

This change was verified to have positive effects in the execution speed
of the INFO command.
 2014-03-24 10:23:55 +01:00
antirez 571d6b01de Cache uname() output across INFO calls. 
Uname was profiled to be a slow syscall. It produces always the same
output in the context of a single execution of Redis, so calling it at
every INFO output generation does not make too much sense.

The uname utsname structure was modified as a static variable. At the
same time a static integer was added to check if we need to call uname
the first time.
 2014-03-24 10:02:55 +01:00
antirez 06b2a389eb sdscatvprintf(): guess buflen using format length. 
sdscatvprintf() uses a loop where it tries to output the formatted
string in a buffer of the initial length, if there was not enough room,
a buffer of doubled size is tried and so forth.

The initial guess for the buffer length was very poor, an hardcoded
"16". This caused the printf to be processed multiple times without a
good reason. Given that printf functions are already not fast, the
overhead was significant.

The new heuristic is to use a buffer 4 times the length of the format
buffer, and 32 as minimal size. This appears to be a good balance for
typical uses of the function inside the Redis code base.

This change improved INFO command performances 3 times.
 2014-03-24 09:44:49 +01:00
antirez fd70c68c97 Add test-lru.rb to utils. 
This is a program useful to evaluate the Redis LRU algorithm behavior.
 2014-03-21 09:57:02 +01:00
antirez 24c350245d Use getLRUClock() instead of server.lruclock to create objects. 
Thanks to Matt Stancliff for noticing this error. It was in the original
code but somehow I managed to remove the change from the commit...
 2014-03-21 09:57:02 +01:00
antirez 7f274e3bfe The default maxmemory policy is now noeviction. 
This is safer as by default maxmemory should just set a memory limit
without any key to be deleted, unless the policy is set to something
more relaxed.
 2014-03-21 09:57:02 +01:00
antirez a219339ee0 Use 24 bits for the lru object field and improve resolution. 
There were 2 spare bits inside the Redis object structure that are now
used in order to enlarge 4x the range of the LRU field.

At the same time the resolution was improved from 10 to 1 second: this
still provides 194 days before the LRU counter overflows (restarting from
zero).

This is not a problem since it only causes lack of eviction precision for
objects not touched for a very long time, and the lack of precision is
only temporary.
 2014-03-21 09:57:02 +01:00
antirez 9928b53102 Default LRU samples is now 5. 2014-03-21 09:57:02 +01:00
antirez fd5e8c0132 Use new dictGetRandomKeys() API to get samples for eviction. 
The eviction quality degradates a bit in my tests, but since the API is
faster, it allows to raise the number of samples, and overall is a win.
 2014-03-21 09:57:02 +01:00
antirez 10c8d86242 struct dictEntry -> dictEntry. 2014-03-21 09:57:02 +01:00
antirez 26292670fa Added dictGetRandomKeys() to dict.c: mass get random entries. 
This new function is useful to get a number of random entries from an
hash table when we just need to do some sampling without particularly
good distribution.

It just jumps at a random place of the hash table and returns the first
N items encountered by scanning linearly.

The main usefulness of this function is to speedup Redis internal
sampling of the key space, for example for key eviction or expiry.
 2014-03-21 09:57:02 +01:00
antirez c641074a21 LRU eviction pool implementation. 
This is an improvement over the previous eviction algorithm where we use
an eviction pool that is persistent across evictions of keys, and gets
populated with the best candidates for evictions found so far.

It allows to approximate LRU eviction at a given number of samples
better than the previous algorithm used.
 2014-03-21 09:57:02 +01:00
antirez c9ac817cc0 Fix OBJECT IDLETIME return value converting to seconds. 
estimateObjectIdleTime() returns a value in milliseconds now, so we need
to scale the output of OBJECT IDLETIME to seconds.
 2014-03-21 09:57:02 +01:00
antirez 205c2ccc0c Obtain LRU clock in a resolution dependent way. 
For testing purposes it is handy to have a very high resolution of the
LRU clock, so that it is possible to experiment with scripts running in
just a few seconds how the eviction algorithms works.

This commit allows Redis to use the cached LRU clock, or a value
computed on demand, depending on the resolution. So normally we have the
good performance of a precomputed value, and a clock that wraps in many
days using the normal resolution, but if needed, changing a define will
switch behavior to an high resolution LRU clock.
 2014-03-21 09:57:02 +01:00
antirez 561e793472 Specify lruclock in redisServer structure via REDIS_LRU_BITS. 
The padding field was totally useless: removed.
 2014-03-21 09:57:02 +01:00
antirez 8f0b74910a Specify LRU resolution in milliseconds. 2014-03-21 09:57:02 +01:00
antirez 63aacbe86c Set LRU parameters via REDIS_LRU_BITS define. 2014-03-21 09:57:02 +01:00
antirez 8b6a674a5d Unify stats reset for CONFIG RESETSTAT / initServer(). 
Now CONFIG RESETSTAT makes sure to reset all the fields, and in the
future it will be simpler to avoid missing new fields.
 2014-03-21 09:57:02 +01:00
antirez 128dcee43d Sentinel: sentinelRefreshInstanceInfo() minor refactoring. 
Test sentinel.tilt condition on top and return if it is true.
This allows to remove the check for the tilt condition in the remaining
code paths of the function.
 2014-03-21 09:57:02 +01:00
antirez b104015fb3 Sentinel test: 02 unit better coverage + refactoring. 2014-03-21 09:57:02 +01:00
antirez f5f281f906 Sentinel test: foreach_instance_id implements 'break'. 2014-03-21 09:57:02 +01:00
antirez f308f67703 Sentinel: instance_is_killed proc added to sentinel.tcl. 2014-03-21 09:57:02 +01:00
antirez 0774d4928a Sentinel: propagate down-after-ms changes to slaves and sentinels. 2014-03-21 09:57:02 +01:00
antirez a86e24de93 Sentinel: down-after-milliseconds is not master-specific. 
addReplySentinelRedisInstance() modified so that this field is displayed
for all the kind of instances: Sentinels, Masters, Slaves.
 2014-03-21 09:57:02 +01:00
antirez 9997b51ff8 Sentinel failure detection implementation improved. 
Failure detection in Sentinel is ping-pong based. It used to work by
remembering the last time a valid PONG reply was received, and checking
if the reception time was too old compared to the current current time.

PINGs were sent at a fixed interval of 1 second.

This works in a decent way, but does not scale well when we want to set
very small values of "down-after-milliseconds" (this is the node
timeout basically).

This commit reiplements the failure detection making a number of
changes. Some changes are inspired to Redis Cluster failure detection
code:

* A new last_ping_time field is added in representation of instances.
  If non zero, we have an active ping that was sent at the specified
  time. When a valid reply to ping is received, the field is zeroed
  again.
* last_ping_time is not reset when we reconnect the link or send a new
  ping, so from our point of view it represents the time we started
  waiting for the instance to reply to our pings without receiving a
  reply.
* last_ping_time is now used in order to check if the instance is
  timed out. This means that we can have a node timeout of 100
  milliseconds and yet the system will work well since the new check is
  not bound to the period used to send pings.
* Pings are now sent every second, or often if the value of
  down-after-milliseconds is less than one second. With a lower limit of
  10 HZ ping frequency.
* Link reconnection code was improved. This is used in order to try to
  reconnect the link when we are at 50% of the node timeout without a
  valid reply received yet. However the old code triggered unnecessary
  reconnections when the node timeout was very small. Now that should be
  ok.

The new code passes the tests but more testing is needed and more unit
tests stressing the failure detector, so currently this is merged only
in the unstable branch.
 2014-03-21 09:57:02 +01:00
antirez 04cd7f773d Sentinel: use CLIENT SETNAME when connecting to Redis. 
This makes debugging / monitoring of Sentinels simpler since you can
identify sentinels in CLIENT LIST output of Redis instances.
 2014-03-21 09:57:02 +01:00
antirez 32f188cdf7 redis-trib: call MIGRATE via r.client.call as fix for redis-rb API changes. 
See issue #1593.

Thanks to @badboy for suggesting the direct client.call fix.
 2014-03-15 11:02:59 +01:00
Matt Stancliff 0c8f0079f9 Fix segfault from accessing array out of bounds 
argc == 2; argv[2] == crash
 2014-03-14 22:57:29 +01:00
antirez 668a95a583 Sentinel: be safe under crash-recovery assumptions. 
Sentinel's main safety argument is that there are no two configurations
for the same master with the same version (configuration epoch).

For this to be true Sentinels require to be authorized by a majority.
Additionally Sentinels require to do two important things:

* Never vote again for the same epoch.
* Never exchange an old vote for a fresh one.

The first prerequisite, in a crash-recovery system model, requires to
persist the master->leader_epoch on durable storage before to reply to
messages. This was not the case.

We also make sure to persist the current epoch in order to never reply
to stale votes requests from other Sentinels, after a recovery.

The configuration is persisted by making use of fsync(), this is
considered in the context of this code a good enough guarantee that
after a restart our durable state is restored, however this may not
always be the case depending on the kind of hardware and operating
system used.
 2014-03-14 22:57:25 +01:00
antirez ca2a3b3240 Sentinel: fake PUBLISH command to receive HELLO messages. 
Now the way HELLO messages are received is unified.
Now it is no longer needed for Sentinels to converge to the higher
configuration for a master to be able to chat via some Redis instance,
the are able to directly exchanges configurations.

Note that this commit does not include the (trivial) change needed to
send HELLO messages to Sentinel instances as well, since for an error I
committed the change in the previous commit that refactored hello
messages processing into a separated function.
 2014-03-14 11:09:32 +01:00
antirez ee513ee958 Sentinel: HELLO processing refactored into sentinelProcessHelloMessage(). 2014-03-14 11:09:32 +01:00
antirez 9e82461af0 Linenoise updated, multiline mode enabled in redis-cli. 2014-03-13 15:11:41 +01:00
antirez e0ef510630 Redis 2.9.51 (Redis 3.0.0 beta-2). 2014-03-11 16:13:39 +01:00
antirez e917de12d0 Cluster: flag the transaction as dirty for the new redirections. 2014-03-11 15:19:00 +01:00
antirez 07036ad45e redis-trib: new subcommand 'call'. Exec command in all nodes. 
Example:

./redis-trib.rb call 192.168.1.11:7000 config get cluster-node-timeout
 2014-03-11 15:02:41 +01:00
antirez a02424e29b redis-trib: create subcommand is now able to assign spare slaves. 
Example: if the user will try to configure a cluster with 9 nodes,
asking for 1 slave for master, redis-trib will configure a 4 masters
cluster with 1 slave each as usually, but this time will assign the
spare node as a slave of one of the masters.
 2014-03-11 15:02:41 +01:00
antirez 412a759a83 Cluster: update node configEpoch on UPDATE messages. 
The UPDATE message contains the configEpoch of the node configuration
advertised in the packet. Update it if needed.
 2014-03-11 15:02:41 +01:00
antirez 48702e0011 Cluster: set slot error if we receive an update for a busy slot. 
By manually modifying nodes configurations in random ways, it is possible
to create the following scenario:

A is serving keys for slot 10
B is manually configured to serve keys for slot 10

A receives an update from B (or another node) where it is informed that
the slot 10 is now claimed by B with a greater configuration epoch,
however A still has keys from slot 10.

With this commit A will put the slot in error setting it in IMPORTING
state, so that redis-trib can detect the issue.
 2014-03-11 15:02:41 +01:00
antirez efd0346ea3 Cluster: clarified a comment in clusterUpdateSlotsConfigWith(). 2014-03-11 15:02:41 +01:00
antirez 47e3f1f16c Cluster: flush importing/migrating state when master is turned into slave. 2014-03-11 15:02:41 +01:00
antirez 117557192e Cluster: clusterCloseAllSlots() added. 2014-03-11 15:02:41 +01:00
antirez 01eee56f4f DEBUG ERROR implemented. 
The new "error" subcommand of the DEBUG command can reply with an user
selected error, specified as its sole argument:

    DEBUG ERROR "LOADING please wait..."

The error is generated just prefixing the command argument with a "-"
character, and replacing newlines with spaces (since error replies can't
include newlines).

The goal of the command is to help in Client libraries unit tests by
making simple to simulate a command call triggering a given error.
 2014-03-11 11:10:33 +01:00
antirez ab8e1bbcdc DEBUG CMDKEYS: provide some guarantee to getKeysFromCommand(). 
getKeysFromCommand() is designed to be called with the command arguments
passing the basic arity checks described in the command table.

DEBUG CMDKEYS must provide the same guarantees for calling
getKeysFromCommand() to be safe.
 2014-03-11 11:10:33 +01:00
antirez 94129415bf Cluster: make sortGetKeys() able to handle multiple STORE options. 
It does not make sense to pass multiple store options, so, better to
handle it ;-)
 2014-03-11 11:10:33 +01:00
antirez db0d9f4326 DEBUG CMDKEYS added for getKeysFromCommand() testing. 
Examples:

    redis 127.0.0.1:6379> debug cmdkeys set foo bar
    1) "foo"
    redis 127.0.0.1:6379> debug cmdkeys mget a b c
    1) "a"
    2) "b"
    3) "c"
    redis 127.0.0.1:6379> debug cmdkeys zunionstore foo 2 a b
    1) "a"
    2) "b"
    3) "foo"
    redis 127.0.0.1:6379> debug cmdkeys ping
    (empty list or set)
 2014-03-11 11:10:33 +01:00
antirez 3b80e0a41d Cluster: don't allow BY option of SORT as well. 
There is the exception of a "constant" BY pattern that is used in order
to signal to don't sort at all. In this case no lookup is needed so it
is possible to support this case in Cluster mode.
 2014-03-11 11:10:33 +01:00
antirez 399fca8f45 Cluster: SORT get keys helper implemented. 2014-03-11 11:10:33 +01:00
antirez 0c38b3c934 Cluster: evalGetKeys() fixed: was not setting keys count. 2014-03-11 11:10:33 +01:00
antirez d7451a0110 Cluster: don't allow GET option in cluster mode. 
The commit also refactors a bit the error handling during SORT option
parsing.
 2014-03-11 11:10:33 +01:00
antirez c3aff0c20a Fixed memory leak in SORT LIMIT option argument parsing on error. 2014-03-11 11:10:33 +01:00
antirez fdf737e132 Cluster: getKeysFromCommand() top comment improved. 2014-03-11 11:10:33 +01:00
antirez 81efa0d296 Cluster: evalGetKey() added for EVAL/EVALSHA. 
Previously we used zunionInterGetKeys(), however after this function was
fixed to account for the destination key (not needed when the API was
designed for "diskstore") the two set of commands can no longer be served
by an unique keys-extraction function.
 2014-03-11 11:10:09 +01:00
antirez 0618d26b89 Cluster: getKeysFromCommand() and related: top-comments added. 2014-03-11 11:10:09 +01:00
antirez a2a72b87e0 Cluster: getKeysFromCommand() API cleaned up. 
This API originated from the "diskstore" experiment, not for Redis
Cluster itself, so there were legacy/useless things trying to
differentiate between keys that are going to be overwritten and keys
that need to be fetched from disk (preloaded).

All useless with Cluster, so removed with the result of code
simplification.
 2014-03-11 11:10:09 +01:00
antirez c8485703f5 Cluster: some zunionInterGetKeys() comment trimmed. 
Everything was pretty clear again from the initial statements.
 2014-03-11 11:10:09 +01:00
antirez d610d2343d Cluster: abort on port too high error. 
It also fixes multi-line comment style to be consistent with the rest of
the code base.

Related to #1555.
 2014-03-11 11:10:09 +01:00
antirez 2a951ce502 Cluster: be explicit about passing NULL as bind addr for connect. 
The code was already correct but it was using that bindaddr[0] is set to
NULL as a side effect of current implementation if no bind address is
configured. This is not guarnteed to hold true in the future.
 2014-03-11 11:10:09 +01:00
antirez 9c9914d779 Cluster: log error when anetTcpNonBlockBindConnect() fails. 2014-03-11 11:10:09 +01:00
antirez 3119f4f694 Cluster: better timeout and retry time for failover. 
When node-timeout is too small, in the order of a few milliseconds,
there is no way the voting process can terminate during that time, so we
set a lower limit for the failover timeout of two seconds.

The retry time is set to two times the failover timeout time, so it is
at least 4 seconds.
 2014-03-11 11:10:09 +01:00
Matt Stancliff 14f77b343a Fix key extraction for z{union,inter}store 
The previous implementation wasn't taking into account
the storage key in position 1 being a requirement (it
was only counting the source keys in positions 3 to N).

Fixes antirez/redis#1581
 2014-03-11 11:10:09 +01:00
antirez 63fc5dc8b1 Typo in sentinel.conf, exists -> exits. 2014-03-11 11:10:09 +01:00
antirez afe28cfd75 Cluster: fix conditional generating TRYAGAIN error. 2014-03-11 11:10:09 +01:00
antirez aa5898f53e Redis Cluster: support for multi-key operations. 2014-03-11 11:10:09 +01:00
Matt Stancliff c0915ad1a0 Reset op_sec_last_sample_ops when reset requested 
This value needs to be set to zero (in addition to
stat_numcommands) or else people may see
a negative operations per second count after they
run CONFIG RESETSTAT.

Fixes antirez/redis#1577
 2014-03-11 11:10:09 +01:00
Matt Stancliff 4b3c87a027 Remove redundant IP length definition 
REDIS_CLUSTER_IPLEN had the same value as
REDIS_IP_STR_LEN.  They were both #define'd
to the same INET6_ADDRSTRLEN.
 2014-03-11 11:10:09 +01:00
Matt Stancliff 7c8964a8cf Remove some redundant code 
Function nodeIp2String in cluster.c is exactly
anetPeerToString with a pre-extracted fd.
 2014-03-11 11:09:37 +01:00
Matt Stancliff 7c359449d5 Fix return value check for anetTcpAccept 
anetTcpAccept returns ANET_ERR, not AE_ERR.

This isn't a physical error since both ANET_ERR
and AE_ERR are -1, but better to be consistent.
 2014-03-11 11:09:37 +01:00
Jan-Erik Rediger 6766fc561e Small typo fixed 2014-03-11 11:09:37 +01:00
Matt Stancliff 9a7cf31960 Bind source address for cluster communication 
The first address specified as a bind parameter
(server.bindaddr[0]) gets used as the source IP
for cluster communication.

If no bind address is specified by the user, the
behavior is unchanged.

This patch allows multiple Redis Cluster instances
to communicate when running on the same interface
of the same host.
 2014-03-11 11:09:37 +01:00
zhanghailei 503938022f refer to updateLRUClock's comment REDIS_LRU_CLOCK_MAX is 22 bits,but #define REDIS_LRU_CLOCK_MAX ((1<<21)-1) only 21 bits 2014-03-11 11:09:37 +01:00
zhanghailei 7eec424953 FIXED a typo more thank should be more than 2014-03-11 11:09:37 +01:00
zhanghailei 0abe98cb4d According to context,the size should be 16 rather than 64 2014-03-11 11:09:37 +01:00
Matt Stancliff a0ea8f235e Cluster: error out quicker if port is unusable 
The default cluster control port is 10,000 ports higher than
the base Redis port.  If Redis is started on a too-high port,
Cluster can't start and everything will exit later anyway.
 2014-03-11 11:09:37 +01:00
Matt Stancliff 6f4b5ef6d5 Fix "can't bind to address" error reporting. 
Report the actual port used for the listening attempt instead of
server.port.

Originally, Redis would just listen on server.port.
But, with clustering, Redis uses a Cluster Port too,
so we can't say server.port is always where we are listening.

If you tried to launch Redis with a too-high port number (any
port where Port+10000 > 65535), Redis would refuse to start, but
only print an error saying it can't connect to the Redis port.

This patch fixes much confusions.
 2014-03-11 11:09:37 +01:00
antirez 4d5ba5962c Cast saveparams[].seconds to long for %ld format specifier. 2014-03-05 11:26:46 +01:00
antirez ef4e465316 Sentinel test: set less time sensitive defaults. 
This commit sets the failover timeout to 30 seconds instead of the 180
seconds default, and allows to reconfigure multiple slaves at the same
time.

This makes tests less sensible to timing, with the result that there are
less false positives due to normal behaviors that require time to
succeed or to be retried.

However the long term solution is probably some way in order to detect
when a test failed because of timing issues (for example split brain
during leader election) and retry it.
 2014-03-05 10:22:07 +01:00
antirez 313f8831ed Sentinel: more aggressive failover start desynchronization. 
Sentinel needs to avoid split brain conditions due to multiple sentinels
trying to get voted at the exact same time.

So far some desynchronization was provided by fluctuating server.hz,
that is the frequency of the timer function call. However the
desynchonization provided in this way was not enough when using many
Sentinel instances, especially when a large quorum value is used in
order to force a greater degree of agreement (more than N/2+1).

It was verified that it was likely to trigger a split brain
condition, forcing the system to try again after a timeout.
Usually the system will succeed after a few retries, but this is not
optimal.

This commit desynchronizes instances in a more effective way to make it
likely that the first attempt will be successful.
 2014-03-05 10:22:07 +01:00
antirez 5ee2394474 CONFIG REWRITE should be logged at WARNING level. 2014-03-05 10:22:07 +01:00
antirez fa8aca5236 Sentinel test: debugging console improved. 2014-03-05 10:22:07 +01:00
antirez cf746c398c Sentinel test: initial debugging console. 2014-03-05 10:22:07 +01:00
antirez cbf51b60a2 Sentinel test: be more patient in create_redis_master_slave_cluster. 2014-03-05 10:22:07 +01:00
antirez 23a12bb8c1 Sentiel test: add test start time in output. 2014-03-05 10:22:07 +01:00
antirez d2e16801f0 Cluster: invalidate current transaction on redirections. 2014-03-05 10:22:07 +01:00
antirez 21557a0d8b Sentinel test: use 1000 as retry in initial 00 unit test. 2014-03-05 10:22:07 +01:00
antirez 290a5d7ff6 Sentinel test: initial tests in 03 unit. 2014-03-05 10:22:07 +01:00
antirez b2668466f4 Sentinel test: foreach_instance_id now supports 'continue'. 2014-03-05 10:22:07 +01:00
antirez a76bca16e6 Sentienl test: fixed typo in unit 03 top comment. 2014-03-05 10:22:07 +01:00
antirez 1d9eb47f9d Document why we update peak memory in INFO. 2014-03-05 10:22:07 +01:00
antirez e4833ed8bf Fix configEpoch assignment when a cluster slot gets "closed". 
This is still code to rework in order to use agreement to obtain a new
configEpoch when a slot is migrated, however this commit handles the
special case that happens when the nodes are just started and everybody
has a configEpoch of 0. In this special condition to have the maximum
configEpoch is not enough as the special epoch 0 is not unique (all the
others are).

This does not fixes the intrinsic race condition of a failover happening
while we are resharding, that will be addressed later.
 2014-03-05 10:22:07 +01:00
Matt Stancliff 7c092b679f Force INFO used_memory_peak to match peak memory 
used_memory_peak only updates in serverCron every server.hz,
but Redis can use more memory and a user can request memory
INFO before used_memory_peak gets updated in the next
cron run.

This patch updates used_memory_peak to the current
memory usage if the current memory usage is higher
than the recorded used_memory_peak value.

(And it only calls zmalloc_used_memory() once instead of
twice as it was doing before.)
 2014-03-05 10:22:07 +01:00
michael-grunder 23addbb5a3 Improved bigkeys with progress, pipelining and summary 
This commit reworks the redis-cli --bigkeys command to provide more
information about our progress as well as output summary information
when we're done.

 - We now show an approximate percentage completion as we go
 - Hiredis pipelining is used for TYPE and SIZE retreival
 - A summary of keyspace distribution and overall breakout at the end
 2014-03-05 10:22:07 +01:00
antirez a46811693d Sentinel test: Makefile target added. 2014-02-28 16:00:14 +01:00
antirez 32db37c3c4 BITPOS fuzzy testing. 2014-02-27 15:52:43 +01:00
antirez 1d62f83364 Basic BITPOS tests. 2014-02-27 15:52:43 +01:00
antirez 9104f1e672 warnigns -> warnings in redisBitpos(). 2014-02-27 15:52:43 +01:00
antirez eacc0951a2 More consistent BITPOS behavior with bit=0 and ranges. 
With the new behavior it is possible to specify just the start in the
range (the end will be assumed to be the first byte), or it is possible
to specify both start and end.

This is useful to change the behavior of the command when looking for
zeros inside a string.

1) If the user specifies both start and end, and no 0 is found inside
   the range, the command returns -1.

2) If instead no range is specified, or just the start is given, even
   if in the actual string no 0 bit is found, the command returns the
   first bit on the right after the end of the string.

So for example if the string stored at key foo is "\xff\xff":

    BITPOS foo (returns 16)
    BITPOS foo 0 -1 (returns -1)
    BITPOS foo 0 (returns 16)

The idea is that when no end is given the user is just looking for the
first bit that is zero and can be set to 1 with SETBIT, as it is
"available". Instead when a specific range is given, we just look for a
zero within the boundaries of the range.
 2014-02-27 15:52:43 +01:00
antirez 1f8005ca09 Initial implementation of BITPOS. 
It appears to work but more stress testing, and both unit tests and
fuzzy testing, is needed in order to ensure the implementation is sane.
 2014-02-27 15:52:43 +01:00
antirez cbeaf17374 Added two more BITCOUNT tests stressing misaligned access. 2014-02-27 15:52:43 +01:00
antirez 9c206cc8e8 BITCOUNT fuzzy test with random start/end added. 
It was verified in practice that this test is able to stress much more
the implementation by introducing errors that were only trivially to
detect with different offsets but impossible to detect starting always
at zero and counting bits the full length of the string.
 2014-02-27 15:52:43 +01:00
antirez 24265edb6c Fix misaligned word access in redisPopcount(). 2014-02-27 15:52:43 +01:00
Matt Stancliff 7e274194bf Fix IP representation in clusterMsgDataGossip 2014-02-27 15:52:43 +01:00
antirez d242b698f4 Sentinel test: add stub for unit 04. 2014-02-25 15:55:21 +01:00
antirez cd8b149dff Sentinel test: added TODO items in 02 unit. 2014-02-25 15:55:21 +01:00
antirez c12ebb4633 Sentinel test: check role at end of unit 01. 2014-02-25 15:09:46 +01:00
michael-grunder a2b3f2eae2 Update --bigkeys to use SCAN 
This commit changes the findBigKeys() function in redis-cli.c to use the new
SCAN command for iterating the keyspace, rather than RANDOMKEY.  Because we
can know when we're done using SCAN, it will exit after exhausting the keyspace.
 2014-02-25 15:09:46 +01:00
antirez cb7ccc9ad0 Sentinel test: kill masters instead of using DEBUG SLEEP in all tests. 2014-02-25 15:09:46 +01:00
antirez 48fa34bf9e redis-cli: also remove useless uint8_t. 2014-02-25 15:09:46 +01:00
antirez ff20d05650 redis-cli: don't use uint64_t where actually not needed. 
The computation is just something to take the CPU busy, no need to use a
specific type. Since stdint.h was not included this prevented
compilation on certain systems.
 2014-02-25 15:09:46 +01:00
antirez 6a95ddb248 redis-cli: check argument existence for --pattern. 2014-02-25 15:09:46 +01:00
antirez 2431b63ff4 redis-cli: --intrinsic-latency run mode added. 2014-02-25 15:09:46 +01:00
antirez 68e9597e8a redis-cli: added comments to split program in parts. 2014-02-25 15:09:46 +01:00
antirez 31ed0911fa Sentinel test: restart instances left killed by previous unit. 
An unit can abort in the middle for an error. The next unit should not
assume that the instances are in a clean state, and must restart what
was left killed.
 2014-02-25 10:24:16 +01:00
antirez edf48d17a2 Sentinel test: jump to next unit on test failure. 
Sentinel tests are designed to be dependent on the previous tests in the
same unit, so usually we can't continue with the next test in the same
unit if a previous test failed.
 2014-02-25 10:24:16 +01:00
antirez 7f521d764f Sentinel test: test majority crashing Sentinels. 
The test was previously performed by removing the master from the
Sentinel monitored masters. The test with the Sentinels crashed is
more similar to real-world partitions / failures.
 2014-02-25 10:24:16 +01:00
antirez 491bcbaf73 Sentinel test: restart_instance should refresh pid attrib. 
Also kill_instance was modified to warn when a test will try to kill the
same instance multiple times for error.
 2014-02-25 10:24:16 +01:00
antirez 55643361f1 Sentinel test: more stuff mored from 00-base to init. 
The area a number of mandatory tests to craete a stable setup for
testing that is not too sensitive to timing issues. All those tests
moved to includes/init-tests, and marked as (init).
 2014-02-25 10:24:16 +01:00
antirez 4f6ed3412e Sentinel: log quorum with +monitor event. 2014-02-25 10:24:16 +01:00
antirez 55a2e10b03 Sentinel test: removed useless code to set SDOWN timeout. 
The new common initialization code used to start a new unit already set
the timeout to 2000 milliseconds.
 2014-02-25 10:24:16 +01:00
antirez 56e18ba4f6 Sentinel: generate +monitor events at startup. 2014-02-25 10:24:16 +01:00
antirez cd68a1d45a Sentinel: log +monitor and +set events. 
Now that we have a runtime configuration system, it is very important to
be able to log how the Sentinel configuration changes over time because
of API calls.
 2014-02-25 10:24:16 +01:00
antirez 4af2acf2b0 Sentinel: added missing exit(1) after checking for config file. 2014-02-25 10:24:16 +01:00
antirez fecfd721df Sentinel test: tmp dir and gitignore added. 2014-02-25 10:24:16 +01:00
antirez b18109e351 Sentinel test: minor fixes to --pause-on-error. 2014-02-25 10:24:16 +01:00
antirez 2ed707bcaa Sentinel test: --pause-on-error option added. 
Pause the test with running instances available for state inspection on
error.
 2014-02-25 10:24:16 +01:00
antirez 2666f737b0 Sentinel test: added empty units to fill later. 2014-02-25 10:24:16 +01:00
antirez 6e2e6d5b8c Sentinel: IDONTKNOW error removed. 
This error was conceived for the older version of Sentinel that worked
via master redirection and that was not able to get configuration
updates from other Sentinels via the Pub/Sub channel of masters or
slaves.

This reply does not make sense today, every Sentinel should reply with
the best information it has currently. The error will make even more
sense in the future since the plan is to allow Sentinels to update the
configuration of other Sentinels via gossip with a direct chat without
the prerequisite that they have at least a monitored instance in common.
 2014-02-25 10:24:16 +01:00
antirez ec593b582e Sentinel test: framework improved and conf-update unit added. 
It is now possible to kill and restart sentinel or redis instances for
more real-world testing.

The 01 unit tests the capability of Sentinel to update the configuration
of Sentinels rejoining the cluster, however the test is pretty trivial
and more tests should be added.
 2014-02-25 10:24:16 +01:00
Matt Stancliff 5a8c9f94a6 Add cluster or sentinel to proc title 
If you launch redis with `redis-server --sentinel` then
in a ps, your output only says "redis-server IP:Port" — this
patch changes the proc title to include [sentinel] or
[cluster] depending on the current server mode:
e.g.  "redis-server IP:Port [sentinel]"
      "redis-server IP:Port [cluster]"
 2014-02-25 10:24:16 +01:00
antirez ef01bbe1f5 Sentinel test: move init tests as includes. 
Most units will start with these two basic tests to create an
environment where the real tests are ran.
 2014-02-25 10:24:16 +01:00
antirez 2b4bc4f900 Sentinel test: ability to run just a subset of test files. 2014-02-25 10:24:16 +01:00
Matt Stancliff 21a7f9e7ef Auto-enter slaveMode when SYNC from redis-cli 
If someone asks for SYNC or PSYNC from redis-cli,
automatically enter slaveMode (as if they ran
redis-cli --slave) and continue printing the replication
stream until either they Ctrl-C or the master gets disconnected.
 2014-02-25 10:24:16 +01:00
antirez 48d74f2039 Sentinel: report instances role switch events. 
This is useful mostly for debugging of issues.
 2014-02-20 12:28:18 +01:00
antirez 8a6774e209 Sentinel test: some reliability fixes to 00-base tests. 2014-02-20 12:28:18 +01:00
antirez 6bb42b5bfe Sentinel test: check that role matches at end of 00-base. 2014-02-20 12:28:18 +01:00
antirez 37be5987e6 Sentinel test: ODOWN and agreement. 2014-02-20 12:28:18 +01:00
antirez b037c897ae Sentinel test: check reconfig of slaves and old master. 2014-02-20 12:28:18 +01:00
antirez f7cb83778b Sentinel test: basic failover tested. Framework improvements. 2014-02-20 12:28:18 +01:00
antirez a11749f9f6 Sentinel test: basic tests for MONITOR and auto-discovery. 2014-02-20 12:28:18 +01:00
antirez 8292dd50cd Sentinel test: info fields, master-slave setup, fixes. 2014-02-20 12:28:18 +01:00
antirez 68a3d5602e Prefix test file names with numbers to force exec order. 2014-02-20 12:28:18 +01:00
antirez bbbe68aa0d Sentinel test: provide basic commands to access instances. 2014-02-20 12:28:18 +01:00
antirez 6e4662e479 Sentinel: SENTINEL_SLAVE_RECONF_RETRY_PERIOD -> RECONF_TIMEOUT 
Rename define to match the new meaning.
 2014-02-18 10:30:47 +01:00
antirez bd31fcf16e Sentinel: fix slave promotion timeout. 
If we can't reconfigure a slave in time during failover, go forward as
anyway the slave will be fixed by Sentinels in the future, once they
detect it is misconfigured.

Otherwise a failover in progress may never terminate if for some reason
the slave is uncapable to sync with the master while at the same time
it is not disconnected.
 2014-02-18 10:30:47 +01:00
antirez 2a35abcf12 Sentinel: initial testing framework. 
Nothing tested at all so far... Just the infrastructure spawning N
Sentinels and N Redis instances that the test will use again and again.
 2014-02-17 17:39:23 +01:00
antirez c1786693b1 Test: colorstr moved to util.tcl. 2014-02-17 17:39:19 +01:00
antirez 48af4d4f26 Test: code to test server availability refactored. 
Some inline test moved into server_is_up procedure.
Also find_available_port was moved into util since it is going
to be used for the Sentinel test as well.
 2014-02-17 17:39:15 +01:00
antirez 58b6dd9beb Get absoulte config file path before processig 'dir'. 
The code tried to obtain the configuration file absolute path after
processing the configuration file. However if config file was a relative
path and a "dir" statement was processed reading the config, the absolute
path obtained was wrong.

With this fix the absolute path is obtained before processing the
configuration while the server is still in the original directory where
it was executed.
 2014-02-17 17:39:12 +01:00
antirez c36a5dce54 Sentinel: better specify startup errors due to config file. 
Now it logs the file name if it is not accessible. Also there is a
different error for the missing config file case, and for the non
writable file case.
 2014-02-17 17:39:09 +01:00
antirez 3c1672da7d Update cached time in rdbLoad() callback. 
server.unixtime and server.mstime are cached less precise timestamps
that we use every time we don't need an accurate time representation and
a syscall would be too slow for the number of calls we require.

Such an example is the initialization and update process of the last
interaction time with the client, that is used for timeouts.

However rdbLoad() can take some time to load the DB, but at the same
time it did not updated the time during DB loading. This resulted in the
bug described in issue #1535, where in the replication process the slave
loads the DB, creates the redisClient representation of its master, but
the timestamp is so old that the master, under certain conditions, is
sensed as already "timed out".

Thanks to @yoav-steinberg and Redis Labs Inc for the bug report and
analysis.
 2014-02-13 15:13:38 +01:00
antirez 116617c5e7 Log when CONFIG REWRITE goes bad. 2014-02-13 14:33:53 +01:00
antirez 767846dc5b Test: regression for issue #1549. 
It was verified that reverting the commit that fixes the bug, the test
no longer passes.
 2014-02-13 12:27:10 +01:00
antirez 14143fbede Fix script cache bug in the scripting engine. 
This commit fixes a serious Lua scripting replication issue, described
by Github issue #1549. The root cause of the problem is that scripts
were put inside the script cache, assuming that slaves and AOF already
contained it, even if the scripts sometimes produced no changes in the
data set, and were not actaully propagated to AOF/slaves.

Example:

    eval "if tonumber(KEYS[1]) > 0 then redis.call('incr', 'x') end" 1 0

Then:

    evalsha <sha1 step 1 script> 1 0

At this step sha1 of the script is added to the replication script cache
(the script is marked as known to the slaves) and EVALSHA command is
transformed to EVAL. However it is not dirty (there is no changes to db),
so it is not propagated to the slaves. Then the script is called again:

    evalsha <sha1 step 1 script> 1 1

At this step master checks that the script already exists in the
replication script cache and doesn't transform it to EVAL command. It is
dirty and propagated to the slaves, but they fail to evaluate the script
as they don't have it in the script cache.

The fix is trivial and just uses the new API to force the propagation of
the executed command regardless of the dirty state of the data set.

Thank you to @minus-infinity on Github for finding the issue,
understanding the root cause, and fixing it.
 2014-02-13 12:16:31 +01:00
antirez 0296aab6da AOF write error: retry with a frequency of 1 hz. 2014-02-12 16:57:08 +01:00
antirez dd73a7bf43 AOF: don't abort on write errors unless fsync is 'always'. 
A system similar to the RDB write error handling is used, in which when
we can't write to the AOF file, writes are no longer accepted until we
are able to write again.

For fsync == always we still abort on errors since there is currently no
easy way to avoid replying with success to the user otherwise, and this
would violate the contract with the user of only acknowledging data
already secured on disk.
 2014-02-12 16:57:04 +01:00
antirez 910b6d34f2 Redis 2.9.50 (Redis 3.0.0 beta-1) 2014-02-11 10:47:54 +01:00
antirez 0725988a07 Cluster: clusterDelNode(): remove node from master's slaves. 2014-02-11 10:34:14 +01:00
antirez 4513d8fcd4 Cluster: UPDATE messages are the norm and verbose. 
Logging them at WARNING level was of little utility and of sure disturb.
 2014-02-11 10:22:05 +01:00
antirez 9721255178 Cluster: redis-trib fix: handling of another trivial case. 2014-02-11 10:22:02 +01:00
antirez 6d550f2de4 Cluster: configEpoch assignment in SETNODE improved. 
Avoid to trash a configEpoch for every slot migrated if this node has
already the max configEpoch across the cluster.

Still work to do in this area but this avoids both ending with a very
high configEpoch without any reason and to flood the system with fsyncs.
 2014-02-11 10:21:58 +01:00
antirez 585e9fb886 Cluster: clusterSetStartupEpoch() made more generally useful. 
The actual goal of the function was to get the max configEpoch found in
the cluster, so make it general by removing the assignment of the max
epoch to currentEpoch that is useful only at startup.
 2014-02-11 10:21:55 +01:00
antirez 8b5196addf Cluster: always increment the configEpoch in SETNODE after import. 
Removed a stale conditional preventing the configEpoch from incrementing
after the import in certain conditions. Since the master got a new slot
it should always claim a new configuration.
 2014-02-11 10:21:52 +01:00
antirez 2e3f6b0fb3 Cluster: on resharding upgrade version of receiving node. 
The node receiving the hash slot needs to have a version that wins over
the other versions in order to force the ownership of the slot.

However the current code is far from perfect since a failover can happen
during the manual resharding. The fix is a work in progress but the
bottom line is that the new version must either be voted as usually,
set by redis-trib manually after it makes sure can't be used by other
nodes, or reserved configEpochs could be used for manual operations (for
example odd versions could be never used by slaves and are always used
by CLUSTER SETSLOT NODE).
 2014-02-11 00:39:24 +01:00
antirez a221ae5ce2 Cluster: fsync at every SETSLOT command puts too pressure on disks. 
During slots migration redis-trib can send a number of SETSLOT commands.
Fsyncing every time is a bit too much in production as verified
empirically.

To make sure configs are fsynced on all nodes after a resharding
redis-trib may send something like CLUSTER CONFSYNC.

In this case fsyncs were not providing too much value since anyway
processes can crash in the middle of the resharding of an hash slot, and
redis-trib should be able to recover from this condition anyway.
 2014-02-11 00:39:20 +01:00
antirez 77c6fa65f1 Cluster: conditions to clear "migrating" on slot for SETSLOT ... NODE changed. 
If the slot is manually assigned to another node, clear the migrating
status regardless of the fact it was previously assigned to us or not,
as long as we no longer have keys for this slot.

This avoid a race during slots migration that may leave the slot in
migrating status in the source node, since it received an update message
from the destination node that is already claiming the slot.

This way we are sure that redis-trib at the end of the slot migration is
always able to close the slot correctly.
 2014-02-11 00:39:14 +01:00
antirez f406a09495 Cluster: remove debugging xputs from redis-trib. 2014-02-11 00:39:11 +01:00
antirez 01de246843 Cluster: redis-trib fix: cover new case of open slot. 
The case is the trivial one a single node claiming the slot as
migrating, without nodes claiming it as importing.
 2014-02-11 00:39:08 +01:00
antirez 8f25428772 redis-trib: log event after we have reference to 'master'. 2014-02-11 00:39:05 +01:00
antirez cc97305ec3 Cluster: don't update slave's master if we don't know it. 
There is no way we can update the slave's node->slaveof pointer if we
don't know the master (no node with such an ID in our tables).
 2014-02-11 00:39:02 +01:00
antirez fa6f4f21c3 Cluster: ignore slot config changes if we are importing it. 2014-02-11 00:38:59 +01:00
antirez 30214fff3e Cluster: update configEpoch after manually messing with slots. 2014-02-11 00:38:56 +01:00
antirez cfcb09bf76 Cluster: redis-trib, more info about open slots error. 2014-02-11 00:38:53 +01:00
antirez 8e12fae05e Cluster: fixed inverted arguments in logging function call. 2014-02-10 17:21:17 +01:00
antirez 6a01545744 Cluster: clear the FAIL status for masters without slots. 
Masters without slots don't participate to the cluster but just do
redirections, no need to take them in FAIL state if they are back
reachable.
 2014-02-10 17:19:16 +01:00
antirez 969a4f1db3 Cluster: replica migration should only work for masters serving slots. 2014-02-10 17:08:47 +01:00
antirez cb92a1ef08 Cluster: redis-trib del-node variable typo fixed. 2014-02-10 16:59:17 +01:00
antirez 6987a95952 Cluster: clusterReadHandler() fixed to work with new message header. 2014-02-10 16:28:44 +01:00
antirez ad85f520f8 Cluster: don't propagate PUBLISH two times. 
PUBLISH both published messages via Cluster bus and replication when
cluster was enabled, resulting in duplicated message in the slave.
 2014-02-10 16:05:26 +01:00
antirez b82b66b51d Cluster: signature changed to "RCmb" (Redis Cluster message bus). 
Sounds better after all.
 2014-02-10 16:05:22 +01:00
antirez b6e04f5584 Cluster: discard bus messages with version != 0. 2014-02-10 16:05:18 +01:00
antirez 0ee1a78c86 Cluster: added signature + version in bus packets. 2014-02-10 16:05:15 +01:00
antirez ff6a75a0c1 3.0 release notes added. 2014-02-10 15:36:02 +01:00
antirez b1c3c6e518 Old Changelog file removed from unstable branch. 2014-02-10 15:19:12 +01:00
antirez dca95f241c Cluster: redis-trib: options table entry for add-node fixed. 2014-02-10 12:34:21 +01:00
antirez 6df4ffe639 Don't count time to feed MONITORs in SLOWLOG. 2014-02-07 18:29:20 +01:00
antirez 142281dc79 Cluster: keys slot computation now supports hash tags. 
Currently this is marginally useful, only to make sure two keys are in
the same hash slot when the cluster is stable (no rehashing in
progress).

In the future it is possible that support will be added to run
mutli-keys operations with keys in the same hash slot.
 2014-02-07 17:39:01 +01:00
antirez 2d6eb68993 Sentinel: allow SHUTDOWN command in Sentinel mode. 2014-02-07 11:22:24 +01:00
antirez 970de3e9c0 Check for EAGAIN in sendBulkToSlave(). 
Sometime an osx master with a Linux server over a slow link caused
a strange error where osx called the writable function for
the socket but actually apparently there was no room in the socket
buffer to accept the write: write(2) call returned an EAGAIN error,
that was not checked, so we considered write(2) == 0 always as a connection
reset, which was unfortunate since the bulk transfer has to start again.

Also more errors are logged with the WARNING level in the same code path
now.
 2014-02-05 16:38:10 +01:00
antirez 04fe000bf8 Cluster: fixed MF condition in clusterHandleSlaveFailover(). 
For manual failover we need a manual failover in progress, and that
mf_can_start is true (master offset received and matched).
 2014-02-05 16:01:56 +01:00
antirez c6f02fd67a Cluster: CLUSTER FAILOVER replies with OK and logs the event. 2014-02-05 15:52:38 +01:00
antirez c72449af30 Cluster: check that a MF is in progress in manualFailoverCheckTimeout(). 
Otherwise it is always detected as a manual failover timed out.
 2014-02-05 15:45:24 +01:00
antirez b7402bcad5 Cluster: force AUTH ACK on manual failover. 
When a slave requests masters vote for a manual failover, the
REQUEST_AUTH message is flagged in a special way in order to force the
masters to give the authorization even if the master is not marked as
failing.
 2014-02-05 13:10:03 +01:00
antirez 4cf0cd5719 Cluster: manual failover initial implementation. 2014-02-05 13:01:24 +01:00
antirez 4919a13f50 CLIENT PAUSE and related API implemented. 
The API is one of the bulding blocks of CLUSTER FAILOVER command that
executes a manual failover in Redis Cluster. However exposed as a
command that the user can call directly, it makes much simpler to
upgrade a standalone Redis instance using a slave in a safer way.

The commands works like that:

    CLIENT PAUSE <milliesconds>

All the clients that are not slaves and not in MONITOR state are paused
for the specified number of milliesconds. This means that slaves are
normally served in the meantime.

At the end of the specified amount of time all the clients are unblocked
and will continue operations normally. This command has no effects on
the population of the slow log, since clients are not blocked in the
middle of operations but only when there is to process new data.

Note that while the clients are unblocked, still new commands are
accepted and queued in the client buffer, so clients will likely not
block while writing to the server while the pause is active.
 2014-02-04 16:16:09 +01:00
antirez b089ba98cc Scripting: expire keys in scripts only at first access. 
Keys expiring in the middle of the execution of Lua scripts are to
create inconsistencies in masters and / or AOF files. See the following
example:

    if redis.call("exists",KEYS[1]) == 1
    then
        redis.call("incr","mycounter")
    end

    if redis.call("exists",KEYS[1]) == 1
    then
        return redis.call("incr","mycounter")
    end

The script executes two times the same *if key exists then incrementcounter*
logic. However the two executions will work differently in the master and
the slaves, provided some unlucky timing happens.

In the master the first time the key may still exist, while the second time
the key may no longer exist. This will result in the key incremented just one
time. However as a side effect the master will generate a synthetic
`DEL` command in the replication channel in order to force the slaves to
expire the key (given that key expiration is master-driven).

When the same script will run in the slave, the key will no longer be
there, so the script will not increment the key.

The key idea used to implement the expire-at-first-lookup semantics was
provided by Marc Gravell.
 2014-02-03 16:15:53 +01:00
antirez b770079f2c Allow CONFIG and SHUTDOWN while in stale-slave state. 2014-02-03 15:51:03 +01:00
antirez 89884e8f6e Scripting: use mstime() and mstime_t for lua_time_start. 
server.lua_time_start is expressed in milliseconds. Use mstime_t instead
of long long, and populate it with mstime() instead of ustime()/1000.

Functionally identical but more natural.
 2014-02-03 15:45:40 +01:00
Salvatore Sanfilippo b0335287ac Merge pull request #1534 from gdi2290/patch-1 
update copyright year
 2014-02-03 02:18:24 -08:00
PatrickJS 79270cae64 update copyright year 2014-02-03 02:10:54 -08:00
antirez 331de6930a Test: fixed osx msg passing issue in testing framework. 
The Redis test uses a server-clients model in order to parallelize the
execution of different tests. However in recent versions of osx not
setting the channel to a binary encoding caused issues even if AFAIK no
binary data is really sent via this channel. However now the channels
are deliberately set to a binary encoding and this solves the issue.

The exact issue was the test not terminating and giving the impression
of running forever, since test clients or servers were unable to
exchange the messages to continue.
 2014-01-31 16:27:03 +01:00
antirez e40d3e28e8 Redis.conf comment about tcp-backlog option improved. 2014-01-31 14:59:50 +01:00
antirez 7be946fde2 Option "backlog" renamed "tcp-backlog". 
This is especially important since we already have a concept of backlog
(the replication backlog).
 2014-01-31 14:56:10 +01:00
Nenad Merdanovic d76aa96d1a Add support for listen(2) backlog definition 
In high RPS environments, the default listen backlog is not sufficient, so
giving users the power to configure it is the right approach, especially
since it requires only minor modifications to the code.
 2014-01-31 14:52:10 +01:00
antirez b8bfbf46c5 Cluster: fix an error in migration-barrier comment in redis.conf. 2014-01-31 11:31:50 +01:00
antirez cbfdd13b3a Cluster: barrer -> barrier in redis.conf. 2014-01-31 11:30:49 +01:00
antirez a7d30681c9 Cluster: configurable replicas migration barrier. 
It is possible to configure the min number of additional working slaves
a master should be left with, for a slave to migrate to an orphaned
master.
 2014-01-31 11:26:36 +01:00
antirez 3ff1bb4b2e Sentinel: check arity for SENTINEL MASTER command. 
This fixes issue #1530.
 2014-01-31 10:13:38 +01:00
antirez 6c9359add1 Cluster: perform orphaned masters check before continue statements. 
The check was placed in a way that conflicted with the continue
statements used by the node hearth beat code later that needs to skip
the current node sometimes. Moved at the start of the function so that's
always executed.
 2014-01-30 18:23:31 +01:00
antirez c2507b0ff6 Cluster: replica migration implementation. 
This feature allows slaves to migrate to orphaned masters (masters
without working slaves), as long as a set of conditions are met,
including the fact that the migrating slave needs to be in a
master-slaves ring with at least another slave working.
 2014-01-30 18:05:11 +01:00
antirez 5b4020fb42 Cluster: swap two code blocks to have a more obvious flow. 2014-01-30 16:34:23 +01:00
antirez 4beaaff8ea Cluster: remove not needed return statement breaking failover. 2014-01-29 17:28:46 +01:00
antirez 3582054982 Cluster: broadcast pong to other slaves in the same ring. 
When we schedule a failover, broadcast a PONG to the slaves.
The other slaves that plan to get elected will do the same too, this way
it is likely that every slave will have a good picture of its own rank.

Note that this is N*N messages where N is the number of slaves for the
failing master, however usually even large clusters have many master
nodes but a limited number of replicas per node, so this is harmless.
 2014-01-29 17:19:55 +01:00
antirez e2b59621a8 Cluster: log offset when announcing the failover election delay. 2014-01-29 17:16:10 +01:00
antirez 940531e9b7 Cluster: added progressive election delay according to slave rank. 
Note that when we compute the initial delay, there are probably still
more up to date information to receive from slaves with new offsets, so
the delay is recomputed when new data is available.
 2014-01-29 16:53:45 +01:00
antirez 6f54032080 Cluster: function clusterGetSlaveRank() added. 
Return the number of slaves for the same master having a better
replication offset of the current slave, that is, the slave "rank" used
to pick a delay before the request for election.
 2014-01-29 16:39:04 +01:00
antirez 40cd38f0c4 Cluster: update node replication offset from bus packets headers. 2014-01-29 16:01:00 +01:00
antirez 9d4ded7ec6 Cluster: refactoring: new macros to check node flags. 2014-01-29 12:17:16 +01:00
antirez 099bd336db Cluster: use myself instead of server->cluster.myself. 2014-01-29 11:38:14 +01:00
antirez e36bd8b43e Cluster: added a global myself pointer in cluster.c. 
Accessing to the 'myself' node, the node representing the currently
running instance, is handy without the need to type
server.cluster->myself every time.
 2014-01-29 11:22:22 +01:00
antirez f1e09d8c41 Cluster: clusterBroadcastPong() improved with target selection. 
Now we can broadcast a pong to all the instances or just the local
slaves (that is useful for replication offset propagation).
 2014-01-29 11:08:52 +01:00
antirez befcf6259e Cluster: broadcast master/slave replication offset in bus header. 2014-01-28 16:51:50 +01:00
antirez 8b32bd483a Cluster: limit cluster.h to 80 cols. 2014-01-28 16:34:23 +01:00
antirez 0b1b25c51c Cluster: introduced repl_offset fields in clusterNode. 
The two fields are used in order to remember the latest known
replication offset and the time we received it from other slave nodes.

This will be used by slaves in order to start the election procedure
with a delay that is proportional to the rank of the slave among the
other slaves for this master, when sorted for replication offset.

Usually this allows the slave with the most updated offset to win the
election and replace the failing master in the cluster.
 2014-01-28 16:28:07 +01:00
antirez 72f1715e45 Fixed inverted if condition in MISCONF error code path. 2014-01-28 10:11:12 +01:00
antirez 23f4e9f0d9 Don't log MONITOR clients as disconnecting slaves. 2014-01-25 11:53:53 +01:00
antirez 40377fa522 Cluster: redis-trib set-timeout implemented. 2014-01-24 15:06:01 +01:00
antirez 0f9422d575 Cluster: update slaves lists in clusterSetMaster(). 2014-01-22 18:46:53 +01:00
antirez 5383ab0bc6 Cluster: CLUSTER SLAVES subcommand added. 2014-01-22 18:38:42 +01:00
antirez 603e480fd5 Cluster: clusterGenNodesDescription() refactored into two functions. 2014-01-22 18:36:12 +01:00
antirez 1cf532dc37 redis-cli --help output improved with --scan and periods. 2014-01-22 12:07:42 +01:00
antirez 994c5b26dd redis-cli: support for --scan option. 2014-01-22 12:04:08 +01:00
antirez 172f14d48c Use fflush() before fsync() in rio.c. 
Incremental flushing in rio.c is only used to avoid huge kernel buffers
synched to slow disks creating big latency spikes, so this fix has no
durability implications, however it is certainly more correct to make
sure that the FILE buffers are flushed to the kernel before calling
fsync on the file descriptor.

Thanks to Li Shao Kai for reporting this issue in the Redis mailing
list.
 2014-01-22 09:54:55 +01:00
antirez 80e80668f4 Cluster: master nodes wait before rejoining the cluster after reboot. 
One of the simple heuristics used by Redis Cluster in order to avoid
losing data in the typical failure modes created by the asynchronous
replication with the slaves (a master is unable, when accepting a
write, to immediately tell if it should be really accepted or refused
because of a configuration change), is to wait some time before to
rejoin the cluster after being partitioned away from the majority of
instances.

A similar condition happens when a master is restarted. It does not know
if it was already failed over, nor if all the clients have already an
updated configuration about the cluster map, so it is possible that
clients will try to write to stale masters that were restarted.

In a similar way this commit changes masters behavior so they wait
2000 milliseconds before accepting writes after a reboot. There is
nothing special about 2 seconds if not to be a value supposedly larger
a few orders of magnitude compared to the cluster bus communication
latencies.
 2014-01-20 11:52:52 +01:00
antirez e6970e204f Cluster: debug printf statemets removed. 
These were committed for error after being inserted in order to fix an
issue.
 2014-01-20 11:19:04 +01:00
antirez e4a5605c9a Cluster: don't rewrite slaveof config directive in cluster mode. 2014-01-20 11:10:42 +01:00
antirez 437fc2cb56 Cluster: fix error reporting when slaveof is found in config. 2014-01-20 11:08:14 +01:00
antirez ac3850cabd Cluster: allow CLUSTER REPLICATE to switch master. 
The code was doing checks for slaves that should be done only when the
instance is currently a master. Switching a slave from a master to
another one should just work.
 2014-01-17 18:22:35 +01:00
antirez abd6308d27 Set server.repl_down_since to 0 when changing master. 
When an instance is potentially set to replicate with another master, it
is conceptually disconnected forever, since we have no old copy of the
dataset for this master in memory.
 2014-01-17 18:20:31 +01:00
antirez 36c24bcca0 Cluster: redis-trib shows number of replicas of masters. 2014-01-17 17:56:45 +01:00
antirez 27ed9da383 Cluster: redis-trib help output format modified. 2014-01-17 12:32:49 +01:00
antirez a68c9ba97e Cluster: redis-trib shows what a slave replicates + fixes. 
Also the :replicates info field in the node object is now correctly
populated. This also fixes the :replicas field computation.
 2014-01-17 12:06:18 +01:00
antirez b451176734 Cluster: redis-trib addnode is now able to add replicas. 2014-01-17 11:48:42 +01:00
antirez 30d9c1dc32 Cluster: fix redis-trib help subcommand. 2014-01-17 10:29:40 +01:00
antirez 17d0c3e85a Cluster: redis-trib delnode implementation. 2014-01-16 18:22:03 +01:00
antirez 3d455393a6 Cluster: don't let a node forget its own master. 
redis-trib should make sure to reconfigure slaves of a node to remove
from the cluster to replicate with other nodes before sending CLUSTER
FORGET.
 2014-01-16 17:49:35 +01:00
antirez 9531c84807 Cluster: redis-trib help output improved. 
Show options if any. Clarify that for some command any node address is
ok.
 2014-01-16 16:23:33 +01:00
antirez 0c373207fa Cluster: don't forget yourself with CLUSTER FORGET. 2014-01-16 09:46:23 +01:00
antirez 3e948970fe Cluster: use the node blacklist in CLUSTER FORGET. 
CLUSTER FORGET is not useful if we can't remove a node from all the
nodes of our cluster because of the Gossip protocol that keeps adding
a given node to nodes where we already tried to remove it.

So now CLUSTER FORGET implements a nodes blacklist that is set and
checked by the Gossip section processing function. This way before a
node is re-added at least 60 seconds must elapse since the FORGET
execution.

This means that redis-trib has some time to remove a node from a whole
cluster. It is possible that in the future it will be uesful to raise
the 60 sec figure to something bigger.
 2014-01-15 16:50:45 +01:00
antirez ccf268fa17 Cluster: fix clusterBlacklistAddNode() by setting right expire time. 
The hash table value should be set to now + 60 seconds otherwise it
expires immediately.
 2014-01-15 16:49:31 +01:00
antirez 4e1861155f Cluster: clusterBlacklistAddNode() key lookup fixed. 
We can't lookup by node->name that's not an SDS string but a plain C
array in the node structure.
 2014-01-15 16:45:07 +01:00
antirez b51be7b34f Cluster: clusterBlacklistExists() requires blacklist cleanup before lookup. 2014-01-15 16:06:54 +01:00
antirez a81340abaf Cluster: set a minimum rejoin delay if node_timeout is too small. 
The rejoin delay usually is the node timeout. However if the node
timeout is too small, we set it to 500 milliseconds, that is a value
chosen to be greater than most setups RTT / instances latency figures
so that likely communication with other nodes happen before rejoining.
 2014-01-15 12:34:33 +01:00
antirez a687cbc19c Cluster: periodically call clusterUpdateState() when cluster is down. 
Usually we update the cluster state (to understand if we should accept
queries or reply with an error) only when there is a change in the state
of the nodes. However for the "delayed rejoin" feature to work, that is,
for a master to wait some time before accepting queries again after it
rejoins the majority, we need to periodically update the last time when
the node was partitioned away from the majority.

With this commit if the cluster is down we update the state ten times
per second.
 2014-01-15 12:26:12 +01:00
antirez 25ddefdea3 Cluster: range checking in getSlotOrReply() fixed. 
See issue #1426 on Github.
 2014-01-15 11:33:46 +01:00
antirez fb659cd334 Cluster: ignore empty lines in nodes.conf. 
Even without the user messing manually with the file, it is still
possible to have blank lines (just a single "\n" per line) because of
how the nodes.conf update/write process works.
 2014-01-15 11:23:41 +01:00
antirez 6c63df3031 Cluster: atomic update of nodes.conf file. 
The way the file was generated was unsafe and leaded to nodes.conf file
corruption (zero length file) on server stop/crash during the creation
of the file.

The previous file update method was as simple as open with O_TRUNC
followed by the write call. While the write call was a single one with
the full payload, ensuring no half-written files for POSIX semantics,
stopping the server just after the open call resulted into a zero-length
file (all the nodes information lost!).
 2014-01-15 10:31:20 +01:00
antirez 28273394cb Cluster: support to read from slave nodes. 
A client can enter a special cluster read-only mode using the READONLY
command: if the client read from a slave instance after this command,
for slots that are actually served by the instance's master, the queries
will be processed without redirection, allowing clients to read from
slaves (but without any kind fo read-after-write guarantee).

The READWRITE command can be used in order to exit the readonly state.
 2014-01-14 16:33:16 +01:00
antirez aacbba2607 Fix typo in aofRewriteBufferAppend() comment. 2014-01-14 15:37:49 +01:00
antirez 5189485625 Set REDIS_AOF_REWRITE_MIN_SIZE to 64mb. 
64mb is the default value in redis.conf. For some reason instead the
hard-coded default was 1mb that is too small.
 2014-01-14 11:27:28 +01:00
antirez d5763dceaf SENTINEL SET master quorum implemented. 2014-01-14 09:23:26 +01:00
antirez fe86f890b0 SENTINEL SET: error on bad option name + flush config on error. 2014-01-13 11:55:57 +01:00
antirez f822516e43 SENTINEL SET implemented. 
The new command allows to change master-specific configurations
at runtime. All the settable parameters can be retrivied via the
SENTINEL MASTER command, so there is no equivalent "GET" command.
 2014-01-13 11:53:29 +01:00
antirez 3cdcaff069 Sentinel: fix wrong arity error message. 2014-01-13 11:05:13 +01:00
antirez 964f6b17e9 Sentinel: SENTINEL REMOVE command added. 
The command totally removes a monitored master.
 2014-01-10 15:39:36 +01:00
antirez cf2835519e Sentinel: releaseSentinelRedisInstance() top comment fixed. 
The claim about unlinking the instance from the connected hash tables
was the opposite of the reality. Also the current actual behavior is
safer in most cases, so it is better to manually unlink when needed.
 2014-01-10 15:33:42 +01:00
antirez 9d0f46c6f5 Sentinel: flush config on disk when new master is added. 2014-01-10 15:22:06 +01:00
antirez d4f296bc1d anetResolveIP() prototype added to anet.h. 2014-01-10 15:18:41 +01:00
antirez 39f9f449b0 Sentinel: SENTINEL MONITOR command implemented. 
It allows to add new masters to monitor at runtime.
 2014-01-10 15:18:24 +01:00
antirez 774f0bd45e anetResolveIP() added to anet.c. 
The new function is used when we want to normalize an IP address without
performing a DNS lookup if the string to resolve is not a valid IP.

This is useful every time only IPs are valid inputs or when we want to
skip DNS resolution that is slow during runtime operations if we are
required to block.
 2014-01-10 15:02:39 +01:00
antirez c42e4bd0b6 Sentinel: added SENTINEL MASTER <name> command. 
With SENTINEL MASTERS it was already possible to list all the configured
masters, but not a specific one.
 2014-01-10 14:41:52 +01:00
antirez 2bb9cd464e Add all the configurable fields to addReplySentinelRedisInstance(). 
Note: the auth password with the master is voluntarily not exposed.
 2014-01-10 14:31:41 +01:00
antirez 5a7d04ee7b Trip comment to 80 cols in SentinelCommand(). 2014-01-10 14:13:04 +01:00
antirez 5d08413b13 Test: regression for issues #1483. 2014-01-09 11:19:03 +01:00
antirez 58c8a071a5 Fix RESTORE ttl handling in 32 bit archs. 
long was used instead of long long in order to handle a 64 bit
resolution millisecond timestamp.

This fixes issue #1483.
 2014-01-09 11:09:23 +01:00
antirez e1ab2991c3 Fix keyspace events flags-to-string conversion. 
Fixes issue #1491 on Github.
 2014-01-08 17:18:34 +01:00
antirez a2f9947827 Test: stress events flags to/from string conversion. 2014-01-08 17:18:30 +01:00
antirez 90a81b4ebb Don't send REPLCONF ACK to old masters. 
Masters not understanding REPLCONF ACK will reply with errors to our
requests causing a number of possible issues.

This commit detects a global replication offest set to -1 at the end of
the replication, and marks the client representing the master with the
REDIS_PRE_PSYNC flag.

Note that this flag was called REDIS_PRE_PSYNC_SLAVE but now it is just
REDIS_PRE_PSYNC as it is used for both slaves and masters starting with
this commit.

This commit fixes issue #1488.
 2014-01-08 14:28:16 +01:00
antirez 3f92e05637 Clarify a comment in slaveTryPartialResynchronization(). 2014-01-08 14:28:13 +01:00
antirez fdf50e1e3d Log disconnection with slave only when ip:port is available. 2013-12-25 18:41:53 +01:00
antirez 2041882286 anetPeerToString / SockName: port can be NULL on errors too. 2013-12-25 18:41:49 +01:00
antirez a2a900356e anetTcpGenericConnect() bug introduced in 9d19977 fixed. 
Durign a refactoring I mispelled _port for port.
This is one of the reasons I never used _varname myself.
 2013-12-25 18:41:45 +01:00
antirez cb23d510f4 Remove useless goto from anetTcpGenericConnect(). 2013-12-25 18:41:41 +01:00
antirez 491f681088 anetTcpGenericConnect() code improved + 1 bug fix. 
Now the socket is closed if anetNonBlock() fails, and in general the
code structure makes it harder to introduce this kind of bugs in the
future.

Reference: pull request #1059.
 2013-12-25 18:15:28 +01:00
antirez f510549044 Cluster: clusterProcessPacket() was not 80 cols friendly. 
The function actually needs to be split into sub-functions at some
point in the future.
 2013-12-25 17:57:36 +01:00
antirez e789384255 Fix CONFIG REWRITE handling of unknown options. 
There were two problems with the implementation.

1) "save" was not correctly processed when no save point was configured,
   as reported in issue #1416.
2) The way the code checked if an option existed in the "processed"
   dictionary was wrong, as we add the element with as a key associated
   with a NULL value, so dictFetchValue() can't be used to check for
   existance, but dictFind() must be used, that returns NULL only if the
   entry does not exist at all.
 2013-12-23 12:50:27 +01:00
antirez 7e9433cee1 Configuring port to 0 disables IP socket as specified. 
This was no longer the case with 2.8 becuase of a bug introduced with
the IPv6 support. Now it is fixed.

This fixes issue #1287 and #1477.
 2013-12-23 11:31:35 +01:00
antirez 94e8c9e77e Make new masters inherit replication offsets. 
Currently replication offsets could be used into a limited way in order
to understand, out of a set of slaves, what is the one with the most
updated data. For example this comparison is possible of N slaves
were replicating all with the same master.

However the replication offset was not transferred from master to slaves
(that are later promoted as masters) in any way, so for instance if
there were three instances A, B, C, with A master and B and C
replication from A, the following could happen:

C disconnects from A.
B is turned into master.
A is switched to master of B.
B receives some write.

In this context there was no way to compare the offset of A and C,
because B would use its own local master replication offset as
replication offset to initialize the replication with A.

With this commit what happens is that when B is turned into master it
inherits the replication offset from A, making A and C comparable.
In the above case assuming no inconsistencies are created during the
disconnection and failover process, A will show to have a replication
offset greater than C.

Note that this does not mean offsets are always comparable to understand
what is, in a set of instances, since in more complex examples the
replica with the higher replication offset could be partitioned away
when picking the instance to elect as new master. However this in
general improves the ability of a system to try to pick a good replica
to promote to master.
 2013-12-22 11:43:25 +01:00
antirez ba5eb44d14 Slave disconnection is an event worth logging. 2013-12-22 10:15:35 +01:00
antirez 66ec1412fe Redis Cluster: add repl_ping_slave_period to slave data validity time. 
When the configured node timeout is very small, the data validity time
(maximum data age for a slave to try a failover) is too little (ten
times the configured node timeout) when the replication link with the
master is mostly idle. In this case we'll receive some data from the
master only every server.repl_ping_slave_period to refresh the last
interaction with the master.

This commit adds to the max data validity time the slave ping period to
avoid this problem of slaves sensing too old data without a good reason.
However this max data validity time is likely a setting that should be
configurable by the Redis Cluster user in a way completely independent
from the node timeout.
 2013-12-22 10:05:16 +01:00
antirez b2dedd9da8 Log when a slave lose the connection with its master. 2013-12-21 00:23:37 +01:00
antirez 658aff9d29 Redis Cluster: move node failure reports logging from VERBOSE to NOTICE level. 2013-12-21 00:04:53 +01:00
antirez 5a404c87c1 Redis Cluster: remove no longer relevant comment. 2013-12-20 14:40:11 +01:00
antirez fda4cba912 Redis Cluster: reconfigure replication when master changes address. 2013-12-20 12:47:22 +01:00
antirez d7374032c0 Redis Cluster: handshake code refactoring + Gossip IP switch detection. 
This commit makes it simple to start an handshake with a specific node
address, and uses this in order to detect a node IP change and start a
new handshake in order to fix the IP if possible.
 2013-12-20 12:38:03 +01:00
antirez a2c938c834 Redis Cluster: delay state change when in the majority again. 
As specified in the Redis Cluster specification, when a node can reach
the majority again after a period in which it was partitioend away with
the minorty of masters, wait some time before accepting queries, to
provide a reasonable amount of time for other nodes to upgrade its
configuration.

This lowers the probabilities of both a client and a master with not
updated configuration to rejoin the cluster at the same time, with a
stale master accepting writes.
 2013-12-20 09:56:18 +01:00
antirez 305d7f29f3 Clarify include directive behavior in example redis.conf. 2013-12-19 16:02:31 +01:00
antirez b3632319a4 CONFIG REWRITE: no special handling or include and rename-command. 
CONFIG REWRITE is now wiser and does not touch what it does not
understand inside redis.conf.
 2013-12-19 15:57:11 +01:00
Yubao Liu 7da423f79f CONFIG REWRITE: don't throw some options on config rewrite 
Those options will be thrown without this patch:
  include, rename-command, min-slaves-to-write, min-slaves-max-lag,
appendfilename.
 2013-12-19 15:56:48 +01:00
antirez 3b9cf3ed3a CONFIG REWRITE: old development comments removed. 2013-12-19 15:30:06 +01:00
antirez b221e13dac CONFIG REWRITE: don't wipe unknown options. 
With this commit options not explicitly rewritten by CONFIG REWRITE are
not touched at all. These include new options that may not have support
for REWRITE, and other special cases like rename-command and include.
 2013-12-19 15:25:45 +01:00
antirez 6d184e02be Example redis.conf formatted to better show appendfilename option. 2013-12-19 10:18:45 +01:00
antirez 7a666ac419 Cluster: set n->slaves to NULL in clusterNodeResetSlaves(). 
The value was otherwise undefined, so next time the node was promoted
again from slave to master, adding a slave to the list of slaves
would likely crash the server or result into undefined behavior.
 2013-12-17 14:50:24 +01:00
antirez fda91dbde3 Cluster: check link is valid before sending UPDATE. 2013-12-17 12:28:37 +01:00
antirez f57bb36ce7 Cluster: initialize todo_before_sleep flags to 0. 2013-12-17 12:22:02 +01:00
antirez c70c0c6db7 Cluster: use proper type mstime_t for ping delay var. 2013-12-17 10:27:36 +01:00
antirez 7c1cbdceb2 Cluster: use an hardcoded 60 sec timeout in redis-trib connections. 
Later this should be configurable from the command line but at least now
we use something more appropriate for our use case compared to the
redis-rb default timeout.
 2013-12-17 10:00:33 +01:00
antirez 47815d38e0 Fixed clearNodeFailureIfNeeded() time type to mstime_t. 
This prevented 32bit cluster instances from clearing the FAIL flag when
needed.
 2013-12-17 09:45:52 +01:00
antirez e88e6a6334 Cluster: use long long for timestamps in clusterGenNodesDescription(). 
Ping sent and pong received fields need to be casted to long long to be
printed correctly into 32 bit systems.
 2013-12-17 09:38:11 +01:00
antirez 2dfc5e35a9 Makefile.dep updated. 2013-12-13 13:10:05 +01:00
antirez b1ba58f341 SDIFF iterator misuse bug regression test added. 
See commit c00453d for more info about the bug.
 2013-12-13 11:37:13 +01:00
antirez c00453da1d SDIFF iterator misuse fixed in diff algorithm #1. 
The bug could be easily triggered by:

    SADD foo a b c 1 2 3 4 5 6
    SDIFF foo foo

When the key was the same in two sets, an unsafe iterator was used to
check existence of elements in the same set we were iterating.
Usually this would just result into a wrong output, however with the
dict.c API misuse protection we have in place, the result was actually
an assertion failed that was triggered by the CI test, while creating
random datasets for the "MASTER and SLAVE consistency" test.
 2013-12-13 11:34:21 +01:00
antirez 5320148883 Sentinel: dead code removed. 2013-12-13 11:01:13 +01:00
antirez 452dea30f6 Makefile: remove odd syntax not compatible with some make versions. 
See issue #1448.
 2013-12-12 15:19:39 +01:00
Salvatore Sanfilippo 62e4956936 Merge pull request #1415 from Dieken/fix-typo 
fix typo in redis.conf and sentinel.conf
 2013-12-12 02:30:11 -08:00
Salvatore Sanfilippo a99c751d6c Merge pull request #1460 from codeeply/simplify2 
comment mistake fixed
 2013-12-12 02:23:44 -08:00
codeeply 0f06f8df07 comment mistake fixed 2013-12-12 16:33:29 +08:00
antirez a5ec247f13 Replication: publish the slave_repl_offset when disconnected from master. 
When a slave was disconnected from its master the replication offset was
reported as -1. Now it is reported as the replication offset of the
previous master, so that failover can be performed using this value in
order to try to select a slave with more processed data from a set of
slaves of the old master.
 2013-12-11 15:23:15 +01:00
Salvatore Sanfilippo 0a89d9a0b1 Merge pull request #1451 from yossigo/unbalanced-quotes-fix 
Return proper error on requests with an unbalanced number of quotes.
 2013-12-11 03:06:18 -08:00
Yossi Gottlieb 88a5cede88 Fix wrong repldboff type which causes dropped replication in rare cases. 2013-12-11 11:38:02 +01:00
Yubao Liu 6d5fa2e06c fix typo in redis.conf and sentinel.conf 2013-12-11 15:46:42 +08:00
antirez 11120689c4 Slaves heartbeats during sync improved. 
The previous fix for false positive timeout detected by master was not
complete. There is another blocking stage while loading data for the
first synchronization with the master, that is, flushing away the
current data from the DB memory.

This commit uses the newly introduced dict.c callback in order to make
some incremental work (to send "\n" heartbeats to the master) while
flushing the old data from memory.

It is hard to write a regression test for this issue unfortunately. More
support for debugging in the Redis core would be needed in terms of
functionalities to simulate a slow DB loading / deletion.
 2013-12-10 18:47:31 +01:00
antirez 2eb781b35b dict.c: added optional callback to dictEmpty(). 
Redis hash table implementation has many non-blocking features like
incremental rehashing, however while deleting a large hash table there
was no way to have a callback called to do some incremental work.

This commit adds this support, as an optiona callback argument to
dictEmpty() that is currently called at a fixed interval (one time every
65k deletions).
 2013-12-10 18:46:24 +01:00
antirez 2c4ab8a534 Log empty DB + Loading data into two separated messages. 2013-12-10 18:43:25 +01:00
antirez 7c531eb5ad Don't send more than 1 newline/sec while loading RDB. 2013-12-10 18:43:19 +01:00
antirez 27db38d069 Slaves heartbeat while loading RDB files. 
Starting with Redis 2.8 masters are able to detect timed out slaves,
while before 2.8 only slaves were able to detect a timed out master.

Now that timeout detection is bi-directional the following problem
happens as described "in the field" by issue #1449:

1) Master and slave setup with big dataset.
2) Slave performs the first synchronization, or a full sync
   after a failed partial resync.
3) Master sends the RDB payload to the slave.
4) Slave loads this payload.
5) Master detects the slave as timed out since does not receive back the
   REPLCONF ACK acknowledges.

Here the problem is that the master has no way to know how much the
slave will take to load the RDB file in memory. The obvious solution is
to use a greater replication timeout setting, but this is a shame since
for the 0.1% of operation time we are forced to use a timeout that is
not what is suited for 99.9% of operation time.

This commit tries to fix this problem with a solution that is a bit of
an hack, but that modifies little of the replication internals, in order
to be back ported to 2.8 safely.

During the RDB loading time, we send the master newlines to avoid
being sensed as timed out. This is the same that the master already does
while saving the RDB file to still signal its presence to the slave.

The single newline is used because:

1) It can't desync the protocol, as it is only transmitted all or
nothing.
2) It can be safely sent while we don't have a client structure for the
master or in similar situations just with write(2).
 2013-12-09 20:26:00 +01:00
antirez eaf1bfb88b Handle inline requested terminated with just \n. 2013-12-09 13:28:39 +01:00
Yossi Gottlieb 6e70c01148 Return proper error on requests with an unbalanced number of quotes. 2013-12-08 12:58:12 +02:00
antirez c590549e40 Sentinel: fix reported role info sampling. 
The way the role change was recoded was not sane and too much
convoluted, causing the role information to be not always updated.

This commit fixes issue #1445.
 2013-12-06 12:46:56 +01:00
antirez 2b414a4b5f Sentinel: fix reported role fields when master is reset. 
When there is a master address switch, the reported role must be set to
master so that we have a chance to re-sample the INFO output to check if
the new address is reporting the right role.

Otherwise if the role was wrong, it will be sensed as wrong even after
the address switch, and for enough time according to the role change
time, for Sentinel consider the master SDOWN.

This fixes isue #1446, that describes the effects of this bug in
practice.
 2013-12-06 11:37:46 +01:00
antirez 8534a290d3 Fixed typo in redis.conf. 2013-12-06 10:48:46 +01:00
Salvatore Sanfilippo 2ef57f8d47 Merge pull request #1439 from AnuragRamdasan/patch-3 
Grammar fix.
 2013-12-05 09:53:45 -08:00
Anurag Ramdasan 839ed7a60b Grammar fix. 2013-12-05 23:15:47 +05:30
Salvatore Sanfilippo 026e561446 Merge pull request #1438 from AnuragRamdasan/patch-2 
fixed typo
 2013-12-05 08:18:20 -08:00
Anurag Ramdasan fb6b9b14bd fixed typo 2013-12-05 21:47:17 +05:30
Salvatore Sanfilippo cbaad0b26f Merge pull request #1437 from AnuragRamdasan/patch-1 
Fixed grammar: 'usually' to 'usual'
 2013-12-05 07:42:05 -08:00
Anurag Ramdasan 74431b80a3 Fixed grammar: 'usually' to 'usual' 2013-12-05 21:09:31 +05:30
antirez 11e81a1e9a Fixed grammar: before H the article is a, not an. 2013-12-05 16:35:32 +01:00
antirez 74da4a574f Fixed typos in redis.conf file. 2013-12-05 16:28:35 +01:00
antirez 58713c6b13 Fix clients timeout handling. 
During the refactoring of blocking operations, commit
82b672f633, a bug was introduced where
a milliseconds time is compared to a seconds time, so all the clients
always appear to timeout if timeout is set to non-zero value.

Thanks to Jonathan Leibiusky for finding the bug and helping verifying
the cause and fix.
 2013-12-05 14:55:07 +01:00
antirez c5618e7fdd WAIT command: synchronous replication for Redis. 2013-12-04 16:20:03 +01:00
antirez c2f305545a blocked.c API commented. 2013-12-03 18:03:15 +01:00
antirez 82b672f633 BLPOP blocking code refactored to be generic & reusable. 2013-12-03 17:43:53 +01:00
antirez 2e027c48e5 Removed old comments and dead code from freeClient(). 2013-12-03 13:54:06 +01:00
antirez e4025ea926 Grammar fix in freeClient(). 2013-12-03 13:40:41 +01:00
antirez f80cf7363a Sentinel: don't write HZ when flushing config. 
See issue #1419.
 2013-12-02 15:56:10 +01:00
antirez dffebbc904 Sentinel: better time desynchronization. 
Sentinels are now desynchronized in a better way changing the time
handler frequency between 10 and 20 HZ. This way on average a
desynchronization of 25 milliesconds is produced that should be larger
enough compared to network latency, avoiding most split-brain condition
during the vote.

Now that the clocks are desynchronized, to have larger random delays when
performing operations can be easily achieved in the following way.
Take as example the function that starts the failover, that is
called with a frequency between 10 and 20 HZ and will start the
failover every time there are the conditions. By just adding as an
additional condition something like rand()%4 == 0, we can amplify the
desynchronization between Sentinel instances easily.

See issue #1419.
 2013-12-02 12:29:42 +01:00
antirez 6fa42b7507 Cluster: nodes re-addition blacklist API. 2013-12-02 11:12:23 +01:00
antirez 8f18345ef0 Cluster: basic data structures for nodes black list. 2013-11-29 17:37:06 +01:00
antirez 3db825fde4 Cluster: some code about clusterHandleSlaveFailover() marginally improved. 
80 cols friendly, some minor change to the code to make it simpler.
 2013-11-29 16:17:05 +01:00
antirez 55f90b11c9 Stop writes on MISCONF only if instance is a master. 
From the point of view of the slave not accepting writes from the master
can only create a bigger consistency issue.
 2013-11-28 16:29:26 +01:00
antirez 60817bb262 Reply to PING with error when there is a MISCONF state. 2013-11-28 16:17:10 +01:00
antirez 0addf8aff1 Sentinel: log vote received from other Sentinels. 2013-11-28 15:23:46 +01:00
Salvatore Sanfilippo c7136269c6 Merge pull request #1401 from huangz1990/fix_sentinel_bug 
fix a bug in sentinel.c about pub/sub link
 2013-11-26 06:15:25 -08:00
huangz1990 86a540a66e fix a bug in sentinel.c about pub/sub link 2013-11-26 19:55:51 +08:00
antirez 6f4fd55762 Sentinel: fixes inverted strcmp() test preventing config updates. 
The result of this one-char bug was pretty serious, if the new master
had the same port of the previous master, but just a different IP
address, non-leader Sentinels would not be able to recognize the
configuration change.

This commit fixes issue #1394.

Many thanks to @shanemadden that reported the bug and helped
investigating it.
 2013-11-25 10:59:53 +01:00
antirez 8d547ebd56 Sentinel: fix type specifier for Hello msg generation. 
This fixes issue #1395.
 2013-11-25 10:24:34 +01:00
antirez fcebd9b0f9 Fix false positive in memory efficiency test. 
Fixes issue #1298.
 2013-11-25 10:21:46 +01:00
antirez 37b43c8a24 Sentinel: example sentinel.conf updated. 2013-11-21 17:07:00 +01:00
antirez cc6053681f Sentinel: different comments updated to new implementation. 2013-11-21 16:22:59 +01:00
antirez 685e79998c Sentinel: cleanup around SENTINEL_INFO_VALIDITY_TIME. 2013-11-21 16:05:41 +01:00
antirez 489d889726 Sentinel: removed mem leak and useless code. 2013-11-21 15:43:55 +01:00
antirez 1ca56ea796 Merge branch 'newsentinel' into unstable 2013-11-21 15:01:25 +01:00
antirez f55ad3038f Sentinel: manual failover works again. 2013-11-21 12:39:47 +01:00
antirez 297de1ab26 Sentinel: test for writable config file. 
This commit introduces a funciton called when Sentinel is ready for
normal operations to avoid putting Sentinel specific stuff in redis.c.
 2013-11-21 12:28:15 +01:00
antirez d920177f8d Sentinel: check for disconnected links in sentinelSendHello(). 
Does not fix any bug as the test is performed by the caller, but better
to have the check.
 2013-11-21 11:35:50 +01:00
antirez 8810167d13 Sentinel: Hello message sending code refactored. 2013-11-21 11:31:06 +01:00
antirez 0101c2bcfe Sentinel: select slave with best (greater) replication offset. 2013-11-20 16:05:36 +01:00
antirez a6ebd910d8 Sentinel: take the replication offset in slaves state. 2013-11-20 15:53:21 +01:00
antirez b1f5a0b3ec CONFIG REWRITE: don't add the signature if it already exists. 
At the end of the file, CONFIG REWRITE adds a comment line that:

    # Generated by CONFIG REWRITE

Followed by the additional config options required. However this was
added again and again at every rewrite in praticular conditions (when a
given set of options change in a given time during the time).

Now if it was alrady encountered, it is not added a second time.

This is especially important for Sentinel that rewrites the config at
every state change.
 2013-11-19 17:58:11 +01:00
antirez 37a51a2568 Sentinel: distinguish between is-master-down-by-addr requests. 
Some are just to know if the master is down, and in this case the runid
in the request is set to "*", others are actually in order to seek for a
vote and get elected. In the latter case the runid is set to the runid
of the instance seeking for the vote.
 2013-11-19 16:50:04 +01:00
antirez b22d1beea0 Sentinel: various fixes to leader election implementation. 2013-11-19 16:20:42 +01:00
antirez 1f9728cb20 Sentinel: failover script execution fixed. 2013-11-19 12:34:46 +01:00
antirez 90635488ce Sentinel: no longer used defines removed. 2013-11-19 11:24:36 +01:00
antirez 0a35f65301 Sentinel: when writing config on disk, remember sentinels runid. 2013-11-19 11:11:43 +01:00
antirez 5450833d02 Sentinel: arity of known-sentinel/slave is 4 not 3. 2013-11-19 11:03:47 +01:00
antirez b8a94463b7 Sentinel: rewriteConfigSentinelOption() sub-iterators var typo fixed. 2013-11-19 10:59:50 +01:00
antirez 16237d78c8 Sentinel: call sentinelFlushConfig() to persist state when needed. 
Also the sentinel configuration rewriting was modified in order to
account for failover in progress, where we need to provide the promoted
slave address as master address, and the old master address as one of
the slaves address.
 2013-11-19 10:55:43 +01:00
antirez e257ab2bfe Sentinel: sentinelFlushConfig() to CONFIG REWRITE + fsync. 2013-11-19 10:13:04 +01:00
antirez 5998769c28 Sentinel: CONFIG REWRITE support for Sentinel config. 2013-11-19 09:48:12 +01:00
antirez 47df12d5d9 Sentinel: can-failover option removed, many comments fixed. 2013-11-19 09:28:47 +01:00
antirez cd4ff9992b Fix typo 'configuraiton' in rewriteConfigRewriteLine() comment. 2013-11-18 18:18:10 +01:00
antirez 232cdb95ab Sentinel: added config options useful to take state on config rewrite. 
We'll use CONFIG REWRITE (internally) in order to store the new
configuration of a Sentinel after the internal state changes. In order
to do so, we need configuration options (that usually the user will not
touch at all) about config epoch of the master, Sentinels and Slaves
known for this master, and so forth.
 2013-11-18 16:03:03 +01:00
antirez 3a374b0511 Sentinel: failover abort function simplified. 2013-11-18 11:43:35 +01:00
antirez e0750acf11 Sentinel: slaves reconfig delay modified. 
The time Sentinel waits since the slave is detected to be configured to
the wrong master, before reconfiguring it, is now the failover_timeout
time as this makes more sense in order to give the Sentinel performing
the failover enoung time to reconfigure the slaves slowly (if required
by the configuration).

Also we now PUBLISH more frequently the new configuraiton as this allows
to switch the reapprearing master back to slave faster.
 2013-11-18 11:37:24 +01:00
antirez 83316f515c Sentinel: failover restart time is now multiple of failover timeout. 
Also defaulf failover timeout changed to 3 minutes as the failover is a
fairly fast procedure most of the times, unless there are a very big
number of slaves and the user picked to configure them sequentially (in
that case the user should change the failover timeout accordingly).
 2013-11-18 11:30:08 +01:00
antirez 3a56013acb Sentinel: state machine and timeouts simplified. 2013-11-18 11:12:58 +01:00
antirez 4be53b1c5d Sentinel: election timeout define. 2013-11-18 10:08:06 +01:00
antirez 69d826a354 Sentinel: fix address of master in Hello messages. 
Once we switched configuration during a failover, we should advertise
the new address.

This was a serious race condition as the Sentinel performing the
failover for a moment advertised the old address with the new
configuration epoch: once trasmitted to the other Sentinels the broken
configuration would remain there forever, until the next failover
(because a greater configuration epoch is required to overwrite an older
one).
 2013-11-14 10:25:55 +01:00
antirez e4c65e72c6 Sentinel: master address selection in get-master-address refactored. 2013-11-14 10:23:54 +01:00
antirez c0d7229364 Sentinel: fix conditional to only affect slaves with wrong master. 2013-11-14 10:23:05 +01:00
antirez dfbd9c5aeb Sentinel: simplify and refactor slave reconfig code. 2013-11-14 00:36:43 +01:00
antirez 64ad6648a8 Sentinel: reconfigure slaves to right master. 2013-11-14 00:29:38 +01:00
antirez 3e27d678da Sentinel: remember last time slave changed master. 2013-11-14 00:20:15 +01:00
antirez 8297745fa6 Sentinel: redirect-to-master is not ok with new algorithm. 
Now Sentinel believe the current configuration is always the winner and
should be applied by Sentinels instead of trying to adapt our view of
the cluster based on what we observe.

So the only way to modify what a Sentinel believe to be the truth is to
win an election and advertise the new configuration via Pub / Sub with a
greater configuration epoch.
 2013-11-13 17:03:48 +01:00
antirez 76a88f56e5 Sentinel: safer slave reconfig, master reported role should match. 2013-11-13 17:02:09 +01:00
antirez ddaad9fe2d Sentinel: role reporting fixed and added in SENTINEL output. 2013-11-13 16:39:57 +01:00
antirez a0afa66f4b Sentinel: being a master and reporting as slave is considered SDOWN. 2013-11-13 16:28:56 +01:00
antirez 17718fdcba Sentinel: make sure role_reported is always updated. 2013-11-13 16:21:58 +01:00
antirez 46a053d34b Sentinel: track role change time. Wait before reconfigurations. 2013-11-13 16:18:23 +01:00
antirez 9e40c46f5e Sentinel: fix no-down check in master->slave conversion code. 2013-11-13 13:43:59 +01:00
antirez ae35b7e240 Sentinel: readd slaves back after a master reset. 2013-11-13 13:01:11 +01:00
antirez 6bd4f6bffe Sentinel: sentinelResetMaster() new flag to avoid removing set of sentinels. 
This commit also removes some dead code and cleanup generic flags.
 2013-11-13 10:30:45 +01:00
antirez 1569af1f23 Sentinel: receive Pub/Sub messages from slaves. 2013-11-12 23:07:33 +01:00
antirez dfa5f8b777 Sentinel: change event name when converting master to slave. 2013-11-12 23:00:17 +01:00
antirez 24158d1488 Sentinel: added config-epoch to SENTINEL masters output. 2013-11-12 17:22:04 +01:00
antirez d2bc6dc39a Sentinel: new failover algo, desync slaves and update config epoch. 2013-11-12 17:07:31 +01:00
antirez 4a128b949d Sentinel: when starting failover seek for votes ASAP. 2013-11-12 16:38:02 +01:00
antirez e6b9d5e97e Sentinel: +new-epoch events. 2013-11-12 13:35:25 +01:00
antirez 54c447be52 Sentinel: wait some time between failover attempts. 2013-11-12 13:30:31 +01:00
antirez ab4b2ec88f Sentinel: allow to vote for myself. 2013-11-12 11:32:40 +01:00
antirez b6b65b29c0 Sentinel: fix PUBLISH to masters and slaves. 2013-11-12 11:12:48 +01:00
antirez 90ab62fd5e Sentinel: epoch introduced in leader vote. 2013-11-12 11:09:35 +01:00
antirez 8c1bf9a2bd Sentinel: leadership handling changes WIP. 
Changes to leadership handling.

Now the leader gets selected by every Sentinel, for a specified epoch,
when the SENTINEL is-master-down-by-addr is sent.

This command now includes the runid and the currentEpoch of the instance
seeking for a vote. The Sentinel only votes a single time in a given
epoch.

Still a work in progress, does not even compile at this stage.
 2013-11-11 18:30:14 +01:00
antirez 0bac36d0a1 Sentinel: handle Hello messages received via slaves correctly. 
Even when messages are received via the slave, we should perform
operations (like adding a new Sentinel) in the context of the master.
 2013-11-11 17:12:27 +01:00
antirez 9e1b27d49e Sentinel: remove code not useful in the new design. 2013-11-11 12:06:11 +01:00
antirez b93b0adc89 Sentinel: epoch introduced. 
Sentinel state now includes the idea of current epoch and config epoch.
In the Hello message, that is now published both on masters and slaves,
a Sentinel no longer just advertises itself but also broadcasts its
current view of the configuration: the master name / ip / port and its
current epoch.

Sentinels receiving such information switch to the new master if the
configuration epoch received is newer and the ip / port of the master
are indeed different compared to the previos ones.
 2013-11-11 11:05:58 +01:00
antirez b2f834390c Log to what master a slave is going to connect to. 2013-11-11 09:25:36 +01:00
antirez a5e7358a12 Cluster: removed not needed newline at end of redisLog() msg. 2013-11-08 17:28:02 +01:00
antirez 28071caf38 Cluster: send a single UPDATE packet for now. 2013-11-08 17:25:49 +01:00
antirez d289c628b1 Cluster: replace hardcoded 4096 for bus msg len with sizeof(). 2013-11-08 17:19:19 +01:00
antirez 94a07d5901 Cluster: slots update refactored + UPDATE msg processing. 
Now there is a function that handles the update of the local slot
configuration every time we have some new info about a node and its set
of served slots and configEpoch.

Moreoever the UPDATE packets are now processed when received (it was a
work in progress in the previous commit).
 2013-11-08 17:02:10 +01:00
antirez dc43f66eac Cluster: UPDATE msg data structure and sending function. 2013-11-08 16:26:50 +01:00
antirez 6c6572be95 Cluster: refactoring of slots update code and more. 
The commit also introduces detection of nodes publishing not updated
configuration. More work in progress to send an UPDATE packet to inform
of the config change.
 2013-11-08 10:32:16 +01:00
antirez d3588dc194 Fix broken rdbWriteRaw() return value check in rdb.c. 
Thanks to @PhoneLi for reporting.
 2013-11-07 23:53:18 +01:00
antirez d75ab87226 redis-trib: fixed slot allocation when --replicas is used. 2013-11-07 16:12:06 +01:00
antirez e1a1d904de Merge branch 'unstable' of github.com:/antirez/redis into unstable 2013-11-06 11:30:44 +01:00
antirez 80da056c29 Sentinel: sentinelSendSlaveOf() was missing a var and the prototype. 2013-11-06 11:23:53 +01:00
antirez 23800d9e49 Sentinel: increment pending_commands counter in two more places. 
AUTH and SCRIPT KILL were sent without incrementing the pending commands
counter. Clearly this needs some kind of wrapper doing it for the caller
in order to be less bug prone.
 2013-11-06 11:21:44 +01:00
antirez 671c1dfb56 Sentinel: always send CONFIG REWRITE when changing instance role. 
This change makes Sentinel less fragile about a number of failure modes.

This commit also fixes a different bug as a side effect, SLAVEOF command
was sent multiple times without incrementing the pending commands count.
 2013-11-06 11:13:27 +01:00
Salvatore Sanfilippo b5ec3ede7a Merge pull request #1365 from yaauie/deprecate-redis-copy-script 
Deprecate utils/redis-copy.rb in favor of redis-copy gem
 2013-11-06 00:42:03 -08:00
Ryan Biesemeyer ae8642fb79 Deprecate utils/redis-copy.rb in favor of redis-copy gem 2013-11-06 08:31:57 +00:00
antirez 5fd6dee0e6 Fixed typo in parseScanCursorOrReply(): ad -> at. 
Thanks to @badboy for reporting it.
 2013-11-05 17:23:11 +01:00
antirez ebcb6251e6 SCAN code refactored to parse cursor first. 
The previous implementation of SCAN parsed the cursor in the generic
function implementing SCAN, SSCAN, HSCAN and ZSCAN.

The actual higher-level command implementation only checked for empty
keys and return ASAP in that case. The result was that inverting the
arguments of, for instance, SSCAN for example and write:

    SSCAN 0 key

Instead of

    SSCAN key 0

Resulted into no error, since 0 is a non-existing key name very likely.
Just the iterator returned no elements at all.

In order to fix this issue the code was refactored to extract the
function to parse the cursor and return the error. Every higher level
command implementation now parses the cursor and later checks if the key
exist or not.
 2013-11-05 15:47:50 +01:00
antirez b4048dfec0 SCAN: when iterating ziplists or intsets always return cursor of 0. 
The previous implementation assumed that the first call always happens
with cursor set to 0, this may not be the case, and we want to return 0
anyway otherwise the (broken) client code will loop forever.
 2013-11-05 15:32:25 +01:00
antirez 101d4bf867 Use strtoul() instead of sscanf() in SCAN implementation. 2013-11-05 15:30:21 +01:00
antirez bf79c0cda2 Added tests for [SHZ]SCAN with MATCH. 2013-11-05 15:19:44 +01:00
antirez f56f78d159 HSCAN/ZSCAN: skip value when matching. 
This fixes issue #1360 and #1362.
 2013-11-05 12:16:29 +01:00
antirez 1a0cea33a0 Cluster: initialize senderConfigEpoch and senderCurrentEpoch for warnings suppression. 2013-11-05 12:01:07 +01:00
antirez eb95d28898 Pass int64_t to intsetGet() instead of long long. 2013-11-05 11:57:30 +01:00
antirez 20fb91fd31 removed not used vars in dictScan(). 2013-11-05 11:56:11 +01:00
antirez 6cf230ea91 Initial support for --replicas in redis-trib. 2013-11-05 11:24:24 +01:00
antirez 9fba193a59 SSCAN with integer encoded object test improved. 2013-10-31 10:37:27 +01:00
antirez c4ca5f99bf Inverted variable boolean value and name after scanGenericCommand() refactoring. 2013-10-31 10:35:56 +01:00
antirez 8fc85a1218 scanGenericCommand() refactoring and handling of integer encoded elements. 
This commit fixes issue #1354.
 2013-10-31 10:32:39 +01:00
antirez e6bb738b87 Regression test added for [SHZ]SCAN issue #1354. 2013-10-31 09:43:21 +01:00
antirez 0b8a0ca4bc Test: added a SCAN test trying to trigger HT resize. 2013-10-30 16:50:25 +01:00
antirez 6cb52256c9 Test: added ZSCAN test. 2013-10-30 16:25:53 +01:00
antirez 1b960378e8 Test: added HSCAN test. 2013-10-30 16:24:39 +01:00
antirez ddc80e026f Test: added SSCAN test. 2013-10-30 11:58:04 +01:00
antirez 5fdb94fd98 SCAN test keys sorting turned into more idiomatic Tcl. 2013-10-30 11:36:12 +01:00
antirez a881827b6e SCAN: tests moved to unit/scan.tcl. 2013-10-30 11:34:01 +01:00
antirez 3c7a0677b0 redis-benchmark: update help for new __rand_int__ form. 2013-10-28 18:13:42 +01:00
antirez 9b2d44e63f Aesthetic fix (missing space) into HSCAN and ZSCAN implementations. 
Thanks to @badboy for reporting.
 2013-10-28 13:20:11 +01:00
antirez 2c643ffa8d ZSCAN implemented. 2013-10-28 11:36:42 +01:00
antirez e50090aa06 HSCAN implemented. 2013-10-28 11:35:26 +01:00
antirez 4a1f1cc0d7 SSCAN implemented. 2013-10-28 11:17:32 +01:00
antirez dfeaa84d46 dictScan(): empty hash table requires special handling. 2013-10-28 11:17:18 +01:00
antirez cd8cb49dc4 SCAN is a random command and does not require output sorting. 
Sorting the output helps when we want to turn a non-deterministic into a
deterministic command, in that case this is not possible.
 2013-10-28 11:13:43 +01:00
antirez 442ae833b2 SCAN: refactored into scanGenericCommand. 
The new implementation is capable of iterating the keyspace but also
sets, hashes, and sorted sets, and can be used to implement SSCAN, ZSCAN
and HSCAN.
 2013-10-28 11:11:34 +01:00
antirez 7bd45659b9 Fixed typos in dictScan() comment. 2013-10-25 17:05:55 +02:00
antirez 34c207227c dictScan() algorithm documented. 2013-10-25 17:01:30 +02:00
antirez 48ea8a0a11 SCAN: stay inside 80 cols. 2013-10-25 12:01:49 +02:00
antirez df13adb037 Revert "Fixed typo in SCAN comment. iff -> if." 
Probably here Pieter means "if and only if".

This reverts commit 43fdf3b404.
 2013-10-25 12:00:13 +02:00
antirez a25fe0b28d SCAN: simplify keys list cleanup using listSetFreeMethod(). 2013-10-25 11:58:03 +02:00
antirez 0471b90844 SCAN: Fix test after option renamed from PATTERN to MATCH. 2013-10-25 11:55:28 +02:00
antirez fd1b0ad07c SCAN: improve variable names for readability. 2013-10-25 11:54:45 +02:00
antirez 908eba5a8f SCAN: remove additional newlines to conform to Redis code base. 2013-10-25 11:51:08 +02:00
antirez 6e55e543c6 SCAN: remove useless assertion, already enforced by command table. 2013-10-25 11:49:08 +02:00
antirez 64722b0946 SCAN: use define REDIS_LONGSTR_SIZE instead of fixed len. 2013-10-25 11:48:18 +02:00
antirez 43fdf3b404 Fixed typo in SCAN comment. iff -> if. 2013-10-25 11:46:02 +02:00
antirez 9c88ace927 SCAN option name changed: pattern -> match. 2013-10-25 11:45:32 +02:00
Pieter Noordhuis b63fbea5e4 Fix error in scan algorithm 
The irrelevant bits shouldn't be masked to 1. This can result in slots being
skipped when the hash table is resized between calls to the iterator.
 2013-10-25 10:50:03 +02:00
Pieter Noordhuis 7a6cfb18f3 SCAN requires at least 1 argument 2013-10-25 10:49:56 +02:00
Pieter Noordhuis 7f490b197f Add SCAN command 2013-10-25 10:49:48 +02:00
antirez 85f52ebcd8 Cluster: rough support for sub-command options in redis-trib. 2013-10-11 17:33:19 +02:00
antirez 0c9f60a628 Cluster: there is a lower limit for the handshake timeout. 2013-10-11 10:34:32 +02:00
antirez 1447d28c0f Cluster: data_age conversion to milliseconds fixed. 2013-10-09 16:36:06 +02:00
antirez 573c2fea91 Cluster: clusterCron() freq is now 10h. Still ping 1 node every sec. 
After the change in clusterCron() frequency of call, we still want to
ping just one random node every second.
 2013-10-09 16:29:17 +02:00
antirez e5864c73b6 Cluster: example redis.conf updated from sec to ms for cluster-node-timeout. 2013-10-09 16:21:27 +02:00
antirez ba42428633 Cluster: time switched from seconds to milliseconds. 
All the internal state of cluster involving time is now using mstime_t
and mstime() in order to use milliseconds resolution.

Also the clusterCron() function is called with a 10 hz frequency instead
of 1 hz.

The cluster node_timeout must be also configured in milliseconds by the
user in redis.conf.
 2013-10-09 16:19:26 +02:00
antirez 929b6a4480 Cluster: cluster stuff moved from redis.h to cluster.h. 2013-10-09 15:38:05 +02:00
antirez 6fa9b1a420 Merge branch 'bettercluster' into unstable 2013-10-08 13:04:33 +02:00
antirez ae2763f564 Cluster: masters don't vote for a slave with stale config. 
When a slave requests our vote, the configEpoch he claims for its master
and the set of served slots must be greater or equal to the configEpoch
of the nodes serving these slots in the current configuraiton of the
master granting its vote.

In other terms, masters don't vote for slaves having a stale
configuration for the slots they want to serve.
 2013-10-08 12:45:35 +02:00
antirez f7d6ad4366 Cluster: fix slave data age computation when master is still connected. 2013-10-07 16:07:13 +02:00
antirez 2c3301b9f5 Cluster: log message improved when FAIL is cleared from a slave node. 2013-10-07 15:44:58 +02:00
antirez 72f38cd70f Cluster: slave nodes advertise master slots bitmap and configEpoch. 2013-10-07 11:31:12 +02:00
antirez 0150c70b2b Replication: install the write handler when reusing a cached master. 
Sometimes when we resurrect a cached master after a successful partial
resynchronization attempt, there is pending data in the output buffers
of the client structure representing the master (likely REPLCONF ACK
commands).

If we don't reinstall the write handler, it will never be installed
again by addReply*() family functions as they'll assume that if there is
already data pending, the write handler is already installed.

This bug caused some slaves after a successful partial sync to never
send REPLCONF ACK, and continuously being detected as timing out by the
master, with a disconnection / reconnection loop.
 2013-10-04 16:14:54 +02:00
antirez 1461422ce6 Replication: install the write handler when reusing a cached master. 
Sometimes when we resurrect a cached master after a successful partial
resynchronization attempt, there is pending data in the output buffers
of the client structure representing the master (likely REPLCONF ACK
commands).

If we don't reinstall the write handler, it will never be installed
again by addReply*() family functions as they'll assume that if there is
already data pending, the write handler is already installed.

This bug caused some slaves after a successful partial sync to never
send REPLCONF ACK, and continuously being detected as timing out by the
master, with a disconnection / reconnection loop.
 2013-10-04 16:12:25 +02:00
antirez 6d8c2a4848 Replication: fix master timeout. 
Since we started sending REPLCONF ACK from slaves to masters, the
lastinteraction field of the client structure is always refreshed as
soon as there is room in the socket output buffer, so masters in timeout
are detected with too much delay (the socket buffer takes a lot of time
to be filled by small REPLCONF ACK <number> entries).

This commit only counts data received as interactions with a master,
solving the issue.
 2013-10-04 13:01:45 +02:00
antirez b41570f719 Replication: fix master timeout. 
Since we started sending REPLCONF ACK from slaves to masters, the
lastinteraction field of the client structure is always refreshed as
soon as there is room in the socket output buffer, so masters in timeout
are detected with too much delay (the socket buffer takes a lot of time
to be filled by small REPLCONF ACK <number> entries).

This commit only counts data received as interactions with a master,
solving the issue.
 2013-10-04 12:59:24 +02:00
antirez d62ae1ec05 PSYNC: safer handling of PSYNC requests. 
There was a bug that over-esteemed the amount of backlog available,
however this could only happen when a slave was asking for an offset
that was in the "future" compared to the master replication backlog.

Now this case is handled well and logged as an incident in the master
log file.
 2013-10-04 12:27:30 +02:00
antirez 4f9a69399b Add REWRITE to CONFIG subcommands help message. 2013-10-04 12:27:26 +02:00
antirez 37e06bd952 PSYNC: safer handling of PSYNC requests. 
There was a bug that over-esteemed the amount of backlog available,
however this could only happen when a slave was asking for an offset
that was in the "future" compared to the master replication backlog.

Now this case is handled well and logged as an incident in the master
log file.
 2013-10-04 12:25:09 +02:00
antirez 7afc0dd59a Cluster: new clusterDoBeforeSleep() API. 
The new API is able to remember operations to perform before returning
to the event loop, such as checking if there is the failover quorum for
a slave, save and fsync the configuraiton file, and so forth.

Because this operations are performed before returning on the event
loop we are sure that messages that are sent in the same event loop run
will be delivered *after* the configuration is already saved, that is a
requirement sometimes. For instance we want to publish a new epoch only
when it is already stored in nodes.conf in order to avoid returning back
in the logical clock when a node is restarted.

This new API provides a big performance advantage compared to saving and
possibly fsyncing the configuration file multiple times in the same
event loop run, especially in the case of big clusters with tens or
hundreds of nodes.
 2013-10-03 09:58:06 +02:00
antirez 211dcbe339 Cluster: update cluster config when slave changes master. 2013-10-02 12:27:12 +02:00
antirez 6c4d904baf Cluster: bus messages stats in CLUSTER info. 2013-10-02 10:10:08 +02:00
antirez abe81781ae Cluster: FAIL messages from unknown senders are handled better. 
Previously the event was not logged but instead the node reported an
unknown packet type received.
 2013-10-02 09:42:45 +02:00
antirez 7970ebd80a Cluster: senderCurrentEpoch == node currentEpoch was too strict. 
We can accept a vote as long as its epoch is >= the epoch at which we
started the voting process. There is no need for it to be exactly the
same.
 2013-10-01 17:21:28 +02:00
antirez f1bfd8233b Cluster: fix typo in clusterProcessPacket() comment. 2013-10-01 15:40:20 +02:00
antirez 1dedf9aa36 Cluster: time field removed from cluster messages header. 
The new algorithm does not check replies time as checking for the
currentEpoch in the reply ensures that the reply is about the current
election process.
 2013-09-30 16:19:44 +02:00
antirez 2b93a19537 Add REWRITE to CONFIG subcommands help message. 2013-09-30 11:53:18 +02:00
antirez 2d0844ee37 Cluster: log message shortened. 2013-09-30 11:51:58 +02:00
antirez 707ff0f714 Make clear that runids are not cluster node IDs. 2013-09-30 11:48:09 +02:00
antirez 4dc247eb31 Cluster: detect cluster reconfiguration when master slots drop to 0. 
The old algorithm used a PROMOTED flag and explicitly checks about
slave->master convertions. Wit the new cluster meta-data propagation
algorithm we just look at the configEpoch to check if we need to
reconfigure slots, then:

1) If a node is a master but it reaches zero served slots becuase of
reconfiguration.
2) If a node is a slave but the master reaches zero served slots because
of a reconfiguration.

We switch as a replica of the new slots owner.
 2013-09-30 11:45:26 +02:00
antirez 62b1591439 Cluster: re-order failover operations to make it safer. 
We need to:

1) Increment the configEpoch.
2) Save it to disk and fsync the file.
3) Broadcast the PONG with the new configuration.

If other nodes will receive the updated configuration we need to be sure
to restart with this new config in the event of a crash.
 2013-09-30 10:16:48 +02:00
antirez b187517719 Cluster: when upading the configEpoch for a node, save config on disk ASAP. 2013-09-30 10:16:25 +02:00
antirez 03ca903983 Cluster: fsync data when saving the cluster config. 2013-09-30 10:13:07 +02:00
antirez 026e63392e Cluster: update the node configEpoch when newer is detected. 2013-09-27 09:55:41 +02:00
antirez 7c4b8f29e7 Cluster: react faster when a slave wins an election. 2013-09-26 16:54:43 +02:00
antirez 42fa46e49a Cluster: removed an old source of delay to start the slave failover. 2013-09-26 13:28:19 +02:00
antirez a445aa30a0 Cluster: master node now uses new protocol to vote. 2013-09-26 13:00:41 +02:00
antirez fb9b76fe14 Cluster: slave node now uses the new protocol to get elected. 2013-09-26 11:13:17 +02:00
Michel Martens 347ab78e90 Document the redis-cli --csv option. 2013-09-26 10:12:46 +02:00
antirez 656c3ffe4a Cluster: fix redis-trib node config fingerprinting for new nodes format. 2013-09-25 12:58:06 +02:00
antirez 341ed1d1a8 Cluster: fix redis-trib for added configEpoch field in CLUSTER NODES. 2013-09-25 12:44:56 +02:00
antirez 32b5410af9 Cluster: add currentEpoch to CLUSTER INFO. 2013-09-25 12:38:36 +02:00
antirez 6ec795d2cf Cluster: update our currentEpoch when a greater one is seen. 2013-09-25 12:36:29 +02:00
antirez d426ada891 Cluster: broadcast currentEpoch and configEpoch in packets header. 2013-09-25 11:53:35 +02:00
antirez 12483b0061 Cluster: configEpoch added in cluster nodes description. 2013-09-25 11:47:13 +02:00
antirez da257afe57 htonu64() and ntohu64 added to endianconv.h. 2013-09-25 09:26:36 +02:00
antirez 3c9bb8751a Cluster: PFAIL -> FAIL transition allowed for slaves. 
First change: now there is no need to be a master in order to detect a
failure, however the majority of masters signaling PFAIL or FAIL is needed.

This change is important because it allows slaves rejoining the cluster
after a partition to sense the FAIL condition so that eventually all the
nodes agree on failures.
 2013-09-20 11:26:44 +02:00
antirez 925ea9f858 Cluster: added time field in cluster bus messages. 
The time is sent in requests, and copied back in reply packets.
This way the receiver can compare the time field in a reply with its
local clock and check the age of the request associated with this reply.

This is an easy way to discard delayed replies. Note that only a clock
is used here, that is the one of the node sending the packet. The
receiver only copies the field back into the reply, so no
synchronization is needed between clocks of different hosts.
 2013-09-20 09:22:21 +02:00
antirez 7bec743e66 Allow AUTH / PING when disconnected from slave and serve-stale-data is no. 2013-09-17 09:46:06 +02:00
antirez d0e327413b Cluster: don't add an handshake node for the same ip:port pair multiple times. 2013-09-04 15:52:16 +02:00
antirez 72587e6cc5 Cluster: free HANDSHAKE nodes after node_timeout. 
Handshake nodes should turn into normal nodes or be freed in a
reasonable amount of time, otherwise they'll keep accumulating if the
address they are associated with is not reachable for some reason.
 2013-09-04 12:41:21 +02:00
antirez 2debce325b redis-cli: fix big keys search when the key no longer exist. 
The code freed a reply object that was never created, resulting in a
segfault every time randomkey returned a key that was deleted before we
queried it for size.
 2013-09-04 10:35:53 +02:00
antirez 8eff339ca4 Cluster: CLUSTER SAVECONFIG command added. 2013-09-04 10:33:00 +02:00
antirez 528201ad6c Cluster: don't save HANDSHAKE nodes in nodes.conf. 2013-09-04 10:25:26 +02:00
antirez e5d5da6f7c Cluster: always use safe iteartors to iterate server.cluster->nodes. 2013-09-04 10:07:50 +02:00
Maxim Zakharov 43fe2d0d97 mistype fixed 2013-09-03 15:15:51 +02:00
Maxim Zakharov 70e82e5c79 A mistype fixed 2013-09-03 15:15:48 +02:00
antirez 354a5de270 Cluster: clusterReadHandler() reworked to be more correct and simpler to follow. 2013-09-03 11:43:52 +02:00
antirez 1036b4b21b Cluster: use non-blocking I/O for the cluster bus. 2013-09-03 11:43:52 +02:00
antirez f6efb6cdec Cluster: fixed a bug in clusterSendPublish() due to inverted statements. 
The code used to copy the header *after* the 'hdr' pointer was already
switched to the new buffer. Of course we need to do the reverse.
 2013-09-03 11:43:43 +02:00
antirez e9d97b453e Test: Lua stack leak regression test added. 2013-08-30 08:59:11 +02:00
antirez f79b1cb49e Test: added a memory efficiency test. 2013-08-29 16:23:57 +02:00
antirez 41d3147344 Fixed critical memory leak from EVAL. 
Multiple missing calls to lua_pop prevented the error handler function
pushed on the stack for lua_pcall() to be popped before returning,
causing a memory leak in almost all the code paths of EVAL (both
successful calls and calls returning errors).

This caused two issues: Lua leaking memory (and this was very visible
from INFO memory output, as the 'used_memory_lua' field reported an
always increasing amount of memory used), and as a result slower and
slower GC cycles resulting in all the CPU being used.

Thanks to Tanguy Le Barzic for noticing something was wrong with his 2.8
slave, and for creating a testing EC2 environment where I was able to
investigate the issue.
 2013-08-29 11:54:03 +02:00
yihuang 9bb91d198c fix lua_cmsgpack pack map as array 2013-08-27 15:19:25 +02:00
antirez e29e426b43 Fix an hypothetical issue in processMultibulkBuffer(). 2013-08-27 13:00:06 +02:00
antirez 8811d49009 Remove useful check from tryObjectEncoding(). 
We are sure the string is large, since when the sds optimization branch
is entered it means that it was not possible to encode it as EMBSTR for
size concerns.
 2013-08-27 12:36:52 +02:00
antirez 38b2e65e15 tryObjectEncoding(): optimize sds strings if possible. 
When no encoding is possible, at least try to reallocate the sds string
with one that does not waste memory (with free space at the end of the
buffer) when the string is large enough.
 2013-08-27 12:15:13 +02:00
antirez 6672bc8b3b tryObjectEncoding(): don't call stringl2() for too big strings. 
We are sure that a string that is longer than 21 chars cannot be
represented by a 64 bit signed integer, as -(2^64) is 21 chars:

strlen(-18446744073709551616) => 21
 2013-08-27 11:56:47 +02:00
antirez ff9d66c4a9 Don't over-allocate the sds string for large bulk requests. 
The call to sdsMakeRoomFor() did not accounted for the amount of data
already present in the query buffer, resulting into over-allocation.
 2013-08-27 11:54:38 +02:00
antirez e21803348a DEBUG SDSLEN added. 
This command is only useful for low-level debugging of memory issues due
to sds wasting memory as empty buffer at the end of the string.
 2013-08-27 11:53:49 +02:00
antirez b34126e378 Update server.lastbgsave_status when fork() fails. 2013-08-27 10:16:29 +02:00
antirez 003cc8a4f5 Only run the fast active expire cycle if master & enabled. 2013-08-27 09:31:55 +02:00
antirez 303dde3757 Don't update node pong time via gossip. 
This feature was implemented in the initial days of the Redis Cluster
implementaiton but is not a good idea at all.

1) It depends on clocks to be synchronized, that is already very bad.
2) Moreover it adds a bug where the pong time is updated via gossip so
no new PING is ever sent by the current node, with the effect of no PONG
received, no update of tables, no clearing of PFAIL flag.

In general to trust other nodes about the reachability of other nodes is
a broken distributed programming model.
 2013-08-26 16:16:25 +02:00
antirez 6ae37b0e1d Cluster: set event handler in cluster bus listening socket. 
The commit using listenToPort() introduced this bug by no longer
creating the event handler to handle incoming messages from the cluster
bus.
 2013-08-22 14:53:53 +02:00
antirez 81a6a9639a Use listenToPort() in cluster.c as well. 2013-08-22 14:05:07 +02:00
antirez 4f310e05c0 Opening TCP listening ports refactored into a function. 2013-08-22 14:01:16 +02:00
antirez 0f0cc88589 Print error message when can't bind * on any address. 2013-08-22 13:02:59 +02:00
antirez 042776aff7 Cluster: fix CLUSTER MEET ip address validation. 
This was broken by the IPv6 support patches.
 2013-08-22 11:54:28 +02:00
antirez 9cf30132cc Cluster: process MEET packets as PING packets. 
Somewhat a previous commit broken this so CLUSTER MEET was no longer
working.
 2013-08-22 11:53:28 +02:00
antirez b804afcf01 Use a safe dict.c iterator in clusterCron(). 2013-08-21 15:51:15 +02:00
antirez 35a977c499 Fix for issue #1214 simplified. 2013-08-21 11:36:09 +02:00
Salvatore Sanfilippo 038e356dbc Merge pull request #1214 from kaoshijuan/unstable 
fixed initServer fail problem
 2013-08-21 02:18:41 -07:00
antirez 7e9929e12e Use printf %zu specifier to print private_dirty. 2013-08-20 12:04:57 +02:00
antirez 1c08302edc dictFingerprint(): cast pointers to integer of same size. 2013-08-20 11:49:55 +02:00
antirez 00ddc3500c Revert "Fixed type in dict.c comment: 265 -> 256." 
This reverts commit 6253180abd.
 2013-08-19 17:25:48 +02:00
antirez 6253180abd Fixed type in dict.c comment: 265 -> 256. 2013-08-19 15:10:37 +02:00
antirez 1c75408457 assert.h replaced with redisassert.h when appropriate. 
Also a warning was suppressed by including unistd.h in redisassert.h
(needed for _exit()).
 2013-08-19 15:01:21 +02:00
antirez ca294c6b1e Added redisassert.h as drop in replacement for assert.h. 
By using redisassert.h version of assert() you get stack traces in the
log instead of a process disappearing on assertions.
 2013-08-19 15:01:15 +02:00
antirez 905d4822da dictFingerprint() fingerprinting made more robust. 
The previous hashing used the trivial algorithm of xoring the integers
together. This is not optimal as it is very likely that different
hash table setups will hash the same, for instance an hash table at the
start of the rehashing process, and at the end, will have the same
fingerprint.

Now we hash N integers in a smarter way, by summing every integer to the
previous hash, and taking the integer hashing again (see the code for
further details). This way it is a lot less likely that we get a
collision. Moreover this way of hashing explicitly protects from the
same set of integers in a different order to hash to the same number.

This commit is related to issue #1240.
 2013-08-19 15:01:12 +02:00
antirez 3039e806d8 Fix comments for correctness in zunionInterGenericCommand(). 
Related to issue #1240.
 2013-08-19 15:01:05 +02:00
antirez cfb9d025c6 Properly init/release iterators in zunionInterGenericCommand(). 
This commit does mainly two things:

1) It fixes zunionInterGenericCommand() by removing mass-initialization
of all the iterators used, so that we don't violate the unsafe iterator
API of dictionaries. This fixes issue #1240.

2) Since the zui* APIs required the allocator to be initialized in the
zsetopsrc structure in order to use non-iterator related APIs, this
commit fixes this strict requirement by accessing objects directly via
the op->subject->ptr pointer we have to the object.
 2013-08-19 15:01:01 +02:00
antirez 48cde3fe47 dict.c iterator API misuse protection. 
dict.c allows the user to create unsafe iterators, that are iterators
that will not touch the dictionary data structure in any way, preventing
copy on write, but at the same time are limited in their usage.

The limitation is that when itearting with an unsafe iterator, no call
to other dictionary functions must be done inside the iteration loop,
otherwise the dictionary may be incrementally rehashed resulting into
missing elements in the set of the elements returned by the iterator.

However after introducing this kind of iterators a number of bugs were
found due to misuses of the API, and we are still finding
bugs about this issue. The bugs are not trivial to track because the
effect is just missing elements during the iteartion.

This commit introduces auto-detection of the API misuse. The idea is
that an unsafe iterator has a contract: from initialization to the
release of the iterator the dictionary should not change.

So we take a fingerprint of the dictionary state, xoring a few important
dict properties when the unsafe iteartor is initialized. We later check
when the iterator is released if the fingerprint is still the same. If it
is not, we found a misuse of the iterator, as not allowed API calls
changed the internal state of the dictionary.

This code was checked against a real bug, issue #1240.

This is what Redis prints (aborting) when a misuse is detected:

Assertion failed: (iter->fingerprint == dictFingerprint(iter->d)),
function dictReleaseIterator, file dict.c, line 587.
 2013-08-19 15:00:57 +02:00
antirez 7398930756 Use precomptued objects for bulk and mbulk prefixes. 2013-08-12 12:50:49 +02:00
antirez c06de115af replicationFeedSlaves() func name typo: feedReplicationBacklogWithObject -> feedReplicationBacklog. 2013-08-12 12:50:45 +02:00
antirez dcc48a8143 replicationFeedSlave() reworked for correctness and speed. 
The previous code using a static buffer as an optimization was lame:

1) Premature optimization, actually it was *slower* than naive code
   because resulted into the creation / destruction of the object
   encapsulating the output buffer.
2) The code was very hard to test, since it was needed to have specific
   tests for command lines exceeding the size of the static buffer.
3) As a result of "2" the code was bugged as the current tests were not
   able to stress specific corner cases.

It was replaced with easy to understand code that is safer and faster.
 2013-08-12 12:50:29 +02:00
antirez aa05128f51 Fix a PSYNC bug caused by a variable name typo. 2013-08-12 11:51:35 +02:00
antirez 610224d0d0 Fix sdsempty() prototype in sds.h. 2013-08-12 11:38:21 +02:00
antirez 89ffba9133 Replication: better way to send a preamble before RDB payload. 
During the replication full resynchronization process, the RDB file is
transfered from the master to the slave. However there is a short
preamble to send, that is currently just the bulk payload length of the
file in the usual Redis form $..length..<CR><LF>.

This preamble used to be sent with a direct write call, assuming that
there was alway room in the socket output buffer to hold the few bytes
needed, however this does not scale in case we'll need to send more
stuff, and is not very robust code in general.

This commit introduces a more general mechanism to send a preamble up to
2GB in size (the max length of an sds string) in a non blocking way.
 2013-08-12 10:29:14 +02:00
antirez db862e8ef0 redis-benchmark: changes to random arguments substitution. 
Before this commit redis-benchmark supported random argumetns in the
form of :rand:000000000000. In every string of that form, the zeros were
replaced with a random number of 12 digits at every command invocation.

However this was far from perfect as did not allowed to generate simply
random numbers as arguments, there was always the :rand: prefix.

Now instead every argument in the form __rand_int__ is replaced with a
12 digits number. Note that "__rand_int__" is 12 characters itself.

In order to implement the new semantic, it was needed to change a few
thigns in the internals of redis-benchmark, as new clients are created
cloning old clients, so without a stable prefix such as ":rand:" the old
way of cloning the client was no longer able to understand, from the old
command line, what was the position of the random strings to substitute.

Now instead a client structure is passed as a reference for cloning, so
that we can directly clone the offsets inside the command line.
 2013-08-08 16:42:08 +02:00
antirez 92ab77f8d5 redis-benchmark: replace snprintf()+memcpy with faster code. 
This change was profiler-driven, but the actual effect is hard to
measure in real-world redis benchmark runs.
 2013-08-08 14:33:14 +02:00
Salvatore Sanfilippo de698798bc Merge pull request #1234 from badboy/patch-2 
Little typo
 2013-08-07 07:08:04 -07:00
Jan-Erik Rediger 74c0af21f5 Little typo 2013-08-07 16:05:09 +02:00
antirez 36a0947185 redis-benchmark: fix memory leak introduced by 346256f 2013-08-07 16:00:18 +02:00
antirez 346256f933 redis-benchmark: max pipeline length hardcoded limit removed. 2013-08-07 15:58:58 +02:00
antirez 6cbfdd9520 redis-benchmark: fix db selection when :rand: feature is used. 2013-08-06 19:01:54 +02:00
antirez d52c9b6cdb redis-benchmark: ability to SELECT a specifid db number. 2013-08-06 18:50:54 +02:00
antirez 112fa47978 Add per-db average TTL information in INFO output. 
Example:

db0:keys=221913,expires=221913,avg_ttl=655

The algorithm uses a running average with only two samples (current and
previous). Keys found to be expired are considered at TTL zero even if
the actual TTL can be negative.

The TTL is reported in milliseconds.
 2013-08-06 15:00:43 +02:00
antirez 4befe73b60 activeExpireCycle(): fix about fast cycle early start. 
We don't want to repeat a fast cycle too soon, the previous code was
broken, we need to wait two times the period *since* the start of the
previous cycle in order to avoid there is an even space between cycles:

.-> start                   .-> second start
|                           |
+-------------+-------------+--------------+
| first cycle |    pause    | second cycle |
+-------------+-------------+--------------+

The second and first start must be PERIOD*2 useconds apart hence the *2
in the new code.
 2013-08-06 12:59:04 +02:00
antirez 6500fabfb8 Some activeExpireCycle() refactoring. 2013-08-06 12:55:49 +02:00
antirez d398f38879 Remove dead code and fix comments for new expire code. 2013-08-06 12:36:13 +02:00
antirez 66a26471dc Darft #2 for key collection algo: more improvements. 
This commit makes the fast collection cycle time configurable, at
the same time it does not allow to run a new fast collection cycle
for the same amount of time as the max duration of the fast
collection cycle.
 2013-08-05 16:14:28 +02:00
antirez b09ea1bd90 Draft #1 of a new expired keys collection algorithm. 
The main idea here is that when we are no longer to expire keys at the
rate the are created, we can't block more in the normal expire cycle as
this would result in too big latency spikes.

For this reason the commit introduces a "fast" expire cycle that does
not run for more than 1 millisecond but is called in the beforeSleep()
hook of the event loop, so much more often, and with a frequency bound
to the frequency of executed commnads.

The fast expire cycle is only called when the standard expiration
algorithm runs out of time, that is, consumed more than
REDIS_EXPIRELOOKUPS_TIME_PERC of CPU in a given cycle without being able
to take the number of already expired keys that are yet not collected
to a number smaller than 25% of the number of keys.

You can test this commit with different loads, but a simple way is to
use the following:

Extreme load with pipelining:

redis-benchmark -r 100000000 -n 100000000  \
        -P 32 set ele:rand:000000000000 foo ex 2

Remove the -P32 in order to avoid the pipelining for a more real-world
load.

In another terminal tab you can monitor the Redis behavior with:

redis-cli -i 0.1 -r -1 info keyspace

and

redis-cli --latency-history

Note: this commit will make Redis printing a lot of debug messages, it
is not a good idea to use it in production.
 2013-08-05 12:05:22 +02:00
antirez cd0ea1f202 Test: regression test for issue #1221. 2013-07-29 17:39:28 +02:00
antirez c151eb6d92 Fix replicationFeedSlaves() off-by-one bug. 
This fixes issue #1221.
 2013-07-28 12:49:34 +02:00
antirez bf56948fd0 Remove dead variable bothsds from object.c. 
Thanks to @run and @badboy for spotting this.
Triva: clang was not able to provide me a warning about that when
compiling.

This closes #1024 and #1207, committing the change myself as the pull
requests no longer apply cleanly after other changes to the same
function.
 2013-07-28 11:00:09 +02:00
antirez 320fa02b9b Use latest sds.c in the hiredis library under deps. 2013-07-25 10:32:19 +02:00
Allan a0e986d7f2 fixed initServer fail while having no IPv6 nor IPv4 2013-07-25 15:36:00 +08:00
Allan cba7a4e69a fixed initServer failed if no IPV4 or no IPV6 2013-07-25 15:28:33 +08:00
antirez 3ccc17a4a5 Ignore sdsrange return value. 2013-07-24 18:59:54 +02:00
Allan 1e7cff23b3 fixed bug issue of #1213 2013-07-24 21:34:55 +08:00
antirez 6ea8e0949c sdsrange() does not need to return a value. 
Actaully the string is modified in-place and a reallocation is never
needed, so there is no need to return the new sds string pointer as
return value of the function, that is now just "void".
 2013-07-24 11:21:39 +02:00
antirez 75b760a72d Inline protocol improved to accept quoted strings. 2013-07-24 10:37:55 +02:00
antirez 076c2623ac Every function inside sds.c is now commented. 2013-07-23 16:35:55 +02:00
antirez ec7f480e11 getStringObjectSdsUsedMemory() function added. 
Now that EMBSTR encoding exists we calculate the amount of memory used
by the SDS part of a Redis String object in two different ways:

1) For raw string object, the size of the allocation is considered.
2) For embstr objects, the length of the string itself is used.

The new function takes care of this logic.
 2013-07-23 11:50:17 +02:00
antirez ddccd0ed58 Test: regression test for issue #1208. 2013-07-22 23:40:48 +02:00
antirez dbaa5b0b9a Fix setDeferredMultiBulkLength() c->reply_bytes handling with EMBSTR 
This function missed proper handling of reply_bytes when gluing to the
previous object was used. The issue was introduced with the EMBSTR new
string object encoding.

This fixes issue #1208.
 2013-07-22 23:29:12 +02:00
antirez 7ed7652846 Fixed a possible bug in client->reply_bytes computation. 2013-07-22 11:05:55 +02:00
antirez a31693417d Fix replicationFeedSlaves() to use sdsEncodedObject() macro. 2013-07-22 10:36:27 +02:00
antirez 894eba07c8 Introduction of a new string encoding: EMBSTR 
Previously two string encodings were used for string objects:

1) REDIS_ENCODING_RAW: a string object with obj->ptr pointing to an sds
stirng.

2) REDIS_ENCODING_INT: a string object where the obj->ptr void pointer
is casted to a long.

This commit introduces a experimental new encoding called
REDIS_ENCODING_EMBSTR that implements an object represented by an sds
string that is not modifiable but allocated in the same memory chunk as
the robj structure itself.

The chunk looks like the following:

+--------------+-----------+------------+--------+----+
| robj data... | robj->ptr | sds header | string | \0 |
+--------------+-----+-----+------------+--------+----+
                     |                       ^
                     +-----------------------+

The robj->ptr points to the contiguous sds string data, so the object
can be manipulated with the same functions used to manipulate plan
string objects, however we need just on malloc and one free in order to
allocate or release this kind of objects. Moreover it has better cache
locality.

This new allocation strategy should benefit both the memory usage and
the performances. A performance gain between 60 and 70% was observed
during micro-benchmarks, however there is more work to do to evaluate
the performance impact and the memory usage behavior.
 2013-07-22 10:31:38 +02:00
antirez b9cc90a119 addReplyDouble(): format infinite in a libc agnostic way. 
There are systems that when printing +/- infinte with printf-family
functions will not use the usual "inf" "-inf", but different strings.
Handle that explicitly.

Fixes issue #930.
 2013-07-17 15:05:25 +02:00
antirez f590dd82ce Fixed typo in rio.h, simgle -> single. 2013-07-16 15:43:36 +02:00
yoav 63d15dfc87 Chunked loading of RDB to prevent redis from stalling reading very large keys. 2013-07-16 15:41:24 +02:00
antirez 9d520a7f70 Make sure that ZADD can accept the full range of double values. 
This fixes issue #1194, that contains many details.

However in short, it was possible for ZADD to not accept as score values
that was however possible to obtain with multiple calls to ZINCRBY, like
in the following example:

redis 127.0.0.1:6379> zadd k 2.5e-308 m
(integer) 1
redis 127.0.0.1:6379> zincrby k -2.4e-308 m
"9.9999999999999694e-310"
redis 127.0.0.1:6379> zscore k m
"9.9999999999999694e-310"
redis 127.0.0.1:6379> zadd k 9.9999999999999694e-310 m1
(error) ERR value is not a valid float

The problem was due to strtod() returning ERANGE in the following case
specified by POSIX:

"If the correct value would cause an underflow, a value whose magnitude
is no greater than the smallest normalized positive number in the return
type shall be returned and errno set to [ERANGE].".

Now instead the returned value is accepted even when ERANGE is returned
as long as the return value of the function is not negative or positive
HUGE_VAL or zero.
 2013-07-16 15:05:13 +02:00
Salvatore Sanfilippo 34e2065830 Merge pull request #1193 from tnm/patch-1 
Make sure the log standardizes on 'timeout'
 2013-07-12 14:12:04 -07:00
Ted Nyman f39a0bdb77 Make sure the log standardizes on 'timeout' 2013-07-12 14:06:27 -07:00
antirez 123b221dc9 Use the environment locale for strcoll() collation. 2013-07-12 13:38:43 +02:00
antirez cf1579a798 SORT ALPHA: use collation instead of binary comparison. 
Note that we only do it when STORE is not used, otherwise we want an
absolutely locale independent and binary safe sorting in order to ensure
AOF / replication consistency.

This is probably an unexpected behavior violating the least surprise
rule, but there is currently no other simple / good alternative.
 2013-07-12 12:02:36 +02:00
antirez 81e55ec0f3 Fixed compareStringObject() and introduced collateStringObject(). 
compareStringObject was not always giving the same result when comparing
two exact strings, but encoded as integers or as sds strings, since it
switched to strcmp() when at least one of the strings were not sds
encoded.

For instance the two strings "123" and "123\x00456", where the first
string was integer encoded, would result into the old implementation of
compareStringObject() to return 0 as if the strings were equal, while
instead the second string is "greater" than the first in a binary
comparison.

The same compasion, but with "123" encoded as sds string, would instead
return a value < 0, as it is correct. It is not impossible that the
above caused some obscure bug, since the comparison was not always
deterministic, and compareStringObject() is used in the implementation
of skiplists, hash tables, and so forth.

At the same time, collateStringObject() was introduced by this commit, so
that can be used by SORT command to return sorted strings usign
collation instead of binary comparison. See next commit.
 2013-07-12 11:56:52 +02:00
Salvatore Sanfilippo 2ed64e1f7a Merge pull request #1192 from badboy/ipv6-rediscli 
Wrap IPv6 in brackets in the prompt.
 2013-07-12 01:55:09 -07:00
Jan-Erik Rediger 1b696dc07c Wrap IPv6 in brackets in the prompt. 2013-07-11 17:47:55 +02:00
antirez 73ae8558c1 Sentinel: embed IPv6 address into [] when naming slave/sentinel instance. 2013-07-11 16:38:40 +02:00
antirez 3fc7f324d2 Sentinel: use comma as separator to publish hello messages. 
We use comma to play well with IPv6 addresses, but the implementation is
still able to parse the old messages separated by colons.
 2013-07-11 16:37:47 +02:00
antirez 4735f03c5a hiredis: minimal IPv6 support. 2013-07-11 13:11:27 +02:00
antirez 5c5ebb0b9a Sentinel: make sure published addr/id buffer is large enough. 
With ipv6 support we need more space, so we account for the IP address
max size plus what we need for the Run ID, port, flags.
 2013-07-10 14:44:38 +02:00
antirez bd3c897e9f anet.c: save some vertical space. 2013-07-10 14:37:13 +02:00
antirez d83dca4c79 anet.c: use SO_REUSEADDR when creating listening sockets. 
It used to be ok, but the socket option was removed when adding IPv6
support.
 2013-07-10 14:34:58 +02:00
antirez d1cbad6d14 Use getClientPeerId() for MONITOR implementation. 2013-07-09 16:21:21 +02:00
antirez d0001fe810 getClientPeerId() refactored into two functions. 2013-07-09 15:46:34 +02:00
antirez e4c019e7a8 getClientPeerId() now reports errors. 
We now also use it in CLIENT KILL implementation.
 2013-07-09 15:28:30 +02:00
antirez 5cdc5da990 getClientPeerID introduced. 
The function returns an unique identifier for the client, as ip:port for
IPv4 and IPv6 clients, or as path:0 for Unix socket clients.

See the top comment in the function for more info.
 2013-07-09 12:49:20 +02:00
antirez 631d656a94 All IP string repr buffers are now REDIS_IP_STR_LEN bytes. 2013-07-09 11:32:52 +02:00
antirez f19e267e9a IPv6: bind IPv4 and IPv6 interfaces by default. 2013-07-09 10:47:17 +02:00
antirez 90038906f4 Fix old anetPeerToString() API call in replication.c 2013-07-08 16:11:52 +02:00
antirez c76df73010 Revert "Document port6 and bind6 config options." 
IPv6 support is not going to use IPv6 specific options, just it will be
possible to specify all the ipv4 / ipv6 addresses of the interfaces to
bind, otherwise connections will be accepted from all the interfaces in
both IPv4 and IPv6 addresses.

This reverts commit 93570e179e.
 2013-07-08 16:08:43 +02:00
Geoff Garside e04fdf26fe Add IPv6 support to sentinel.c. 
This has been done by exposing the anetSockName() function anet.c
to be used when the sentinel is publishing its existence to the masters.

This implementation is very unintelligent as it will likely break if used
with IPv6 as the nested colons will break any parsing of the PUBLISH string
by the master.
 2013-07-08 16:08:36 +02:00
Geoff Garside a68e3d4c6a Cleanup main() and BACKTRACE mistaken pulled while rebasing. 2013-07-08 16:07:26 +02:00
Geoff Garside 1ca4008d14 Fix calls to anetPeerToString() missing buffer size. 2013-07-08 16:07:26 +02:00
Geoff Garside 93570e179e Document port6 and bind6 config options. 
Add commented port6 and bind6 options to default redis.conf file.
 2013-07-08 16:06:44 +02:00
Geoff Garside 56df827198 Add anetTcp6Server() function. 
Refactor the common code from anetTcpServer into internal function which
can be used by both anetTcpServer and anetTcp6Server.
 2013-07-08 15:58:14 +02:00
Geoff Garside 72a3922617 Add static anetV6Only() function. 
This function sets the IPV6_V6ONLY option to 1 to use separate stack
IPv6 sockets.
 2013-07-08 15:58:14 +02:00
Geoff Garside 62a3b7e3d9 Change anetTcpGenericConnect to use AF_UNSPEC. 
This allows anetTcpGenericConnect to try to connect to AF_INET6
addresses in addition to any resolved AF_INET addresses.
 2013-07-08 15:58:14 +02:00
Geoff Garside ca78446c55 Mark places that might want changing for IPv6. 
Any places which I feel might want to be updated to work differently
with IPv6 have been marked with a comment starting "IPV6:".

Currently the only comments address places where an IP address is
combined with a port using the standard : separated form. These may want
to be changed when printing IPv6 addresses to wrap the address in []
such as

	[2001:db8::c0:ffee]:6379

instead of

	2001:db8::c0:ffee:6379

as the latter format is a technically valid IPv6 address and it is hard
to distinguish the IPv6 address component from the port unless you know
the port is supposed to be there.
 2013-07-08 15:58:14 +02:00
Geoff Garside 96b02dc055 Expand ip char buffers which are too small for v6. 
Increase the size of character buffers being used to store printable IP
addresses so that they can safely store IPv6 addresses.
 2013-07-08 15:58:14 +02:00
Geoff Garside f7d9a92d4e Mark ip string buffers which could be reduced. 
In two places buffers have been created with a size of 128 bytes which
could be reduced to INET6_ADDRSTRLEN to still hold a full IP address.
These places have been marked as they are presently big enough to handle
the needs of storing a printable IPv6 address.
 2013-07-08 15:57:23 +02:00
Geoff Garside e6bf4c2676 Update clusterCommand to handle AF_INET6 addresses 
Changes the sockaddr_in to a sockaddr_storage. Attempts to convert the
IP address into an AF_INET or AF_INET6 before returning an "Invalid IP
address" error. Handles converting the sockaddr from either AF_INET or
AF_INET6 back into a string for storage in the clusterNode ip field.
 2013-07-08 15:57:23 +02:00
Geoff Garside 5be83eecac Update node2IpString to handle AF_INET6 addresses. 
Change the sockaddr_in to sockaddr_storage which is capable of storing
both AF_INET and AF_INET6 sockets. Uses the sockaddr_storage ss_family
to correctly return the printable IP address and port.

Function makes the assumption that the buffer is of at least
REDIS_CLUSTER_IPLEN bytes in size.
 2013-07-08 15:57:23 +02:00
Geoff Garside 6181455ac6 Update REDIS_CLUSTER_IPLEN to INET6_ADDRSTRLEN. 
Change REDIS_CLUSTER_IPLEN to INET6_ADDRSTRLEN so that the clusterNode
ip character buffer is big enough to hold an IPv6 address.
 2013-07-08 15:57:23 +02:00
Geoff Garside 23f4d905ce Update anetPeerToString to handle AF_INET6 addrs. 
Change the sockaddr_in to sockaddr_storage which is capable of storing
both AF_INET and AF_INET6 sockets. Uses the sockaddr_storage ss_family
to correctly return the printable IP address and port.
 2013-07-08 15:57:22 +02:00
Geoff Garside fa723d98d6 Update anetTcpAccept to handle AF_INET6 addresses. 
Change the sockaddr_in to sockaddr_storage which is capable of storing
both AF_INET and AF_INET6 sockets. Uses the sockaddr_storage ss_family
to correctly return the printable IP address and port.
 2013-07-08 15:57:22 +02:00
Geoff Garside e7b34e8dc3 Update anetResolve to resolve AF_INET6 as well. 
Change the getaddrinfo(3) hints family from AF_INET to AF_UNSPEC to
allow resolution of IPv6 addresses as well as IPv4 addresses. The
function will return the IP address of whichever address family is
preferenced by the operating system. Most current operating systems
will preference AF_INET6 over AF_INET.

Unfortunately without attempting to establish a connection to the
remote address we can't know if the host is capable of using the
returned IP address. It might be desirable to have anetResolve
accept an additional argument specifying the AF_INET/AF_INET6 address
the caller would like to receive. Currently though it does not appear
as though the anetResolve function is ever used within Redis.
 2013-07-08 15:57:22 +02:00
Geoff Garside 2345cee335 Update calls to anetResolve to include buffer size 2013-07-08 15:57:22 +02:00
Geoff Garside ee5a6df101 Update calls to anetPeerToString to include ip_len. 2013-07-08 15:57:22 +02:00
Geoff Garside b39e827d22 Add missing includes for getpeername. 
getpeername(2) requires <sys/socket.h> which on some systems also
requires <sys/types.h>. Include both to avoid compilation warnings.
 2013-07-08 15:55:39 +02:00
Geoff Garside 9cfa02fe73 Add macro to define clusterNode.ip buffer size. 
Add REDIS_CLUSTER_IPLEN macro to define the size of the clusterNode ip
character array. Additionally use this macro in inet_ntop(3) calls where
the size of the array was being defined manually.

The REDIS_CLUSTER_IPLEN is defined as INET_ADDRSTRLEN which defines the
correct size of a buffer to store an IPv4 address in. The
INET_ADDRSTRLEN macro itself is defined in the <netinet/in.h> header
file and should be portable across the majority of systems.
 2013-07-08 15:55:39 +02:00
Geoff Garside 6e894f02cf Fix cluster.c inet_ntop use of sizeof(n->ip). 
Using sizeof with an array will only return expected results if the
array is created in the scope of the function where sizeof is used. This
commit changes the inet_ntop calls so that they use the fixed buffer
value as defined in redis.h which is 16.
 2013-07-08 15:51:37 +02:00
Geoff Garside 693b640510 Use inet_pton(3) in clusterCommand. 
Replace inet_aton(3) call with the more future proof inet_pton(3)
function which is capable of handling additional address families.
 2013-07-08 15:51:37 +02:00
Geoff Garside a6ea707cec Use inet_ntop(3) in nodeIp2String & clusterCommand 
Replace inet_ntoa(3) calls with the more future proof inet_ntop(3)
function which is capable of handling additional address families.
 2013-07-08 15:51:37 +02:00
Geoff Garside f5494a427e Update anetTcpAccept & anetPeerToString calls. 
Add the additional ip buffer length argument to function calls of
anetTcpAccept and anetPeerToString in network.c and cluster.c
 2013-07-08 15:51:37 +02:00
Geoff Garside ef839f9006 Use inet_ntop(3) in anet. #apichange 
Replace inet_ntoa(3) calls with the more future proof inet_ntop(3)
function which is capable of handling additional address families.

API Change: anetTcpAccept() & anetPeerToString() additional argument
  additional argument required to specify the length of the character
  buffer the IP address is written to in order to comply with
  inet_ntop(3) function semantics.
 2013-07-08 15:50:15 +02:00
Geoff Garside e0cb24351c Use getaddrinfo(3) in a anetTcpServer. 
Change anetTcpServer() function to use getaddrinfo(3) to perform
address resolution, socket creation and binding. Resolved addresses
are limited to those reachable by the AF_INET address family.
 2013-07-08 15:49:22 +02:00
Geoff Garside 0e01ce1b13 Use getaddrinfo(3) in anetTcpGenericConnect. 
Change anetTcpGenericConnect() function to use getaddrinfo(3) to
perform address resolution, socket creation and connection. Resolved
addresses are limited to those reachable by the AF_INET family.
 2013-07-08 15:49:22 +02:00
Geoff Garside 580b7dce9b Add anetSetReuseAddr(err, fd) static function. 
Extract setting SO_REUSEADDR socket option into separate function
so the same code can be more easily used by anetCreateSocket and
other functions.
 2013-07-08 15:49:22 +02:00
Geoff Garside 071963c855 Use getaddrinfo(3) in anetResolve. #apichange 
Change anetResolve() function to use getaddrinfo(3) to resolve hostnames.
Resolved hostnames are limited to those reachable by the AF_INET address
family.

API Change: anetResolve requires additional argument.
  additional argument required to specify the length of the character
  buffer the IP address is written to in order to comply with
  inet_ntop(3) function semantics. inet_ntop(3) replaces inet_ntoa(3)
  as it has been designed to be compatible with more address families.
 2013-07-08 15:47:57 +02:00
antirez 64b3b9b1d4 Example redis.conf: bind to multiple interfaces documented. 2013-07-08 10:42:22 +02:00
antirez 98eecb70eb Binding multiple IPs done properly with multiple sockets. 2013-07-05 11:47:20 +02:00
antirez 2160effc78 Revert "Cluster: use new anet.c listening socket creation API." 
This reverts commit 016ac38a21.
 2013-07-05 11:08:44 +02:00
antirez c978b864f7 Revert "anet.c: Allow creation of TCP listening sockets bound to N addresses." 
Bind() can't be called multiple times against the same socket, multiple
sockets are required to bind multiple interfaces, silly me.

This reverts commit bd234d62bb.
 2013-07-05 11:07:55 +02:00
antirez 90b0d66cce Ability to bind multiple addresses. 2013-07-04 18:50:15 +02:00
antirez 016ac38a21 Cluster: use new anet.c listening socket creation API. 2013-07-04 18:49:49 +02:00
antirez bd234d62bb anet.c: Allow creation of TCP listening sockets bound to N addresses. 2013-07-04 18:48:46 +02:00
antirez 585b0a61ce sds.c: new function sdsjoin() to join strings. 2013-07-04 18:30:59 +02:00
antirez 1135e9faa2 redis-cli: introduced --pipe-timeout. 
When in --pipe mode, after all the data transfer to the server is
complete, now redis-cli waits at max the specified amount of
seconds (30 by default, use 0 to wait forever) without receiving any
reply at all from the server. After this time limit the operation is
aborted with an error.

That's related to issue #681.
 2013-07-03 12:22:03 +02:00
antirez fbb97c6b13 redis-cli --pipe: send final ECHO in a safer way. 
If the protocol read from stdin happened to contain grabage (invalid
random chars), in the previous implementation it was possible to end
with something like:

dksfjdksjflskfjl*2\r\n$4\r\nECHO....

That is invalid as the *2 should start into a new line. Now we prefix
the ECHO with a CRLF that has no effects on the server but prevents this
issues most of the times.

Of course if the offending wrong sequence is something like:

$3248772349\r\n

No one is going to save us as Redis will wait for data in the context of
a big argument, so this fix does not cover all the cases.

This partially fixes issue #681.
 2013-07-03 11:59:44 +02:00
antirez 7e63167d27 pqsort.c: remove the "switch to insertion sort" optimization. 
It causes catastrophic performance for certain inputs.

Relevant NetBSD commit:

http://cvsweb.netbsd.org/bsdweb.cgi/src/lib/libc/stdlib/qsort.c?rev=1.20&content-type=text/x-cvsweb-markup&only_with_tag=MAIN

This fixes issue #968.
 2013-07-02 17:47:32 +02:00
Salvatore Sanfilippo 7d626d4975 Merge pull request #776 from charsyam/ziplist-bug 
fix randstring bug in ziplist.c
 2013-07-02 03:18:18 -07:00
antirez 1b10522a08 Only allow basenames for dbfilename and appendfilename. 
This fixes issue #1094.
 2013-07-02 12:14:28 +02:00
antirez 6978aeb3bf pathIsBaseName() added to utils.c 
The function is used to test that the specified string looks like just
as the basename of a path, without any absolute or relative path.
 2013-07-02 12:08:07 +02:00
antirez 0781ad6899 getAbsolutePath() moved into utils.c 2013-07-02 11:56:52 +02:00
antirez de9a221749 CONFIG SET maxclients. 2013-06-28 17:08:03 +02:00
antirez 8e2d082066 ae.c event loop: API to resize the fd set size on the run. 2013-06-28 16:39:49 +02:00
antirez 3130670b97 Allow SHUTDOWN in loading state. 2013-06-27 12:18:29 +02:00
antirez 13585dd677 function renamed: popcount_binary -> redisPopcount. 2013-06-26 15:19:06 +02:00
Salvatore Sanfilippo bae60ede1d Merge pull request #1111 from yamt/netbsd3 
netbsd support
 2013-06-26 06:17:02 -07:00
antirez 8ca265cdb7 Don't disconnect pre PSYNC replication clients for timeout. 
Clients using SYNC to replicate are older implementations, such as
redis-cli --slave, and are not designed to acknowledge the master with
REPLCONF ACK commands, so we don't have any feedback and should not
disconnect them on timeout.
 2013-06-26 10:11:20 +02:00
antirez d5f1e4b0b5 Test: add some AOF testing to EVALSHA replication test. 2013-06-25 15:49:07 +02:00
antirez 73d7955c6f Flush the replication script cache after SCRIPT FLUSH. 2013-06-25 15:36:48 +02:00
antirez 882e36366f Test: EVALSHA replication. 2013-06-25 15:35:48 +02:00
antirez d1f2d0733c Test: randomInt() behavior commented. 2013-06-25 15:32:37 +02:00
antirez 7e5be50cbf Test: replication-3 test speedup in master-slave setup. 2013-06-25 15:13:14 +02:00
antirez fb67468813 Force propagation of SCRIPT LOAD to AOF. 2013-06-25 12:49:56 +02:00
antirez e27b136069 SCRIPT FLUSH comment minor pedantic improvement. 2013-06-25 10:56:59 +02:00
antirez 82ea1c6f5d Move Replication Script Cache initialization in safer place. 
It should be called just one time at startup and not every time the Lua
scripting engine is re-initialized, otherwise memory is leaked.
 2013-06-24 19:27:49 +02:00
antirez f0bf5fd8c7 Use the RSC to replicate EVALSHA unmodified. 
This commit uses the Replication Script Cache in order to avoid
translating EVALSHA into EVAL whenever possible for both the AOF and
slaves.
 2013-06-24 18:57:31 +02:00
antirez 94ec7db470 Replication of scripts as EVALSHA: sha1 caching implemented. 
This code is only responsible to take an LRU-evicted fixed length cache
of SHA1 that we are sure all the slaves received.

In this commit only the implementation is provided, but the Redis core
does not use it to actually send EVALSHA to slaves when possible.
 2013-06-24 10:26:04 +02:00
antirez 515a26bbc1 New API to force propagation. 
The old REDIS_CMD_FORCE_REPLICATION flag was removed from the
implementation of Redis, now there is a new API to force specific
executions of a command to be propagated to AOF / Replication link:

    void forceCommandPropagation(int flags);

The new API is also compatible with Lua scripting, so a script that will
execute commands that are forced to be propagated, will also be
propagated itself accordingly even if no change to data is operated.

As a side effect, this new design fixes the issue with scripts not able
to propagate PUBLISH to slaves (issue #873).
 2013-06-21 12:07:53 +02:00
Salvatore Sanfilippo b0c2cdd6a7 Merge pull request #1167 from badboy/patch-1 
Initialize char* to NULL to remove compiler warning
 2013-06-20 07:56:31 -07:00
Jan-Erik Rediger 5ac7ca9c94 Initialize char* to NULL to remove compiler warning 2013-06-20 17:53:35 +03:00
antirez 519c9e11d1 Allow PUBSUB NUMSUB without channels. 
The result is an empty list but it is handy to call it programmatically.
 2013-06-20 15:34:56 +02:00
antirez 455563faec PUBSUB command implemented. 
Currently it implements three subcommands:

PUBSUB CHANNELS [<pattern>]    List channels with non-zero subscribers.
PUBSUB NUMSUB [channel_1 ...]  List number of subscribers for channels.
PUBSUB NUMPAT                  Return number of subscribed patterns.
 2013-06-20 15:32:00 +02:00
antirez 4c0f8c4e5a Sentinel: parse new INFO replication output correctly. 
Sentinel was not able to detect slaves when connected to a very recent
version of Redis master since a previos non-backward compatible change
to INFO broken the parsing of the slaves ip:port INFO output.

This fixes issue #1164
 2013-06-20 10:23:23 +02:00
antirez b02bb47e67 Test: regression test for #1163. 2013-06-19 18:53:07 +02:00
antirez f8ba3b5668 Fix comment typo in integration/aof.tcl. 2013-06-19 18:31:33 +02:00
antirez d363299af3 Allow writes from scripts called by AOF loading in read-only slaves. 
This fixes issue #1163
 2013-06-19 18:25:03 +02:00
antirez 338cd4835d Fix logStackTrace() when logging to stdout. 
When the semantics changed from logfile = NULL to logfile = "" to log
into standard output, no proper change was made to logStackTrace() to
make it able to work with the new setup.

This commit fixes the issue.
 2013-06-19 14:44:40 +02:00
antirez 9c2c878e45 Lua script errors format more unified. 
lua_pcall error handler now formats errors in a way more similar to
luaPushError() so that errors generated in different contexts look alike.
 2013-06-18 19:30:56 +02:00
antirez 51adc6e1bc Lua scripting: improve error reporting. 
When calling Lua scripts we try to report not just the error but
information about the code line causing the error.
 2013-06-18 17:33:35 +02:00
Salvatore Sanfilippo b96ba52cfa Merge pull request #1124 from ioddly/fix-issue-1121 
Try to report source of bad Lua API calls
 2013-06-17 03:14:31 -07:00
Salvatore Sanfilippo 82d782d3b5 Merge pull request #819 from Keruspe/unstable 
test-server: only listen to 127.0.0.1
 2013-06-17 03:08:14 -07:00
Marc-Antoine Perennou edd7eb9b7d test-server: only listen to 127.0.0.1 
Signed-off-by: Marc-Antoine Perennou <Marc-Antoine@Perennou.com>
 2013-06-12 22:49:32 +02:00
antirez dfc98dccf4 Cluster: detect nodes address change. 2013-06-12 10:50:07 -07:00
antirez d427373f01 clusterProcessPacket() comments improved for correctness. 2013-06-11 21:34:34 +02:00
antirez cf71b82111 Binary safe dump of object content in redisLogObjectDebugInfo(). 2013-06-04 15:53:53 +02:00
antirez b6a2878aa5 CONFIG SET: accept slave-priority zero, it is valid. 2013-05-31 19:31:36 +02:00
antirez ca35de1d1f Test: avoid a false positive in min-slaves test. 2013-05-31 11:43:30 +02:00
antirez 434a86dbd8 Tests added for min-slaves feature. 2013-05-30 18:54:28 +02:00
antirez 915c06a96c Refresh good slaves count after CONFIG SET min-slaves-... 
This way just after the CONFIG SET enabling the min-slaves feature it is
possible to write to the database without delays.
 2013-05-30 12:23:41 +02:00
antirez 88441bf18f New INFO field "min_slaves_good_slaves". 
When min-slaves-to-write feature is active, this field reports the
number of slaves considered good (online state, lag within the specified
range).
 2013-05-30 12:18:31 +02:00
antirez 1a54d5963e Refresh good slaves count when setting slave state as online. 2013-05-30 12:13:25 +02:00
antirez d64d2e21c9 Make tests compatible with new INFO replication output. 2013-05-30 11:43:43 +02:00
antirez 2ec7875cbf min-replicas-to-write: only deny write commands. 
I guess I needed another coffee...
 2013-05-30 11:30:09 +02:00
antirez ed599d3aca min-slaves-to-write: don't accept writes with less than N replicas. 
This feature allows the user to specify the minimum number of
connected replicas having a lag less or equal than the specified
amount of seconds for writes to be accepted.
 2013-05-30 11:30:04 +02:00
antirez cbdb2153c5 min-slaves-to-write: initial description of the feature in redis.conf 2013-05-30 11:29:58 +02:00
antirez 888400ebd5 repl_offset field in INFO replication is now just offset. 2013-05-29 19:56:33 +02:00
antirez 37c29e037b Slaves list in INFO output: lag added, format changed. 
There is a new 'lag' information in the list of slaves, in the
"replication" section of the INFO output.

Also the format was changed in a backward incompatible way in order to
make it more easy to parse if new fields are added in the future, as the
new format is comma separated but has named fields (no longer positional
fields).
 2013-05-29 19:54:44 +02:00
antirez 091ed386f7 Accept REPLCONF in any state. 2013-05-28 15:26:20 +02:00
antirez a864cae2a5 A comment about BLPOP timeout did not reflected actual behavior. 2013-05-27 19:34:14 +02:00
antirez 1909b8753d Version bumped to 2.9.11 2013-05-27 11:44:04 +02:00
antirez c9b55a29d6 redis.conf updated: repl-timeout now uesd by masters as well. 2013-05-27 11:42:45 +02:00
antirez 3c82c85fcf Close connection with timedout slaves. 
Now masters, using the time at which the last REPLCONF ACK was received,
are able to explicitly disconnect slaves that are no longer responding.

Previously the only chance was to see a very long output buffer, that
was highly suboptimal.
 2013-05-27 11:42:42 +02:00
antirez e06a560466 Send ACK to master once every second. 
ACKs can be also used as a base for synchronous replication. However in
that case they'll be explicitly requested by the master when the client
sends a request that needs to be replicated synchronously.
 2013-05-27 11:42:38 +02:00
antirez efd87031d0 Don't ACK the master after every command. 
Sending an ACK is now moved into the replicationSendAck() function.
 2013-05-27 11:42:35 +02:00
antirez dd0adbb777 Make sure that REPLCONF ACK really has no return value. 2013-05-27 11:42:30 +02:00
antirez 0292c5f7ae Replication: send REPLCONF ACK to master. 2013-05-27 11:42:25 +02:00
antirez 6b4635f4f5 REPLCONF ACK command. 
This special command is used by the slave to inform the master the
amount of replication stream it currently consumed.

it does not return anything so that we not need to consume additional
bandwidth needed by the master to reply something.

The master can do a number of things knowing the amount of stream
processed, such as understanding the "lag" in bytes of the slave, verify
if a given command was already processed by the slave, and so forth.
 2013-05-27 11:42:17 +02:00
antirez a950f48906 Fixed a bug in no queueing replies to master. 2013-05-25 01:00:07 +02:00
antirez dd937030f0 Replication: don't even queue replies to master commands. 
When master send commands, there is no need for the slave to reply.
Redis used to queue the reply in the output buffer and discard the reply
later, this is a waste of work and it is not clear why it was this way
(I sincerely don't remember).

This commit changes it in order to don't queue the reply at all.

All tests passing.
 2013-05-24 18:58:57 +02:00
antirez c853239a5e Top comment for prepareClientToWrite() clarified. 
We don't write the output buffer to the client socket for slaves only if
the slave is not online.
 2013-05-24 18:41:43 +02:00
ioddly c3229c33b1 Try to report source of bad Lua API calls 2013-05-22 18:17:58 -05:00
antirez 5d0f408d42 Don't stop reading redis.conf if line has no args. 
Should be "continue" and was "return".

This fixes issue #1110
 2013-05-18 16:21:52 +02:00
YAMAMOTO Takashi 9fcead7a59 don't assume time_t == long 
time_t is always 64bit on recent versions of NetBSD.
 2013-05-17 17:22:39 +09:00
YAMAMOTO Takashi ae82867664 use nanosleep instead of usleep 
SUSv3 says that:
	The useconds argument shall be less than one million. If the value of
	useconds is 0, then the call has no effect.
and actually NetBSD's implementation rejects such a value with EINVAL.
use nanosleep which has no such a limitation instead.
 2013-05-17 17:21:28 +09:00
YAMAMOTO Takashi b2dd0849ce rename popcount to popcount_binary to avoid a conflict with NetBSD libc 
NetBSD-current's libc has a function named popcount.
hiding these extensions using feature macros is not possible because
redis uses other extensions covered by the same feature macro.
eg. inet_aton
 2013-05-17 17:21:28 +09:00
YAMAMOTO Takashi 0a20ad40de don't define _XOPEN_SOURCE for NetBSD 
on NetBSD, defining _XOPEN_SOURCE hides extensions
like inet_aton, strcasecmp, etc.
 2013-05-17 17:19:02 +09:00
antirez 92f18d04a0 Merge branch 'config-rewrite' into unstable 2013-05-15 11:58:21 +02:00
antirez 78167807be Use memtoll() when parsing the backlog-size option. 2013-05-15 11:55:14 +02:00
antirez 25f2be6267 CONFIG REWRITE: backlog size is a bytes option. 2013-05-15 11:39:29 +02:00
antirez cd48e4fc40 CONFIG REWRITE: bindaddr -> bind. 2013-05-15 11:38:43 +02:00
antirez 8a44e6c490 CONFIG REWRITE: when rewriting amount of bytes use GB, MB, KB if possible. 2013-05-15 11:33:02 +02:00
antirez c590e18d15 CONFIG REWRITE: correctly escape the notify-keyspace-events option. 2013-05-15 11:15:31 +02:00
antirez 328843849f CONFIG REWRITE: "active-rehashing" -> "activerehashing". 2013-05-15 11:09:19 +02:00
antirez 402a0f1ca0 CONFIG REWRITE: fixed typo in AOF fsync policy. 2013-05-15 11:06:56 +02:00
antirez 72e980231c CONFIG REWRITE: repl-disable-tcp-nodelay is a boolean option. 2013-05-15 11:04:53 +02:00
antirez 86a6aad984 Test default config modified to set logfile to empty string. 
Required because of recent changes in the way logfile is set to standard
output.
 2013-05-15 10:19:14 +02:00
antirez 310dbba01c Added a define for most configuration defaults. 
Also the logfile option was modified to always have an explicit value
and to log to stdout when an empty string is used as log file.

Previously there was special handling of the string "stdout" that set
the logfile to NULL, this always required some special handling.
 2013-05-15 10:12:29 +02:00
antirez 9e74d216e1 CONFIG REWRITE: Use sane perms when creating redis.conf from scratch. 2013-05-14 12:45:04 +02:00
antirez e252045454 CONFIG REWRITE: actually rewrite the config file, atomically. 2013-05-14 12:32:25 +02:00
antirez 25e049cc68 redis-cli: help.h updated. 2013-05-14 11:23:16 +02:00
antirez 78f94d90eb CONFIG REWRITE: remove orphaned lines. 2013-05-14 11:17:18 +02:00
antirez ee721f1e5c CONFIG REWRITE: strip multiple empty lines. 2013-05-14 10:22:55 +02:00
antirez c184f36d21 CONFIG REWRITE: support for client-output-buffer-limit. 2013-05-13 18:34:18 +02:00
antirez d95592b116 CONFIG REWRITE: support for dir and slaveof. 2013-05-13 11:26:49 +02:00
antirez 2bc31e55ee CONFIG REWRITE: support for save and syslog-facility. 2013-05-13 11:11:45 +02:00
antirez 7e049fafd3 CONFIG REWRITE: Initial support code and design. 2013-05-13 11:11:12 +02:00
antirez 5947f170f9 Obtain absoute path of configuration file, expose it in INFO. 2013-05-09 16:57:59 +02:00
antirez e938575220 Test: more PSYNC tests (backlog TTL). 2013-05-09 12:52:04 +02:00
antirez a7486a6592 Version bumped to 2.9.10 2013-05-08 15:24:40 +02:00
antirez 16f2c3ea14 Test: check that replication partial sync works if we break the link. 
The test checks both successful syncs and unsuccessful ones by changing
the backlog size.
 2013-05-08 13:01:44 +02:00
antirez c87dd0fe49 Test: various issues with the replication-4.tcl test fixed. 
The test actually worked, but vars for master and slave were inverted
and sometimes used incorrectly.
 2013-05-08 11:58:26 +02:00
antirez 0ae1b5b0a1 Revert "use long long instead of size_t make it more safe" 
This reverts commit 2c75f2cf1a.

After further analysis, it is very unlikely that we'll raise the
string size limit to > 512MB, and at the same time such big strings
will be used in 32 bit systems.

Better to revert to size_t so that 32 bit processors will not be
forced to use a 64 bit counter in normal operations, that is currently
completely useless.
 2013-05-08 10:01:27 +02:00
Salvatore Sanfilippo c1ae7d2377 Merge pull request #1093 from july2993/unstable 
in 32bit machine, popcount don't work with a input string length up to 5...
 2013-05-08 00:49:33 -07:00
antirez 5885a34b67 Added more info about 32 bit builds into README. 2013-05-08 09:48:38 +02:00
Jiahao Huang 2c75f2cf1a use long long instead of size_t make it more safe 2013-05-07 23:37:22 +08:00
Jiahao Huang e3ed78d43b in 32bit machine, popcount don't work with a input string length up to 512 MB, 
bitcount commant may return negtive integer with string length more than 256 MB
 2013-05-07 19:55:57 +08:00
antirez 5c9f6d4f55 Cluster: link reconnection on delayed PONG reply. 
When the PONG delay is half the cluster node timeout, the link gets
disconnected (and later automatically reconnected) in order to ensure
that it's not just a dead connection issue.

However this operation is only performed if the link is old enough, in
order to avoid to disconnect the same link again and again (and among
the other problems, never receive the PONG because of that).

Note: when the link is reconnected, the 'ping_sent' field is not updated
even if a new ping is sent using the new connection, so we can still
reliably detect a node ping timeout.
 2013-05-03 15:43:03 +02:00
antirez 1315b9f246 Cluster: restore PING sent time on reconnections. 2013-05-03 15:42:59 +02:00
antirez ae71731019 Cluster: PING/PONG handling redesigned. 2013-05-03 15:42:38 +02:00
antirez a120560f70 Cluster: process config from PING packets as we do for PONG. 
Also clusterBroadcastPing() was renamed into clusterBroadcastPong()
that's what the function is actually doing.
 2013-05-03 15:41:34 +02:00
antirez 8a51c067ad Cluster: createClusterLink() comment fixed for grammar. 2013-05-03 15:41:29 +02:00
antirez 0d9c1f536b CONFIG SET server.masterauth aesthetic change. 
This is just to make the code exactly like the above instance used for
requirepass. No actual change nor the original code violated the Redis
coding style.
 2013-05-02 17:20:49 +02:00
Michel Martens 649b304e0f Reset masterauth if an empty string is configured. 2013-05-02 17:19:11 +02:00
charsyam 9cd06e4406 Fix AOF bug: expire could be removed from key on AOF rewrite. 
There was a race condition in the AOF rewrite code that, with bad enough
timing, could cause a volatile key just about to expire to be turned
into a non-volatile key. The bug was never reported to cause actualy
issues, but was found analytically by an user in the Redis mailing list:

https://groups.google.com/forum/?fromgroups=#!topic/redis-db/Kvh2FAGK4Uk

This commit fixes issue #1079.
 2013-05-02 15:35:59 +02:00
antirez e5ef85c444 Sentinel: changes to tilt mode. 
Tilt mode was too aggressive (not processing INFO output), this
resulted in a few problems:

1) Redirections were not followed when in tilt mode. This opened a
   window to misinform clients about the current master when a Sentinel
   was in tilt mode and a fail over happened during the time it was not
   able to update the state.

2) It was possible for a Sentinel exiting tilt mode to detect a false
   fail over start, if a slave rebooted with a wrong configuration
   about at the same time. This used to happen since in tilt mode we
   lose the information that the runid changed (reboot).

   Now instead the Sentinel in tilt mode will still remove the instance
   from the list of slaves if it changes state AND runid at the same
   time.

Both are edge conditions but the changes should overall improve the
reliability of Sentinel.
 2013-04-30 15:08:29 +02:00
antirez ef05a78e7e Sentinel: more sensible delay in master demote after tilt. 2013-04-30 15:08:22 +02:00
antirez 214adc5051 Lua updated to version 5.1.5. 2013-04-29 12:00:54 +02:00
antirez 48ede0d84d Sentinel: only demote old master into slave under certain conditions. 
We used to always turn a master into a slave if the DEMOTE flag was set,
as this was a resurrecting master instance.

However the following race condition is possible for a Sentinel that
got partitioned or internal issues (tilt mode), and was not able to
refresh the state in the meantime:

1) Sentinel X is running, master is instance "A".
3) "A" fails, sentinels will promote slave "B" as master.
2) Sentinel X goes down because of a network partition.
4) "A" returns available, Sentinels will demote it as a slave.
5) "B" fails, other Sentinels will promote slave "A" as master.
6) At this point Sentinel X comes back.

When "X" comes back he thinks that:

"B" is the master.
"A" is the slave to demote.

We want to avoid that Sentinel "X" will demote "A" into a slave.
We also want that Sentinel "X" will detect that the conditions changed
and will reconfigure itself to monitor the right master.

There are two main ways for the Sentinel to reconfigure itself after
this event:

1) If "B" is reachable and already configured as a slave by other
sentinels, "X" will perform a redirection to "A".
2) If there are not the conditions to demote "A", the fact that "A"
reports to be a master will trigger a failover detection in "X", that
will end into a reconfiguraiton to monitor "A".

However if the Sentinel was not reachable, its state may not be updated,
so in case it titled, or was partiitoned from the master instance of the
slave to demote, the new implementation waits some time (enough to
guarantee we can detect the new INFO, and new DOWN conditions).

If after some time still there are not the right condiitons to demote
the instance, the DEMOTE flag is cleared.
 2013-04-26 17:02:13 +02:00
antirez 1965e22aa1 Sentinel: always redirect on master->slave transition. 
Sentinel redirected to the master if the instance changed runid or it
was the first time we got INFO, and a role change was detected from
master to slave.

While this is a good idea in case of slave->master, since otherwise we
could detect a failover without good reasons just after a reboot with a
slave with a wrong configuration, in the case of master->slave
transition is much better to always perform the redirection for the
following reasons:

1) A Sentinel may go down for some time. When it is back online there is
no other way to understand there was a failover.
2) Pointing clients to a slave seems to be always the wrong thing to do.
3) There is no good rationale about handling things differently once an
instance is rebooted (runid change) in that case.
 2013-04-24 11:30:17 +02:00
antirez d264122f6a Config option to turn AOF rewrite incremental fsync on/off. 2013-04-24 10:57:07 +02:00
antirez 336d722fba AOF: sync data on disk every 32MB when rewriting. 
This prevents the kernel from putting too much stuff in the output
buffers, doing too heavy I/O all at once. So the goal of this commit is
to split the disk pressure due to the AOF rewrite process into smaller
spikes.

Please see issue #1019 for more information.
 2013-04-24 10:26:31 +02:00
antirez 91f4213ddf rio.c: added ability to fdatasync() from time to time while writing. 2013-04-24 10:26:30 +02:00
antirez c4656119b6 Test: fix RDB test checking file permissions. 
When the test is executed using the root account, setting the permission
to 222 does not work as expected, as root can read files with 222
permission.

Now we skip the test if root is detected.

This fixes issue #1034 and the duplicated #1040 issue.

Thanks to Jan-Erik Rediger (@badboy on Github) for finding a way to reproduce the issue.
 2013-04-23 14:16:50 +02:00
antirez 4ed2581a92 Test: split conceptually unrelated comments in RDB test. 2013-04-22 11:25:49 +02:00
antirez 8e222c888f Sentinel: turn old master into a slave when it comes back. 2013-04-19 16:47:24 +02:00
antirez 9d823fc222 More explicit panic message on out of memory. 2013-04-19 15:11:34 +02:00
xiaost7 ecdbaf4695 Cluster: fix clusterNode.name print format on debug message. 
It was %40s instead of %.40s, and since the string is not null
terminated it caused random garbage to be displayed, and possibly a
crash.
 2013-04-19 09:53:43 +02:00
antirez d04afd62d6 Redis/Jemalloc Gitignore were too aggressive. 
Redis gitignore was too aggressive since simply broken.

Jemalloc gitignore was too agressive because it is conceived to just
keep the files that allow to generate all the rest in development
environments (so for instance the "configure" file is excluded).
 2013-04-18 16:23:15 +02:00
antirez f8ae70cf7c redis-cli: raise error on bad command line switch. 
Previously redis-cli never tried to raise an error when an unrecognized
switch was encountered, as everything after the initial options is to be
transmitted to the server.

However this is too liberal, as there are no commands starting with "-".
So the new behavior is to produce an error if there is an unrecognized
switch starting with "-". This should not break past redis-cli usages
but should prevent broken options to be silently discarded.

As far the first token not starting with "-" is encountered, all the
rest is considered to be part of the command, so you cna still use
strings starting with "-" as values, like in:

    redis-cli --port 6380 set foo --my-value
 2013-04-11 13:17:25 +02:00
antirez 0280c2f252 redis-cli: --latency-history mode implemented. 2013-04-11 13:11:41 +02:00
antirez b84570dece Cluster: reconfigure additonal slaves on failover. 2013-04-09 12:13:26 +02:00
antirez d1aee359c0 Cluster: CONFIG SET cluster-node-timeout. 2013-04-09 11:29:51 +02:00
antirez 68cf249f81 Cluster: use server.cluster_node_timeout directly. 
We used to copy this value into the server.cluster structure, however this
was not necessary.

The reason why we don't directly use server.cluster->node_timeout is
that things that can be configured via redis.conf need to be directly
available in the server structure as server.cluster is allocated later
only if needed in order to reduce the memory footprint of non-cluster
instances.
 2013-04-09 11:24:18 +02:00
antirez ef4f25ff6e Cluster: configdigest field no longer used. Removed. 2013-04-09 11:07:25 +02:00
antirez f09b2508f4 Cluster: properly send ping to nodes not pinged foro too much time. 
In commit d728ec6 it was introduced the concept of sending a ping to
every node not receiving a ping since node_timeout/2 seconds.
However the code was located in a place that was not executed because of
a previous conditional causing the loop to re-iterate.

This caused false positives in nodes availability detection.

The current code is still not perfect as a node may be detected to be in
PFAIL state even if it does not reply for just node_timeout/2 seconds
that is not correct. There is a plan to improve this code ASAP.
 2013-04-08 19:40:20 +02:00
antirez ace86cfaef Test: remove useless statements and comments from test default config. 2013-04-04 14:30:05 +02:00
antirez d5b383477e Cluster: move REDIS_CLUSTER_FAILOVER_DELAY near other timing defines. 2013-04-04 14:23:34 +02:00
antirez 3cc6e7d01d Cluster: CONFIG GET cluster-node-timeout. 2013-04-04 14:21:01 +02:00
antirez 05fa4f4034 Cluster: node timeout is now configurable. 2013-04-04 12:29:10 +02:00
antirez 00bab23c41 Cluster: turn hardcoded node timeout multiplicators into defines. 
Most Redis Cluster time limits are expressed in terms of the configured
node timeout. Turn them into defines.
 2013-04-04 12:04:11 +02:00
antirez 8419397665 Make rio.c comment 80-columns friendly. 2013-04-03 12:41:14 +02:00
antirez b237de33d1 Throttle BGSAVE attempt on saving error. 
When a BGSAVE fails, Redis used to flood itself trying to BGSAVE at
every next cron call, that is either 10 or 100 times per second
depending on configuration and server version.

This commit does not allow a new automatic BGSAVE attempt to be
performed before a few seconds delay (currently 5).

This avoids both the auto-flood problem and filling the disk with
logs at a serious rate.

The five seconds limit, considering a log entry of 200 bytes, will use
less than 4 MB of disk space per day that is reasonable, the sysadmin
should notice before of catastrofic events especially since by default
Redis will stop serving write queries after the first failed BGSAVE.

This fixes issue #849
 2013-04-02 14:05:50 +02:00
antirez b14fda7deb Version bumped to 2.9.9. 2013-04-02 11:55:23 +02:00
Salvatore Sanfilippo 63457d1a5d Merge pull request #1017 from jbergstroem/build-improvements 
Build improvements
 2013-04-02 02:24:52 -07:00
charsyam 86d87e3554 Support for case unsensitive SET options. 2013-03-29 10:33:52 +01:00
antirez dfc31a1ee8 Test: Extended SET tests. 2013-03-28 16:25:24 +01:00
antirez 30d5d416e6 Extended SET command implemented (issue #931). 2013-03-28 15:40:19 +01:00
antirez 140260409e EXPIRE should not resurrect keys. Issue #1026. 2013-03-28 12:45:07 +01:00
antirez 601cea665d Test: regression test for issue #1026. 2013-03-28 11:46:14 +01:00
antirez 3816e7bda9 Better DEBUG error message when num of arguments is wrong. 2013-03-28 11:39:29 +01:00
antirez 1b13adf388 Test: verify that lazy-expire works. 2013-03-28 11:36:49 +01:00
antirez 32a83c8206 DEBUG set-active-expire added. 
We need the ability to disable the activeExpireCycle() (active
expired key collection) call for testing purposes.
 2013-03-27 17:55:02 +01:00
antirez 252cf3052d Test: test replication of MULTI/EXEC. 2013-03-27 11:44:50 +01:00
antirez 40b692e822 Test: Restore DB back to 9 after testing MULTI/EXEC with DB 5. 2013-03-27 11:30:23 +01:00
antirez 7572dc47da Test: new functions to capture and analyze the replication stream. 2013-03-27 11:29:47 +01:00
antirez 4fa30b781e redis-trib: ClusterNode #info_string output modified. 
The hope is that the new one is more readable.
 2013-03-26 18:16:03 +01:00
antirez df69155e8a Allow SELECT while loading the DB. 
Fixes issue #1024.
 2013-03-26 13:51:17 +01:00
antirez 873f328fd8 TTL / PTTL commands: two bugs fixed. 
This commit fixes two corner cases for the TTL command.

1) When the key was already logically expired (expire time older
than current time) the command returned -1 instead of -2.

2) When the key was existing and the expire was found to be exactly 0
(the key was just about to expire), the command reported -1 (that is, no
expire) instead of a TTL of zero (that is, about to expire).
 2013-03-26 11:45:22 +01:00
antirez 8bb5eb7357 Flag PUBLISH as read-only in the command table. 2013-03-26 11:09:22 +01:00
antirez 71f3e74317 Transactions: propagate MULTI/EXEC only when needed. 
MULTI/EXEC is now propagated to the AOF / Slaves only once we encounter
the first command that is not a read-only one inside the transaction.

The old behavior was to always propagate an empty MULTI/EXEC block when
the transaction was composed just of read only commands, or even
completely empty. This created two problems:

1) It's a bandwidth waste in the replication link and a space waste
   inside the AOF file.

2) We used to always increment server.dirty to force the propagation of
   the EXEC command, resulting into triggering RDB saves more often
   than needed.

Note: even read-only commands may also trigger writes that will be
propagated, when we access a key that is found expired and Redis will
synthesize a DEL operation. However there is no need for this to stay
inside the transaction itself, but only to be ordered.

So for instance something like:

    MULTI
    GET foo
    SET key zap
    EXEC

May be propagated into:

    DEL foo
    MULTI
    SET key zap
    EXEC

While the DEL is outside the transaction, the commands are delivered in
the right order and it is not possible for other commands to be inserted
between DEL and MULTI.
 2013-03-26 10:58:10 +01:00
antirez 02c269e27e Transactions: use discardTransaction() in EXEC implementation. 2013-03-26 10:48:15 +01:00
antirez 2f49734029 Transactions: use the propagate() API to propagate MULTI. 
The behavior is the same, but the code is now cleaner and uses the
proper interface instead of dealing directly with AOF/replication
functions.
 2013-03-26 10:27:45 +01:00
Salvatore Sanfilippo afce010639 Merge pull request #1003 from NanXiao/patch-3 
Update config.c
 2013-03-25 11:30:59 -07:00
antirez c39e34d007 Cluster: when slave changes master, remove it from the old master. 2013-03-25 15:01:25 +01:00
antirez 70091a2f63 redis-trib: wait cluster join after cluster creation. 2013-03-25 13:14:17 +01:00
antirez 34c1871e9f Cluster: set node role on successful handshake. 2013-03-25 13:03:01 +01:00
antirez cda0cdfb70 redis-trib: Don't use colorization if TERM != xterm. 2013-03-25 12:51:53 +01:00
antirez 4286d8b979 redis-trib: initial output colorization 2013-03-25 12:50:38 +01:00
antirez 797d98e906 Test: obuf-limits test false positive removed. 
Fixes #621.
 2013-03-25 11:56:34 +01:00
antirez 09aa55a334 redis-cli --stat, stolen from redis-tools. 
Redis-tools is a connection of tools no longer mantained that was
intented as a way to economically make sense of Redis in the pre-vmware
sponsorship era. However there was a nice redis-stat utility, this
commit imports one of the functionalities of this tool here in redis-cli
as it seems to be pretty useful.

Usage: redis-cli --stat

The output is similar to vmstat in the format, but with Redis specific
stuff of course.

From the point of view of the monitored instance, only INFO is used in
order to grab data.
 2013-03-22 17:54:32 +01:00
antirez c195289e5e redis-trib: All output wrapped by a specific function. 
This is needed in order to colorize it as next step.
We use conventions in output messages such as

>>> This is an action
*** This is a warning
[ERR] This is an error
[OK] That's fine

And so forth, so that a color will be associated checking the first
three chars.
 2013-03-22 17:39:43 +01:00
antirez be7bdd376e redis-trib: fix open slot correction. 
Slot zero was hardcoded (!)
 2013-03-22 13:03:33 +01:00
antirez 813d7cbdd1 redis-trib: added cluster_error method to add errors. 2013-03-22 12:59:18 +01:00
antirez d7eae8d8d7 redis-trib: fixed ClusterNode migrating/importing slots detection. 2013-03-22 12:54:04 +01:00
antirez 1baa153028 redis-trib: added some more output for check. 2013-03-22 12:47:49 +01:00
antirez 77bba91b48 redis-trib: fixed type has_flags? -> has_flag. 2013-03-22 12:28:06 +01:00
antirez 4dded0c187 redis-trib: ignore slaves when resharding. 2013-03-21 18:17:06 +01:00
antirez 47da76576e redis-trib: fix conditional otherwise always true. 2013-03-21 17:22:14 +01:00
antirez a89d435d8e Cluster: move slotToKeyFlush() to emptyDb(). 
This way we are sure to destroy the slot->key map every time we destroy
the DB, for instance when reloading a DB due to replication.
 2013-03-21 17:13:08 +01:00
antirez ed47f77977 redis-trib: initial support to fix "open" slots. 
Open slots are slots found in importing or migrating slot when a
cluster check is performed.
 2013-03-21 17:11:54 +01:00
antirez 1a6df1049d redis-trib: load info about importing/migrating slots from node. 2013-03-21 16:31:53 +01:00
antirez a8b09faf3d Cluster: comment no longer in sync with code removed. 2013-03-21 10:47:10 +01:00
Johan Bergström 8080e1cfbe use install as default installer (except on SunOS) 2013-03-21 13:32:08 +11:00
antirez 8c1bc8e865 Cluster: clear the PROMOTED slave directly into clusterSetMaster(). 
This way we make sure every time a master is turned into a replica
the flag will be cleared.
 2013-03-20 11:51:44 +01:00
antirez e006407fd0 Cluster: master node must clear its hash slots when turning into a slave. 
When a master turns into a slave after a failover event, make sure to
clear the assigned slots before setting up the replication, as a slave
should never claim slots in an explicit way, but just take over the
master slots when replacing its master.
 2013-03-20 11:32:35 +01:00
antirez 506f9a42b0 Cluster: new flag PROMOTED introduced. 
A slave node set this flag for itself when, after receiving authorization
from the majority of nodes, it turns itself into a master.

At the same time now this flag is tested by nodes receiving a PING
message before reconfiguring after a failover event. This makes the
system more robust: even if currently there is no way to manually turn
a slave into a master it is possible that we'll have such a feature in
the future, or that simply because of misconfiguration a node joins the
cluster as master while others believe it's a slave. This alone is now
no longer enough to trigger reconfiguration as other nodes will check
for the PROMOTED flag.

The PROMOTED flag is cleared every time the node is turned back into a
replica of some other node.
 2013-03-20 10:48:42 +01:00
antirez 026b9483db Cluster: add sender flags in cluster bus messages header. 
Sender flags were not propagated for the sender, but only for nodes in
the gossip section. This is odd and in the next commits we'll need to
get updated flags for the sender node, so this commit adds a new field
in the cluster messages header.

The message header is the same size as we reused some free space that
was marked as 'unused' because of alignment concerns.
 2013-03-20 10:32:00 +01:00
antirez d15b027d91 Cluster: turn old master into a replica of node that failed over. 
So when the failing master node is back in touch with the cluster,
instead of remaining unused it is converted into a replica of the
new master, ready to perform the fail over if the new master node
will fail at some point.

Note that as a side effect clients with stale configuration are now
not an issue as well, as the node converted into a slave will not
accept queries but will redirect clients accordingly.
 2013-03-20 00:30:47 +01:00
antirez 4d62623015 Cluster: node replication role change handle improved. 
The code handling a master that turns into a slave or the contrary was
improved in order to avoid repeating the same operations. Also
the readability and conceptual simplicity was improved.
 2013-03-19 16:01:30 +01:00
antirez 88221f88c0 Cluster: new command CLUSTER FLUSHSLOTS. 
It's just a simpler way to CLUSTER DELSLOTS with all the slots as
arguments, in order to obtain a node without assigned slots for
reconfiguration.
 2013-03-19 09:58:05 +01:00
antirez 7d3e32d526 redis-trib: don't load cluster config from nodes in FAIL state. 2013-03-19 09:46:12 +01:00
Johan Bergström dc4003be71 Silence mkdir output 2013-03-17 18:37:38 +11:00
Johan Bergström 348be19b5f Only pass -rdynamic as linker option 2013-03-17 17:49:57 +11:00
Johan Bergström 33a4bc2c70 Remove extra spaces 2013-03-17 17:23:45 +11:00
Johan Bergström 978c895b69 make check is a common naming convention for tests 2013-03-16 18:40:22 +11:00
Johan Bergström a66195629b Inherit CC for Lua 2013-03-16 18:38:37 +11:00
Johan Bergström ada7aa7ac9 Spaces to tabs 2013-03-16 18:35:20 +11:00
Johan Bergström bea60bec75 Slightly refactor CFLAGS/LDFLAGS/LIBS 
This way, we can avoid -rdynamic and -pthread warnings on darwin.
 2013-03-16 18:33:42 +11:00
antirez e28e61e839 Cluster: when failing over claim master slots. 2013-03-15 16:53:41 +01:00
antirez b8127e337a Version incremented to 2.9.8 after major cluster progresses. 2013-03-15 16:45:45 +01:00
antirez dd091661d4 Cluster: log when a slave asks for failover authorization. 2013-03-15 16:44:08 +01:00
antirez 1375b0611b Cluster: slaves start failover with a small delay. 
Redis Cluster can cope with a minority of nodes not informed about the
failure of a master in time for some reason (netsplit or node not
functioning properly, blocked, ...) however to wait a few seconds before
to start the failover will make most "normal" failovers simpler as the
FAIL message will propagate before the slave election happens.
 2013-03-15 16:39:49 +01:00
antirez d512a09c20 Cluster: a bit more serious node role change handling. 2013-03-15 16:35:16 +01:00
antirez 004fbef847 Cluster: remove node from master slaves when it turns into a master. 
Also, a few nearby comments improved.
 2013-03-15 16:16:19 +01:00
antirez 44c92f5aeb Cluster: slave failover implemented. 2013-03-15 16:11:34 +01:00
antirez 1d8f302e0d Cluster: election -> promotion in two comments. 2013-03-15 15:44:49 +01:00
antirez bf82195467 Cluster: added function to broadcast pings. 
See the function top-comment for info why this is useful sometimes.
 2013-03-15 15:43:58 +01:00
antirez 892e98548a Cluster: don't broadcast messages to HANDSHAKE nodes. 
Also don't check for NOADDR as we check that node->link is not NULL
that's enough.
 2013-03-15 15:36:36 +01:00
antirez 76a3954f4a Cluster: fix clusterHandleSlaveFailover() conditional: quorum is enough. 2013-03-15 13:20:34 +01:00
antirez 90e99a2082 Cluster: two lame bugs fixed in FAILOVER AUTH messages generation. 2013-03-14 21:27:12 +01:00
antirez aeacaa57e6 Cluster: code to process messages moved in the right if-else chain. 2013-03-14 21:21:58 +01:00
antirez 35f05c66b6 Cluster: handle FAILOVER_AUTH_ACK messages. 
That's trivial as we just need to increment the count of masters that
received with an ACK.
 2013-03-14 16:43:13 +01:00
antirez c2595500ac Cluster: request failover authorization, log if we have quorum. 
However the failover is yet not really performed.
 2013-03-14 16:39:02 +01:00
antirez 7fa42b801d Cluster: clusterSendFailoverAuth() implementation. 2013-03-14 16:31:57 +01:00
NanXiao 79a13b46fb Update config.c 
Fix bug in configGetCommand function: get correct masterauth value.
 2013-03-14 13:52:43 +08:00
antirez f59ff6fe61 Cluster: clusterSendFailoverAuthIfNeeded() work in progress. 2013-03-13 19:08:03 +01:00
antirez 44f6fdab60 Cluster: handle FAILOVER_AUTH_REQUEST in clusterProcessPacket(). 
However currently the control is passed to a function doing nothing at
all.
 2013-03-13 18:38:08 +01:00
antirez ece95b2dea Cluster: sanity check FAILOVER_AUTH_REQUEST messages for proper length. 2013-03-13 17:31:26 +01:00
antirez 66144337bf Cluster: use 'else if' for mutually exclusive conditionals. 2013-03-13 17:27:06 +01:00
antirez db7c17e969 Cluster: FAILOVER_AUTH_REQUEST message type introduced. 
This message is sent by a slave that is ready to failover its master to
other nodes to get the authorization from the majority of masters.
 2013-03-13 17:21:20 +01:00
antirez 575cbc9990 Cluster: clusterHandleSlaveFailover() stub. 2013-03-13 13:10:49 +01:00
antirez 1902a9c532 Replication: master_link_down_since_seconds initial value should be huge. 
server.repl_down_since used to be initialized to the current time at
startup. This is wrong since the replication never started. Clients
testing this filed to check if data is uptodate should never believe
data is recent if we never ever connected to our master.
 2013-03-13 12:54:48 +01:00
antirez 3d448bda39 Cluster: call clusterHandleSlaveFailover() when our master is down. 2013-03-13 12:44:02 +01:00
antirez ae94fe7e48 Test: make sure broken RDB checksum is detected. 2013-03-13 11:12:45 +01:00
antirez 82b0eae690 Test: more RDB loading checks. 
A test for issue #1001 is included.
 2013-03-13 10:04:33 +01:00
antirez 0b74a85678 Test: check that Redis starts empty without an RDB file. 2013-03-12 19:55:33 +01:00
antirez 79a6844e44 rdbLoad(): rework code to save vertical space. 2013-03-12 19:46:33 +01:00
Salvatore Sanfilippo 9925c7c670 Merge pull request #1001 from djanowski/fatal-errors-rdb-load 
Abort when opening the RDB file results in an error other than ENOENT.
 2013-03-12 11:40:36 -07:00
Damian Janowski 4178a80282 Abort when opening the RDB file results in an error other than ENOENT. 
This fixes cases where the RDB file does exist but can't be accessed for
any reason. For instance, when the Redis process doesn't have enough
permissions on the file.
 2013-03-12 14:37:50 -03:00
antirez 215bfaea16 Set default for stop_writes_on_bgsave_err in initServerConfig(). 
It was placed for error in initServer() that's called after the
configuation is already loaded, causing issue #1000.
 2013-03-12 18:34:08 +01:00
antirez 91d3b487e7 redis-cli --bigkeys: don't crash with empty DBs. 2013-03-12 09:58:00 +01:00
antirez 2d851333a6 activeExpireCycle() smarter with many DBs and under expire pressure. 
activeExpireCycle() tries to test just a few DBs per iteration so that
it scales if there are many configured DBs in the Redis instance.
However this commit makes it a bit smarter when one a few of those DBs
are under expiration pressure and there are many many keys to expire.

What we do is to remember if in the last iteration had to return because
we ran out of time. In that case the next iteration we'll test all the
configured DBs so that we are sure we'll test again the DB under
pressure.

Before of this commit after some mass-expire in a given DB the function
tested just a few of the next DBs, possibly empty, a few per iteration,
so it took a long time for the function to reach again the DB under
pressure. This resulted in a lot of memory being used by already expired
keys and never accessed by clients.
 2013-03-11 11:10:33 +01:00
antirez 08b107e405 In databasesCron() never test more DBs than we have. 2013-03-11 10:51:03 +01:00
antirez 4b1ccdfd49 Make comment name match var name in activeExpireCycle(). 2013-03-11 10:42:14 +01:00
antirez 1f7d2c1e27 Optimize inner loop of activeExpireCycle() for no-expires case. 2013-03-09 11:48:54 +01:00
antirez 5f5aa487f9 REDIS_DBCRON_DBS_PER_SEC -> REDIS_DBCRON_DBS_PER_CALL 2013-03-09 11:44:20 +01:00
antirez db29d71a30 activeExpireCycle(): process only a small number of DBs per iteration. 
This small number of DBs is set to 16 so actually in the default
configuraiton Redis should behave exactly like in the past.
However the difference is that when the user configures a very large
number of DBs we don't do an O(N) operation, consuming a non trivial
amount of CPU per serverCron() iteration.
 2013-03-08 17:48:58 +01:00
antirez 40a2da159c Use unsigned integers for DB ids, for defined wrap-to-zero. 2013-03-08 17:41:20 +01:00
antirez 7ac3b3a486 Only resize/rehash a few databases per cron iteration. 
This is the first step to lower the CPU usage when many databases are
configured. The other is to also process a limited number of DBs per
call in the active expire cycle.
 2013-03-08 14:01:12 +01:00
antirez dfd732dff3 Actually call databasesCron() inside serverCron(). 2013-03-08 13:59:50 +01:00
antirez cd9dcd1835 Move Redis databases background processing to databasesCron(). 2013-03-08 12:34:05 +01:00
antirez f0b807cd47 Cluster: update cluster state on PFAIL flag set/cleared on nodes. 2013-03-07 15:40:53 +01:00
antirez 299b8f76c2 Cluster: mark cluster state as fail of majority of masters is unreachable. 2013-03-07 15:36:59 +01:00
antirez abf06fd5ff Cluster: log global cluster state change. 2013-03-07 15:22:32 +01:00
antirez 3dad8196b7 Cluster: clusterUpdateState() function simplified. 
Also the NEEDHELP Cluster state was removed as it will no longer be
used by Redis Cluster.
 2013-03-06 18:25:40 +01:00
Gengliang Wang 042ed270c8 Removed useless "return" statements in pubsub.c 
(original commit message edited)
 2013-03-06 16:49:20 +01:00
antirez 7b190a08cf API to lookup commands with their original name. 
A new server.orig_commands table was added to the server structure, this
contains a copy of the commant table unaffected by rename-command
statements in redis.conf.

A new API lookupCommandOrOriginal() was added that checks both tables,
new first, old later, so that rewriteClientCommandVector() and friends
can lookup commands with their new or original name in order to fix the
client->cmd pointer when the argument vector is renamed.

This fixes the segfault of issue #986, but does not fix a wider range of
problems resulting from renaming commands that actually operate on data
and are registered into the AOF file or propagated to slaves... That is
command renaming should be handled with care.
 2013-03-06 16:28:26 +01:00
antirez 4d62912619 Add a warning about command renaming in redis.conf. 2013-03-06 15:47:16 +01:00
antirez bfa25441e7 Handle a non-impossible empty argv in loadServerConfigFromString(). 
Usually this does not happens since we trim for " \t\r\n", but if there
are other chars that return true with isspace(), we may end with an
empty argv. Better to handle the condition in an explicit way.
 2013-03-06 12:40:48 +01:00
antirez 8c193af696 redis-cli: use sdsfreesplitres() instead of hand-coding it. 2013-03-06 12:38:32 +01:00
antirez 011fa89ac9 Cluster: sdssplitargs_free() -> sdsfreesplitres(). 2013-03-06 12:38:06 +01:00
antirez 729a3432ba sds.c: sdssplitargs_free() removed as it was a duplicate. 2013-03-06 12:38:06 +01:00
antirez cf4d7737bb More specific error message in loadServerConfigFromString(). 2013-03-06 12:24:12 +01:00
antirez 4ea89e64c0 sdssplitargs(): on error set *argc to 0. 
This makes programs not checking the return value for NULL much safer
since with this change:

1) It is still possible to iterate the zero-length result without
crashes.
2) sdssplitargs_free will work against NULL and 0 count.
 2013-03-06 12:21:31 +01:00
antirez 5cabae84e6 sdssplitargs(): now returns NULL only on error. 
An empty input string also resulted into the function returning NULL
making it harder for the caller to distinguish between error and empty
string without checking the original input string length.
 2013-03-06 12:21:21 +01:00
charsyam 1303f02be6 Don't segfault on unbalanced quotes. 2013-03-06 11:54:02 +01:00
antirez 304ef5e283 Allow AUTH while loading the DB in memory. 
While Redis is loading the AOF or RDB file in memory only a subset of
commands are allowed. This commit adds AUTH to this subset.
 2013-03-06 11:50:38 +01:00
antirez 1025dd7786 Cluster: connect to our master ASAP after startup if we are a slave node. 2013-03-05 16:12:08 +01:00
antirez bac57ad14b Cluster: more robust FAIL flag cleaup. 
If we have a master in FAIL state that's reachable again, and apparently
no one is going to serve its slots, clear the FAIL flag and let the
cluster continue with its operations again.
 2013-03-05 15:05:32 +01:00
antirez 1a02b7440a Cluster: new node field fail_time. 
This is the unix time at which we set the FAIL flag for the node.
It is only valid if FAIL is set.

The idea is to use it in order to make the cluster more robust, for
instance in order to revert a FAIL state if it is long-standing but
still slots are assigned to this node, that is, no one is going to fix
these slots apparently.
 2013-03-05 13:15:05 +01:00
antirez d3b4662347 Cluster: don't check keys hash slots when the source is our master. 
Usually we redirect clients to the right hash slot, however we don't
want to do that with our master, we want just to mirror it.
 2013-03-05 13:02:44 +01:00
antirez 31ac376051 Cluster: slaveof not allowed in redis.conf as well. 2013-03-05 12:58:22 +01:00
antirez b7d085fc0d Cluster: SLAVEOF command not allowed in cluster mode. 2013-03-05 12:39:41 +01:00
antirez e4b481a5f6 Cluster: A comment updated in clusterCron(). 2013-03-05 12:17:30 +01:00
antirez d728ec6dee Cluster: send a ping to every node we never contacted in timeout/2 seconds. 
Usually we try to send just 1 ping every second, however when we detect
we are going to have unreliable failure detection because we can't ping
some node in time, send an additional ping.

This should only happen with very large clusters or when the the node
timeout is set to a very low value.
 2013-03-05 12:16:02 +01:00
antirez e7628be2a7 Cluster: set node->slaveof correctly when a node state is updated. 2013-03-05 11:50:11 +01:00
antirez d6457577d4 Cluster: don't perform startup slots sanity check for slaves. 
If we are a cluster node the DB content will not match our configured
slots. Don't do the check at all.
 2013-03-04 19:47:00 +01:00
antirez d334897e80 Cluster: fix maximum line length when loading config. 
There are pathological cases where the line can be even longer a single
node may contain all the slots in importing/migrating state.
 2013-03-04 19:45:36 +01:00
antirez 3be893123f Make sure replicationSetMaster() works when ip argument is not an sds. 2013-03-04 15:39:55 +01:00
antirez b8a28bf442 Cluster: actually setup replication in CLUSTER REPLICATE. 2013-03-04 15:27:58 +01:00
antirez 7bead003e2 SLAVEOF command refactored into a proper API. 
We now have replicationSetMaster() and replicationUnsetMaster() that can
be called in other contexts (for instance Redis Cluster).
 2013-03-04 13:22:21 +01:00
antirez 0c01088b51 Cluster: REPLICATE subcommand and stub for clusterSetMaster(). 2013-03-04 13:15:09 +01:00
charsyam bc84c399f8 adding check error code 
adding check error code
 2013-03-04 11:20:11 +01:00
antirez 3b3974410e redis-cli: use keepalive socket option. 
This should improve things in two ways:

1) Prevent timeouts caused by the execution of long commands.
2) Improve detection of real connection errors.

This is mostly effective only on Linux because of the bogus default
keepalive settings. In Linux we have OS-specific calls to set the
keepalive interval to reasonable values.
 2013-03-04 11:14:32 +01:00
Salvatore Sanfilippo 174e51cb75 Merge pull request #967 from 0x20h/fix-git-diff 
suppress external diff program when using git diff.
 2013-03-04 01:57:01 -08:00
Stam He d783860457 point 2 of slave-serve-stale-data miss '-' between 'stale' and 'data' 2013-03-04 10:49:07 +01:00
antirez caf9b24a7d Cluster: don't set the slot as unassigned because of PONG info. 
As stated in the comment this is usually due to a resharding in progress
so the client should be still redirected to the old node that will
handle the redirection elsewhere.
 2013-02-28 15:54:29 +01:00
antirez 0d77440b26 Cluster: better handling of slots changes in PONG packets. 
The new code makes sure that the node slots bitmap is always consistent
with the cluster->slots array.
 2013-02-28 15:41:54 +01:00
antirez 5f8fd27ace Cluster: refactoring of clusterNode*Bit to use helper bitmap functions. 2013-02-28 15:23:09 +01:00
antirez d21d6b666f Cluster: use node->numslots instead of popcount() where possible. 2013-02-28 15:13:32 +01:00
antirez 4521115b17 Cluster: new field in cluster node structure, "numslots". 
Before a relatively slow popcount() operation was needed every time we
needed to get the number of slots served by a given cluster node.
Now we just need to check an integer that is taken in sync with the
bitmap.
 2013-02-28 15:11:05 +01:00
antirez a2566d6618 Cluster: don't gossip about nodes that are not useful to the cluster. 2013-02-28 15:00:09 +01:00
antirez bc922dc688 redis-trib: skip nodes without slots when creating the config signature. 2013-02-28 13:12:56 +01:00
antirez 64942fca01 redis-trib help. 2013-02-27 18:02:22 +01:00
antirez d45d184118 Cluster: CLUSTER FORGET implemented. 2013-02-27 17:55:59 +01:00
antirez d2b8281b3f Cluster: added a missing return on CLUSTER SETSLOT. 2013-02-27 17:53:48 +01:00
antirez 7ddc0fe652 redis-trib: skip noaddr and disconnected nodes while loading cluster info. 2013-02-27 17:23:11 +01:00
antirez d20dea3eb7 Cluster: blank node address when flagging it as NOADDR. 2013-02-27 17:09:33 +01:00
antirez 2dcb5ab72b Cluster: add comments in sub-sections of CLUSTER command. 2013-02-27 16:12:59 +01:00
antirez 96dd210970 redis-trib: initial implementation of addnode command. 2013-02-27 15:58:41 +01:00
antirez f7dac639a9 Remove warning when printing redisBuildId(). 2013-02-27 12:33:27 +01:00
antirez e4682c82e3 Remove too agressive/spamming log in rdb.c. 2013-02-27 12:29:20 +01:00
antirez f9b5ca29fd Use GCC printf format attribute for redisLog(). 
This commit also fixes redisLog() statements producing warnings.
 2013-02-27 12:27:15 +01:00
antirez c35b065a64 Better panic message for failed time event creation. 2013-02-27 12:00:11 +01:00
Stam He e431a97660 add a check for aeCreateTimeEvent 
1) Add a check for aeCreateTimeEvent in function initServer.
 2013-02-27 11:57:35 +01:00
Stam He 9c8be6cab9 Set proctitle: avoid the use of __attribute__((constructor)). 
This cased a segfault in some Linux system and was GCC-specific.

Commit modified by @antirez:

1) Stripped away the part to set the proc title via config for now.
2) Handle initialization of setproctitle only when the replacement
   is used.
3) Don't require GCC now that the attribute constructor is no
   longer used.
 2013-02-27 11:50:35 +01:00
antirez deb1f4d841 setproctitle.c: declar tmp as static so valgrind will not detect a leak. 2013-02-26 15:51:33 +01:00
antirez d0992d6e8b Cluster: a few random fixes to the new failure detection. 2013-02-26 15:15:44 +01:00
antirez f288b07563 Cluster: log the event when we clear the FAIL flag. 2013-02-26 15:03:38 +01:00
antirez 97ffcd351b Cluster: use the failure report API to reimplement failure detection. 
The new system detects a failure only when there is quorum from masters.
 2013-02-26 14:58:39 +01:00
antirez 6356cf6808 Set process name in ps output to make operations safer. 
This commit allows Redis to set a process name that includes the binding
address and the port number in order to make operations simpler.

Redis children processes doing AOF rewrites or RDB saving change the
name into redis-aof-rewrite and redis-rdb-bgsave respectively.

This in general makes harder to kill the wrong process because of an
error and makes simpler to identify saving children.

This feature was suggested by Arnaud GRANAL in the Redis Google Group,
Arnaud also pointed me to the setproctitle.c implementation includeed in
this commit.

This feature should work on all the Linux, OSX, and all the three major
BSD systems.
 2013-02-26 11:52:12 +01:00
antirez 1b1b3f6c06 Cluster: invert two functions declarations in more natural order. 2013-02-26 11:19:48 +01:00
antirez d5e8b0a47f Cluster: cleanup idle failure reports every time we remove one. 
This is not very important as anyway when the function counting the
number of reports is called the cleanup is performed. However with this
change if only part of the nodes that reported the failure will report
the node is back ok, we'll cleanup the older entries ASAP. In complex
split net split scenarios, and when we are dealing with clusters having
nodes in the order of ~ 1000, this can save some CPU.
 2013-02-26 11:15:18 +01:00
antirez 9cb578ced0 Cluster: new function clusterNodeDelFailureReport() for failure reports. 
This is the missing part of the API that will be used to reimplement
failure detection of Cluster nodes.
 2013-02-25 19:13:22 +01:00
Salvatore Sanfilippo b26bb768bc Merge pull request #969 from serphen/unstable 
Fix error "repl-backlog-size must be 1 or greater"
 2013-02-25 09:38:15 -08:00
Arnaud Granal 3d588b37ce Fix error "repl-backlog-size must be 1 or greater" 
The parameter repl-backlog-size is not parsed correctly in the configuration file. argv[0] is parsed instead of argv[1].
 2013-02-25 19:25:48 +02:00
antirez 18f537083a Cluster: no limits for the count parameter of CLUSTER GETKEYSINSLOT. 
Not sure why I set a limit to 1 million keys, there is no reason for
this artificial limit, and anyway this is s a stupid limit because it is
already high enough to create latency issues. So let's the users shoot
on their feet because maybe they just actually know what they are doing.
 2013-02-25 12:41:13 +01:00
antirez 544bbe5387 Cluster: validate slot number in CLUSTER COUNTKEYSINSLOT. 2013-02-25 12:40:32 +01:00
antirez 996a643752 Cluster: use O(log(N)) algo for countKeysInSlot(). 2013-02-25 12:37:50 +01:00
antirez d4fa40655d Cluster: new sub-command CLUSTER COUNTKEYSINSLOT. 
The new sub-command uses the new countKeysInSlot() API and allows a
cluster client to get the number of keys for a given hashslot.
 2013-02-25 12:04:31 +01:00
antirez a517c89321 Cluster: verifyClusterConfigWithData() implemented. 2013-02-25 11:43:49 +01:00
antirez 9947b0d96d A comment in main() clarified. 2013-02-25 11:40:21 +01:00
antirez d2154254be Cluster: fix case for getKeysInSlot() and countKeysInSlot(). 
Redis functions start in low case. A few functions about cluster were
capitalized the wrong way.
 2013-02-25 11:25:40 +01:00
antirez c2eb4a606f Cluster: use CountKeysInSlot() when we just need the count. 2013-02-25 11:23:04 +01:00
antirez ad3bca1fdf Cluster: added stub for verifyClusterConfigWithData(). 
See the top-comment for the function in this commit for details about
what the function is supposed to do.
 2013-02-25 11:20:17 +01:00
antirez 1abce14611 Cluster: added new API countKeysInSlot(). 
This is similar to getKeysInSlot() but just returns the number of keys
found in a given hash slot, without returning the keys.
 2013-02-25 11:15:03 +01:00
0x20h 977035b7b5 suppress external diff program when using git diff. 2013-02-24 18:17:46 +01:00
antirez 825e07f2fd Cluster: if no previous config exists, create the myself node as master. 2013-02-22 19:24:01 +01:00
antirez f4093753e4 Cluster: add cluster_size field in CLUSTER INFO output. 2013-02-22 19:20:38 +01:00
antirez d218a4e244 Cluster: new state information, cluster size. 
The definition of cluster size is: the number of known nodes in the
cluster that are masters and serving at least an hash slot.
 2013-02-22 19:18:30 +01:00
antirez 5c55ed9388 Cluster: remove warning adding clusterNodeSetSlotBit() prototype. 2013-02-22 17:45:49 +01:00
antirez 974929770b Cluster: introduced a failure reports system. 
A §Redis Cluster node used to mark a node as failing when itself
detected a failure for that node, and a single acknowledge was received
about the possible failure state.

The new API will be used in order to possible to require that N other
nodes have a PFAIL or FAIL state for a given node for a node to set it
as failing.
 2013-02-22 17:43:35 +01:00
antirez 36af851550 redis-trib: check that all the nodes agree about the slots configuration. 2013-02-22 12:25:16 +01:00
antirez 51b5058d04 redis-trib: skeleton of coverage fix for "keys in multiple nodes" case. 2013-02-22 11:33:10 +01:00
antirez a81c598f95 redis-trib: handle slot coverage fix in the "no nodes with keys" case. 2013-02-22 10:23:53 +01:00
antirez 392e0fa7eb Cluster: fix case of slotToKey...() functions. 2013-02-22 10:16:21 +01:00
antirez d04770988d Cluster: empty the internal sorted set mapping keys to slots on FLUSHDB/ALL. 2013-02-22 10:15:32 +01:00
antirez efe51dfff5 redis-trib: specify single node address when fixing coverage. 2013-02-22 10:05:07 +01:00
antirez 915e81335e redis-trib: ability to fix uncovered slots for the trivial case. 2013-02-21 18:10:06 +01:00
antirez 619d3945f8 redis-trib: fixed typo in method name. 2013-02-21 16:58:27 +01:00
antirez 07b6322735 Cluster: more correct update of slots state when PONG is received. 2013-02-21 16:52:06 +01:00
antirez c6da9d9fac Call clusterUpdateState() after CLUSTER SETSLOT too. 2013-02-21 16:31:22 +01:00
antirez 3a99d1228a Aesthetic change to make a line more 80-cols friendly. 2013-02-21 16:24:48 +01:00
antirez 5fd9f701da redis-trib: move instance vars in the right class. 2013-02-21 13:06:59 +01:00
antirez 7898bf4b7e redis-trib: some refactoring and skeleton of the "fix" command. 2013-02-21 13:00:41 +01:00
antirez dc4af60628 Cluster: clusterAddSlot() was not doing what stated in the comment. 2013-02-21 11:51:17 +01:00
antirez fdb57233e2 Cluster: always use cluster(Add|Del)Slot to modify the cluster slots table. 2013-02-21 11:44:58 +01:00
antirez 786b8d6c87 Use RESTORE-ASKING for MIGRATE when in cluster mode. 2013-02-20 17:36:54 +01:00
antirez ea7fc82a4a Cluster: new command flag forcing implicit ASKING. 
Also using this new flag the RESTORE-ASKING command was implemented that
will be used by MIGRATE.
 2013-02-20 17:28:35 +01:00
antirez 9ec1b709f5 Cluster: ASKING command fixed, state was not retained. 2013-02-20 17:07:52 +01:00
antirez b8d8b9ec41 redis-trib: set the migrating slot in the correct way when resharding. 2013-02-20 15:29:53 +01:00
antirez 9a04e12cc0 Cluster: I/O errors are now logged at DEBUG level. 2013-02-20 13:18:51 +01:00
antirez 917dd53216 redis-trib: make a few comments 80-cols friendly. 2013-02-15 17:11:55 +01:00
antirez 455da35c7f Cluster: specific error code for cluster down condition. 2013-02-15 16:53:24 +01:00
antirez 522c3255db Merge branch 'unstable' of github.com:antirez/redis into unstable 2013-02-15 16:45:04 +01:00
antirez 02796ba7a7 Cluster: sanity checks on the cluster bus message length. 2013-02-15 16:44:39 +01:00
antirez 8853698a6f Removed useless newlines from hashTypeCurrentObject(). 2013-02-15 13:12:55 +01:00
antirez 6b9c661838 Cluster: make valgrind happy initializing all the bytes of the node IP. 2013-02-15 12:58:35 +01:00
antirez 7371d5e248 Remove wrong decrRefCount() from getNodeByQuery(). 
This fixes issue #607.
 2013-02-15 11:57:53 +01:00
antirez 20f52b5b78 Top comment for getNodeByQuery() improved. 2013-02-15 11:50:54 +01:00
antirez 6b641f3aeb redis-cli: update prompt on cluster redirection. 2013-02-14 18:49:08 +01:00
antirez e0e15bd06d Cluster: with 16384 slots we need bigger buffers. 2013-02-14 15:36:33 +01:00
antirez 1a32d99b28 Cluster: move cluster config file out of config state. 
This makes us able to avoid allocating the cluster state structure if
cluster is not enabled, but still we can handle the configuration
directive that sets the cluster config filename.
 2013-02-14 15:20:02 +01:00
antirez 1649e509c3 Cluster: the cluster state structure is now heap allocated. 2013-02-14 13:20:56 +01:00
antirez 9dfd11c3da Cluster: Initialize ip and port in createClusterNode(). 2013-02-14 13:01:28 +01:00
antirez a26690e8b5 Cluster: redis-trib updated to use 16384 hash slots. 2013-02-14 12:55:34 +01:00
antirez ebd666db47 Cluster: from 4096 to 16384 hash slots. 2013-02-14 12:49:16 +01:00
antirez 072c91fe13 PSYNC: another change to unexpected reply from PSYNC. 2013-02-13 18:43:40 +01:00
antirez 0e1be5347b PSYNC: More robust handling of unexpected reply to PSYNC. 2013-02-13 18:33:33 +01:00
antirez 7404b95833 Avoid compiler warning by casting to match printf() specifier. 2013-02-13 13:38:20 +01:00
antirez 3419c8ce70 Replication: more strict error checking for master PING reply. 2013-02-12 16:53:27 +01:00
antirez dc24a6b132 Return a specific NOAUTH error if authentication is required. 2013-02-12 16:25:41 +01:00
antirez 24f258360b Replication: added new stats counting full and partial resynchronizations. 2013-02-12 15:33:54 +01:00
antirez ac8c89cb20 Test: avoid false positives in CLIENT SETNAME closed connection test. 2013-02-12 13:27:24 +01:00
antirez 7f9bc42f17 Merge remote-tracking branch 'origin/unstable' into unstable 2013-02-12 12:59:15 +01:00
antirez 04bdb3a2a4 Add missing bracket removed for error after rebase of PSYNC. 2013-02-12 12:56:32 +01:00
antirez 3af478e9ef PSYNC: debugging printf() calls are now logs at DEBUG level. 2013-02-12 12:52:22 +01:00
antirez 89b48f0825 Remove harmless warning in slaveTryPartialResynchronization(). 2013-02-12 12:52:21 +01:00
antirez 0ed6daa48b PSYNC: don't use the client buffer to send +CONTINUE and +FULLRESYNC. 
When we are preparing an handshake with the slave we can't touch the
connection buffer as it'll be used to accumulate differences between
the sent RDB file and what arrives next from clients.

So in short we can't use addReply() family functions.

However we just use write(2) because we know that the socket buffer is
empty, since a prerequisite for SYNC to work is that the static buffer
and the output list are empty, and in general it is not expected that a
client SYNCs after doing some heavy I/O with the master.

However a short write connection is explicitly handled to avoid
fragility (we simply close the connection and the slave will retry).
 2013-02-12 12:52:21 +01:00
antirez d2a0348a49 SYNC not allowed with pending data on the static output buffer. 2013-02-12 12:52:21 +01:00
antirez da315d3325 Log the unexpected string received in place of the SYNC payload length. 2013-02-12 12:52:21 +01:00
antirez 41d64a7516 After SLAVEOF <newslave> don't allow chained slaves to PSYNC. 2013-02-12 12:52:21 +01:00
antirez 078882025e PSYNC: work in progress, preview #2, rebased to unstable. 2013-02-12 12:52:21 +01:00
antirez e34a35a511 Use the new unified protocol to send SELECT to slaves. 
SELECT was still transmitted to slaves using the inline protocol, that
is conceived mostly for humans to type into telnet sessions, and is
notably not understood by redis-cli --slave.

Now the new protocol is used instead.
 2013-02-12 12:50:28 +01:00
antirez 4b83ad4e1f Use replicationFeedSlaves() to send PING to slaves. 
A Redis master sends PING commands to slaves from time to time: doing
this ensures that even if absence of writes, the master->slave channel
remains active and the slave can feel the master presence, instead of
closing the connection for timeout.

This commit changes the way PINGs are sent to slaves in order to use the
standard interface used to replicate all the other commands, that is,
the function replicationFeedSlaves().

With this change the stream of commands sent to every slave is exactly
the same regardless of their exact state (Transferring RDB for first
synchronization or slave already online). With the previous
implementation the PING was only sent to online slaves, with the result
that the output stream from master to slaves was not identical for all
the slaves: this is a problem if we want to implement partial resyncs in
the future using a global replication stream offset.

TL;DR: this commit should not change the behaviour in practical terms,
but is just something in preparation for partial resynchronization
support.
 2013-02-12 12:50:28 +01:00
antirez 7465ac7ab1 Emit SELECT to slaves in a centralized way. 
Before this commit every Redis slave had its own selected database ID
state. This was not actually useful as the emitted stream of commands
is identical for all the slaves.

Now the the currently selected database is a global state that is set to
-1 when a new slave is attached, in order to force the SELECT command to
be re-emitted for all the slaves.

This change is useful in order to implement replication partial
resynchronization in the future, as makes sure that the stream of
commands received by slaves, including SELECT commands, are exactly the
same for every slave connected, at any time.

In this way we could have a global offset that can identify a specific
piece of the master -> slaves stream of commands.
 2013-02-12 12:50:28 +01:00
Salvatore Sanfilippo 3043f890a5 Merge pull request #941 from svnpenn/patch-1 
Format to fit 80 columns
 2013-02-12 01:30:39 -08:00
antirez 1a27d41156 Makefile: valgrind target added (forces -O0 and libc malloc). 2013-02-11 12:11:28 +01:00
antirez 93ae95de74 TCP keep-alive. Better documentation in redis.conf. 2013-02-09 01:17:59 +01:00
Steven Penny 3244d8bc0c Format to fit 80 columns 
This makes it readable on GitHub and editors without auto wrapping.
 2013-02-08 12:11:06 -06:00
antirez 76a5b21c3a Tcp keep-alive: send three probes before detectin an error. 
Otherwise we end with less reliable connections because it's too easy
that a single packet gets lost.
 2013-02-08 17:06:01 +01:00
antirez 98b1a852b4 tcp-keepalive option documented in redis.conf. 2013-02-08 17:03:11 +01:00
antirez 124a635bc5 Set SO_KEEPALIVE on client sockets if configured to do so. 2013-02-08 16:40:59 +01:00
antirez ee21c18e5d Add SO_KEEPALIVE support to anet.c. 2013-02-08 16:30:26 +01:00
antirez a6c2f9012f Make all WATCHers dirty when the slave reloads the DB. 2013-02-08 10:26:19 +01:00
antirez 46dd4c62b3 LASTSAVE is a "random" command. 2013-02-07 19:13:00 +01:00
antirez b70b459b0e TCP_NODELAY after SYNC: changes to the implementation. 2013-02-05 12:04:30 +01:00
charsyam c85647f354 Turn off TCP_NODELAY on the slave socket after SYNC. 
Further details from @antirez:

It was reported by @StopForumSpam on Twitter that the Redis replication
link was strangely using multiple TCP packets for multiple commands.
This wastes a lot of bandwidth and is due to the TCP_NODELAY option we
enable on the socket after accepting a new connection.

However the master -> slave channel is a one-way channel since Redis
replication is asynchronous, so there is no point in trying to reduce
the latency, we should aim to reduce the bandwidth. For this reason this
commit introduces the ability to disable the nagle algorithm on the
socket after a successful SYNC.

This feature is off by default because the delay can be up to 40
milliseconds with normally configured Linux kernels.
 2013-02-05 12:04:25 +01:00
antirez fd7350b084 Test: No clients timeout while testing. 2013-02-05 12:02:46 +01:00
Salvatore Sanfilippo bb7aa1772c Merge pull request #906 from jbergstroem/tclsh-agnostic 
Make redis test suite tcl version agnostic, v2
 2013-02-05 02:43:56 -08:00
Salvatore Sanfilippo 58463e0505 Merge pull request #933 from RockLi/unstable 
retval doesn't initalized
 2013-02-05 01:48:49 -08:00
Rock Li 8063155cd0 retval doesn't initalized 
If each if conditions are all fail, variable retval will under uninitlized
 2013-02-05 15:56:04 +08:00
Salvatore Sanfilippo f95b393603 Merge pull request #928 from davidcelis/unstable 
Fix a few typos and improve grammar of redis.conf
 2013-02-04 01:32:31 -08:00
Salvatore Sanfilippo 74f4f29880 Merge pull request #929 from ltnwgl/unstable 
Fix a bug in srandmemberWithCountCommand()
 2013-02-04 01:21:18 -08:00
Gengliang Wang 002747336a Fix a bug in srandmemberWithCountCommand() 
In CASE 2, the call sunionDiffGenericCommand will involve the string "srandmember" 
> sadd foo one
(integer 1)
> sadd srandmember two
(integer 2)
> srandmember foo 3
1)"one"
2)"two"
 2013-02-04 14:01:08 +08:00
David Celis 81144645b4 Fix a few typos and improve grammar of redis.conf 
Make several edits to the example redis.conf configuration file for
improved flow and grammar.

Signed-off-by: David Celis <me@davidcel.is>
 2013-02-03 11:40:07 -08:00
antirez 089cbe643f Sentinel: advertise the promoted slave address only after successful setup. 2013-01-31 17:19:21 +01:00
Salvatore Sanfilippo aca005c246 Merge pull request #914 from catwell/unstable 
fix comments forgotten in #285 (zipmap -> ziplist)
 2013-01-31 03:37:48 -08:00
antirez ad297a1a67 Z*STORE event fixed: generate del only if resulting sorted set is empty. 2013-01-29 13:50:01 +01:00
antirez e41d1d77e3 Generate del events when S*STORE commands delete the destination key. 2013-01-29 13:43:13 +01:00
antirez d2b27f1d96 Tests for keyspace notifications. 2013-01-28 13:15:22 +01:00
antirez 4dfb5752e0 Send 'expired' events when a key expires by lookup. 2013-01-28 13:15:19 +01:00
antirez 562b2bd6a7 Keyspace notifications: fixed a leak and a bug introduced in the latest commit. 2013-01-28 13:15:16 +01:00
antirez fce016d31b Keyspace events: it is now possible to select subclasses of events. 
When keyspace events are enabled, the overhead is not sever but
noticeable, so this commit introduces the ability to select subclasses
of events in order to avoid to generate events the user is not
interested in.

The events can be selected using redis.conf or CONFIG SET / GET.
 2013-01-28 13:15:12 +01:00
antirez 1c0c551776 decrRefCount -> decrRefCountVoid in list constructor. 2013-01-28 13:15:08 +01:00
antirez 40f822da73 Enable keyspace events notification when testing. 2013-01-28 13:15:01 +01:00
antirez da04e6ed44 Keyspace events added for more commands. 2013-01-28 13:14:56 +01:00
antirez 8766e81079 Fix decrRefCount() prototype from void to robj pointer. 
decrRefCount used to get its argument as a void* pointer in order to be
used as destructor where a 'void free_object(void*)' prototype is
expected. However this made simpler to introduce bugs by freeing the
wrong pointer. This commit fixes the argument type and introduces a new
wrapper called decrRefCountVoid() that can be used when the void*
argument is needed.
 2013-01-28 13:14:53 +01:00
antirez 2d20e68fe4 notifyKeyspaceEvent(): release channel names using the right pointers. 2013-01-28 13:14:49 +01:00
antirez 5b9357a6b3 Initial test events for the new keyspace notification API. 2013-01-28 13:14:46 +01:00
antirez 2ea9518a53 Fixed over-80-cols comment in db.c 2013-01-28 13:14:42 +01:00
antirez 08d8bb57df Two fixes to initial keyspace notifications API. 2013-01-28 13:14:39 +01:00
antirez 4cdbce341e Keyspace events notification API. 2013-01-28 13:14:36 +01:00
Pierre Chapuis 50d43a9823 fix comments forgotten in #285 (zipmap -> ziplist) 2013-01-28 11:07:17 +01:00
Johan Bergström 1154283577 Use info nameofexectuable to find current executable 2013-01-24 09:37:18 +11:00
Johan Bergström acb046773d Enforce tcl 8.5 or newer 2013-01-24 09:36:59 +11:00
Johan Bergström a405979f5f Check available tcl versions 2013-01-24 09:25:47 +11:00
antirez 61853a9c75 Lua struct library updated to version 0.2. 
There was a bug in the previous version of this library that caused a
crash under the circumstances described in issue #901.

The newer version of the library appears to be fixed (I tested it
manually with valgrind and everything seems fine now).

For more information about this library please visit this web site:

    http://www.inf.puc-rio.br/~roberto/struct/
 2013-01-23 10:50:48 +01:00
antirez 88015b89a0 redis-cli --bigkeys output is now simpler to understand. 2013-01-21 19:15:58 +01:00
antirez 2039f1a38a UNSUBSCRIBE and PUNSUBSCRIBE: always provide a reply. 
UNSUBSCRIBE and PUNSUBSCRIBE commands are designed to mass-unsubscribe
the client respectively all the channels and patters if called without
arguments.

However when these functions are called without arguments, but there are
no channels or patters we are subscribed to, the old behavior was to
don't reply at all.

This behavior is broken, as every command should always reply.
Also it is possible that we are no longer subscribed to a channels but we
are subscribed to patters or the other way around, and the client should
be notified with the correct number of subscriptions.

Also it is not pretty that sometimes we did not receive a reply at all
in a redis-cli session from these commands, blocking redis-cli trying
to read the reply.

This fixes issue #714.
 2013-01-21 19:02:26 +01:00
antirez 93f61bb2a4 Fixed a bug in memtest progress bar, that had no actual effects. 
This closes issue #859, thanks to @erbenmo.
 2013-01-21 12:34:22 +01:00
Salvatore Sanfilippo 442235fe40 Merge pull request #869 from bilalhusain/patch-2 
s/adiacent/adjacent/
 2013-01-21 03:19:02 -08:00
Salvatore Sanfilippo 6db93e6740 Merge pull request #883 from deepsolo9/fix-typos-deepsolo9 
fixed some typos in Redis.c
 2013-01-21 03:12:03 -08:00
Salvatore Sanfilippo 4698f6756b Merge pull request #884 from kaerast/patch-1 
Typo in sentinel.conf
 2013-01-21 03:11:25 -08:00
antirez cd892d015d Not every __sun has backtrace(). 
I don't know how to test for Open Solaris that has support for
backtrace() so for now removing the #ifdef that breaks compilation under
other Solaris flavors.
 2013-01-21 12:06:54 +01:00
antirez e50cdbe461 Additionally two typos fixed thanks to @jodal 2013-01-19 13:46:14 +01:00
antirez 79a0ef62db Whitelist SIGUSR1 to avoid auto-triggering errors. 
This commit fixes issue #875 that was caused by the following events:

1) There is an active child doing BGSAVE.
2) flushall is called (or any other condition that makes Redis killing
the saving child process).
3) An error is sensed by Redis as the child exited with an error (killed
by a singal), that stops accepting write commands until a BGSAVE happens
to be executed with success.

Whitelisting SIGUSR1 and making sure Redis always uses this signal in
order to kill its own children fixes the issue.
 2013-01-19 13:30:38 +01:00
antirez ab247fc176 Clear server.shutdown_asap on failed shutdown. 
When a SIGTERM is received Redis schedules a shutdown. However if it
fails to perform the shutdown it must be clear the shutdown_asap flag
otehrwise it will try again and again possibly making the server
unusable.
 2013-01-19 13:19:41 +01:00
antirez 08d200baeb Slowlog: don't log EXEC but just the executed commands. 
The Redis Slow Log always used to log the slow commands executed inside
a MULTI/EXEC block. However also EXEC was logged at the end, which is
perfectly useless.

Now EXEC is no longer logged and a test was added to test this behavior.

This fixes issue #759.
 2013-01-19 12:53:21 +01:00
guiquanz 9d09ce3981 Fixed many typos. 2013-01-19 10:59:44 +01:00
Salvatore Sanfilippo 61dfc2e521 Merge pull request #887 from charsyam/redis-cli-prompt 
redis-cli prompt bug fix
 2013-01-19 01:32:28 -08:00
Salvatore Sanfilippo 74f137308e Merge pull request #895 from badboy/catch_con_error 
redis-cli: always exit if connection fails.
 2013-01-19 01:27:56 -08:00
Salvatore Sanfilippo 947f1689d3 Merge pull request #896 from bitterb/fix-client-command-error-message 
Fix an error reply for CLIENT command
 2013-01-19 01:25:01 -08:00
bitterb c2dc172a9d Fix an error reply for CLIENT command 2013-01-19 14:11:33 +09:00
Salvatore Sanfilippo 21b3ff9ba0 Merge pull request #893 from nparry/fix-redis-cli-rdb 
redis-cli --rdb fails if server sends a ping
 2013-01-18 02:41:29 -08:00
Jan-Erik Rediger 59046a7373 Always exit if connection fails. 
This avoids unnecessary core dumps. Fixes antirez/redis#894
 2013-01-18 10:13:10 +01:00
Nathan Parry 1920cab3bc redis-cli --rdb fails if server sends a ping 
Redis pings slaves in "pre-synchronization stage" with newlines. (See
https://github.com/antirez/redis/blob/2.6.9/src/replication.c#L814)
However, redis-cli does not expect this - it sees the newline as the end
of the bulk length line, and ends up returning 0 as bulk the length.
This manifests as the following when running redis-cli:

    $ ./src/redis-cli --rdb some_file
    SYNC sent to master, writing 0 bytes to 'some_file'
    Transfer finished with success.

With this commit, we just ignore leading newlines while reading the bulk
length line.

To reproduce the problem, load enough data into Redis so that the
preparation of the RDB snapshot takes long enough for a ping to occur
while redis-cli is waiting for the data.
 2013-01-18 00:10:58 -05:00
charsyam e396236651 redis-cli prompt bug fix 2013-01-16 17:20:54 -08:00
antirez a0c24821e2 redis-cli: save an RDB dump from remote server to local file. 2013-01-16 19:44:37 +01:00
Alice Kærast 8937f5e500 Typo in sentinel.conf 
Fixes a minor typo in sentinel.conf
 2013-01-16 15:02:06 +00:00
Hu Weiwei 1dc2d6dc47 fixed some typos. 2013-01-16 18:04:43 +08:00
antirez ea1f503cfe Tests for CLIENT GETNAME/SETNAME. 2013-01-15 13:34:17 +01:00
antirez 9b89ab06c4 Typo fixed, ASCI -> ASCII. 2013-01-15 13:34:13 +01:00
antirez 1971740f0c CLIENT GETNAME and CLIENT SETNAME introduced. 
Sometimes it is much simpler to debug complex Redis installations if it
is possible to assign clients a name that is displayed in the CLIENT
LIST output.

This is the case, for example, for "leaked" connections. The ability to
provide a name to the client makes it quite trivial to understand what
is the part of the code implementing the client not releasing the
resources appropriately.

Behavior:

    CLIENT SETNAME: set a name for the client, or remove the current
                    name if an empty name is set.
    CLIENT GETNAME: get the current name, or a nil.
    CLIENT LIST: now displays the client name if any.

Thanks to Mark Gravell for pushing this idea forward.
 2013-01-15 13:34:10 +01:00
antirez ef99e146a8 Undo slave-master handshake when SLAVEOF sets a new slave. 
Issue #828 shows how Redis was not correctly undoing a non-blocking
connection attempt with the previous master when the master was set to a
new address using the SLAVEOF command.

This was also a result of lack of refactoring, so now there is a
function to cancel the non blocking handshake with the master.
The new function is now used when SLAVEOF NO ONE is called or when
SLAVEOF is used to set the master to a different address.
 2013-01-15 13:33:24 +01:00
antirez a33869c330 Makefile.dep updated. 2013-01-11 23:50:32 +01:00
antirez f5fa6824db Comment in the call() function clarified a bit. 2013-01-10 11:19:40 +01:00
antirez a5cc063c17 Test: added regression for issue #872. 2013-01-10 11:10:31 +01:00
antirez baee565029 Multiple fixes for EVAL (issue #872). 
1) The event handler was no restored after a timeout condition if the
   command was eventually executed with success.
2) The command was not converted to EVAL in case of errors in the middle
   of the execution.
3) Terrible duplication of code without any apparent reason.
 2013-01-10 10:46:05 +01:00
Bilal Husain 717e5ffb45 s/adiacent/adjacent/ 
fixed typo in a comment (step 2 memcheck)
 2013-01-09 21:46:58 +05:30
antirez d7740fc8f3 Better error reporting when fd event creation fails. 2013-01-03 14:29:34 +01:00
antirez e9261b1e2d ae.c: set errno when error is not a failing syscall. 
In this way the caller is able to perform better error checking or to
use strerror() without the risk of meaningless error messages being
displayed.
 2013-01-03 14:29:20 +01:00
antirez 4dc1e0dd30 Fix overflow in mstime() in redis-cli and benchmark. 
The problem does not exist in the Redis server implementation of mstime()
but is only limited to redis-cli and redis-benchmark.

Thix fixes issue #839.
 2012-12-20 15:20:55 +01:00
antirez ebbc4ebb26 CONTRIBUTING updated with request to add BSD license. 2012-12-17 11:13:54 +01:00
antirez d01aad329c Added missing license and copyright in deps/hiredis. 2012-12-17 11:11:09 +01:00
antirez f1481d4a03 serverCron() frequency is now a runtime parameter (was REDIS_HZ). 
REDIS_HZ is the frequency our serverCron() function is called with.
A more frequent call to this function results into less latency when the
server is trying to handle very expansive background operations like
mass expires of a lot of keys at the same time.

Redis 2.4 used to have an HZ of 10. This was good enough with almost
every setup, but the incremental key expiration algorithm was working a
bit better under *extreme* pressure when HZ was set to 100 for Redis
2.6.

However for most users a latency spike of 30 milliseconds when million
of keys are expiring at the same time is acceptable, on the other hand a
default HZ of 100 in Redis 2.6 was causing idle instances to use some
CPU time compared to Redis 2.4. The CPU usage was in the order of 0.3%
for an idle instance, however this is a shame as more energy is consumed
by the server, if not important resources.

This commit introduces HZ as a runtime parameter, that can be queried by
INFO or CONFIG GET, and can be modified with CONFIG SET. At the same
time the default frequency is set back to 10.

In this way we default to a sane value of 10, but allows users to
easily switch to values up to 500 for near real-time applications if
needed and if they are willing to pay this small CPU usage penalty.
 2012-12-14 17:10:40 +01:00
Salvatore Sanfilippo a4d68dc541 Merge pull request #824 from ptjm/unstable 
Define _XOPEN_SOURCE appropriately on NetBSD.
 2012-12-12 09:34:41 -08:00
Patrick TJ McPhee 289942b625 Define _XOPEN_SOURCE appropriately on NetBSD. 2012-12-12 10:49:12 -05:00
antirez 705874e31d Fix config.h endianess detection to work on Linux / PPC64. 
Config.h performs endianess detection including OS-specific headers to
define the endianess macros, or when this is not possible, checking the
processor type via ifdefs.

Sometimes when the OS-specific macro is included, only __BYTE_ORDER is
defined, while BYTE_ORDER remains undefined. There is code at the end of
config.h endianess detection in order to define the macros without the
underscore, but it was not working correctly.

This commit fixes endianess detection fixing Redis on Linux / PPC64 and
possibly other systems.
 2012-12-11 17:01:00 +01:00
Salvatore Sanfilippo bf0852e5ed Merge pull request #807 from bmcmanus/issue-804 
Resolution for Issue #804 Add Default-Start and Default-Stop LSB tags for RedHat startup...
 2012-12-03 03:30:13 -08:00
antirez ab2924cff3 Memory leak fixed: release client's bpop->keys dictionary. 
Refactoring performed after issue #801 resolution (see commit
2f87cf8b01) introduced a memory leak that
is fixed by this commit.

I simply forgot to free the new allocated dictionary in the client
structure trusting the output of "make test" on OSX.

However due to changes in the "leaks" utility the test was no longer
testing memory leaks. This problem was also fixed.

Fortunately the CI test running at ci.redis.io spotted the bug in the
valgrind run.

The leak never ended into a stable release.
 2012-12-03 12:12:53 +01:00
antirez a18ca73681 Test: fixed osx "leaks" support in test. 
Due to changes in recent releases of osx leaks utility, the osx leak
detection no longer worked. Now it is fixed in a way that should be
backward compatible.
 2012-12-03 12:06:38 +01:00
Brian J. McManus bb58e517a9 Issue 804 Add Default-Start and Default-Stop LSB tags for RedHat startup and update-rc.d compatability. 2012-12-02 21:46:37 -07:00
antirez 2f87cf8b01 Blocking POP: use a dictionary to store keys clinet side. 
To store the keys we block for during a blocking pop operation, in the
case the client is blocked for more data to arrive, we used a simple
linear array of redis objects, in the blockingState structure:

    robj **keys;
    int count;

However in order to fix issue #801 we also use a dictionary in order to
avoid to end in the blocked clients queue for the same key multiple
times with the same client.

The dictionary was only temporary, just to avoid duplicates, but since
we create / destroy it there is no point in doing this duplicated work,
so this commit simply use a dictionary as the main structure to store
the keys we are blocked for. So instead of the previous fields we now
just have:

    dict *keys;

This simplifies the code and reduces the work done by the server during
a blocking POP operation.
 2012-12-02 20:43:15 +01:00
antirez c135b856c6 Test: regression for issue #801. 2012-12-02 20:43:11 +01:00
antirez 4e6dd7bc86 Client should not block multiple times on the same key. 
Sending a command like:

BLPOP foo foo foo foo 0

Resulted into a crash before this commit since the client ended being
inserted in the waiting list for this key multiple times.
This resulted into the function handleClientsBlockedOnLists() to fail
because we have code like that:

    if (de) {
        list *clients = dictGetVal(de);
        int numclients = listLength(clients);

        while(numclients--) {
            listNode *clientnode = listFirst(clients);

            /* server clients here... */
        }
    }

The code to serve clients used to remove the served client from the
waiting list, so if a client is blocking multiple times, eventually the
call to listFirst() will return NULL or worse will access random memory
since the list may no longer exist as it is removed by the function
unblockClientWaitingData() if there are no more clients waiting for this
list.

To avoid making the rest of the implementation more complex, this commit
modifies blockForKeys() so that a client will be put just a single time
into the waiting list for a given key.

Since it is Saturday, I hope this fixes issue #801.
 2012-12-02 20:43:07 +01:00
antirez f50e658455 SDIFF is now able to select between two algorithms for speed. 
SDIFF used an algorithm that was O(N) where N is the total number
of elements of all the sets involved in the operation.

The algorithm worked like that:

ALGORITHM 1:

1) For the first set, add all the members to an auxiliary set.
2) For all the other sets, remove all the members of the set from the
auxiliary set.

So it is an O(N) algorithm where N is the total number of elements in
all the sets involved in the diff operation.

Cristobal Viedma suggested to modify the algorithm to the following:

ALGORITHM 2:

1) Iterate all the elements of the first set.
2) For every element, check if the element also exists in all the other
remaining sets.
3) Add the element to the auxiliary set only if it does not exist in any
of the other sets.

The complexity of this algorithm on the worst case is O(N*M) where N is
the size of the first set and M the total number of sets involved in the
operation.

However when there are elements in common, with this algorithm we stop
the computation for a given element as long as we find a duplicated
element into another set.

I (antirez) added an additional step to algorithm 2 to make it faster,
that is to sort the set to subtract from the biggest to the
smallest, so that it is more likely to find a duplicate in a larger sets
that are checked before the smaller ones.

WHAT IS BETTER?

None of course, for instance if the first set is much larger than the
other sets the second algorithm does a lot more work compared to the
first algorithm.

Similarly if the first set is much smaller than the other sets, the
original algorithm will less work.

So this commit makes Redis able to guess the number of operations
required by each algorithm, and select the best at runtime according
to the input received.

However, since the second algorithm has better constant times and can do
less work if there are duplicated elements, an advantage is given to the
second algorithm.
 2012-11-30 16:36:42 +01:00
antirez b4abbaf755 redis-benchmark: seed the PRNG with time() at startup. 2012-11-30 15:41:09 +01:00
antirez 395d663d29 SDIFF fuzz test added. 2012-11-30 01:35:34 +01:00
antirez 925090f476 Make an EXEC test more latency proof. 2012-11-29 16:12:14 +01:00
antirez 2f62c9663c Introduced the Build ID in INFO and --version output. 
The idea is to be able to identify a build in a unique way, so for
instance after a bug report we can recognize that the build is the one
of a popular Linux distribution and perform the debugging in the same
environment.
 2012-11-29 14:20:08 +01:00
antirez b1b602a928 On crash memory test rewrote so that it actaully works. 
1) We no longer test location by location, otherwise the CPU write cache
completely makes our business useless.
2) We still need a memory test that operates in steps from the first to
the last location in order to never hit the cache, but that is still
able to retain the memory content.

This was tested using a Linux box containing a bad memory module with a
zingle bit error (always zero).

So the final solution does has an error propagation step that is:

1) Invert bits at every location.
2) Swap adiacent locations.
3) Swap adiacent locations again.
4) Invert bits at every location.
5) Swap adiacent locations.
6) Swap adiacent locations again.

Before and after these steps, and after step 4, a CRC64 checksum is computed.
If the three CRC64 checksums don't match, a memory error was detected.
 2012-11-29 10:24:35 +01:00
antirez 7383c3b129 Jemalloc updated to version 3.2.0. 2012-11-28 18:39:35 +01:00
charsyam dee0b939fc Remove unnecessary condition in _dictExpandIfNeeded (dict.c) 2012-11-28 11:44:39 +01:00
antirez c87a40897c Merge remote-tracking branch 'origin/unstable' into unstable 2012-11-28 11:41:27 +01:00
Matt Arsenault 504e5072eb It's a watchdog, not a watchdong. 2012-11-28 11:35:19 +01:00
Salvatore Sanfilippo 23d517d2c4 Merge pull request #787 from charsyam/remove-warning-bio 
remove compile warning bioKillThreads
 2012-11-23 03:44:18 -08:00
charsyam d7c7ac4a57 remove compile warning bioKillThreads 2012-11-23 05:52:39 +08:00
antirez 95f68f7b0f 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 15:50:00 +01:00
antirez cceb0c5b4a Fix integer overflow in zunionInterGenericCommand(). 
This fixes issue #761.
 2012-11-22 15:28:28 +01:00
antirez 65606b3bc6 Test: MULTI state is cleared after EXECABORT error. 2012-11-22 10:32:20 +01:00
antirez 4977ab79af Test: make sure EXEC fails after previous transaction errors. 2012-11-22 10:32:16 +01:00
antirez 9c00f07897 Test: MULTI/EXEC tests moved into multi.tcl. 2012-11-22 10:32:12 +01:00
antirez 3d1391272a 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 10:32:07 +01:00
antirez 7536991726 Make bio.c threads killable ASAP if needed. 
We use this new bio.c feature in order to stop our I/O threads if there
is a memory test to do on crash. In this case we don't want anything
else than the main thread to run, otherwise the other threads may mess
with the heap and the memory test will report a false positive.
 2012-11-22 10:12:11 +01:00
antirez 5a9e3f5842 Fast memory test on Redis crash. 2012-11-21 13:24:44 +01:00
antirez 3cb432837c Use more fine grained HAVE macros instead of HAVE_PROCFS. 2012-11-21 13:17:38 +01:00
charsyam 52b52a3508 fix randstring bug 2012-11-20 02:50:31 +08:00
antirez 49b6452351 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.
 2012-11-19 12:02:08 +01:00
antirez 3bfeb9c1a7 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.
 2012-11-19 11:47:35 +01:00
antirez af0b220756 zmalloc: kill unused __size parameter in update_zmalloc_stat_alloc() macro. 2012-11-14 12:52:38 +01:00
antirez a779b7e901 Merge branch 'migrate-cache' into unstable 2012-11-14 12:21:23 +01:00
antirez 989a7820ca Test: more MIGRATE tests. 2012-11-14 12:12:52 +01:00
antirez 2feef47aa1 MIGRATE: retry one time on I/O error. 
Now that we cache connections, a retry attempt makes sure that the
operation don't fail just because there is an existing connection error
on the socket, like the other end closing the connection.

Unfortunately this condition is not detectable using
getsockopt(SO_ERROR), so the only option left is to retry.

We don't retry on timeouts.
 2012-11-14 11:30:24 +01:00
antirez 17411f7afd Test: check if MIGRATE is caching connections. 2012-11-14 10:58:34 +01:00
antirez aa2bf6ba8b TTL API change: TTL returns -2 for non existing keys. 
The previous behavior was to return -1 if:

1) Existing key but without an expire set.
2) Non existing key.

Now the second case is handled in a different, and TTL will return -2
if the key does not exist at all.

PTTL follows the same behavior as well.
 2012-11-12 23:04:36 +01:00
antirez 05705bc8bb 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 18:54:35 +01:00
antirez c8852ebf19 MIGRATE count of cached sockets in INFO output. 2012-11-12 14:01:56 +01:00
antirez 149b527a74 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 14:01:02 +01:00
antirez e23d281e48 MIGRATE TCP connections caching. 
By caching TCP connections used by MIGRATE to chat with other Redis
instances a 5x performance improvement was measured with
redis-benchmark against small keys.

This can dramatically speedup cluster resharding and other processes
where an high load of MIGRATE commands are used.
 2012-11-12 00:47:24 +01:00
antirez 5513397de7 Copyright date fixed in COPYING file. 2012-11-08 19:14:42 +01:00
antirez 399669fc6f Make clear that contributing code to the Redis project means to release it under the terms of the BSD license. 2012-11-08 18:43:20 +01:00
antirez 4365e5b2d3 BSD license added to every C source and header file. 2012-11-08 18:31:32 +01:00
antirez 1237d71c4e COPY and REPLACE options for MIGRATE. 
With COPY now MIGRATE does not remove the key from the source instance.
With REPLACE it uses RESTORE REPLACE on the target host so that even if
the key already eixsts in the target instance it will be overwritten.

The options can be used together.
 2012-11-07 15:32:27 +01:00
antirez e5b5763f56 REPLACE option for RESTORE. 
The REPLACE option deletes an existing key with the same name (if any)
and materializes the new one. The default behavior without RESTORE is to
return an error if a key already exists.
 2012-11-07 10:57:23 +01:00
antirez c4b0b6854e Type mismatch errors are now prefixed with WRONGTYPE. 
So instead to reply with a generic error like:

-ERR ... wrong kind of value ...

now it replies with:

-WRONGTYPE ... wrong kind of value ...

This makes this particular error easy to check without resorting to
(fragile) pattern matching of the error string (however the error string
used to be consistent already).

Client libraries should return a specific exeption type for this error.

Most of the commit is about fixing unit tests.
 2012-11-06 20:25:34 +01:00
Salvatore Sanfilippo 06851a93de Merge pull request #741 from Run/typo 
fix a typo in redis.h line 595 comment
 2012-11-02 04:10:47 -07:00
antirez 05d8e2c938 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 22:39:39 +01:00
Yecheng Fu f0266532fc fix typo in comments (redis.c, networking.c) 2012-11-01 22:26:46 +01:00
antirez 2ea41242f6 Unix socket clients properly displayed in MONITOR and CLIENT LIST. 
This also fixes issue #745.
 2012-11-01 22:10:45 +01:00
antirez d36fb95a10 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 15:40:48 +01:00
Runzhen Wang c23c657cdd fix a typo in redis.h line 595 comment 2012-11-01 02:14:22 +08:00
Salvatore Sanfilippo 021d16b643 Merge pull request #726 from yamt/typo 
fix a typo in a comment
 2012-10-31 01:29:04 -07:00
antirez 973cb21a01 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 09:23:05 +01:00
antirez b16e423430 No longer used macro rdbIsOpcode() removed. 2012-10-30 19:10:46 +01:00
antirez 08a4f12e76 help.h update (adds bitop, bitcount, evalsha...) 2012-10-30 18:57:20 +01:00
antirez c5cca554d9 Ctrl+w support in linenoise. 2012-10-26 16:06:25 +02:00
antirez 89423052ca Marginally more robust glibc version test for sync_file_range detection. 2012-10-26 15:55:12 +02:00
charsyam 4800331bf8 patch config.h for sync_file_range 2012-10-26 04:27:58 +08:00
antirez d2f83d4a76 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 22:01:20 +02:00
antirez 68fc64afd4 Update memory peak stats while loading RDB / AOF. 2012-10-24 12:21:41 +02:00
YAMAMOTO Takashi 164d57c60d fix a typo in a comment 2012-10-24 17:47:56 +09:00
antirez 89e74abfb6 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 19:21:47 +02:00
Salvatore Sanfilippo ecd82f59fe Merge pull request #693 from ghurrell/dict-h-typos 
Fix (cosmetic) typos in dict.h
 2012-10-22 02:55:23 -07:00
Schuster e5f794ff3a 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 11:44:20 +02:00
antirez c2661ed761 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 10:43:39 +02:00
antirez acfe3675e3 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 10:31:28 +02:00
antirez a1b1c1ea3a 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 17:35:50 +02:00
antirez be6cbd3a6e 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 18:34:05 +02:00
Salvatore Sanfilippo d10a01bb6d Merge pull request #707 from NanXiao/patch-1 
Update src/redis-benchmark.c
 2012-10-10 02:18:14 -07:00
NanXiao 9eb3a7bc6b 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).
 2012-10-10 17:08:43 +08:00
antirez 278304cc4d CONTRIBUTING file updated. 2012-10-06 12:04:27 +02:00
dvir volk 34f37fb7d3 fixed server install script to rewrite the default configuration file and not a template, and removed the old config template 
Conflicts:

	utils/redis.conf.tpl
 2012-10-05 12:32:58 +02:00
antirez da920e75d4 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 11:20:13 +02:00
antirez c43aea7e9f 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 11:16:22 +02:00
antirez 2b73b3509a Include time.h in ae.c as we now use time(). 2012-10-05 10:10:43 +02:00
Jokea b7b2a1cc5e 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 19:30:42 +02:00
antirez f0b9f80345 "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 11:52:16 +02:00
antirez 9a914a632d "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 14:54:43 +02:00
Greg Hurrell 4b1f6ad3e7 Fix (cosmetic) typos in dict.h 2012-10-02 22:01:26 -07:00
antirez ece77037e9 Revert "Scripting: redis.NIL to return nil bulk replies." 
This reverts commit e061d797d7.

Conflicts:

	src/scripting.c
 2012-10-01 10:10:31 +02:00
antirez 9c21b72bb9 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 16:54:57 +02:00
antirez 6dd1693c0e 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 14:26:20 +02:00
antirez db100c4671 Sentinel: Support for AUTH. 2012-09-26 18:59:54 +02:00
antirez 0ee3f05518 Test for SRANDMEMBER with <count>. 2012-09-21 11:55:36 +02:00
antirez 578c94597f 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 11:55:32 +02:00
antirez be90c803e3 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 11:55:28 +02:00
antirez d310fbedab Fix compilation on FreeBSD. Thanks to @koobs on twitter. 2012-09-17 12:46:06 +02:00
antirez b5b2aceaf6 .gitignore modified to be more general with less entries. 2012-09-17 10:49:48 +02:00
antirez 7eb850ef0e 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 10:26:46 +02:00
antirez bfc197c3b6 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 10:32:04 +02:00
antirez 978e5177fd Match printf format with actual type in genRedisInfoString(). 2012-09-10 12:42:55 +02:00
antirez 74e57d0ece 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 17:50:10 +02:00
Haruto Otake 749aac72ad 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 16:19:04 +02:00
Salvatore Sanfilippo 24bc807b5c Merge pull request #576 from saj/fix-slave-ping-period 
Bug fix: slaves being pinged every second
 2012-09-05 06:59:37 -07:00
antirez 36741b2c81 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 01:17:49 +02:00
antirez 9bd0e097aa 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.
 2012-09-04 16:06:53 +02:00
antirez 8bdde086ac 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.
 2012-09-04 15:52:04 +02:00
antirez bb66fc3120 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.
 2012-09-02 12:24:38 +02:00
antirez e323635c2d 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 11:15:02 +02:00
antirez 46c31a150a 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 10:28:13 +02:00
antirez 6276434ad2 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.
 2012-08-30 18:01:52 +02:00
antirez 58186b9dcf Sentinel: INFO command implementation. 2012-08-29 12:44:24 +02:00
antirez 8246e58abe 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.
 2012-08-29 11:44:01 +02:00
antirez 712656e848 Sentinel: added documentation about slave-priority in redis.conf 2012-08-28 17:53:18 +02:00
antirez 3ec701e059 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".
 2012-08-28 17:45:01 +02:00
antirez 169a44cbd6 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.
 2012-08-28 17:20:26 +02:00
antirez c14e0ecafd 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.
 2012-08-28 12:56:05 +02:00
antirez 784b93087c 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 12:47:33 +02:00
antirez 1caa627e4e Fix a forget zmalloc_oom() -> zmalloc_oom_handler() replacement. 2012-08-24 15:40:22 +02:00
antirez 6fdc635447 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 12:55:37 +02:00
antirez 850789ce73 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.
 2012-08-24 12:29:54 +02:00
antirez 01477753e6 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.
 2012-08-24 12:10:24 +02:00
Salvatore Sanfilippo 724371d748 Merge pull request #628 from pietern/unstable-zip 
Fix ziplist edge case
 2012-08-22 02:32:27 -07:00
antirez 227b429364 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).
 2012-08-21 17:31:44 +02:00
antirez d6704c9bd0 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.
 2012-08-21 17:27:01 +02:00
Pieter Noordhuis 3cc2904e3e Set p to its new offset before modifying it 2012-08-13 14:13:09 -07:00
Pieter Noordhuis 5117c20ab5 Add ziplist test for deleting next to last entries 2012-08-13 14:09:40 -07:00
antirez cada7f9671 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.
 2012-08-03 12:41:27 +02:00
antirez 6275004ca6 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.
 2012-08-02 18:40:30 +02:00
Salvatore Sanfilippo 2990c366e8 Merge pull request #613 from tobstarr/patch-1 
Fix version numbers
 2012-08-02 05:37:10 -07:00
Tobias Schwab 2dccf1e4ce Fix version numbers 2012-08-02 15:10:51 +03:00
antirez fd92b366b0 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.
 2012-07-31 17:11:26 +02:00
antirez 91c15ed1b5 Sentinel: sentinelRefreshInstanceInfo() comments improved a bit. 2012-07-31 16:18:15 +02:00
Michael Parker f1d187bb3e 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 11:48:00 +02:00
antirez ed2a691aba Sentinel: sentinel.conf self-documenation improved. 2012-07-31 11:05:21 +02:00
antirez 75084e057d 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.
 2012-07-31 10:19:34 +02:00
antirez 7f5bdba434 Merge remote-tracking branch 'origin/unstable' into unstable 2012-07-28 20:55:17 +02:00
antirez 3f194a9d25 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.
 2012-07-28 20:54:27 +02:00
Jeremy Zawodny 78b606acc2 comment fix 
improve English a bit. :-)
 2012-07-28 20:54:26 +02:00
Salvatore Sanfilippo f3a6143199 Merge pull request #606 from badboy/patch-1 
Include sys/wait.h to avoid compiler warning
 2012-07-28 02:51:56 -07:00
Jan-Erik Rediger c6c19c8372 Include sys/wait.h to avoid compiler warning 
gcc warned about an implicit declaration of function 'wait3'. 
Including this header fixes this.
 2012-07-28 12:33:01 +03:00
Salvatore Sanfilippo ed7b308c1c Merge pull request #587 from saj/truncate-short-write-from-aof 
Truncate short write from the AOF
 2012-07-27 03:56:48 -07:00
Salvatore Sanfilippo 04950a9e4d Merge pull request #586 from saj/aof_last_bgrewrite_status 
New in INFO: aof_last_bgrewrite_status
 2012-07-27 03:55:20 -07:00
antirez ce7b838fb9 Sentinel: don't start a failover as leader if there is no good slave. 2012-07-26 12:09:40 +02:00
Salvatore Sanfilippo a3593ee4d0 Merge pull request #604 from jzawodn/patch-1 
comment fix
 2012-07-25 08:31:57 -07:00
Jeremy Zawodny 7617a6f272 comment fix 
improve English a bit. :-)
 2012-07-25 08:29:11 -07:00
antirez baace5fc42 Sentinel: ability to execute notification scripts. 2012-07-25 16:33:37 +02:00
Salvatore Sanfilippo 42c571864e Merge pull request #603 from mrb/fix_sentinel_config_warning 
Fix warning in redis.c for sentinel config load
 2012-07-25 07:15:53 -07:00
Salvatore Sanfilippo 7c15dd1596 Merge pull request #602 from mrb/sentinel_docs 
Some cleanup in sentinel.conf
 2012-07-25 07:15:02 -07:00
mrb f1c8661e74 Fix warning in redis.c for sentinel config load 2012-07-25 09:55:53 -04:00
mrb 23023fc626 Some cleanup in sentinel.conf 2012-07-25 09:52:26 -04:00
antirez 672102c2ce 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.
 2012-07-25 11:32:19 +02:00
antirez 9e5bef38e6 Sentinel: reset pending_commands in a more generic way. 2012-07-24 18:57:26 +02:00
antirez a23a5b6c7d 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.
 2012-07-24 18:46:04 +02:00
antirez d918e6f127 Sentinel: debugging message removed. 2012-07-24 18:20:05 +02:00
antirez 75fb6e5b8a 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.
 2012-07-24 18:15:44 +02:00
antirez 2179c26916 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.
 2012-07-24 16:37:57 +02:00
antirez d876d6feac 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.
 2012-07-24 12:42:40 +02:00
antirez 1bbdf1709f Fixed an error in the example sentinel.conf. 2012-07-23 15:08:36 +02:00
antirez be2be3d903 Typo. 2012-07-23 15:06:55 +02:00
antirez 6b5daa2df2 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
 2012-07-23 13:14:44 +02:00
antirez 03f412ddef Merge remote-tracking branch 'origin/unstable' into unstable 2012-07-22 17:18:42 +02:00
antirez 5d73073f6e 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 17:18:16 +02:00
Salvatore Sanfilippo cddf5946fc Merge pull request #593 from steevel/unstable 
Check that we have connection before enabling pipe mode
 2012-07-21 11:12:35 -07:00
antirez b62bdf1c64 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 12:04:58 +02:00
Saj Goonatilleke 55302e9e28 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.
 2012-07-18 10:35:17 +10:00
Saj Goonatilleke 48553a29e8 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.)
 2012-07-18 09:54:55 +10:00
Steeve Lennmark e9828cb6f7 Check that we have connection before enabling pipe mode 2012-07-15 14:35:02 +02:00
Salvatore Sanfilippo a2db8e4801 Merge pull request #569 from jokea/unstable 
mark fd as writable when EPOLLERR or EPOLLHUP is returned by epoll_wait.
 2012-07-09 03:14:08 -07:00
Saj Goonatilleke 9edfe63553 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.
 2012-07-05 14:29:27 +10:00
jokea 93b0075d33 mark fd as writable when EPOLLERR or EPOLLHUP is returned by epoll_wait. 2012-06-29 12:06:38 +08:00
antirez 36def8fd9a Typo in comment. 2012-06-27 11:26:44 +02:00
antirez 3a32897856 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.
 2012-06-27 09:43:57 +02:00
antirez 5410168c6e Fixed comment typo into time_independent_strcmp(). 2012-06-21 14:25:53 +02:00
antirez 31a1439bfd 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 11:50:01 +02:00
antirez 5b63ccce6c 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 10:03:25 +02:00
antirez ba779119b8 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.
 2012-06-14 16:01:27 +02:00
Salvatore Sanfilippo 96b8ff3760 Merge pull request #552 from tnm/unstable 
Standardize punctuation in redis-cli help.
 2012-06-13 01:25:52 -07:00
Ted Nyman d665dd0865 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.
 2012-06-12 22:35:00 -07:00
antirez 84d9ef4f31 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 15:21:54 +02:00
antirez ee789e157c 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 00:41:48 +02:00
antirez c0de45924c 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 15:19:46 +02:00
antirez 80e808b6d6 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 23:29:57 +02:00
Alex Mitrofanov 51857c7e5c 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 10:24:27 +02:00
antirez c7a25200e2 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 10:21:57 +02:00
antirez 1419406e8d 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 21:52:47 +02:00
antirez bc70b8e5f4 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 15:20:59 +02:00
antirez 33e1db36fa 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 12:11:30 +02:00
antirez d866803818 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 15:20:20 +02:00
antirez fa4a5d5922 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 15:20:16 +02:00
antirez 7c34643f15 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 15:20:11 +02:00
antirez 80f8028e3c 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 15:20:06 +02:00
antirez 01d3a7e736 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 15:20:02 +02:00
antirez 343d3bd287 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 15:19:58 +02:00
antirez dbbbe49ef5 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 15:19:55 +02:00
antirez 760e776526 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 15:19:51 +02:00
antirez a3f2b4895b 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 15:19:48 +02:00
antirez 0bd6d68e34 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 15:19:43 +02:00
antirez 6f05a65336 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 15:19:18 +02:00
antirez 47ca4b6e28 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 15:19:15 +02:00
antirez ef37997608 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 11:35:21 +02:00
jokea e150ce3ce6 Set fd to writable when poll(2) detects POLLERR or POLLHUP event. 2012-05-23 11:33:32 +02:00
antirez 348ee1a40a 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 11:12:43 +02:00
antirez 2bcd18a2e9 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.
 2012-05-22 13:13:24 +02:00
antirez 5a55999304 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.
 2012-05-21 16:50:05 +02:00
antirez d44a36e006 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.
 2012-05-20 23:49:08 +02:00
antirez d165f78561 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 12:23:45 +02:00
antirez ad4c0b4117 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 11:09:45 +02:00
Pieter Noordhuis 3c72c94aae Whitespace 2012-05-15 11:19:01 +02:00
Dave Pacheco 6d31277048 use port_getn instead of port_get 2012-05-15 11:18:58 +02:00
Dave Pacheco 05da63da0c first cut at event port support 2012-05-15 11:18:54 +02:00
Pieter Noordhuis f16d090c72 Fix install target on OSX (see #495) 2012-05-15 11:18:50 +02:00
antirez e9f0419c99 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 17:35:51 +02:00
antirez b3624f5a16 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 16:04:41 +02:00
antirez 61daf8914d 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.
 2012-05-13 21:52:35 +02:00
antirez 9434349236 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.
 2012-05-13 16:40:29 +02:00
antirez f333788fbc Comment improved so that the code goal is more clear. Thx to @agladysh. 2012-05-11 22:33:28 +02:00
antirez 1dcc95d081 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.
 2012-05-11 19:17:31 +02:00
antirez 27737964c4 If the computer running the Redis test is slow, we revert to --clients 1 to avoid false positives. 2012-05-11 17:26:16 +02:00
antirez ea66be6080 redis-cli pipe mode: handle EINTR properly as well so that SIGSTOP/SIGCONT are handled correctly. 2012-05-11 16:08:57 +02:00
antirez f6bd9122c0 redis-cli pipe mode: handle EAGAIN while writing to socket. 2012-05-11 10:45:12 +02:00
antirez 8c6301462b Fix PREFIX typo in Makefile. 2012-05-09 20:45:19 +02:00
antirez b26808a532 Allow PREFIX to be overrided in Makefile. 2012-05-09 10:34:52 +02:00
antirez 088c508abc redis-cli --pipe for mass import. 2012-05-07 16:37:18 +02:00
Pieter Noordhuis bf21941686 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.
 2012-05-06 10:06:21 +02:00
Salvatore Sanfilippo abf54ee73c Merge pull request #494 from quiver/init-fixes 
fix several bugs of init.d scripts
 2012-05-04 04:41:02 -07:00
quiver c4286feabe fix several bugs of init.d scripts 
- PIDFILE environ variable was not properly retrieved
- chkconfig command failed
 2012-05-04 20:07:00 +09:00
antirez af3853c3bf 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 22:41:50 +02:00
antirez 299290d3a4 Remove useless trailing space in SYNC command sent to master. 2012-05-02 21:47:53 +02:00
antirez ae62d29d1d 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 20:57:55 +02:00
antirez 92997d06ea Test "Turning off AOF kills the background writing child if any" is now more reliable. 2012-05-02 11:40:46 +02:00
Pieter Noordhuis cc4f65fea4 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.
 2012-05-01 10:52:03 +02:00
Harmen 2ac546e00c Show problem with 'keys' command with specific command sequence. 2012-05-01 10:51:59 +02:00
antirez d9241b35e5 Properly wait the slave to sync with master in BRPOPLPUSH test. 2012-04-30 10:55:03 +02:00
antirez 2d4b55214f A more lightweight implementation of issue 141 regression test. 2012-04-29 17:16:44 +02:00
antirez ffe003dcbe memtest.c fixed to actually use v1 and v2 in memtest_fill_value(). 2012-04-27 16:29:44 +02:00
antirez 841048f2c8 redis-cli commands description in help.h updated. 2012-04-27 15:57:27 +02:00
antirez 0ad10db220 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 11:41:25 +02:00
antirez a3eb7ac87d 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.
 2012-04-26 16:53:11 +02:00
antirez 11bd247d2b Produce the stack trace in an async safe way. 2012-04-26 16:28:54 +02:00
antirez 3ada43e732 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.
 2012-04-26 16:21:19 +02:00
antirez 28ccb53008 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.
 2012-04-26 11:25:13 +02:00
antirez 57356b80cd Merge remote-tracking branch 'origin/unstable' into unstable 2012-04-26 11:17:47 +02:00
antirez 5080e625d3 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.
 2012-04-26 11:16:52 +02:00
Salvatore Sanfilippo d445988c68 Merge pull request #481 from dtran320/unstable 
Spelling: s/synchrnonization/synchronization
 2012-04-25 13:39:13 -07:00
David Tran 31788f50b7 Spelling: s/synchrnonization/synchronization 2012-04-25 12:21:56 -07:00
antirez d9237055ba Explicitly use bash for install_server.sh. Fixes issue #397 2012-04-24 19:14:03 +02:00
antirez 29c8cf0116 redis-check-dump now is RDB version 6 ready. 2012-04-24 19:05:27 +02:00
antirez 7dbc514de7 Spurious debugging printf removed. 2012-04-24 17:15:21 +02:00
antirez a76b9063d0 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 17:04:00 +02:00
antirez 053d56a1fa rdbLoad() should check REDIS_RDB_VERSION instead of hardcoded number. 2012-04-24 12:53:30 +02:00
antirez f22cff43a6 ziplist.c: added comments about the new 24 bit encoding. 2012-04-24 12:52:36 +02:00
Grisha Trubetskoy 5a86ab4799 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 12:02:19 +02:00
antirez 21661d7acc Fixed a bug in propagation of PUBLISH via the cluster bus. 
This bug was spotted by clang on FreeBSD.
 2012-04-24 11:28:10 +02:00
antirez e54fe9a79f A few compiler warnings suppressed. 2012-04-24 11:11:55 +02:00
antirez a66a496349 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 11:11:35 +02:00
antirez c4a4755286 Ziplist encoding now tested with negative integers as well. 2012-04-23 17:27:46 +02:00
Salvatore Sanfilippo d344228734 Merge pull request #421 from ErikDubbelboer/sds-const 
Added consts keyword where possible
 2012-04-23 02:34:19 -07:00
Salvatore Sanfilippo c22e0eecd5 Merge pull request #461 from schlenk/unstable 
Replace some unnecessary calls to echo and cat in tests
 2012-04-23 02:07:21 -07:00
antirez 4e7cc35f1a New time limit for protocol desync test set to 30 seconds to reduce false positives. 2012-04-23 10:57:43 +02:00
antirez a3f990818d Remove loadfile() access from the scripting engine. 2012-04-23 10:43:24 +02:00
antirez eb6e7eb94d Even inside #if 0 comments are comments. 2012-04-21 21:49:21 +02:00
antirez 92cc20ecfe Merge remote-tracking branch 'origin/unstable' into unstable 2012-04-21 20:35:51 +02:00
antirez d3701d2714 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 20:34:45 +02:00
Salvatore Sanfilippo 7d3ee4172f Merge pull request #440 from ErikDubbelboer/spelling 
Fixed some spelling errors in comments
 2012-04-21 03:31:06 -07:00
Salvatore Sanfilippo 28d1ee2cda Merge pull request #414 from chobits/unstable 
fix typos in tests/assets/default.conf
 2012-04-21 03:27:15 -07:00
antirez fd72fe261d README now makes clear that our support for solaris derived systems is "best effort". 2012-04-21 12:08:26 +02:00
antirez 60e2e5b50d redis.conf AOF section comments improved. 2012-04-20 00:04:07 +02:00
antirez 47db53c3c3 New tests related to scripts max execution time. 2012-04-19 23:49:33 +02:00
antirez e3923a3508 SHUTDOWN NOSAVE now can stop a non returning script. Issue #466. 2012-04-19 23:35:15 +02:00
antirez 1e35ae7486 Currenly not used code in dict.c commented out. 2012-04-18 23:56:07 +02:00
antirez 5498e7bc76 Tests for scripting PRNG. 2012-04-18 23:50:16 +02:00
antirez 5386f72d00 redis-cli --bigkeys output modified to be simpler to read.. 2012-04-18 20:53:37 +02:00
antirez f26761aa10 redis-cli --bigkeys 2012-04-18 20:33:02 +02:00
antirez cddfd67ed4 Test SDIFF with first set empty. 2012-04-18 18:13:31 +02:00
antirez 4264459141 Test SINTER against same integer elements, but different set encoding. 2012-04-18 18:10:48 +02:00
antirez f800942f64 Test SINTER with non existing key. 2012-04-18 18:05:02 +02:00
antirez 88f77a2b7e Added an SMOVE test where src and dest key are the same. 2012-04-18 18:00:12 +02:00
antirez e51a218aaa New hash fuzzing test. 2012-04-18 17:56:17 +02:00
antirez 5ba79bda7f Document mostly dead code in RPOPLPUSH implementation. 2012-04-18 17:38:02 +02:00
antirez 79daddd5ba Explicit RPOP/LPOP tests. 2012-04-18 17:32:48 +02:00
antirez 7aa0dfd014 Test LINSERT syntax error. 2012-04-18 17:22:14 +02:00
antirez 94e5c1d7e5 Test LINDEX out of range index. 2012-04-18 17:17:53 +02:00
Salvatore Sanfilippo d162f9eb64 Merge pull request #456 from pietern/unstable-mkopts 
Persist Makefile flags (unstable)
 2012-04-18 07:25:59 -07:00
antirez acf41c96cb 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 11:37:14 +02:00
antirez 8ec9b03c6b More robust maxclients test. 2012-04-18 11:34:18 +02:00
antirez a5f8341245 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 11:31:24 +02:00
Michael Schlenker 875944a23f 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.
 2012-04-17 22:20:54 +02:00
antirez 6a3f0ac68a Added test for SORT corner case: pattern ending with just "->". 2012-04-17 16:28:59 +02:00
antirez 3c25c4a691 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 13:05:09 +02:00
antirez 0705ff3f04 Less false positives in maxclients test, hopefully. 2012-04-17 10:04:42 +02:00
Pieter Noordhuis 7e7b69fee1 Persist make settings and trigger rebuild if anything changes 2012-04-13 17:52:33 -07:00
Pieter Noordhuis 34c943b395 Don't set flags recursively 2012-04-13 17:52:33 -07:00
Pieter Noordhuis 620357fc8f Remove unused LIBS variable 2012-04-13 17:52:33 -07:00
Pieter Noordhuis caba585121 First set defaults, then do composition 2012-04-13 17:52:33 -07:00
Pieter Noordhuis c04278ba3b Question mark assignment is not needed 2012-04-13 17:52:33 -07:00
Pieter Noordhuis 166cf8a3b8 Ignore gcov/lcov artifacts 2012-04-13 17:52:33 -07:00
Pieter Noordhuis 0342dd7647 The lcov target shouldn't clean 
This is not needed because every change in compiler/linker flags
triggers a cleanup.
 2012-04-13 17:52:30 -07:00
Pieter Noordhuis 631539a5f2 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 17:34:31 -07:00
antirez 93a74949d7 Merge branch 'strict.lua' into unstable 2012-04-13 16:16:13 +02:00
antirez c18405c909 A few var names fixed in Makefile. 
I modified it for error in a previous commit doing search & replace.
 2012-04-13 16:13:56 +02:00
antirez 13a21caae3 New test for scripting engine: DECR_IF_GT. 2012-04-13 15:23:32 +02:00
antirez 6f659f34cf EVAL errors are more clear now. 2012-04-13 15:12:16 +02:00
antirez 3a02140415 Use Lua tostring() before concatenation. 2012-04-13 14:54:49 +02:00
antirez 3cd4ad267c Tests modified to match the new global protection implementation. 2012-04-13 13:40:57 +02:00
antirez d86c4a7bf0 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 13:36:08 +02:00
antirez 6663653f51 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 13:26:59 +02:00
antirez c9edd1b28a Globals protection global() function modified for speed and correctness. 2012-04-13 12:13:02 +02:00
antirez 2fd7c9efde Tests for lua globals protection. 2012-04-13 11:48:45 +02:00
antirez 37b29ef2fa Scripting: globals protection can now be switched on/off. 2012-04-13 11:23:45 +02:00
antirez 6e05f333a2 Print arch bits with redis-server -v 2012-04-12 11:50:18 +02:00
antirez 4c442e9d88 memtest.c: integer overflow fixed. 2012-04-12 11:49:52 +02:00
antirez 91e56965a3 Make gcov fixed. 2012-04-12 11:51:58 +02:00
antirez 9c83aec955 Makefile now introduces Redis-specific CFLAGS / LDFLAGS. Gcov target fixed. Added comments to describe how it works. 2012-04-12 11:09:38 +02:00
Salvatore Sanfilippo 0d83011a11 Merge pull request #449 from ErikDubbelboer/unstable 
added explanation for the magic 511 backlog number
 2012-04-11 09:25:44 -07:00
Erik Dubbelboer 815c06f2f7 added explanation for the magic 511 backlog number 2012-04-11 17:04:31 +02:00
antirez ced068955d Makefile.dep updated. 2012-04-11 12:12:30 +02:00
antirez a470689d2e make dep: redirect output to Makefile.dep. 2012-04-11 12:12:05 +02:00
antirez c44ab51da1 Make inline functions rioRead/Write/Tell static. This fixes issue #447. 2012-04-11 11:58:32 +02:00
antirez d98b03b791 Macros ULONG_ONEZERO / ULONG_ZEROONE were inverted in #ifdef to test 32/64 bit arch. 2012-04-11 11:32:22 +02:00
Pieter Noordhuis 0a08d2b0e5 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 11:24:17 +02:00
Pieter Noordhuis 3f7438ef9b Everything x86 is little endian 2012-04-11 11:03:30 +02:00
antirez f2f2ba1b3a Comment typo fixed. Clusetr -> Cluster. 2012-04-11 10:57:02 +02:00
antirez 0b913c650d Check write(2) return value to avoid warnings, because in this context failing write is not critical. 2012-04-10 16:48:28 +02:00
antirez a3fb7fd4f6 Minor MIGRATE implementation simplification about ttl handling. 2012-04-10 16:46:29 +02:00
antirez 0882715710 Version 2.9.7. 2012-04-10 16:34:33 +02:00
antirez 84bcd3aa24 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 15:47:10 +02:00
antirez 82e32055d8 RDB files now embed a crc64 checksum. Version of RDB bumped to 5. 2012-04-09 22:40:41 +02:00
antirez 8491f1d9fd Fixed compilation of new rio.c changes (typos and so forth.) 2012-04-09 12:36:44 +02:00
antirez 46738646d4 dump/restore fixed to use the new crc64 API. 2012-04-09 12:33:57 +02:00
antirez 736b7c3f04 Add checksum computation to rio.c 2012-04-09 12:33:09 +02:00
antirez 88c1d9550d crc64.c modified for incremental computation. 2012-04-09 12:20:47 +02:00
antirez 5a181d43cb rio.c file somewhat documented so that the casual reader can understand what's going on without reading the code. 2012-04-09 11:11:00 +02:00
antirez a3af8d8e49 Test for maxclients. 2012-04-08 11:16:40 +02:00
Erik Dubbelboer e1d9857b12 Update src/rdb.c 2012-04-07 15:48:30 +03:00
Erik Dubbelboer 8d16e7a3c6 Update src/dict.c 2012-04-07 15:45:53 +03:00
Erik Dubbelboer 65fd32ab0a Fixed some spelling errors in the comments 2012-04-07 14:40:29 +02:00
antirez 69e7958918 Added new test to check that "CONFIG appendonly no" actually kills the background AOF operation in progress if any. 2012-04-07 13:22:04 +02:00
antirez 2cbdab903f For coverage testing use exit() instead of _exit() when termiating saving children. 2012-04-07 12:11:23 +02:00
antirez 4ce22375f4 Tests for MONITOR. 2012-04-07 11:26:24 +02:00
antirez 09f66a0be8 New client info field added to CLIENT LIST output: multi, containing the length of the current pipeline. Test modified accordingly. 2012-04-07 11:14:52 +02:00
antirez 4cb8bb2952 Never used function stringObjectEqualsMs() removed. 2012-04-07 02:10:48 +02:00
antirez f02aa7bb79 Removed dead code: function rdbSaveTime() is no longer used since RDB now saves expires in milliseconds. 2012-04-07 02:03:29 +02:00
antirez 0daf7a2d16 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.
 2012-04-06 23:52:53 +02:00
antirez ff79ab71db redis.tcl: no longer leave unread replies if an error happens during a MULTI/EXEC block. 2012-04-06 23:52:28 +02:00
antirez 618a922957 New INFO field in persistence section: bgrewriteaof_scheduled. 2012-04-06 21:12:50 +02:00
antirez 321a724da5 version bumped to 2.9.6 2012-04-06 12:27:17 +02:00
jokea 3e1e1ac27d implement aeWait using poll(2). Fixes issue #267. 2012-04-06 11:47:17 +02:00
antirez 70381bbf82 expireGenericCommand(): better variable names and a top-comment that describes the function's behavior. 2012-04-05 15:52:08 +02:00
Premysl Hruby c6bf4a0034 for (p)expireat use absolute time, without double recomputation 2012-04-05 15:46:21 +02:00
Premysl Hruby d48d1309c6 fix mstime() ommited while comparing if key is already expired 2012-04-05 15:46:15 +02:00
Premysl Hruby c35b4e845b add support for generation of lcov coverage reports 2012-04-05 12:23:35 +02:00
Salvatore Sanfilippo 0934a4df93 Merge pull request #430 from anydot/f-testsuite 
Two changes of testsuite (killing redis and new option)
 2012-04-05 01:56:12 -07:00
Salvatore Sanfilippo d84f776e87 Merge pull request #426 from anydot/fix-rm-vm-comments 
remove mentions of VM in comments
 2012-04-05 01:54:09 -07:00
Salvatore Sanfilippo 0d5f4ba7cd Merge pull request #431 from anydot/f-signal 
allocate alternate signal stack, change of sigaction flags for sigterm
 2012-04-05 01:52:40 -07:00
Premysl Hruby ebba7b3c92 future-proof version comparison 2012-04-05 10:41:28 +02:00
antirez 459e2975f4 On slow computers, 10 seconds are not enough for this heavy replication test. 2012-04-04 19:54:23 +02:00
antirez 4b8c966140 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 18:32:22 +02:00
antirez 5ad1faa090 New "os" field in INFO output providing information about the operating system. 2012-04-04 15:38:13 +02:00
antirez bb0fbc840d SLAVEOF is not a write command. 2012-04-04 15:11:30 +02:00
antirez 9a322ab730 Print milliseconds of the current second in log lines timestamps. Sometimes precise timing is very important for debugging. 2012-04-04 15:11:17 +02:00
Premysl Hruby c575766202 allocate alternate signal stack, change of sigaction flags for sigterm 2012-04-03 17:40:31 +02:00
antirez 1ad4d50805 redis-cli help.h updated. Script to generate it updated as well. 2012-04-03 15:30:33 +02:00
antirez 15113dcd3d More MIGRATE tests. 2012-04-03 15:10:51 +02:00
antirez 12e91892a0 Another fix for MIGRATE. 2012-04-03 15:10:42 +02:00
Premysl Hruby 9184f8fd00 in kill_server send the signal once, then wait for up to 5sec before sending lethal SIGKILL 2012-04-03 14:20:52 +02:00
Premysl Hruby 7cc17ed8ff new option for choosing number of test clients to run 2012-04-03 14:20:52 +02:00
antirez 84e5684bca Two fixed for MIGRATE: fix TTL propagation and fix transferring of data bigger than 64k. 2012-04-03 12:17:40 +02:00
antirez e7957ca628 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 11:53:45 +02:00
antirez 06e178708e MIGRATE test modified because the implementation changed. 2012-04-02 16:38:59 +02:00
antirez 31f2ecf436 MIGRATE now let the client distinguish I/O errors and timeouts from other erros. 2012-04-02 16:38:24 +02:00
antirez f8ea19e539 DUMP/RESTORE now use CRC64 instead of truncated SHA1. 2012-04-02 13:10:39 +02:00
antirez 9510d65dc8 CRC64 implementation added to Redis code base. 2012-04-02 12:31:44 +02:00
Premysl Hruby 8918de9202 remove mentions of VM in comments 2012-04-02 11:56:03 +02:00
antirez ab8232d022 DUMP, RESTORE, MIGRATE tests. 2012-04-02 11:44:25 +02:00
antirez 70d848e1fa RESTORE ability to set a TTL fixed, bug introduced with millisecond expires. 2012-04-02 11:14:47 +02:00
antirez a149ce6875 Prettify source code of create/verify DUMP payload. 2012-04-02 10:52:39 +02:00
antirez bd04465931 DUMP / RESTORE: store RDB version in little endian. 2012-04-02 10:46:24 +02:00
antirez 4de6c9a055 New DUMP format includes RDB version and truncated SHA1 checksum. 2012-04-01 12:51:40 +02:00
antirez 2281f90c9b Merge remote-tracking branch 'origin/unstable' into unstable 2012-04-01 10:15:07 +02:00
Salvatore Sanfilippo a7138a4cf8 Merge pull request #425 from ThePicard/unstable 
chkconfig in install_server.sh doesn't work because of a typo
 2012-04-01 01:14:52 -07:00
ThePicard b0c6ee1d1d Fixed a typo in install_server.sh 2012-03-31 23:39:58 -07:00
antirez d0ace5a314 Write RDB magic using a REDIS_RDB_VERSION define that is defined inside rdb.h 2012-03-31 17:08:40 +02:00
antirez 609ea08445 redis-trib: fix the MIGRATE call that now has milliseconds timeout (were seconds before). 2012-03-31 11:28:37 +02:00
antirez 9157549fad syncio.c calls in replication.c fixed for the new millisecond timeout API. 2012-03-31 11:23:30 +02:00
antirez 04d360fdcd Better syncio.c with millisecond resolution. 2012-03-31 11:21:45 +02:00
Erik Dubbelboer 0a2ff52eae Added consts keyword where possible 2012-03-30 21:19:51 +02:00
antirez c2672a06cd Purely aesthetic code change. 2012-03-30 10:39:34 +02:00
Joseph Jang f892797e1b 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 10:34:29 +02:00
antirez 22c9c4076b Regression test for issue 417 (memory leak when replicating to DB with id >= 10) 2012-03-30 10:26:07 +02:00
Xiaochen Wang f2305046ef fix typos in tests/assets/default.conf 2012-03-29 18:23:04 +08:00
antirez 0cdecca141 Protect globals access in Lua scripting. 2012-03-29 12:02:28 +02:00
antirez 4cba71e898 Fixed typo in comment: "te" -> "the". 2012-03-29 09:33:29 +02:00
antirez 179e54d2a9 Fix for slaves chains. Force resync of slaves (simply disconnecting them) when SLAVEOF turns a master into a slave. 2012-03-29 09:24:02 +02:00
antirez 6aa2f98938 Test for redis.sha1hex(). 2012-03-28 20:47:50 +02:00
Nathan Fritz 52ae8af807 added redis.sha1hex(string) as lua scripting function. 
(The original implementation was modified by @antirez to conform Redis
coding standards.)
 2012-03-28 20:37:40 +02:00
antirez 5471b8babd Fixes for redisLogFromHandler(). 2012-03-28 13:51:23 +02:00
antirez a7d12cbaf1 Log from signal handlers is now safer. 2012-03-28 13:45:39 +02:00
antirez 1043c8064b Merge branch 'watchdog' into unstable 2012-03-28 13:16:19 +02:00
antirez 0fefb5bbeb 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 10:55:17 +02:00
Salvatore Sanfilippo e4669c045d Merge pull request #389 from huangz1990/patch-1 
fix typo
 2012-03-27 14:00:08 -07:00
Salvatore Sanfilippo 44f26aea0c Merge pull request #407 from anydot/fixes 
Fixes around time usage
 2012-03-27 13:58:24 -07:00
Premysl Hruby b57dbdbba3 remove disk-store related comments 2012-03-27 18:46:51 +02:00
Premysl Hruby 8af9fe841c declare hashDictType as external too 2012-03-27 18:18:57 +02:00
Premysl Hruby d194905449 use server.unixtime instead of time(NULL) where possible (cluster.c not checked though) 2012-03-27 17:39:58 +02:00
Premysl Hruby 024f213b12 fix time() instead of mstime() in expireIfNeeded 2012-03-27 17:31:21 +02:00
antirez d174eed5d7 define zlibc_free() in a way that is not shadowed by jemalloc. 2012-03-27 16:54:53 +02:00
antirez 23c0cdd2ad Produce the watchlog warning log in a way that is safer from a signal handler. Fix a memory leak in the backtrace generation function. 2012-03-27 15:24:33 +02:00
antirez aa96122d96 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.
 2012-03-27 13:48:57 +02:00
antirez a354da9acd Correctly set the SIGARLM timer for the software watchdog. 2012-03-27 12:11:37 +02:00
antirez 39bd025c29 Redis software watchdog. 2012-03-27 11:47:51 +02:00
antirez a323870450 SIGSEGV handler refactored so that we can reuse stack trace and current client logging functionalities in other contexts. 2012-03-27 10:40:07 +02:00
antirez ca09ad4d64 CONFIG RESETSTAT resets two more fields. 2012-03-25 11:43:19 +02:00
antirez c1d01b3c57 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 11:27:35 +02:00
antirez 512f682340 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 11:02:16 +02:00
antirez 0f9997845a Merge remote-tracking branch 'origin/unstable' into unstable 2012-03-24 12:07:58 +01:00
antirez 74f10793c8 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 12:07:14 +01:00
antirez 1b247d1333 Add used allocator in redis-server -v output. 2012-03-24 11:53:03 +01:00
antirez c79373482f convert-zipmap-hash-on-load test enabled 2012-03-24 11:42:20 +01:00
antirez 6c658d5554 Contextualize comment. 2012-03-23 20:24:40 +01:00
antirez 58e34e6cb1 Fixed memory leak in hash loading. 2012-03-23 20:24:34 +01:00
antirez 03116904c3 RDB load of different encodings test added. 2012-03-23 20:24:30 +01:00
antirez bd376d13f8 Big endian fix. The bug was introduced because of a typo. 2012-03-23 12:42:20 +01:00
antirez 6f0e77ca19 Replicate HINCRBYFLOAT as HSET. 2012-03-23 10:22:58 +01:00
antirez 7b558b1d64 Code style hack. 2012-03-22 18:17:09 +01:00
antirez 1f6146df0c 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 18:17:05 +01:00
antirez ba864e09d4 Comments about security of slave-read-only in redis.coinf. 2012-03-21 12:26:05 +01:00
antirez b22eab8faf Correctly create shared.oomerr as an sds string. 2012-03-21 12:11:07 +01:00
antirez 7dcdd281f5 DEBUG should not be flagged as w otherwise we can not call DEBUG DIGEST and other commands against read only slaves. 2012-03-20 17:53:47 +01:00
antirez f3fd419fc9 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.
 2012-03-20 17:32:48 +01:00
antirez 7a0c72f345 redis_init_script template updated. 2012-03-20 13:07:59 +01:00
antirez 0d44d50792 Suppress warnings compiling redis-cli with certain gcc versions. 2012-03-19 19:28:49 +01:00
antirez bb0aadbe21 Read-only flag removed from PUBLISH command. 2012-03-19 19:16:41 +01:00
antirez d4a515c56d Memory addressing test implemented. 2012-03-19 14:02:34 +01:00
antirez d033ccb0af More memory tests implemented. Default number of iterations lowered to a more acceptable value of 50. 2012-03-18 18:03:27 +01:00
antirez a5801142a4 Fixed typo. 2012-03-18 17:27:56 +01:00
antirez 1a197a3c1a 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 17:25:00 +01:00
antirez 525be599a8 On crash suggest to give --test-memory a try. 2012-03-18 11:35:35 +01:00
antirez fb068dc91d Memory test function now less boring thanks to screen-wide progress bar. 2012-03-16 21:19:53 +01:00
antirez 54e0fa1c27 Hem... actual memtest.c file added. 2012-03-16 17:21:49 +01:00
antirez c5166e3fc5 First implementation of --test-memory. Still a work in progress. 2012-03-16 17:17:39 +01:00
antirez c9d3dda29c 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 20:55:14 +01:00
huangz1990 9448ddb0c6 fix typo 2012-03-15 14:27:14 +08:00
antirez ae22bf1ef6 Reclaim space from the client querybuf if needed. 2012-03-14 15:32:30 +01:00
antirez 739803c064 sds.c: sdsAllocSize() function added. 2012-03-14 14:58:26 +01:00
antirez 9555f8f21b 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 10:13:23 +01:00
antirez 529bde82ec Call all the helper functions needed by clientsCron() as clientsCronSomething() for clarity. 2012-03-14 09:56:22 +01:00
antirez 6df450b77c CLIENT LIST test modified to reflect the new output. 2012-03-13 18:06:29 +01:00
antirez d19015be12 Process async client checks like client timeouts and BLPOP timeouts incrementally using a circular list. 2012-03-13 18:05:11 +01:00
antirez bbaeda402c Added a qbuf-free field to CLIENT LIST output. 2012-03-13 13:26:33 +01:00
antirez e74dca73d9 Client creation time in redisClient structure. New age field in CLIENT LIST output. 2012-03-13 13:05:08 +01:00
antirez 573373802e c->bufpos initialization moved for aesthetics. 2012-03-13 12:59:27 +01:00
antirez c3e7441dad 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 11:00:18 +01:00
antirez c3c856228d RDB hashes loading fixed removing the assertion that failed every time an HT-encoded hash was loaded. 2012-03-13 09:49:11 +01:00
Salvatore Sanfilippo 79642420b0 Merge pull request #378 from quiver/unstable 
fix typo of redis.conf
 2012-03-10 04:22:29 -08:00
quiver 96e9f8d5e3 fix typo of redis.conf 2012-03-10 21:09:34 +09:00
antirez 64b4c33c0b Build dependencies updated. 2012-03-10 12:40:03 +01:00
antirez f12d0224f3 RDB4 support in redis-check-dump. 2012-03-10 12:38:42 +01:00
antirez c7d7d0a80f 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 12:35:31 +01:00
392 changed files with 63992 additions and 19748 deletions
Expand all files Collapse all files
.gitignore
+6 -5
View File
@@ -1,12 +1,13 @@
.*.swp
*.o
*.rdb
*.log
redis-cli
redis-server
dump.rdb
redis-benchmark
redis-check-dump
redis-check-aof
redis-check-dump
redis-cli
redis-sentinel
redis-server
doc-tools
release
misc/*
@@ -16,7 +17,6 @@ SHORT_TERM_TODO
release.h
src/transfer.sh
src/configs
src/redis-server.dSYM
redis.ds
src/redis.conf
deps/lua/src/lua
@@ -24,3 +24,4 @@ deps/lua/src/luac
deps/lua/src/liblua.a
.make-*
.prerequisites
*.dSYM
00-RELEASENOTES
+279 -73
View File
@@ -1,83 +1,289 @@
Redis 2.6 release notes
Redis 3.0 release notes
=======================
Migrating from 2.4 to 2.6
WARNING: Redis 3.0 is currently a BETA not suitable for production environments.
--------------------------------------------------------------------------------
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 3.0.0 Beta 7 (version 2.9.56) ] Release date: 30 jun 2014
This is the 7th beta of Redis 3.0.0.
>> General changes
* [FIX] Scripting fixes backported from unstable, see Redis 2.8.12 changelog
for more info. (Salvatore Sanfilippo)
* [FIX] Cancel SHUTDOWN if initial AOF is being written. (Matt Stancliff)
* [NEW] New command: COMMAND, for commands introspection (Matt Stancliff &
Salvatore Sanfilippo)
* [NEW] hiredis: Update to latest version. (Matt Stancliff)
* [NEW] Jemalloc updated to 3.6.0. (Salvatore Sanfilippo)
>> Cluster changes
* [FIX] Cluster: clear NOADDR flag when updating node address.
(Salvatore Sanfilippo)
* [NEW] New CLUSTER SLOTS command to simplify Cluster clients operations.
(Matt Stancliff)
* [NEW] More Cluster tests. (Salvatore Sanfilippo)
* [NEW] Log when failover authorization are granted / denied.
(Salvatore Sanfilippo)
>> Sentinel changes
* [FIX] A few Sentinel bugs fixed and improvements, see Redis 2.8.12
changelog for more info. (Salvatore Sanfilippo & Matt Stancliff)
* [NEW] New Sentinel-Client handshake protocol, ROLE command, CLIENT KILL,
all backported to 3.0 branch. (Salvatore Sanfilippo)
--[ Redis 3.0.0 Beta 6 (version 2.9.55) ] Release date: 9 jun 2014
This is the 6th beta of Redis 3.0.0.
>> General changes
* [FIX] Fix software watchdog signal handler crash due to re-entering.
* [FIX] Better Lua number -> string conversion for Lua scripts.
* [FIX] Serious replication bug when min-slaves-* feature is used in slaves
configuration fixed.
* [FIX] Blocking pop on lists now works when the list is created by commands
other than *PUSH* (for example SORT STORE).
>> Cluster changes
* [FIX] CRITICAL: For a bug in the implementation of CLUSTER SET-CONFIG-EPOCH
introduced with beta-3 (especially beta-4 where the command
is actually used by redis-trib), a configEpoch for a node could
jump backward, breaking the eventual consistency property of the
slots -> nodes mapping in the cluster.
>> Sentinel changes
* No changes for Sentinel in this release.
--[ Redis 3.0.0 Beta 5 (version 2.9.54) ] Release date: 26 may 2014
This is the 5th beta of Redis 3.0.0. It does not include any real
worthwhile change (just three days passed since the previous beta), but
fixes two stupid bugs preventing cluster tests to pass.
--[ Redis 3.0.0 Beta 4 (version 2.9.53) ] Release date: 23 may 2014
This is the fourth beta of Redis 3.0.0.
>> General changes
* [NEW] Scripting engine performances improvements.
* [NEW] Log format slightly changed to report current node role.
* [FIX] Correct the HyperLogLog stale cache flag to prevent unnecessary
computation.
>> Cluster changes
* [NEW] redis-trib: ability to import data from standalone Redis instances.
* [NEW] redis-trib: "fix" subcommand much better at fixing errors.
* [NEW] CLUSTER FAILOVER FORCE implemented.
* [NEW] CLUSTER RESET implemented, it is now possible to completely reset
nodes and create a new cluster without restarting anything.
* [NEW] Slave validity factor (max estimated data age to still failover)
is now under the control of the user, and can be configured via
redis.conf or CONFIG SET. Option name cluster-slave-validity-factor.
* [NEW] Cluster test: failure detection and failover initial tests.
* [NEW] CLUSTER MEET: better error messages when address is invalid.
* [NEW] Bulk-accept new Cluster nodes in the Cluster bus instead of
performing just a single accept() per event fired.
* [FIX] Bypass data_age check for manual failovers.
* [FIX] Fixed data_age computation / check integer overflow.
* [FIX] Various fixes to Tcl client.tcl Redis Cluster client used in tests.
* [FIX] Better handling of stolen slots.
* [FIX] Don't accept cluster bus connections during startup.
>> Sentinel changes
* [NEW] Generate +config-update-from event when a new config is received.
* [NEW] Log when a failover will be re-attempted again.
* [FIX] Sentinel: Add "dir /tmp" directive in example sentinel.conf.
--[ Redis 3.0.0 Beta 3 (version 2.9.52) ] Release date: 5 may 2014
This is the third beta of Redis 3.0.0.
>> General changes
* [NEW] New data structure: the HyperLogLog (see 2.8 release notes).
* [NEW] Lexicograhical range queries in sorted sets (see 2.8 release notes).
* [NEW] LRU algorithm precision greatly improved.
* [FIX] Redis is now much more responsive to reply with LOADING / BUSY errors.
>> Cluster changes
* [NEW] Cluster testing framework and initial tests.
* [NEW] Cluster epochs collision resolution (make Redis Cluster more resilient
to admin and programming errors).
* [NEW] Persist / fsync some global state to ensure correct crash-recovery
semantics.
* [NEW] New command introduced: CLUSTER SET-CONFIG-EPOCH, still not used
by redis-trib. Will be used to speedup the assignment of unique
epochs to different nodes at cluster creation time. For now this is
handled as a side effect of the cluster epochs collision resolution.
* [FIX] Different fixes to redis-trib cluster creation.
* [FIX] Fix an error in the CLUSTER NODES output for nodes slots.
>> Sentinel changes
* [NEW] Sentinels are now able to send update messages in a peer-to-peer
fashion even if no Redis instances are available. Now the Sentinel
liveness property that the most updated configuration in a given
partition is propagated to all the Sentinels is extended to partitions
without reachable instances.
* [NEW] Sentinel safety properties are now ensured in a crash-recovery system
model since some state is persisted on disk before replying to other
nodes, and reloaded at startup.
* [NEW] Sentinel now uses CLIENT SETNAME so that it is easy to identify
Sentinels using CLIENT LIST among other clients.
* [NEW] Sentinel failure detection and reconnection code improved.
--[ Redis 3.0.0 Beta 2 (version 2.9.51) ] Release date: 11 mar 2014
This is the second beta of Redis 3.0.0.
>> General changes
* [FIX] Sometimes the absolute config file path was obtained in a wrong way.
This happened when there was a "dir" directive inside the config file
and at the same time the configuration file was given as a relative
path to redis-server or redis-sentinel executables.
* [FIX] redis-cli: Automatically enter --slave mode when SYNC or PSYNC are
called during an interactive session.
* [FIX] BITCOUNT: fixed unaligned access causing issues in sparc and other
archs not capable of dealing with unaligned accesses. This also makes
the code faster in archs where unaligned accesses are allowed.
* [FIX] Force INFO used_memory_peak to match peak memory. This generated some
confusion among users even if it was not an actual bug.
* [FIX] Fixed an critical EVALSHA script cache bug: scripts executed may not
propagate to AOF / Slaves correctly under certain conditions.
See issue #1549 at Github for more information.
* [FIX] Fixed multiple bugs resulting into closing the link with master or slave
during replication without good reasons. This will result in useless
resynchronizations, or infinite loops where the replication link can't
be established.
* [FIX] Don't count the time needed to populate the buffers of clients waiting
in MONITOR mode when populating the Slow Log entries.
* [NEW] Redis-cli updated to use SCAN instead of random sampling via
RANDOMKEY in order to implement --bigkeys feature. Moreover the
implementation now supports pipelining and reports more information
at the end of the scan. Much faster, much better. A special thank
you to Michael Grunder for this improvement.
* [NEW] redis-cli now supports a new --intrinsic-latency mode that is able
to meter the latency of a system due to kernel / hypervisor.
How to use it is explained at http://redis.io/topics/latency.
* [NEW] New command BITPOS: find first bit set or clear in a bitmap.
* [NEW] CONFIG REWRITE calls are now logged.
* [NEW] AOF write errors (like no space on device) no longer abort Redis if the
fsync policy is none or every second. The database enters a read-only
mode where every write is refused with an error. Normal operations are
restored as soon as Redis is able to append again data to the AOF file.
* [NEW] Sentinel now accepts SHUTDOWN command.
>> Cluster changes
* [FIX] Bind the first interface listed in the "bind" configuration directive
if any, in order to perform outgoing connections. This fixes Cluster
usage when an address is bound but there are multiple interfaces that
may be used to connect with other nodes.
* [FIX] When an "Importing" slot is closed via CLUSTER SETSLOT NODE ...
increment the configEpoch in the special case it is zero.
* [FIX] Current transaction is invalidated on redirection errors.
* [FIX] Abort if port does not allow for a valid cluster bus port that is
always at fixed +10000 offset.
* [FIX] Keys extraction algorithm fixed for ZUNIONSTORE/ZINTERSTORE and SORT.
* [FIX] Better failover timeout and retry times: failover should now work
reliabily when node-timeout is very small (a few milliseconds).
* [FIX] Don't allow SORT GET/BY options in Cluster mode.
* [FIX] Clear importing/migrating state when turning from master to slave role.
* [FIX] Set slot error if we receive an update for a busy slot.
* [FIX] Update node configEpoch on UPDATE messages.
* [NEW] Support multi-key operations as long as keys resolve to the same
hash slot, and the slot is not migrating, or it is migrating but all
the mentioned keys are available.
* [NEW] New DEBUG command CMDKEYS available to debug / test keys identification
in Redis commands.
* [NEW] redis-trib: create subcommand is now able to assign spare slaves.
* [NEW] redis-trib: new subcommand 'call'. Exec command in all nodes.
>> Sentinel changes
* [FIX] Sentinel "IDONTKNOW" error removed as it does not made sense with the
new Sentinel design. This error was actually a fix for a design error
in the first implementation of Sentinel.
* [FIX] Sentinel: added a missing exit() call to abort after config file
checks at startup. This error was introduced with an improvement in
a previous 2.8 release.
* [FIX] Sentinel: better nodes fail over start time desynchronization to avoid
split-brain during the voting process needed to get authorization to
fail over. This means the system is less likely to need to retry
and will fail over faster. No changes in behavior / correctness.
* [NEW] Sentinel unit tests and framework. More tests needed and units must
be improved in order to have less false positives, but it is a start
and features a debugging console that is useful to fix tests or to
inspect bugs causing tests failures.
* [NEW] New Sentinel events: +/-monitor and +set used to monitor when an
instance to monitor is added or removed, or when a configuration
is modified via SENTINEL SET.
--[ Redis 3.0.0 Beta 1 (version 2.9.50) ] Release date: 11 Feb 2014
This is the first beta of Redis 3.0.0.
The following is a list of improvements in Redis 3.0, compared to Redis 2.8.
* [NEW] Redis Cluster: a distributed implementation of a subset of Redis.
* [NEW] New "embedded string" object encoding resulting in less cache
misses. Big speed gain under certain work loads.
* [NEW] WAIT command to block waiting for a write to be transmitted to
the specified number of slaves.
* [NEW] MIGRATE connection caching. Much faster keys migraitons.
* [NEW] MIGARTE new options COPY and REPLACE.
* [NEW] CLIENT PAUSE command: stop processing client requests for a
specified amount of time.
Migrating from 2.8 to 3.0
=========================
Redis 2.4 is mostly a strict subset of 2.6. However there are a few things
that you should be aware of:
Redis 3.0 is mostly a strict subset of 2.8, you should not have any problem
upgrading your application from 2.8 to 3.0. However this is a list of small
non-backward compatible changes introduced in the 3.0 release:
* You can't use .rdb and AOF files generated with 2.6 into a 2.4 instance.
* 2.4 slaves can be attached to 2.6 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.
Also the following redis.conf and CONFIG GET / SET parameters changed name:
* hash-max-zipmap-entries, now replaced by hash-max-ziplist-entries
* hash-max-zipmap-value, now replaced by hash-max-ziplist-value
* glueoutputbuf was no completely removed as it does not make sense
---------
CHANGELOG
---------
What's new in Redis 2.6.0
=========================
UPGRADE URGENCY: We suggest new users to start with 2.6.0, and old users to
upgrade after some testing of the application with the new
Redis version.
* 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, ...).
* Clinets max output buffer soft and hard limits. You can specifiy different
limits for different classes of clients (normal,pubsub,slave).
* 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.
* 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.
* INCRBYFLOAT and HINCRBYFLOAT commands.
* The log format was modified. The prefix of each line included the pid
in the following format [1234]. Now instead it is 1234:? Where
'?' is actually the role of the instance. M for master, S for slave, C
if this process is a saving child (for RDB/AOF), and X for Sentinel.
--------------------------------------------------------------------------------
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.
Credits: Where not specified the implementation and design is done by
Salvatore Sanfilippo. Thanks to Pivotal 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.
CONTRIBUTING
+24 -3
View File
@@ -1,8 +1,29 @@
1. Enter irc.freenode.org #redis and start talking with 'antirez' and/or 'pietern' to check if there is interest for such a feature and to understand the probability of it being merged. We'll try hard to keep Redis simple... so you'll likely encounter high resistance.
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. You will include BSD license in the COPYING file within
each source file that you contribute.
2. Drop a message to the Redis Google Group with a proposal of semantics/API.
# IMPORTANT: HOW TO USE REDIS GITHUB ISSUES
3. If steps 1 and 2 are ok, use the following procedure to submit a patch:
* 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 step 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)
COPYING
+1 -1
View File
@@ -1,4 +1,4 @@
Copyright (c) 2006-2009, Salvatore Sanfilippo
Copyright (c) 2006-2014, 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
-1032
View File
File diff suppressed because it is too large Load Diff
MANIFESTO
+55 -8
View File
@@ -4,17 +4,64 @@
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.
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.
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.
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.
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.
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.
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
+4
View File
@@ -5,3 +5,7 @@ default: all
.DEFAULT:
cd src && $(MAKE) $@
install:
cd src && $(MAKE) $@
.PHONY: install
README
+34 -4
View File
@@ -7,6 +7,13 @@ 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
@@ -19,9 +26,20 @@ 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.
Fixing problems building 32 bit binaries
---------
If after building Redis with a 32 bit target you need to rebuild it
with a 64 bit target, or the other way around, you need to perform a
"make distclean" in the root directory of the Redis distribution.
In case of build errors when trying to build a 32 bit binary of Redis, try
the following steps:
* Install the packages libc6-dev-i386 (also try g++-multilib).
* Try using the following command line instead of "make 32bit":
make CFLAGS="-m32 -march=native" LDFLAGS="-m32"
Allocator
---------
@@ -112,7 +130,7 @@ it the proper way for a production system, we have a script doing this
for Ubuntu and Debian systems:
% cd utils
% ./install_server
% ./install_server.sh
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
@@ -121,4 +139,16 @@ 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
-45
View File
@@ -1,45 +0,0 @@
Redis TODO
----------
WARNING: are you a possible Redis contributor?
Before implementing what is listed in this file
please drop a message in the Redis google group or chat with
antirez or pietern on irc.freenode.org #redis to check if the work
is already in progress and if the feature is still interesting for
us, and *how* exactly this can be implemented to have good changes
of a merge. Otherwise it is probably wasted work! Thank you
CLUSTER
=======
* Implement rehashing and cluster check in redis-trib.
* Reimplement MIGRATE / RESTORE to use just in memory buffers (no disk at
all). This will require touching a lot of the RDB stuff around, but we may
hand with faster persistence for RDB.
* Implement the slave nodes semantics and election.
* Allow redis-trib to create a cluster-wide snapshot (using SYNC).
* Allow redis-trib to restore a cluster-wide snapshot (implement UPLOAD?).
SCRIPTING
=========
* SCRIPT FLUSH or alike to start a fresh interpreter?
OPTIMIZATIONS
=============
* 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.
* Read-only mode for slaves.
* Redis big lists as linked lists of small ziplists?
Possibly a simple heuristic that join near nodes when some node gets smaller than the low_level, and split it into two if gets bigger than high_level.
KNOWN BUGS
==========
* #519: Slave may have expired keys that were never read in the master (so a DEL
is not sent in the replication channel) but are already expired since
a lot of time. Maybe after a given delay that is undoubtably greater than
the replication link latency we should expire this key on the slave on
access?
deps/Makefile
Vendored
+52 -33
View File
@@ -1,17 +1,6 @@
# Redis dependency Makefile
UNAME_S:=$(shell sh -c 'uname -s 2> /dev/null || echo not')
LUA_CFLAGS=-O2 -Wall $(ARCH)
ifeq ($(UNAME_S),SunOS)
# Make isinf() available
LUA_CFLAGS+= -D__C99FEATURES__=1
endif
JEMALLOC_CFLAGS=
ifeq ($(ARCH),-m32)
JEMALLOC_CFLAGS+=CFLAGS="-std=gnu99 -Wall -pipe -g3 -fvisibility=hidden -O3 -funroll-loops -m32"
endif
uname_S:= $(shell sh -c 'uname -s 2>/dev/null || echo not')
CCCOLOR="\033[34m"
LINKCOLOR="\033[34;1m"
@@ -23,37 +12,67 @@ ENDCOLOR="\033[0m"
default:
@echo "Explicit target required"
# Clean everything when ARCH is different
ifneq ($(shell sh -c '[ -f .make-arch ] && cat .make-arch'), $(ARCH))
.make-arch: distclean
else
.make-arch:
.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
.make-arch:
-(echo $(ARCH) > .make-arch)
# Clean everything when LDFLAGS is different
ifneq ($(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-arch)
-(rm -f .make-*)
hiredis: .make-arch
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)hiredis$(ENDCOLOR)
cd hiredis && $(MAKE) static ARCH="$(ARCH)"
.PHONY: distclean
linenoise: .make-arch
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)linenoise$(ENDCOLOR)
cd linenoise && $(MAKE) ARCH="$(ARCH)"
hiredis: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd hiredis && $(MAKE) static
lua: .make-arch
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)lua$(ENDCOLOR)
cd lua && $(MAKE) CFLAGS="$(LUA_CFLAGS)" MYLDFLAGS="$(ARCH)" ansi
.PHONY: hiredis
jemalloc: .make-arch
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)jemalloc$(ENDCOLOR)
cd jemalloc && ./configure $(JEMALLOC_CFLAGS) --with-jemalloc-prefix=je_ --enable-cc-silence && $(MAKE) lib/libjemalloc.a
linenoise: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd linenoise && $(MAKE)
.PHONY: default conditional_clean hiredis linenoise lua jemalloc
.PHONY: linenoise
ifeq ($(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
+1 -1
View File
@@ -1,5 +1,5 @@
/hiredis-test
/hiredis-example*
/examples/hiredis-example*
/*.o
/*.so
/*.dylib
deps/hiredis/.travis.yml
Vendored
+6
View File
@@ -0,0 +1,6 @@
language: c
compiler:
- gcc
- clang
script: make && make check
deps/hiredis/CHANGELOG.md
Vendored
+8
View File
@@ -1,3 +1,11 @@
### 0.11.0
* Increase the maximum multi-bulk reply depth to 7.
* Increase the read buffer size from 2k to 16k.
* Use poll(2) instead of select(2) to support large fds (>= 1024).
### 0.10.1
* Makefile overhaul. Important to check out if you override one or more
deps/hiredis/Makefile
Vendored
+42 -24
View File
@@ -4,11 +4,24 @@
# 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
EXAMPLES=hiredis-example hiredis-example-libevent hiredis-example-libev
TESTS=hiredis-test
LIBNAME=libhiredis
HIREDIS_MAJOR=0
HIREDIS_MINOR=10
HIREDIS_MINOR=11
# redis-server configuration used for testing
REDIS_PORT=56379
REDIS_SERVER=redis-server
define REDIS_TEST_CONFIG
daemonize yes
pidfile /tmp/hiredis-test-redis.pid
port $(REDIS_PORT)
bind 127.0.0.1
unixsocket /tmp/hiredis-test-redis.sock
endef
export REDIS_TEST_CONFIG
# Fallback to gcc when $CC is not in $PATH.
CC:=$(shell sh -c 'type $(CC) >/dev/null 2>/dev/null && echo $(CC) || echo gcc')
@@ -41,12 +54,11 @@ ifeq ($(uname_S),Darwin)
DYLIB_MAKE_CMD=$(CC) -shared -Wl,-install_name,$(DYLIB_MINOR_NAME) -o $(DYLIBNAME) $(LDFLAGS)
endif
all: $(DYLIBNAME) $(BINS)
all: $(DYLIBNAME)
# 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
@@ -61,36 +73,44 @@ 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-libevent: examples/example-libevent.c adapters/libevent.h $(STLIBNAME)
$(CC) -o examples/$@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -I. $< -levent $(STLIBNAME)
hiredis-example-libev: example-libev.c adapters/libev.h $(STLIBNAME)
$(CC) -o $@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -lev example-libev.c $(STLIBNAME)
hiredis-example-libev: examples/example-libev.c adapters/libev.h $(STLIBNAME)
$(CC) -o examples/$@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -I. $< -lev $(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)
hiredis-example-ae: examples/example-ae.c adapters/ae.h $(STLIBNAME)
$(CC) -o examples/$@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -I. -I$(AE_DIR) $< $(AE_DIR)/ae.o $(AE_DIR)/zmalloc.o $(AE_DIR)/../deps/jemalloc/lib/libjemalloc.a -pthread $(STLIBNAME)
endif
hiredis-%: %.o $(STLIBNAME)
ifndef LIBUV_DIR
hiredis-example-libuv:
@echo "Please specify LIBUV_DIR (e.g. ../libuv/)"
@false
else
hiredis-example-libuv: examples/example-libuv.c adapters/libuv.h $(STLIBNAME)
$(CC) -o examples/$@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -I. -I$(LIBUV_DIR)/include $< $(LIBUV_DIR)/.libs/libuv.a -lpthread $(STLIBNAME)
endif
hiredis-example: examples/example.c $(STLIBNAME)
$(CC) -o examples/$@ $(REAL_CFLAGS) $(REAL_LDFLAGS) -I. $< $(STLIBNAME)
examples: $(EXAMPLES)
hiredis-test: test.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 || \
@echo "$$REDIS_TEST_CONFIG" | $(REDIS_SERVER) -
./hiredis-test -h 127.0.0.1 -p $(REDIS_PORT) -s /tmp/hiredis-test-redis.sock || \
( kill `cat /tmp/hiredis-test-redis.pid` && false )
kill `cat /tmp/hiredis-test-redis.pid`
@@ -98,17 +118,15 @@ check: hiredis-test
$(CC) -std=c99 -pedantic -c $(REAL_CFLAGS) $<
clean:
rm -rf $(DYLIBNAME) $(STLIBNAME) $(BINS) hiredis-example* *.o *.gcda *.gcno *.gcov
rm -rf $(DYLIBNAME) $(STLIBNAME) $(TESTS) examples/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)
INSTALL_INCLUDE_PATH= $(PREFIX)/include/hiredis
INSTALL_LIBRARY_PATH= $(PREFIX)/lib
ifeq ($(uname_S),SunOS)
INSTALL?= cp -r
deps/hiredis/README.md
Vendored
+35 -3
View File
@@ -1,3 +1,5 @@
[![Build Status](https://travis-ci.org/redis/hiredis.png)](https://travis-ci.org/redis/hiredis)
# HIREDIS
Hiredis is a minimalistic C client library for the [Redis](http://redis.io/) database.
@@ -44,7 +46,7 @@ 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) {
if (c != NULL && c->err) {
printf("Error: %s\n", c->errstr);
// handle error
}
@@ -66,14 +68,14 @@ When you need to pass binary safe strings in a command, the `%b` specifier can b
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);
reply = redisCommand(context, "SET foo %b", value, (size_t) 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("SET key:%s %s", myid, value);
reply = redisCommand(context, "SET key:%s %s", myid, value);
### Using replies
@@ -320,6 +322,10 @@ The reply parsing API consists of the following functions:
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
@@ -333,6 +339,9 @@ 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).
The parser limits the level of nesting for multi bulk payloads to 7. If the
multi bulk nesting level is higher than this, the parser returns an error.
### Customizing replies
The function `redisReaderGetReply` creates `redisReply` and makes the function
@@ -346,6 +355,29 @@ 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
deps/hiredis/adapters/ae.h
Vendored
+30
View File
@@ -1,3 +1,33 @@
/*
* 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_AE_H__
#define __HIREDIS_AE_H__
#include <sys/types.h>
deps/hiredis/adapters/libev.h
Vendored
+30
View File
@@ -1,3 +1,33 @@
/*
* 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_LIBEV_H__
#define __HIREDIS_LIBEV_H__
#include <stdlib.h>
deps/hiredis/adapters/libevent.h
Vendored
+30
View File
@@ -1,3 +1,33 @@
/*
* 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_LIBEVENT_H__
#define __HIREDIS_LIBEVENT_H__
#include <event.h>
deps/hiredis/adapters/libuv.h
Vendored
+121
View File
@@ -0,0 +1,121 @@
#ifndef __HIREDIS_LIBUV_H__
#define __HIREDIS_LIBUV_H__
#include <uv.h>
#include "../hiredis.h"
#include "../async.h"
#include <string.h>
typedef struct redisLibuvEvents {
redisAsyncContext* context;
uv_poll_t handle;
int events;
} redisLibuvEvents;
int redisLibuvAttach(redisAsyncContext*, uv_loop_t*);
static void redisLibuvPoll(uv_poll_t* handle, int status, int events) {
redisLibuvEvents* p = (redisLibuvEvents*)handle->data;
if (status != 0) {
return;
}
if (events & UV_READABLE) {
redisAsyncHandleRead(p->context);
}
if (events & UV_WRITABLE) {
redisAsyncHandleWrite(p->context);
}
}
static void redisLibuvAddRead(void *privdata) {
redisLibuvEvents* p = (redisLibuvEvents*)privdata;
p->events |= UV_READABLE;
uv_poll_start(&p->handle, p->events, redisLibuvPoll);
}
static void redisLibuvDelRead(void *privdata) {
redisLibuvEvents* p = (redisLibuvEvents*)privdata;
p->events &= ~UV_READABLE;
if (p->events) {
uv_poll_start(&p->handle, p->events, redisLibuvPoll);
} else {
uv_poll_stop(&p->handle);
}
}
static void redisLibuvAddWrite(void *privdata) {
redisLibuvEvents* p = (redisLibuvEvents*)privdata;
p->events |= UV_WRITABLE;
uv_poll_start(&p->handle, p->events, redisLibuvPoll);
}
static void redisLibuvDelWrite(void *privdata) {
redisLibuvEvents* p = (redisLibuvEvents*)privdata;
p->events &= ~UV_WRITABLE;
if (p->events) {
uv_poll_start(&p->handle, p->events, redisLibuvPoll);
} else {
uv_poll_stop(&p->handle);
}
}
static void on_close(uv_handle_t* handle) {
redisLibuvEvents* p = (redisLibuvEvents*)handle->data;
free(p);
}
static void redisLibuvCleanup(void *privdata) {
redisLibuvEvents* p = (redisLibuvEvents*)privdata;
uv_close((uv_handle_t*)&p->handle, on_close);
}
static int redisLibuvAttach(redisAsyncContext* ac, uv_loop_t* loop) {
redisContext *c = &(ac->c);
if (ac->ev.data != NULL) {
return REDIS_ERR;
}
ac->ev.addRead = redisLibuvAddRead;
ac->ev.delRead = redisLibuvDelRead;
ac->ev.addWrite = redisLibuvAddWrite;
ac->ev.delWrite = redisLibuvDelWrite;
ac->ev.cleanup = redisLibuvCleanup;
redisLibuvEvents* p = (redisLibuvEvents*)malloc(sizeof(*p));
if (!p) {
return REDIS_ERR;
}
memset(p, 0, sizeof(*p));
if (uv_poll_init(loop, &p->handle, c->fd) != 0) {
return REDIS_ERR;
}
ac->ev.data = p;
p->handle.data = p;
p->context = ac;
return REDIS_OK;
}
#endif
deps/hiredis/async.c
Vendored
+77 -20
View File
@@ -62,7 +62,8 @@ 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));
return dictGenHashFunction((const unsigned char *)key,
sdslen((const sds)key));
}
static void *callbackValDup(void *privdata, const void *src) {
@@ -76,8 +77,8 @@ static int callbackKeyCompare(void *privdata, const void *key1, const void *key2
int l1, l2;
((void) privdata);
l1 = sdslen((sds)key1);
l2 = sdslen((sds)key2);
l1 = sdslen((const sds)key1);
l2 = sdslen((const sds)key2);
if (l1 != l2) return 0;
return memcmp(key1,key2,l1) == 0;
}
@@ -102,7 +103,12 @@ static dictType callbackDict = {
};
static redisAsyncContext *redisAsyncInitialize(redisContext *c) {
redisAsyncContext *ac = realloc(c,sizeof(redisAsyncContext));
redisAsyncContext *ac;
ac = realloc(c,sizeof(redisAsyncContext));
if (ac == NULL)
return NULL;
c = &(ac->c);
/* The regular connect functions will always set the flag REDIS_CONNECTED.
@@ -142,15 +148,45 @@ static void __redisAsyncCopyError(redisAsyncContext *ac) {
}
redisAsyncContext *redisAsyncConnect(const char *ip, int port) {
redisContext *c = redisConnectNonBlock(ip,port);
redisContext *c;
redisAsyncContext *ac;
c = redisConnectNonBlock(ip,port);
if (c == NULL)
return NULL;
ac = redisAsyncInitialize(c);
if (ac == NULL) {
redisFree(c);
return NULL;
}
__redisAsyncCopyError(ac);
return ac;
}
redisAsyncContext *redisAsyncConnectBind(const char *ip, int port,
const char *source_addr) {
redisContext *c = redisConnectBindNonBlock(ip,port,source_addr);
redisAsyncContext *ac = redisAsyncInitialize(c);
__redisAsyncCopyError(ac);
return ac;
}
redisAsyncContext *redisAsyncConnectUnix(const char *path) {
redisContext *c = redisConnectUnixNonBlock(path);
redisAsyncContext *ac = redisAsyncInitialize(c);
redisContext *c;
redisAsyncContext *ac;
c = redisConnectUnixNonBlock(path);
if (c == NULL)
return NULL;
ac = redisAsyncInitialize(c);
if (ac == NULL) {
redisFree(c);
return NULL;
}
__redisAsyncCopyError(ac);
return ac;
}
@@ -182,6 +218,9 @@ static int __redisPushCallback(redisCallbackList *list, redisCallback *source) {
/* Copy callback from stack to heap */
cb = malloc(sizeof(*cb));
if (cb == NULL)
return REDIS_ERR_OOM;
if (source != NULL) {
memcpy(cb,source,sizeof(*cb));
cb->next = NULL;
@@ -360,7 +399,7 @@ static int __redisGetSubscribeCallback(redisAsyncContext *ac, redisReply *reply,
void redisProcessCallbacks(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
redisCallback cb;
redisCallback cb = {NULL, NULL, NULL};
void *reply = NULL;
int status;
@@ -373,6 +412,11 @@ void redisProcessCallbacks(redisAsyncContext *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;
@@ -381,22 +425,31 @@ void redisProcessCallbacks(redisAsyncContext *ac) {
/* 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 only 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. */
if ( !(c->flags & REDIS_SUBSCRIBED) && ((redisReply*)reply)->type == REDIS_REPLY_ERROR ) {
/*
* 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. */
assert(c->flags & REDIS_SUBSCRIBED);
__redisGetSubscribeCallback(ac,reply,&cb);
/* 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) {
@@ -428,7 +481,7 @@ void redisProcessCallbacks(redisAsyncContext *ac) {
static int __redisAsyncHandleConnect(redisAsyncContext *ac) {
redisContext *c = &(ac->c);
if (redisCheckSocketError(c,c->fd) == REDIS_ERR) {
if (redisCheckSocketError(c) == REDIS_ERR) {
/* Try again later when connect(2) is still in progress. */
if (errno == EINPROGRESS)
return REDIS_OK;
@@ -557,6 +610,10 @@ static int __redisAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void
/* (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
deps/hiredis/async.h
Vendored
+1
View File
@@ -102,6 +102,7 @@ typedef struct redisAsyncContext {
/* Functions that proxy to hiredis */
redisAsyncContext *redisAsyncConnect(const char *ip, int port);
redisAsyncContext *redisAsyncConnectBind(const char *ip, int port, const char *source_addr);
redisAsyncContext *redisAsyncConnectUnix(const char *path);
int redisAsyncSetConnectCallback(redisAsyncContext *ac, redisConnectCallback *fn);
int redisAsyncSetDisconnectCallback(redisAsyncContext *ac, redisDisconnectCallback *fn);
deps/hiredis/example-ae.c → deps/hiredis/examples/example-ae.c
Vendored
+10 -4
View File
@@ -2,9 +2,10 @@
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include "hiredis.h"
#include "async.h"
#include "adapters/ae.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;
@@ -21,17 +22,22 @@ void getCallback(redisAsyncContext *c, void *r, void *privdata) {
void connectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
aeStop(loop);
return;
}
printf("Connected...\n");
}
void disconnectCallback(const redisAsyncContext *c, int status) {
if (status != REDIS_OK) {
printf("Error: %s\n", c->errstr);
aeStop(loop);
return;
}
printf("Disconnected...\n");
aeStop(loop);
}
int main (int argc, char **argv) {
@@ -44,7 +50,7 @@ int main (int argc, char **argv) {
return 1;
}
loop = aeCreateEventLoop();
loop = aeCreateEventLoop(64);
redisAeAttach(loop, c);
redisAsyncSetConnectCallback(c,connectCallback);
redisAsyncSetDisconnectCallback(c,disconnectCallback);
deps/hiredis/example-libev.c → deps/hiredis/examples/example-libev.c
Vendored
+4 -3
View File
@@ -2,9 +2,10 @@
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include "hiredis.h"
#include "async.h"
#include "adapters/libev.h"
#include <hiredis.h>
#include <async.h>
#include <adapters/libev.h>
void getCallback(redisAsyncContext *c, void *r, void *privdata) {
redisReply *reply = r;
deps/hiredis/example-libevent.c → deps/hiredis/examples/example-libevent.c
Vendored
+4 -3
View File
@@ -2,9 +2,10 @@
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include "hiredis.h"
#include "async.h"
#include "adapters/libevent.h"
#include <hiredis.h>
#include <async.h>
#include <adapters/libevent.h>
void getCallback(redisAsyncContext *c, void *r, void *privdata) {
redisReply *reply = r;
deps/hiredis/examples/example-libuv.c
Vendored
+53
View File
@@ -0,0 +1,53 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <signal.h>
#include <hiredis.h>
#include <async.h>
#include <adapters/libuv.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);
uv_loop_t* loop = uv_default_loop();
redisAsyncContext *c = redisAsyncConnect("127.0.0.1", 6379);
if (c->err) {
/* Let *c leak for now... */
printf("Error: %s\n", c->errstr);
return 1;
}
redisLibuvAttach(c,loop);
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");
uv_run(loop, UV_RUN_DEFAULT);
return 0;
}
deps/hiredis/example.c → deps/hiredis/examples/example.c
Vendored
+16 -6
View File
@@ -2,17 +2,24 @@
#include <stdlib.h>
#include <string.h>
#include "hiredis.h"
#include <hiredis.h>
int main(void) {
int main(int argc, char **argv) {
unsigned int j;
redisContext *c;
redisReply *reply;
const char *hostname = (argc > 1) ? argv[1] : "127.0.0.1";
int port = (argc > 2) ? atoi(argv[2]) : 6379;
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);
c = redisConnectWithTimeout(hostname, port, timeout);
if (c == NULL || c->err) {
if (c) {
printf("Connection error: %s\n", c->errstr);
redisFree(c);
} else {
printf("Connection error: can't allocate redis context\n");
}
exit(1);
}
@@ -27,7 +34,7 @@ int main(void) {
freeReplyObject(reply);
/* Set a key using binary safe API */
reply = redisCommand(c,"SET %b %b", "bar", 3, "hello", 5);
reply = redisCommand(c,"SET %b %b", "bar", (size_t) 3, "hello", (size_t) 5);
printf("SET (binary API): %s\n", reply->str);
freeReplyObject(reply);
@@ -64,5 +71,8 @@ int main(void) {
}
freeReplyObject(reply);
/* Disconnects and frees the context */
redisFree(c);
return 0;
}
deps/hiredis/fmacros.h
Vendored
+5 -1
View File
@@ -7,10 +7,14 @@
#if defined(__sun__)
#define _POSIX_C_SOURCE 200112L
#elif defined(__linux__)
#elif defined(__linux__) || defined(__OpenBSD__) || defined(__NetBSD__)
#define _XOPEN_SOURCE 600
#else
#define _XOPEN_SOURCE
#endif
#if __APPLE__ && __MACH__
#define _OSX
#endif
#endif
deps/hiredis/hiredis.c
Vendored
+91 -19
View File
@@ -446,7 +446,7 @@ static int processMultiBulkItem(redisReader *r) {
long elements;
int root = 0;
/* Set error for nested multi bulks with depth > 2 */
/* Set error for nested multi bulks with depth > 7 */
if (r->ridx == 8) {
__redisReaderSetError(r,REDIS_ERR_PROTOCOL,
"No support for nested multi bulk replies with depth > 7");
@@ -564,6 +564,7 @@ redisReader *redisReaderCreate(void) {
r->errstr[0] = '\0';
r->fn = &defaultFunctions;
r->buf = sdsempty();
r->maxbuf = REDIS_READER_MAX_BUF;
if (r->buf == NULL) {
free(r);
return NULL;
@@ -590,9 +591,8 @@ int redisReaderFeed(redisReader *r, const char *buf, size_t len) {
/* Copy the provided buffer. */
if (buf != NULL && len >= 1) {
#if 0
/* Destroy internal buffer when it is empty and is quite large. */
if (r->len == 0 && sdsavail(r->buf) > 16*1024) {
if (r->len == 0 && r->maxbuf != 0 && sdsavail(r->buf) > r->maxbuf) {
sdsfree(r->buf);
r->buf = sdsempty();
r->pos = 0;
@@ -600,7 +600,6 @@ int redisReaderFeed(redisReader *r, const char *buf, size_t len) {
/* r->buf should not be NULL since we just free'd a larger one. */
assert(r->buf != NULL);
}
#endif
newbuf = sdscatlen(r->buf,buf,len);
if (newbuf == NULL) {
@@ -651,7 +650,7 @@ int redisReaderGetReply(redisReader *r, void **reply) {
/* Discard part of the buffer when we've consumed at least 1k, to avoid
* doing unnecessary calls to memmove() in sds.c. */
if (r->pos >= 1024) {
r->buf = sdsrange(r->buf,r->pos,-1);
sdsrange(r->buf,r->pos,-1);
r->pos = 0;
r->len = sdslen(r->buf);
}
@@ -918,7 +917,7 @@ err:
* %b represents a binary safe string
*
* When using %b you need to provide both the pointer to the string
* and the length in bytes. Examples:
* and the length in bytes as a size_t. Examples:
*
* len = redisFormatCommand(target, "GET %s", mykey);
* len = redisFormatCommand(target, "SET %s %b", mykey, myval, myvallen);
@@ -1011,58 +1010,122 @@ void redisFree(redisContext *c) {
free(c);
}
int redisFreeKeepFd(redisContext *c) {
int fd = c->fd;
c->fd = -1;
redisFree(c);
return fd;
}
/* Connect to a Redis instance. On error the field error in the returned
* context will be set to the return value of the error function.
* When no set of reply functions is given, the default set will be used. */
redisContext *redisConnect(const char *ip, int port) {
redisContext *c = redisContextInit();
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->flags |= REDIS_BLOCK;
redisContextConnectTcp(c,ip,port,NULL);
return c;
}
redisContext *redisConnectWithTimeout(const char *ip, int port, struct timeval tv) {
redisContext *c = redisContextInit();
redisContext *redisConnectWithTimeout(const char *ip, int port, const struct timeval tv) {
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->flags |= REDIS_BLOCK;
redisContextConnectTcp(c,ip,port,&tv);
return c;
}
redisContext *redisConnectNonBlock(const char *ip, int port) {
redisContext *c = redisContextInit();
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->flags &= ~REDIS_BLOCK;
redisContextConnectTcp(c,ip,port,NULL);
return c;
}
redisContext *redisConnectUnix(const char *path) {
redisContext *redisConnectBindNonBlock(const char *ip, int port,
const char *source_addr) {
redisContext *c = redisContextInit();
c->flags &= ~REDIS_BLOCK;
redisContextConnectBindTcp(c,ip,port,NULL,source_addr);
return c;
}
redisContext *redisConnectUnix(const char *path) {
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->flags |= REDIS_BLOCK;
redisContextConnectUnix(c,path,NULL);
return c;
}
redisContext *redisConnectUnixWithTimeout(const char *path, struct timeval tv) {
redisContext *c = redisContextInit();
redisContext *redisConnectUnixWithTimeout(const char *path, const struct timeval tv) {
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->flags |= REDIS_BLOCK;
redisContextConnectUnix(c,path,&tv);
return c;
}
redisContext *redisConnectUnixNonBlock(const char *path) {
redisContext *c = redisContextInit();
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->flags &= ~REDIS_BLOCK;
redisContextConnectUnix(c,path,NULL);
return c;
}
redisContext *redisConnectFd(int fd) {
redisContext *c;
c = redisContextInit();
if (c == NULL)
return NULL;
c->fd = fd;
c->flags |= REDIS_BLOCK | REDIS_CONNECTED;
return c;
}
/* Set read/write timeout on a blocking socket. */
int redisSetTimeout(redisContext *c, struct timeval tv) {
int redisSetTimeout(redisContext *c, const struct timeval tv) {
if (c->flags & REDIS_BLOCK)
return redisContextSetTimeout(c,tv);
return REDIS_ERR;
}
/* Enable connection KeepAlive. */
int redisEnableKeepAlive(redisContext *c) {
if (redisKeepAlive(c, REDIS_KEEPALIVE_INTERVAL) != REDIS_OK)
return REDIS_ERR;
return REDIS_OK;
}
/* Use this function to handle a read event on the descriptor. It will try
* and read some bytes from the socket and feed them to the reply parser.
*
@@ -1078,7 +1141,7 @@ int redisBufferRead(redisContext *c) {
nread = read(c->fd,buf,sizeof(buf));
if (nread == -1) {
if (errno == EAGAIN && !(c->flags & REDIS_BLOCK)) {
if ((errno == EAGAIN && !(c->flags & REDIS_BLOCK)) || (errno == EINTR)) {
/* Try again later */
} else {
__redisSetError(c,REDIS_ERR_IO,NULL);
@@ -1115,7 +1178,7 @@ int redisBufferWrite(redisContext *c, int *done) {
if (sdslen(c->obuf) > 0) {
nwritten = write(c->fd,c->obuf,sdslen(c->obuf));
if (nwritten == -1) {
if (errno == EAGAIN && !(c->flags & REDIS_BLOCK)) {
if ((errno == EAGAIN && !(c->flags & REDIS_BLOCK)) || (errno == EINTR)) {
/* Try again later */
} else {
__redisSetError(c,REDIS_ERR_IO,NULL);
@@ -1126,7 +1189,7 @@ int redisBufferWrite(redisContext *c, int *done) {
sdsfree(c->obuf);
c->obuf = sdsempty();
} else {
c->obuf = sdsrange(c->obuf,nwritten,-1);
sdsrange(c->obuf,nwritten,-1);
}
}
}
@@ -1181,7 +1244,7 @@ int redisGetReply(redisContext *c, void **reply) {
* is used, you need to call redisGetReply yourself to retrieve
* the reply (or replies in pub/sub).
*/
int __redisAppendCommand(redisContext *c, char *cmd, size_t len) {
int __redisAppendCommand(redisContext *c, const char *cmd, size_t len) {
sds newbuf;
newbuf = sdscatlen(c->obuf,cmd,len);
@@ -1194,6 +1257,15 @@ int __redisAppendCommand(redisContext *c, char *cmd, size_t len) {
return REDIS_OK;
}
int redisAppendFormattedCommand(redisContext *c, const char *cmd, size_t len) {
if (__redisAppendCommand(c, cmd, len) != REDIS_OK) {
return REDIS_ERR;
}
return REDIS_OK;
}
int redisvAppendCommand(redisContext *c, const char *format, va_list ap) {
char *cmd;
int len;
deps/hiredis/hiredis.h
Vendored
+21 -5
View File
@@ -36,8 +36,8 @@
#include <sys/time.h> /* for struct timeval */
#define HIREDIS_MAJOR 0
#define HIREDIS_MINOR 10
#define HIREDIS_PATCH 1
#define HIREDIS_MINOR 11
#define HIREDIS_PATCH 0
#define REDIS_ERR -1
#define REDIS_OK 0
@@ -76,6 +76,9 @@
/* 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
@@ -83,6 +86,10 @@
#define REDIS_REPLY_STATUS 5
#define REDIS_REPLY_ERROR 6
#define REDIS_READER_MAX_BUF (1024*16) /* Default max unused reader buffer. */
#define REDIS_KEEPALIVE_INTERVAL 15 /* seconds */
#ifdef __cplusplus
extern "C" {
#endif
@@ -122,6 +129,7 @@ typedef struct redisReader {
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 */
@@ -165,13 +173,17 @@ typedef struct redisContext {
} redisContext;
redisContext *redisConnect(const char *ip, int port);
redisContext *redisConnectWithTimeout(const char *ip, int port, struct timeval tv);
redisContext *redisConnectWithTimeout(const char *ip, int port, const struct timeval tv);
redisContext *redisConnectNonBlock(const char *ip, int port);
redisContext *redisConnectBindNonBlock(const char *ip, int port, const char *source_addr);
redisContext *redisConnectUnix(const char *path);
redisContext *redisConnectUnixWithTimeout(const char *path, struct timeval tv);
redisContext *redisConnectUnixWithTimeout(const char *path, const struct timeval tv);
redisContext *redisConnectUnixNonBlock(const char *path);
int redisSetTimeout(redisContext *c, struct timeval tv);
redisContext *redisConnectFd(int fd);
int redisSetTimeout(redisContext *c, const struct timeval tv);
int redisEnableKeepAlive(redisContext *c);
void redisFree(redisContext *c);
int redisFreeKeepFd(redisContext *c);
int redisBufferRead(redisContext *c);
int redisBufferWrite(redisContext *c, int *done);
@@ -182,6 +194,10 @@ int redisBufferWrite(redisContext *c, int *done);
int redisGetReply(redisContext *c, void **reply);
int redisGetReplyFromReader(redisContext *c, void **reply);
/* Write a formatted command to the output buffer. Use these functions in blocking mode
* to get a pipeline of commands. */
int redisAppendFormattedCommand(redisContext *c, const char *cmd, size_t len);
/* 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);
deps/hiredis/net.c
Vendored
+156 -54
View File
@@ -45,6 +45,8 @@
#include <errno.h>
#include <stdarg.h>
#include <stdio.h>
#include <poll.h>
#include <limits.h>
#include "net.h"
#include "sds.h"
@@ -52,8 +54,15 @@
/* Defined in hiredis.c */
void __redisSetError(redisContext *c, int type, const char *str);
static void redisContextCloseFd(redisContext *c) {
if (c && c->fd >= 0) {
close(c->fd);
c->fd = -1;
}
}
static void __redisSetErrorFromErrno(redisContext *c, int type, const char *prefix) {
char buf[128];
char buf[128] = { 0 };
size_t len = 0;
if (prefix != NULL)
@@ -62,11 +71,11 @@ static void __redisSetErrorFromErrno(redisContext *c, int type, const char *pref
__redisSetError(c,type,buf);
}
static int redisSetReuseAddr(redisContext *c, int fd) {
static int redisSetReuseAddr(redisContext *c) {
int on = 1;
if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(on)) == -1) {
if (setsockopt(c->fd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(on)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
@@ -78,23 +87,24 @@ static int redisCreateSocket(redisContext *c, int type) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
return REDIS_ERR;
}
c->fd = s;
if (type == AF_INET) {
if (redisSetReuseAddr(c,s) == REDIS_ERR) {
if (redisSetReuseAddr(c) == REDIS_ERR) {
return REDIS_ERR;
}
}
return s;
return REDIS_OK;
}
static int redisSetBlocking(redisContext *c, int fd, int blocking) {
static int redisSetBlocking(redisContext *c, 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) {
if ((flags = fcntl(c->fd, F_GETFL)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"fcntl(F_GETFL)");
close(fd);
redisContextCloseFd(c);
return REDIS_ERR;
}
@@ -103,84 +113,135 @@ static int redisSetBlocking(redisContext *c, int fd, int blocking) {
else
flags |= O_NONBLOCK;
if (fcntl(fd, F_SETFL, flags) == -1) {
if (fcntl(c->fd, F_SETFL, flags) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"fcntl(F_SETFL)");
close(fd);
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
}
static int redisSetTcpNoDelay(redisContext *c, int fd) {
int redisKeepAlive(redisContext *c, int interval) {
int val = 1;
int fd = c->fd;
if (setsockopt(fd, SOL_SOCKET, SO_KEEPALIVE, &val, sizeof(val)) == -1){
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
val = interval;
#ifdef _OSX
if (setsockopt(fd, IPPROTO_TCP, TCP_KEEPALIVE, &val, sizeof(val)) < 0) {
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
#else
#ifndef __sun
val = interval;
if (setsockopt(fd, IPPROTO_TCP, TCP_KEEPIDLE, &val, sizeof(val)) < 0) {
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
val = interval/3;
if (val == 0) val = 1;
if (setsockopt(fd, IPPROTO_TCP, TCP_KEEPINTVL, &val, sizeof(val)) < 0) {
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
val = 3;
if (setsockopt(fd, IPPROTO_TCP, TCP_KEEPCNT, &val, sizeof(val)) < 0) {
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
#endif
#endif
return REDIS_OK;
}
static int redisSetTcpNoDelay(redisContext *c) {
int yes = 1;
if (setsockopt(fd, IPPROTO_TCP, TCP_NODELAY, &yes, sizeof(yes)) == -1) {
if (setsockopt(c->fd, IPPROTO_TCP, TCP_NODELAY, &yes, sizeof(yes)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(TCP_NODELAY)");
close(fd);
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
}
static int redisContextWaitReady(redisContext *c, int fd, const struct timeval *timeout) {
struct timeval to;
struct timeval *toptr = NULL;
fd_set wfd;
#define __MAX_MSEC (((LONG_MAX) - 999) / 1000)
static int redisContextWaitReady(redisContext *c, const struct timeval *timeout) {
struct pollfd wfd[1];
long msec;
msec = -1;
wfd[0].fd = c->fd;
wfd[0].events = POLLOUT;
/* Only use timeout when not NULL. */
if (timeout != NULL) {
to = *timeout;
toptr = &to;
if (timeout->tv_usec > 1000000 || timeout->tv_sec > __MAX_MSEC) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
redisContextCloseFd(c);
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) {
FD_ZERO(&wfd);
FD_SET(fd, &wfd);
int res;
if (select(FD_SETSIZE, NULL, &wfd, NULL, toptr) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"select(2)");
close(fd);
if ((res = poll(wfd, 1, msec)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "poll(2)");
redisContextCloseFd(c);
return REDIS_ERR;
}
if (!FD_ISSET(fd, &wfd)) {
} else if (res == 0) {
errno = ETIMEDOUT;
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
redisContextCloseFd(c);
return REDIS_ERR;
}
if (redisCheckSocketError(c, fd) != REDIS_OK)
if (redisCheckSocketError(c) != REDIS_OK)
return REDIS_ERR;
return REDIS_OK;
}
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
close(fd);
redisContextCloseFd(c);
return REDIS_ERR;
}
int redisCheckSocketError(redisContext *c, int fd) {
int redisCheckSocketError(redisContext *c) {
int err = 0;
socklen_t errlen = sizeof(err);
if (getsockopt(fd, SOL_SOCKET, SO_ERROR, &err, &errlen) == -1) {
if (getsockopt(c->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) {
int redisContextSetTimeout(redisContext *c, const 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;
@@ -192,10 +253,12 @@ int redisContextSetTimeout(redisContext *c, struct timeval tv) {
return REDIS_OK;
}
int redisContextConnectTcp(redisContext *c, const char *addr, int port, struct timeval *timeout) {
static int _redisContextConnectTcp(redisContext *c, const char *addr, int port,
const struct timeval *timeout,
const char *source_addr) {
int s, rv;
char _port[6]; /* strlen("65535"); */
struct addrinfo hints, *servinfo, *p;
struct addrinfo hints, *servinfo, *bservinfo, *p, *b;
int blocking = (c->flags & REDIS_BLOCK);
snprintf(_port, 6, "%d", port);
@@ -203,33 +266,63 @@ int redisContextConnectTcp(redisContext *c, const char *addr, int port, struct t
hints.ai_family = AF_INET;
hints.ai_socktype = SOCK_STREAM;
/* Try with IPv6 if no IPv4 address was found. We do it in this order since
* in a Redis client you can't afford to test if you have IPv6 connectivity
* as this would add latency to every connect. Otherwise a more sensible
* route could be: Use IPv6 if both addresses are available and there is IPv6
* connectivity. */
if ((rv = getaddrinfo(addr,_port,&hints,&servinfo)) != 0) {
__redisSetError(c,REDIS_ERR_OTHER,gai_strerror(rv));
return REDIS_ERR;
hints.ai_family = AF_INET6;
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)
c->fd = s;
if (redisSetBlocking(c,0) != REDIS_OK)
goto error;
if (source_addr) {
int bound = 0;
/* Using getaddrinfo saves us from self-determining IPv4 vs IPv6 */
if ((rv = getaddrinfo(source_addr, NULL, &hints, &bservinfo)) != 0) {
char buf[128];
snprintf(buf,sizeof(buf),"Can't get addr: %s",gai_strerror(rv));
__redisSetError(c,REDIS_ERR_OTHER,buf);
goto error;
}
for (b = bservinfo; b != NULL; b = b->ai_next) {
if (bind(s,b->ai_addr,b->ai_addrlen) != -1) {
bound = 1;
break;
}
}
if (!bound) {
char buf[128];
snprintf(buf,sizeof(buf),"Can't bind socket: %s",strerror(errno));
__redisSetError(c,REDIS_ERR_OTHER,buf);
goto error;
}
}
if (connect(s,p->ai_addr,p->ai_addrlen) == -1) {
if (errno == EHOSTUNREACH) {
close(s);
redisContextCloseFd(c);
continue;
} else if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
} else {
if (redisContextWaitReady(c,s,timeout) != REDIS_OK)
if (redisContextWaitReady(c,timeout) != REDIS_OK)
goto error;
}
}
if (blocking && redisSetBlocking(c,s,1) != REDIS_OK)
if (blocking && redisSetBlocking(c,1) != REDIS_OK)
goto error;
if (redisSetTcpNoDelay(c,s) != REDIS_OK)
if (redisSetTcpNoDelay(c) != REDIS_OK)
goto error;
c->fd = s;
c->flags |= REDIS_CONNECTED;
rv = REDIS_OK;
goto end;
@@ -248,32 +341,41 @@ end:
return rv; // Need to return REDIS_OK if alright
}
int redisContextConnectUnix(redisContext *c, const char *path, struct timeval *timeout) {
int s;
int redisContextConnectTcp(redisContext *c, const char *addr, int port,
const struct timeval *timeout) {
return _redisContextConnectTcp(c, addr, port, timeout, NULL);
}
int redisContextConnectBindTcp(redisContext *c, const char *addr, int port,
const struct timeval *timeout,
const char *source_addr) {
return _redisContextConnectTcp(c, addr, port, timeout, source_addr);
}
int redisContextConnectUnix(redisContext *c, const char *path, const struct timeval *timeout) {
int blocking = (c->flags & REDIS_BLOCK);
struct sockaddr_un sa;
if ((s = redisCreateSocket(c,AF_LOCAL)) < 0)
if (redisCreateSocket(c,AF_LOCAL) < 0)
return REDIS_ERR;
if (redisSetBlocking(c,s,0) != REDIS_OK)
if (redisSetBlocking(c,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 (connect(c->fd, (struct sockaddr*)&sa, sizeof(sa)) == -1) {
if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
} else {
if (redisContextWaitReady(c,s,timeout) != REDIS_OK)
if (redisContextWaitReady(c,timeout) != REDIS_OK)
return REDIS_ERR;
}
}
/* Reset socket to be blocking after connect(2). */
if (blocking && redisSetBlocking(c,s,1) != REDIS_OK)
if (blocking && redisSetBlocking(c,1) != REDIS_OK)
return REDIS_ERR;
c->fd = s;
c->flags |= REDIS_CONNECTED;
return REDIS_OK;
}
deps/hiredis/net.h
Vendored
+8 -4
View File
@@ -39,9 +39,13 @@
#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);
int redisCheckSocketError(redisContext *c);
int redisContextSetTimeout(redisContext *c, const struct timeval tv);
int redisContextConnectTcp(redisContext *c, const char *addr, int port, const struct timeval *timeout);
int redisContextConnectBindTcp(redisContext *c, const char *addr, int port,
const struct timeval *timeout,
const char *source_addr);
int redisContextConnectUnix(redisContext *c, const char *path, const struct timeval *timeout);
int redisKeepAlive(redisContext *c, int interval);
#endif
deps/hiredis/sds.c
Vendored
+630 -135
View File
File diff suppressed because it is too large Load Diff
deps/hiredis/sds.h
Vendored
+21 -8
View File
@@ -31,6 +31,8 @@
#ifndef __SDS_H
#define __SDS_H
#define SDS_MAX_PREALLOC (1024*1024)
#include <sys/types.h>
#include <stdarg.h>
@@ -58,12 +60,13 @@ sds sdsempty(void);
size_t sdslen(const sds s);
sds sdsdup(const sds s);
void sdsfree(sds s);
size_t sdsavail(sds s);
size_t sdsavail(const sds s);
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 sdscatsds(sds s, const sds t);
sds sdscpylen(sds s, const char *t, size_t len);
sds sdscpy(sds s, const char *t);
sds sdscatvprintf(sds s, const char *fmt, va_list ap);
#ifdef __GNUC__
@@ -73,16 +76,26 @@ sds sdscatprintf(sds s, const char *fmt, ...)
sds sdscatprintf(sds s, const char *fmt, ...);
#endif
sds sdscatfmt(sds s, char const *fmt, ...);
sds sdstrim(sds s, const char *cset);
sds sdsrange(sds s, int start, int end);
void 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);
void sdsclear(sds s);
int sdscmp(const sds s1, const sds s2);
sds *sdssplitlen(const char *s, int len, const char *sep, int seplen, int *count);
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);
sds sdscatrepr(sds s, const char *p, size_t len);
sds *sdssplitargs(const char *line, int *argc);
sds sdsmapchars(sds s, const char *from, const char *to, size_t setlen);
sds sdsjoin(char **argv, int argc, char *sep);
/* Low level functions exposed to the user API */
sds sdsMakeRoomFor(sds s, size_t addlen);
void sdsIncrLen(sds s, int incr);
sds sdsRemoveFreeSpace(sds s);
size_t sdsAllocSize(sds s);
#endif
deps/hiredis/test.c
Vendored
+95 -16
View File
@@ -8,12 +8,14 @@
#include <unistd.h>
#include <signal.h>
#include <errno.h>
#include <limits.h>
#include "hiredis.h"
enum connection_type {
CONN_TCP,
CONN_UNIX
CONN_UNIX,
CONN_FD
};
struct config {
@@ -22,6 +24,7 @@ struct config {
struct {
const char *host;
int port;
struct timeval timeout;
} tcp;
struct {
@@ -62,7 +65,7 @@ static redisContext *select_database(redisContext *c) {
return c;
}
static void disconnect(redisContext *c) {
static int disconnect(redisContext *c, int keep_fd) {
redisReply *reply;
/* Make sure we're on DB 9. */
@@ -73,8 +76,11 @@ static void disconnect(redisContext *c) {
assert(reply != NULL);
freeReplyObject(reply);
/* Free the context as well. */
/* Free the context as well, but keep the fd if requested. */
if (keep_fd)
return redisFreeKeepFd(c);
redisFree(c);
return -1;
}
static redisContext *connect(struct config config) {
@@ -84,11 +90,22 @@ static redisContext *connect(struct config config) {
c = redisConnect(config.tcp.host, config.tcp.port);
} else if (config.type == CONN_UNIX) {
c = redisConnectUnix(config.unix.path);
} else if (config.type == CONN_FD) {
/* Create a dummy connection just to get an fd to inherit */
redisContext *dummy_ctx = redisConnectUnix(config.unix.path);
if (dummy_ctx) {
int fd = disconnect(dummy_ctx, 1);
printf("Connecting to inherited fd %d\n", fd);
c = redisConnectFd(fd);
}
} else {
assert(NULL);
}
if (c->err) {
if (c == NULL) {
printf("Connection error: can't allocate redis context\n");
exit(1);
} else if (c->err) {
printf("Connection error: %s\n", c->errstr);
exit(1);
}
@@ -125,13 +142,13 @@ static void test_format_commands(void) {
free(cmd);
test("Format command with %%b string interpolation: ");
len = redisFormatCommand(&cmd,"SET %b %b","foo",3,"b\0r",3);
len = redisFormatCommand(&cmd,"SET %b %b","foo",(size_t)3,"b\0r",(size_t)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);
len = redisFormatCommand(&cmd,"SET %b %b","foo",(size_t)3,"",(size_t)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);
@@ -177,7 +194,7 @@ static void test_format_commands(void) {
FLOAT_WIDTH_TEST(double);
test("Format command with invalid printf format: ");
len = redisFormatCommand(&cmd,"key:%08p %b",(void*)1234,"foo",3);
len = redisFormatCommand(&cmd,"key:%08p %b",(void*)1234,"foo",(size_t)3);
test_cond(len == -1);
const char *argv[3];
@@ -200,10 +217,33 @@ static void test_format_commands(void) {
free(cmd);
}
static void test_append_formatted_commands(struct config config) {
redisContext *c;
redisReply *reply;
char *cmd;
int len;
c = connect(config);
test("Append format command: ");
len = redisFormatCommand(&cmd, "SET foo bar");
test_cond(redisAppendFormattedCommand(c, cmd, len) == REDIS_OK);
assert(redisGetReply(c, (void*)&reply) == REDIS_OK);
free(cmd);
freeReplyObject(reply);
disconnect(c, 0);
}
static void test_reply_reader(void) {
redisReader *reader;
void *reply;
int ret;
int i;
test("Error handling in reply parser: ");
reader = redisReaderCreate();
@@ -225,12 +265,13 @@ static void test_reply_reader(void) {
strcasecmp(reader->errstr,"Protocol error, got \"@\" as reply type byte") == 0);
redisReaderFree(reader);
test("Set error on nested multi bulks with depth > 2: ");
test("Set error on nested multi bulks with depth > 7: ");
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);
for (i = 0; i < 9; i++) {
redisReaderFeed(reader,(char*)"*1\r\n",4);
}
ret = redisReaderGetReply(reader,NULL);
test_cond(ret == REDIS_ERR &&
strncasecmp(reader->errstr,"No support for",14) == 0);
@@ -284,7 +325,10 @@ static void test_blocking_connection_errors(void) {
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));
strcmp(c->errstr,"Can't resolve: idontexist.local") == 0 ||
strcmp(c->errstr,"nodename nor servname provided, or not known") == 0 ||
strcmp(c->errstr,"No address associated with hostname") == 0 ||
strcmp(c->errstr,"no address associated with name") == 0));
redisFree(c);
test("Returns error when the port is not open: ");
@@ -326,7 +370,7 @@ static void test_blocking_connection(struct config config) {
freeReplyObject(reply);
test("%%b String interpolation works: ");
reply = redisCommand(c,"SET %b %b","foo",3,"hello\x00world",11);
reply = redisCommand(c,"SET %b %b","foo",(size_t)3,"hello\x00world",(size_t)11);
freeReplyObject(reply);
reply = redisCommand(c,"GET foo");
test_cond(reply->type == REDIS_REPLY_STRING &&
@@ -374,7 +418,7 @@ static void test_blocking_connection(struct config config) {
strcasecmp(reply->element[1]->str,"pong") == 0);
freeReplyObject(reply);
disconnect(c);
disconnect(c, 0);
}
static void test_blocking_io_errors(struct config config) {
@@ -428,6 +472,30 @@ static void test_blocking_io_errors(struct config config) {
redisFree(c);
}
static void test_invalid_timeout_errors(struct config config) {
redisContext *c;
test("Set error when an invalid timeout usec value is given to redisConnectWithTimeout: ");
config.tcp.timeout.tv_sec = 0;
config.tcp.timeout.tv_usec = 10000001;
c = redisConnectWithTimeout(config.tcp.host, config.tcp.port, config.tcp.timeout);
test_cond(c->err == REDIS_ERR_IO);
test("Set error when an invalid timeout sec value is given to redisConnectWithTimeout: ");
config.tcp.timeout.tv_sec = (((LONG_MAX) - 999) / 1000) + 1;
config.tcp.timeout.tv_usec = 0;
c = redisConnectWithTimeout(config.tcp.host, config.tcp.port, config.tcp.timeout);
test_cond(c->err == REDIS_ERR_IO);
redisFree(c);
}
static void test_throughput(struct config config) {
redisContext *c = connect(config);
redisReply **replies;
@@ -490,7 +558,7 @@ static void test_throughput(struct config config) {
free(replies);
printf("\t(%dx LRANGE with 500 elements (pipelined): %.3fs)\n", num, (t2-t1)/1000000.0);
disconnect(c);
disconnect(c, 0);
}
// static long __test_callback_flags = 0;
@@ -603,6 +671,7 @@ int main(int argc, char **argv) {
}
};
int throughput = 1;
int test_inherit_fd = 1;
/* Ignore broken pipe signal (for I/O error tests). */
signal(SIGPIPE, SIG_IGN);
@@ -621,6 +690,8 @@ int main(int argc, char **argv) {
cfg.unix.path = argv[0];
} else if (argc >= 1 && !strcmp(argv[0],"--skip-throughput")) {
throughput = 0;
} else if (argc >= 1 && !strcmp(argv[0],"--skip-inherit-fd")) {
test_inherit_fd = 0;
} else {
fprintf(stderr, "Invalid argument: %s\n", argv[0]);
exit(1);
@@ -636,6 +707,8 @@ int main(int argc, char **argv) {
cfg.type = CONN_TCP;
test_blocking_connection(cfg);
test_blocking_io_errors(cfg);
test_invalid_timeout_errors(cfg);
test_append_formatted_commands(cfg);
if (throughput) test_throughput(cfg);
printf("\nTesting against Unix socket connection (%s):\n", cfg.unix.path);
@@ -644,6 +717,12 @@ int main(int argc, char **argv) {
test_blocking_io_errors(cfg);
if (throughput) test_throughput(cfg);
if (test_inherit_fd) {
printf("\nTesting against inherited fd (%s):\n", cfg.unix.path);
cfg.type = CONN_FD;
test_blocking_connection(cfg);
}
if (fails) {
printf("*** %d TESTS FAILED ***\n", fails);
return 1;
deps/hiredis/zmalloc.h
Vendored
+13
View File
@@ -0,0 +1,13 @@
/* Drop in replacement for zmalloc.h in order to just use libc malloc without
* any wrappering. */
#ifndef ZMALLOC_H
#define ZMALLOC_H
#define zmalloc malloc
#define zrealloc realloc
#define zcalloc(x) calloc(x,1)
#define zfree free
#define zstrdup strdup
#endif
deps/jemalloc/.gitignore
+58 -9
View File
@@ -1,23 +1,72 @@
/*.gcov.*
/autom4te.cache/
/bin/jemalloc.sh
/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/jemalloc\.h
/include/jemalloc/jemalloc_defs\.h
/test/jemalloc_test\.h
/include/jemalloc/internal/jemalloc_internal.h
/include/jemalloc/internal/jemalloc_internal_defs.h
/include/jemalloc/internal/private_namespace.h
/include/jemalloc/internal/private_unnamespace.h
/include/jemalloc/internal/public_namespace.h
/include/jemalloc/internal/public_symbols.txt
/include/jemalloc/internal/public_unnamespace.h
/include/jemalloc/internal/size_classes.h
/include/jemalloc/jemalloc.h
/include/jemalloc/jemalloc_defs.h
/include/jemalloc/jemalloc_macros.h
/include/jemalloc/jemalloc_mangle.h
/include/jemalloc/jemalloc_mangle_jet.h
/include/jemalloc/jemalloc_protos.h
/include/jemalloc/jemalloc_protos_jet.h
/include/jemalloc/jemalloc_rename.h
/src/*.[od]
/test/*.[od]
/test/*.out
/test/[a-z]*
!test/*.c
!test/*.exp
/src/*.gcda
/src/*.gcno
/test/test.sh
test/include/test/jemalloc_test.h
test/include/test/jemalloc_test_defs.h
/test/integration/[A-Za-z]*
!/test/integration/[A-Za-z]*.*
/test/integration/*.[od]
/test/integration/*.gcda
/test/integration/*.gcno
/test/integration/*.out
/test/src/*.[od]
/test/src/*.gcda
/test/src/*.gcno
/test/stress/[A-Za-z]*
!/test/stress/[A-Za-z]*.*
/test/stress/*.[od]
/test/stress/*.gcda
/test/stress/*.gcno
/test/stress/*.out
/test/unit/[A-Za-z]*
!/test/unit/[A-Za-z]*.*
/test/unit/*.[od]
/test/unit/*.gcda
/test/unit/*.gcno
/test/unit/*.out
/VERSION
deps/jemalloc/COPYING
Vendored
+4 -28
View File
@@ -1,9 +1,10 @@
Unless otherwise specified, files in the jemalloc source distribution are
subject to the following licenses:
subject to the following license:
--------------------------------------------------------------------------------
Copyright (C) 2002-2010 Jason Evans <jasone@canonware.com>.
Copyright (C) 2002-2014 Jason Evans <jasone@canonware.com>.
All rights reserved.
Copyright (C) 2007-2010 Mozilla Foundation. All rights reserved.
Copyright (C) 2007-2012 Mozilla Foundation. All rights reserved.
Copyright (C) 2009-2014 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:
@@ -24,28 +25,3 @@ 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.
--------------------------------------------------------------------------------
Copyright (C) 2009-2010 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:
* 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 Facebook, Inc. 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/jemalloc/ChangeLog
Vendored
+300 -2
View File
@@ -3,8 +3,306 @@ 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
https://github.com/jemalloc/jemalloc
* 3.6.0 (March 31, 2014)
This version contains a critical bug fix for a regression present in 3.5.0 and
3.5.1.
Bug fixes:
- Fix a regression in arena_chunk_alloc() that caused crashes during
small/large allocation if chunk allocation failed. In the absence of this
bug, chunk allocation failure would result in allocation failure, e.g. NULL
return from malloc(). This regression was introduced in 3.5.0.
- Fix backtracing for gcc intrinsics-based backtracing by specifying
-fno-omit-frame-pointer to gcc. Note that the application (and all the
libraries it links to) must also be compiled with this option for
backtracing to be reliable.
- Use dss allocation precedence for huge allocations as well as small/large
allocations.
- Fix test assertion failure message formatting. This bug did not manifect on
x86_64 systems because of implementation subtleties in va_list.
- Fix inconsequential test failures for hash and SFMT code.
New features:
- Support heap profiling on FreeBSD. This feature depends on the proc
filesystem being mounted during heap profile dumping.
* 3.5.1 (February 25, 2014)
This version primarily addresses minor bugs in test code.
Bug fixes:
- Configure Solaris/Illumos to use MADV_FREE.
- Fix junk filling for mremap(2)-based huge reallocation. This is only
relevant if configuring with the --enable-mremap option specified.
- Avoid compilation failure if 'restrict' C99 keyword is not supported by the
compiler.
- Add a configure test for SSE2 rather than assuming it is usable on i686
systems. This fixes test compilation errors, especially on 32-bit Linux
systems.
- Fix mallctl argument size mismatches (size_t vs. uint64_t) in the stats unit
test.
- Fix/remove flawed alignment-related overflow tests.
- Prevent compiler optimizations that could change backtraces in the
prof_accum unit test.
* 3.5.0 (January 22, 2014)
This version focuses on refactoring and automated testing, though it also
includes some non-trivial heap profiling optimizations not mentioned below.
New features:
- Add the *allocx() API, which is a successor to the experimental *allocm()
API. The *allocx() functions are slightly simpler to use because they have
fewer parameters, they directly return the results of primary interest, and
mallocx()/rallocx() avoid the strict aliasing pitfall that
allocm()/rallocm() share with posix_memalign(). Note that *allocm() is
slated for removal in the next non-bugfix release.
- Add support for LinuxThreads.
Bug fixes:
- Unless heap profiling is enabled, disable floating point code and don't link
with libm. This, in combination with e.g. EXTRA_CFLAGS=-mno-sse on x64
systems, makes it possible to completely disable floating point register
use. Some versions of glibc neglect to save/restore caller-saved floating
point registers during dynamic lazy symbol loading, and the symbol loading
code uses whatever malloc the application happens to have linked/loaded
with, the result being potential floating point register corruption.
- Report ENOMEM rather than EINVAL if an OOM occurs during heap profiling
backtrace creation in imemalign(). This bug impacted posix_memalign() and
aligned_alloc().
- Fix a file descriptor leak in a prof_dump_maps() error path.
- Fix prof_dump() to close the dump file descriptor for all relevant error
paths.
- Fix rallocm() to use the arena specified by the ALLOCM_ARENA(s) flag for
allocation, not just deallocation.
- Fix a data race for large allocation stats counters.
- Fix a potential infinite loop during thread exit. This bug occurred on
Solaris, and could affect other platforms with similar pthreads TSD
implementations.
- Don't junk-fill reallocations unless usable size changes. This fixes a
violation of the *allocx()/*allocm() semantics.
- Fix growing large reallocation to junk fill new space.
- Fix huge deallocation to junk fill when munmap is disabled.
- Change the default private namespace prefix from empty to je_, and change
--with-private-namespace-prefix so that it prepends an additional prefix
rather than replacing je_. This reduces the likelihood of applications
which statically link jemalloc experiencing symbol name collisions.
- Add missing private namespace mangling (relevant when
--with-private-namespace is specified).
- Add and use JEMALLOC_INLINE_C so that static inline functions are marked as
static even for debug builds.
- Add a missing mutex unlock in a malloc_init_hard() error path. In practice
this error path is never executed.
- Fix numerous bugs in malloc_strotumax() error handling/reporting. These
bugs had no impact except for malformed inputs.
- Fix numerous bugs in malloc_snprintf(). These bugs were not exercised by
existing calls, so they had no impact.
* 3.4.1 (October 20, 2013)
Bug fixes:
- Fix a race in the "arenas.extend" mallctl that could cause memory corruption
of internal data structures and subsequent crashes.
- Fix Valgrind integration flaws that caused Valgrind warnings about reads of
uninitialized memory in:
+ arena chunk headers
+ internal zero-initialized data structures (relevant to tcache and prof
code)
- Preserve errno during the first allocation. A readlink(2) call during
initialization fails unless /etc/malloc.conf exists, so errno was typically
set during the first allocation prior to this fix.
- Fix compilation warnings reported by gcc 4.8.1.
* 3.4.0 (June 2, 2013)
This version is essentially a small bugfix release, but the addition of
aarch64 support requires that the minor version be incremented.
Bug fixes:
- Fix race-triggered deadlocks in chunk_record(). These deadlocks were
typically triggered by multiple threads concurrently deallocating huge
objects.
New features:
- Add support for the aarch64 architecture.
* 3.3.1 (March 6, 2013)
This version fixes bugs that are typically encountered only when utilizing
custom run-time options.
Bug fixes:
- Fix a locking order bug that could cause deadlock during fork if heap
profiling were enabled.
- Fix a chunk recycling bug that could cause the allocator to lose track of
whether a chunk was zeroed. On FreeBSD, NetBSD, and OS X, it could cause
corruption if allocating via sbrk(2) (unlikely unless running with the
"dss:primary" option specified). This was completely harmless on Linux
unless using mlockall(2) (and unlikely even then, unless the
--disable-munmap configure option or the "dss:primary" option was
specified). This regression was introduced in 3.1.0 by the
mlockall(2)/madvise(2) interaction fix.
- Fix TLS-related memory corruption that could occur during thread exit if the
thread never allocated memory. Only the quarantine and prof facilities were
susceptible.
- Fix two quarantine bugs:
+ Internal reallocation of the quarantined object array leaked the old
array.
+ Reallocation failure for internal reallocation of the quarantined object
array (very unlikely) resulted in memory corruption.
- Fix Valgrind integration to annotate all internally allocated memory in a
way that keeps Valgrind happy about internal data structure access.
- Fix building for s390 systems.
* 3.3.0 (January 23, 2013)
This version includes a few minor performance improvements in addition to the
listed new features and bug fixes.
New features:
- Add clipping support to lg_chunk option processing.
- Add the --enable-ivsalloc option.
- Add the --without-export option.
- Add the --disable-zone-allocator option.
Bug fixes:
- Fix "arenas.extend" mallctl to output the number of arenas.
- Fix chunk_recycle() to unconditionally inform Valgrind that returned memory
is undefined.
- Fix build break on FreeBSD related to alloca.h.
* 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)
deps/jemalloc/INSTALL
Vendored
+87 -38
View File
@@ -26,6 +26,19 @@ any of the following arguments (not a definitive list) to 'configure':
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:
@@ -42,11 +55,16 @@ any of the following arguments (not a definitive list) to 'configure':
jemalloc overlays the default malloc zone, but makes no attempt to actually
replace the "malloc", "calloc", etc. symbols.
--without-export
Don't export public APIs. This can be useful when building jemalloc as a
static library, or to avoid exporting public APIs when using the zone
allocator on OSX.
--with-private-namespace=<prefix>
Prefix all library-private APIs with <prefix>. For shared libraries,
Prefix all library-private APIs with <prefix>je_. 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).
default, <prefix> is empty, which results in a symbol prefix of je_ .
--with-install-suffix=<suffix>
Append <suffix> to the base name of all installed files, such that multiple
@@ -61,9 +79,28 @@ any of the following arguments (not a definitive list) to 'configure':
--enable-debug
Enable assertions and validation code. This incurs a substantial
performance hit, but is very useful during application development.
Implies --enable-ivsalloc.
--enable-stats
Enable statistics gathering functionality. See the "opt.stats_print"
--enable-code-coverage
Enable code coverage support, for use during jemalloc test development.
Additional testing targets are available if this option is enabled:
coverage
coverage_unit
coverage_integration
coverage_stress
These targets do not clear code coverage results from previous runs, and
there are interactions between the various coverage targets, so it is
usually advisable to run 'make clean' between repeated code coverage runs.
--enable-ivsalloc
Enable validation code, which verifies that pointers reside within
jemalloc-owned chunks before dereferencing them. This incurs a substantial
performance hit.
--disable-stats
Disable statistics gathering functionality. See the "opt.stats_print"
option documentation for usage details.
--enable-prof
@@ -90,51 +127,54 @@ any of the following arguments (not a definitive list) to 'configure':
Statically link against the specified libunwind.a rather than dynamically
linking with -lunwind.
--disable-tiny
Disable tiny (sub-quantum-sized) object support. Technically it is not
legal for a malloc implementation to allocate objects with less than
quantum alignment (8 or 16 bytes, depending on architecture), but in
practice it never causes any problems if, for example, 4-byte allocations
are 4-byte-aligned.
--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-swap
Enable mmap()ed swap file support. When this feature is built in, it is
possible to specify one or more files that act as backing store. This
effectively allows for per application swap files.
--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).
--enable-fill
Enable support for junk/zero filling of memory. See the "opt.junk"/
"opt.zero" option documentation for usage details.
--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()).
--disable-zone-allocator
Disable zone allocator for Darwin. This means jemalloc won't be hooked as
the default allocator on OSX/iOS.
--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-sysv
Enable support for System V semantics, wherein malloc(0) returns NULL
rather than a minimal allocation. See the "opt.sysv" option documentation
for usage details.
--enable-dynamic-page-shift
Under most conditions, the system page size never changes (usually 4KiB or
8KiB, depending on architecture and configuration), and unless this option
is enabled, jemalloc assumes that page size can safely be determined during
configuration and hard-coded. Enabling dynamic page size determination has
a measurable impact on performance, since the compiler is forced to load
the page size from memory rather than embedding immediate values.
--disable-lazy-lock
Disable code that wraps pthread_create() to detect when an application
--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
@@ -181,11 +221,24 @@ PATH="?"
=== 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:
@@ -248,10 +301,6 @@ directory, issue configuration and build commands:
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 any of the following commands:
prior to installation via the following command:
nroff -man -t doc/jemalloc.3
groff -man -t -Tps doc/jemalloc.3 | ps2pdf - doc/jemalloc.3.pdf
(cd doc; groff -man -man-ext -t -Thtml jemalloc.3 > jemalloc.3.html)
deps/jemalloc/Makefile.in
Vendored
+335 -156
View File
@@ -17,130 +17,262 @@ INCLUDEDIR := $(DESTDIR)@INCLUDEDIR@
LIBDIR := $(DESTDIR)@LIBDIR@
DATADIR := $(DESTDIR)@DATADIR@
MANDIR := $(DESTDIR)@MANDIR@
srcroot := @srcroot@
objroot := @objroot@
abs_srcroot := @abs_srcroot@
abs_objroot := @abs_objroot@
# Build parameters.
CPPFLAGS := @CPPFLAGS@ -I@srcroot@include -I@objroot@include
CPPFLAGS := @CPPFLAGS@ -I$(srcroot)include -I$(objroot)include
CFLAGS := @CFLAGS@
ifeq (macho, @abi@)
CFLAGS += -dynamic
endif
LDFLAGS := @LDFLAGS@
EXTRA_LDFLAGS := @EXTRA_LDFLAGS@
LIBS := @LIBS@
RPATH_EXTRA := @RPATH_EXTRA@
ifeq (macho, @abi@)
SO := dylib
WL_SONAME := dylib_install_name
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_code_coverage := @enable_code_coverage@
enable_experimental := @enable_experimental@
enable_zone_allocator := @enable_zone_allocator@
DSO_LDFLAGS = @DSO_LDFLAGS@
SOREV = @SOREV@
PIC_CFLAGS = @PIC_CFLAGS@
CTARGET = @CTARGET@
LDTARGET = @LDTARGET@
MKLIB = @MKLIB@
AR = @AR@
ARFLAGS = @ARFLAGS@
CC_MM = @CC_MM@
ifeq (macho, $(ABI))
TEST_LIBRARY_PATH := DYLD_FALLBACK_LIBRARY_PATH="$(objroot)lib"
else
SO := so
WL_SONAME := soname
endif
REV := 1
ifeq (macho, @abi@)
TEST_LIBRARY_PATH := DYLD_FALLBACK_LIBRARY_PATH=@objroot@lib
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
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/chunk_swap.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/rtree.c @srcroot@src/stats.c @srcroot@src/tcache.c
ifeq (macho, @abi@)
CSRCS += @srcroot@src/zone.c
BINS := $(srcroot)bin/pprof $(objroot)bin/jemalloc.sh
C_HDRS := $(objroot)include/jemalloc/jemalloc$(install_suffix).h
C_SRCS := $(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 $(srcroot)src/tsd.c
ifeq ($(enable_zone_allocator), 1)
C_SRCS += $(srcroot)src/zone.c
endif
STATIC_LIBS := @objroot@lib/libjemalloc@install_suffix@.a
DSOS := @objroot@lib/libjemalloc@install_suffix@.$(SO).$(REV) \
@objroot@lib/libjemalloc@install_suffix@.$(SO) \
@objroot@lib/libjemalloc@install_suffix@_pic.a
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 := $(DOCS_XML:@objroot@%.xml=@srcroot@%.3)
ifeq ($(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/allocated.c @srcroot@test/allocm.c \
@srcroot@test/bitmap.c @srcroot@test/mremap.c \
@srcroot@test/posix_memalign.c @srcroot@test/rallocm.c \
@srcroot@test/thread_arena.c
C_TESTLIB_SRCS := $(srcroot)test/src/math.c $(srcroot)test/src/mtx.c \
$(srcroot)test/src/SFMT.c $(srcroot)test/src/test.c \
$(srcroot)test/src/thd.c
C_UTIL_INTEGRATION_SRCS := $(srcroot)src/util.c
TESTS_UNIT := $(srcroot)test/unit/bitmap.c \
$(srcroot)test/unit/ckh.c \
$(srcroot)test/unit/hash.c \
$(srcroot)test/unit/junk.c \
$(srcroot)test/unit/mallctl.c \
$(srcroot)test/unit/math.c \
$(srcroot)test/unit/mq.c \
$(srcroot)test/unit/mtx.c \
$(srcroot)test/unit/prof_accum.c \
$(srcroot)test/unit/prof_gdump.c \
$(srcroot)test/unit/prof_idump.c \
$(srcroot)test/unit/ql.c \
$(srcroot)test/unit/qr.c \
$(srcroot)test/unit/quarantine.c \
$(srcroot)test/unit/rb.c \
$(srcroot)test/unit/rtree.c \
$(srcroot)test/unit/SFMT.c \
$(srcroot)test/unit/stats.c \
$(srcroot)test/unit/tsd.c \
$(srcroot)test/unit/util.c \
$(srcroot)test/unit/zero.c
TESTS_UNIT_AUX := $(srcroot)test/unit/prof_accum_a.c \
$(srcroot)test/unit/prof_accum_b.c
TESTS_INTEGRATION := $(srcroot)test/integration/aligned_alloc.c \
$(srcroot)test/integration/allocated.c \
$(srcroot)test/integration/mallocx.c \
$(srcroot)test/integration/mremap.c \
$(srcroot)test/integration/posix_memalign.c \
$(srcroot)test/integration/rallocx.c \
$(srcroot)test/integration/thread_arena.c \
$(srcroot)test/integration/thread_tcache_enabled.c \
$(srcroot)test/integration/xallocx.c
ifeq ($(enable_experimental), 1)
TESTS_INTEGRATION += $(srcroot)test/integration/allocm.c \
$(srcroot)test/integration/MALLOCX_ARENA.c \
$(srcroot)test/integration/rallocm.c
endif
TESTS_STRESS :=
TESTS := $(TESTS_UNIT) $(TESTS_INTEGRATION) $(TESTS_STRESS)
.PHONY: all dist doc_html doc_man doc
C_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.$(O))
C_PIC_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.pic.$(O))
C_JET_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.jet.$(O))
C_TESTLIB_UNIT_OBJS := $(C_TESTLIB_SRCS:$(srcroot)%.c=$(objroot)%.unit.$(O))
C_TESTLIB_INTEGRATION_OBJS := $(C_TESTLIB_SRCS:$(srcroot)%.c=$(objroot)%.integration.$(O))
C_UTIL_INTEGRATION_OBJS := $(C_UTIL_INTEGRATION_SRCS:$(srcroot)%.c=$(objroot)%.integration.$(O))
C_TESTLIB_STRESS_OBJS := $(C_TESTLIB_SRCS:$(srcroot)%.c=$(objroot)%.stress.$(O))
C_TESTLIB_OBJS := $(C_TESTLIB_UNIT_OBJS) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(C_TESTLIB_STRESS_OBJS)
TESTS_UNIT_OBJS := $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%.$(O))
TESTS_UNIT_AUX_OBJS := $(TESTS_UNIT_AUX:$(srcroot)%.c=$(objroot)%.$(O))
TESTS_INTEGRATION_OBJS := $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%.$(O))
TESTS_STRESS_OBJS := $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%.$(O))
TESTS_OBJS := $(TESTS_UNIT_OBJS) $(TESTS_UNIT_AUX_OBJS) $(TESTS_INTEGRATION_OBJS) $(TESTS_STRESS_OBJS)
.PHONY: all dist build_doc_html build_doc_man build_doc
.PHONY: install_bin install_include install_lib
.PHONY: install_html install_man install_doc install
.PHONY: install_doc_html install_doc_man install_doc install
.PHONY: tests check clean distclean relclean
.SECONDARY : $(CTESTS:@srcroot@%.c=@objroot@%.o)
.SECONDARY : $(TESTS_OBJS)
# Default target.
all: $(DSOS) $(STATIC_LIBS)
all: build_lib
dist: doc
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/%.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 $<
$(srcroot)doc/%.3 : $(objroot)doc/%.xml $(srcroot)doc/stylesheet.xsl $(objroot)doc/manpages.xsl
$(XSLTPROC) -o $@ $(objroot)doc/manpages.xsl $<
doc_html: $(DOCS_HTML)
doc_man: $(DOCS_MAN3)
doc: $(DOCS)
build_doc_html: $(DOCS_HTML)
build_doc_man: $(DOCS_MAN3)
build_doc: $(DOCS)
#
# Include generated dependency files.
#
-include $(CSRCS:@srcroot@%.c=@objroot@%.d)
-include $(CSRCS:@srcroot@%.c=@objroot@%.pic.d)
-include $(CTESTS:@srcroot@%.c=@objroot@%.d)
ifdef CC_MM
-include $(C_OBJS:%.$(O)=%.d)
-include $(C_PIC_OBJS:%.$(O)=%.d)
-include $(C_JET_OBJS:%.$(O)=%.d)
-include $(C_TESTLIB_OBJS:%.$(O)=%.d)
-include $(TESTS_OBJS:%.$(O)=%.d)
endif
@objroot@src/%.o: @srcroot@src/%.c
$(C_OBJS): $(objroot)src/%.$(O): $(srcroot)src/%.c
$(C_PIC_OBJS): $(objroot)src/%.pic.$(O): $(srcroot)src/%.c
$(C_PIC_OBJS): CFLAGS += $(PIC_CFLAGS)
$(C_JET_OBJS): $(objroot)src/%.jet.$(O): $(srcroot)src/%.c
$(C_JET_OBJS): CFLAGS += -DJEMALLOC_JET
$(C_TESTLIB_UNIT_OBJS): $(objroot)test/src/%.unit.$(O): $(srcroot)test/src/%.c
$(C_TESTLIB_UNIT_OBJS): CPPFLAGS += -DJEMALLOC_UNIT_TEST
$(C_TESTLIB_INTEGRATION_OBJS): $(objroot)test/src/%.integration.$(O): $(srcroot)test/src/%.c
$(C_TESTLIB_INTEGRATION_OBJS): CPPFLAGS += -DJEMALLOC_INTEGRATION_TEST
$(C_UTIL_INTEGRATION_OBJS): $(objroot)src/%.integration.$(O): $(srcroot)src/%.c
$(C_TESTLIB_STRESS_OBJS): $(objroot)test/src/%.stress.$(O): $(srcroot)test/src/%.c
$(C_TESTLIB_STRESS_OBJS): CPPFLAGS += -DJEMALLOC_STRESS_TEST -DJEMALLOC_STRESS_TESTLIB
$(C_TESTLIB_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include
$(TESTS_UNIT_OBJS): CPPFLAGS += -DJEMALLOC_UNIT_TEST
$(TESTS_UNIT_AUX_OBJS): CPPFLAGS += -DJEMALLOC_UNIT_TEST
define make-unit-link-dep
$(1): TESTS_UNIT_LINK_OBJS += $(2)
$(1): $(2)
endef
$(foreach test, $(TESTS_UNIT:$(srcroot)test/unit/%.c=$(objroot)test/unit/%$(EXE)), $(eval $(call make-unit-link-dep,$(test),$(filter $(test:%=%_a.$(O)) $(test:%=%_b.$(O)),$(TESTS_UNIT_AUX_OBJS)))))
$(TESTS_INTEGRATION_OBJS): CPPFLAGS += -DJEMALLOC_INTEGRATION_TEST
$(TESTS_STRESS_OBJS): CPPFLAGS += -DJEMALLOC_STRESS_TEST
$(TESTS_OBJS): $(objroot)test/%.$(O): $(srcroot)test/%.c
$(TESTS_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include
ifneq ($(IMPORTLIB),$(SO))
$(C_OBJS): 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))
$(C_OBJS) $(C_PIC_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS): $(HEADERS)
$(TESTS_OBJS): $(objroot)test/unit/jemalloc_test.h
endif
$(C_OBJS) $(C_PIC_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS): %.$(O):
@mkdir -p $(@D)
$(CC) $(CFLAGS) -c $(CPPFLAGS) -o $@ $<
@$(SHELL) -ec "$(CC) -MM $(CPPFLAGS) $< | sed \"s/\($(subst /,\/,$(notdir $(basename $@)))\)\.o\([ :]*\)/$(subst /,\/,$(strip $(dir $@)))\1.o \2/g\" > $(@:%.o=%.d)"
$(CC) $(CFLAGS) -c $(CPPFLAGS) $(CTARGET) $<
ifdef CC_MM
@$(CC) -MM $(CPPFLAGS) -MT $@ -o $(@:%.$(O)=%.d) $<
endif
@objroot@src/%.pic.o: @srcroot@src/%.c
@mkdir -p $(@D)
$(CC) $(CFLAGS) -fPIC -DPIC -c $(CPPFLAGS) -o $@ $<
@$(SHELL) -ec "$(CC) -MM $(CPPFLAGS) $< | sed \"s/\($(subst /,\/,$(notdir $(basename $(basename $@))))\)\.o\([ :]*\)/$(subst /,\/,$(strip $(dir $@)))\1.pic.o \2/g\" > $(@:%.o=%.d)"
%.$(SO) : %.$(SO).$(REV)
ifneq ($(SOREV),$(SO))
%.$(SO) : %.$(SOREV)
@mkdir -p $(@D)
ln -sf $(<F) $@
@objroot@lib/libjemalloc@install_suffix@.$(SO).$(REV) : $(CSRCS:@srcroot@%.c=@objroot@%.pic.o)
@mkdir -p $(@D)
$(CC) -shared -Wl,-$(WL_SONAME),$(@F) $(RPATH_EXTRA:%=@RPATH@%) -o $@ $+ $(LDFLAGS) $(LIBS)
@objroot@lib/libjemalloc@install_suffix@_pic.a : $(CSRCS:@srcroot@%.c=@objroot@%.pic.o)
@mkdir -p $(@D)
ar crus $@ $+
@objroot@lib/libjemalloc@install_suffix@.a : $(CSRCS:@srcroot@%.c=@objroot@%.o)
@mkdir -p $(@D)
ar crus $@ $+
@objroot@test/%.o: @srcroot@test/%.c
@mkdir -p $(@D)
$(CC) $(CFLAGS) -c $(CPPFLAGS) -I@objroot@test -o $@ $<
@$(SHELL) -ec "$(CC) -MM $(CPPFLAGS) -I@objroot@test $< | sed \"s/\($(subst /,\/,$(notdir $(basename $@)))\)\.o\([ :]*\)/$(subst /,\/,$(strip $(dir $@)))\1.o \2/g\" > $(@:%.o=%.d)"
# Automatic dependency generation misses #include "*.c".
@objroot@test/bitmap.o : @objroot@src/bitmap.o
@objroot@test/%: @objroot@test/%.o \
@objroot@lib/libjemalloc@install_suffix@.$(SO)
@mkdir -p $(@D)
ifneq (@RPATH@, )
$(CC) -o $@ $< @RPATH@@objroot@lib -L@objroot@lib -ljemalloc@install_suffix@ -lpthread
else
$(CC) -o $@ $< -L@objroot@lib -ljemalloc@install_suffix@ -lpthread
endif
$(objroot)lib/$(LIBJEMALLOC).$(SOREV) : $(if $(PIC_CFLAGS),$(C_PIC_OBJS),$(C_OBJS))
@mkdir -p $(@D)
$(CC) $(DSO_LDFLAGS) $(call RPATH,$(RPATH_EXTRA)) $(LDTARGET) $+ $(LDFLAGS) $(LIBS) $(EXTRA_LDFLAGS)
$(objroot)lib/$(LIBJEMALLOC)_pic.$(A) : $(C_PIC_OBJS)
$(objroot)lib/$(LIBJEMALLOC).$(A) : $(C_OBJS)
$(objroot)lib/$(LIBJEMALLOC)_s.$(A) : $(C_OBJS)
$(STATIC_LIBS):
@mkdir -p $(@D)
$(AR) $(ARFLAGS)@AROUT@ $+
$(objroot)test/unit/%$(EXE): $(objroot)test/unit/%.$(O) $(TESTS_UNIT_LINK_OBJS) $(C_JET_OBJS) $(C_TESTLIB_UNIT_OBJS)
@mkdir -p $(@D)
$(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(LDFLAGS) $(filter-out -lm,$(LIBS)) -lm $(EXTRA_LDFLAGS)
$(objroot)test/integration/%$(EXE): $(objroot)test/integration/%.$(O) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB)
@mkdir -p $(@D)
$(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(filter -lpthread,$(LIBS))) -lm $(EXTRA_LDFLAGS)
$(objroot)test/stress/%$(EXE): $(objroot)test/stress/%.$(O) $(C_JET_OBJS) $(C_TESTLIB_STRESS_OBJS) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB)
@mkdir -p $(@D)
$(CC) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(LIBS)) -lm $(EXTRA_LDFLAGS)
build_lib_shared: $(DSOS)
build_lib_static: $(STATIC_LIBS)
build_lib: build_lib_shared build_lib_static
install_bin:
install -d $(BINDIR)
@for b in $(BINS); do \
@@ -150,110 +282,157 @@ done
install_include:
install -d $(INCLUDEDIR)/jemalloc
@for h in $(CHDRS); do \
@for h in $(C_HDRS); do \
echo "install -m 644 $$h $(INCLUDEDIR)/jemalloc"; \
install -m 644 $$h $(INCLUDEDIR)/jemalloc; \
done
install_lib: $(DSOS) $(STATIC_LIBS)
install_lib_shared: $(DSOS)
install -d $(LIBDIR)
install -m 755 @objroot@lib/libjemalloc@install_suffix@.$(SO).$(REV) $(LIBDIR)
ln -sf libjemalloc@install_suffix@.$(SO).$(REV) $(LIBDIR)/libjemalloc@install_suffix@.$(SO)
install -m 755 @objroot@lib/libjemalloc@install_suffix@_pic.a $(LIBDIR)
install -m 755 @objroot@lib/libjemalloc@install_suffix@.a $(LIBDIR)
install -m 755 $(objroot)lib/$(LIBJEMALLOC).$(SOREV) $(LIBDIR)
ifneq ($(SOREV),$(SO))
ln -sf $(LIBJEMALLOC).$(SOREV) $(LIBDIR)/$(LIBJEMALLOC).$(SO)
endif
install_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@; \
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_man:
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_html install_man
install_doc: install_doc_html install_doc_man
install: install_bin install_include install_lib install_doc
tests: $(CTESTS:@srcroot@%.c=@objroot@%)
tests_unit: $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%$(EXE))
tests_integration: $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%$(EXE))
tests_stress: $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%$(EXE))
tests: tests_unit tests_integration tests_stress
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) $${t} @abs_srcroot@ @abs_objroot@ \
> @objroot@$${t}.out 2>&1; \
if test -e "@srcroot@$${t}.exp"; then \
diff -u @srcroot@$${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}"'
check_unit_dir:
@mkdir -p $(objroot)test/unit
check_integration_dir:
@mkdir -p $(objroot)test/integration
check_stress_dir:
@mkdir -p $(objroot)test/stress
check_dir: check_unit_dir check_integration_dir check_stress_dir
check_unit: tests_unit check_unit_dir
$(SHELL) $(objroot)test/test.sh $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%)
check_integration: tests_integration check_integration_dir
$(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%)
check_stress: tests_stress check_stress_dir
$(SHELL) $(objroot)test/test.sh $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%)
check: tests check_dir
$(SHELL) $(objroot)test/test.sh $(TESTS:$(srcroot)%.c=$(objroot)%)
ifeq ($(enable_code_coverage), 1)
coverage_unit: check_unit
$(SHELL) $(srcroot)coverage.sh $(srcroot)src jet $(C_JET_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/src unit $(C_TESTLIB_UNIT_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/unit unit $(TESTS_UNIT_OBJS)
coverage_integration: check_integration
$(SHELL) $(srcroot)coverage.sh $(srcroot)src pic $(C_PIC_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)src integration $(C_UTIL_INTEGRATION_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/src integration $(C_TESTLIB_INTEGRATION_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/integration integration $(TESTS_INTEGRATION_OBJS)
coverage_stress: check_stress
$(SHELL) $(srcroot)coverage.sh $(srcroot)src pic $(C_PIC_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)src jet $(C_JET_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/src stress $(C_TESTLIB_STRESS_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/stress stress $(TESTS_STRESS_OBJS)
coverage: check
$(SHELL) $(srcroot)coverage.sh $(srcroot)src pic $(C_PIC_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)src jet $(C_JET_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)src integration $(C_UTIL_INTEGRATION_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/src unit $(C_TESTLIB_UNIT_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/src integration $(C_TESTLIB_INTEGRATION_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/src stress $(C_TESTLIB_STRESS_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/unit unit $(TESTS_UNIT_OBJS) $(TESTS_UNIT_AUX_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/integration integration $(TESTS_INTEGRATION_OBJS)
$(SHELL) $(srcroot)coverage.sh $(srcroot)test/stress integration $(TESTS_STRESS_OBJS)
endif
clean:
rm -f $(CSRCS:@srcroot@%.c=@objroot@%.o)
rm -f $(CSRCS:@srcroot@%.c=@objroot@%.pic.o)
rm -f $(CSRCS:@srcroot@%.c=@objroot@%.d)
rm -f $(CSRCS:@srcroot@%.c=@objroot@%.pic.d)
rm -f $(CTESTS:@srcroot@%.c=@objroot@%)
rm -f $(CTESTS:@srcroot@%.c=@objroot@%.o)
rm -f $(CTESTS:@srcroot@%.c=@objroot@%.d)
rm -f $(CTESTS:@srcroot@%.c=@objroot@%.out)
rm -f $(C_OBJS)
rm -f $(C_PIC_OBJS)
rm -f $(C_JET_OBJS)
rm -f $(C_TESTLIB_OBJS)
rm -f $(C_OBJS:%.$(O)=%.d)
rm -f $(C_OBJS:%.$(O)=%.gcda)
rm -f $(C_OBJS:%.$(O)=%.gcno)
rm -f $(C_PIC_OBJS:%.$(O)=%.d)
rm -f $(C_PIC_OBJS:%.$(O)=%.gcda)
rm -f $(C_PIC_OBJS:%.$(O)=%.gcno)
rm -f $(C_JET_OBJS:%.$(O)=%.d)
rm -f $(C_JET_OBJS:%.$(O)=%.gcda)
rm -f $(C_JET_OBJS:%.$(O)=%.gcno)
rm -f $(C_TESTLIB_OBJS:%.$(O)=%.d)
rm -f $(C_TESTLIB_OBJS:%.$(O)=%.gcda)
rm -f $(C_TESTLIB_OBJS:%.$(O)=%.gcno)
rm -f $(TESTS_OBJS:%.$(O)=%$(EXE))
rm -f $(TESTS_OBJS)
rm -f $(TESTS_OBJS:%.$(O)=%.d)
rm -f $(TESTS_OBJS:%.$(O)=%.gcda)
rm -f $(TESTS_OBJS:%.$(O)=%.gcno)
rm -f $(TESTS_OBJS:%.$(O)=%.out)
rm -f $(DSOS) $(STATIC_LIBS)
rm -f $(objroot)*.gcov.*
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@
rm -rf $(objroot)autom4te.cache
rm -f $(objroot)bin/jemalloc.sh
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 $(objroot)configure
rm -f $(srcroot)VERSION
rm -f $(DOCS_HTML)
rm -f $(DOCS_MAN3)
#===============================================================================
# Re-configuration rules.
ifeq (@enable_autogen@, 1)
@srcroot@configure : @srcroot@configure.ac
cd ./@srcroot@ && @AUTOCONF@
ifeq ($(enable_autogen), 1)
$(srcroot)configure : $(srcroot)configure.ac
cd ./$(srcroot) && $(AUTOCONF)
@objroot@config.status : @srcroot@configure
./@objroot@config.status --recheck
$(objroot)config.status : $(srcroot)configure
./$(objroot)config.status --recheck
@srcroot@config.stamp.in : @srcroot@configure.ac
echo stamp > @srcroot@config.stamp.in
$(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
$(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
$(cfgoutputs_out) $(cfghdrs_out) : $(objroot)config.stamp
@true
endif
deps/jemalloc/README
Vendored
+11 -7
View File
@@ -1,10 +1,14 @@
jemalloc is a general-purpose scalable concurrent malloc(3) implementation.
This distribution is a stand-alone "portable" implementation that currently
targets Linux and Apple OS X. jemalloc is included as the default allocator in
the FreeBSD and NetBSD operating systems, and it is used by the Mozilla Firefox
web browser on Microsoft Windows-related platforms. Depending on your needs,
one of the other divergent versions may suit your needs better than this
distribution.
jemalloc is a general purpose malloc(3) implementation that emphasizes
fragmentation avoidance and scalable concurrency support. jemalloc first came
into use as the FreeBSD libc allocator in 2005, and since then it has found its
way into numerous applications that rely on its predictable behavior. In 2010
jemalloc development efforts broadened to include developer support features
such as heap profiling, Valgrind integration, and extensive monitoring/tuning
hooks. Modern jemalloc releases continue to be integrated back into FreeBSD,
and therefore versatility remains critical. Ongoing development efforts trend
toward making jemalloc among the best allocators for a broad range of demanding
applications, and eliminating/mitigating weaknesses that have practical
repercussions for real world applications.
The COPYING file contains copyright and licensing information.
deps/jemalloc/VERSION
Vendored
+1 -1
View File
@@ -1 +1 @@
2.2.5-0-gfc1bb70e5f0d9a58b39efa39cc549b5af5104760
3.6.0-0-g46c0af68bd248b04df75e4f92d5fb804c3d75340
deps/jemalloc/bin/jemalloc.sh.in
Vendored
+9
View File
@@ -0,0 +1,9 @@
#!/bin/sh
prefix=@prefix@
exec_prefix=@exec_prefix@
libdir=@libdir@
@LD_PRELOAD_VAR@=${libdir}/libjemalloc.@SOREV@
export @LD_PRELOAD_VAR@
exec "$@"
deps/jemalloc/bin/pprof
Vendored
+674 -195
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/config.guess
Vendored
+607 -505
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/config.sub
Vendored
+368 -124
View File
@@ -1,42 +1,40 @@
#! /bin/sh
# Configuration validation subroutine script.
# Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
# 2000, 2001, 2002, 2003 Free Software Foundation, Inc.
# Copyright 1992-2013 Free Software Foundation, Inc.
timestamp='2004-02-23'
timestamp='2013-10-01'
# This file is (in principle) common to ALL GNU software.
# The presence of a machine in this file suggests that SOME GNU software
# can handle that machine. It does not imply ALL GNU software can.
#
# This file is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330,
# Boston, MA 02111-1307, USA.
# along with this program; if not, see <http://www.gnu.org/licenses/>.
#
# As a special exception to the GNU General Public License, if you
# distribute this file as part of a program that contains a
# configuration script generated by Autoconf, you may include it under
# the same distribution terms that you use for the rest of that program.
# the same distribution terms that you use for the rest of that
# program. This Exception is an additional permission under section 7
# of the GNU General Public License, version 3 ("GPLv3").
# Please send patches to <config-patches@gnu.org>. Submit a context
# diff and a properly formatted ChangeLog entry.
# Please send patches with a ChangeLog entry to config-patches@gnu.org.
#
# Configuration subroutine to validate and canonicalize a configuration type.
# Supply the specified configuration type as an argument.
# If it is invalid, we print an error message on stderr and exit with code 1.
# Otherwise, we print the canonical config type on stdout and succeed.
# You can get the latest version of this script from:
# http://git.savannah.gnu.org/gitweb/?p=config.git;a=blob_plain;f=config.sub;hb=HEAD
# This file is supposed to be the same for all GNU packages
# and recognize all the CPU types, system types and aliases
# that are meaningful with *any* GNU software.
@@ -70,8 +68,7 @@ Report bugs and patches to <config-patches@gnu.org>."
version="\
GNU config.sub ($timestamp)
Copyright (C) 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001
Free Software Foundation, Inc.
Copyright 1992-2013 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE."
@@ -83,11 +80,11 @@ Try \`$me --help' for more information."
while test $# -gt 0 ; do
case $1 in
--time-stamp | --time* | -t )
echo "$timestamp" ; exit 0 ;;
echo "$timestamp" ; exit ;;
--version | -v )
echo "$version" ; exit 0 ;;
echo "$version" ; exit ;;
--help | --h* | -h )
echo "$usage"; exit 0 ;;
echo "$usage"; exit ;;
-- ) # Stop option processing
shift; break ;;
- ) # Use stdin as input.
@@ -99,7 +96,7 @@ while test $# -gt 0 ; do
*local*)
# First pass through any local machine types.
echo $1
exit 0;;
exit ;;
* )
break ;;
@@ -118,11 +115,18 @@ esac
# Here we must recognize all the valid KERNEL-OS combinations.
maybe_os=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\2/'`
case $maybe_os in
nto-qnx* | linux-gnu* | linux-dietlibc | linux-uclibc* | uclinux-uclibc* | uclinux-gnu* | \
kfreebsd*-gnu* | knetbsd*-gnu* | netbsd*-gnu* | storm-chaos* | os2-emx* | rtmk-nova*)
nto-qnx* | linux-gnu* | linux-android* | linux-dietlibc | linux-newlib* | \
linux-musl* | linux-uclibc* | uclinux-uclibc* | uclinux-gnu* | kfreebsd*-gnu* | \
knetbsd*-gnu* | netbsd*-gnu* | \
kopensolaris*-gnu* | \
storm-chaos* | os2-emx* | rtmk-nova*)
os=-$maybe_os
basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`
;;
android-linux)
os=-linux-android
basic_machine=`echo $1 | sed 's/^\(.*\)-\([^-]*-[^-]*\)$/\1/'`-unknown
;;
*)
basic_machine=`echo $1 | sed 's/-[^-]*$//'`
if [ $basic_machine != $1 ]
@@ -145,10 +149,13 @@ case $os in
-convergent* | -ncr* | -news | -32* | -3600* | -3100* | -hitachi* |\
-c[123]* | -convex* | -sun | -crds | -omron* | -dg | -ultra | -tti* | \
-harris | -dolphin | -highlevel | -gould | -cbm | -ns | -masscomp | \
-apple | -axis)
-apple | -axis | -knuth | -cray | -microblaze*)
os=
basic_machine=$1
;;
-bluegene*)
os=-cnk
;;
-sim | -cisco | -oki | -wec | -winbond)
os=
basic_machine=$1
@@ -163,13 +170,17 @@ case $os in
os=-chorusos
basic_machine=$1
;;
-chorusrdb)
os=-chorusrdb
-chorusrdb)
os=-chorusrdb
basic_machine=$1
;;
;;
-hiux*)
os=-hiuxwe2
;;
-sco6)
os=-sco5v6
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco5)
os=-sco3.2v5
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
@@ -186,6 +197,10 @@ case $os in
# Don't forget version if it is 3.2v4 or newer.
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco5v6*)
# Don't forget version if it is 3.2v4 or newer.
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-sco*)
os=-sco3.2v2
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
@@ -203,6 +218,12 @@ case $os in
-isc*)
basic_machine=`echo $1 | sed -e 's/86-.*/86-pc/'`
;;
-lynx*178)
os=-lynxos178
;;
-lynx*5)
os=-lynxos5
;;
-lynx*)
os=-lynxos
;;
@@ -227,57 +248,107 @@ case $basic_machine in
# Some are omitted here because they have special meanings below.
1750a | 580 \
| a29k \
| aarch64 | aarch64_be \
| alpha | alphaev[4-8] | alphaev56 | alphaev6[78] | alphapca5[67] \
| alpha64 | alpha64ev[4-8] | alpha64ev56 | alpha64ev6[78] | alpha64pca5[67] \
| am33_2.0 \
| arc | arm | arm[bl]e | arme[lb] | armv[2345] | armv[345][lb] | avr \
| c4x | clipper \
| arc | arceb \
| arm | arm[bl]e | arme[lb] | armv[2-8] | armv[3-8][lb] | armv7[arm] \
| avr | avr32 \
| be32 | be64 \
| bfin \
| c4x | c8051 | clipper \
| d10v | d30v | dlx | dsp16xx \
| fr30 | frv \
| epiphany \
| fido | fr30 | frv \
| h8300 | h8500 | hppa | hppa1.[01] | hppa2.0 | hppa2.0[nw] | hppa64 \
| hexagon \
| i370 | i860 | i960 | ia64 \
| ip2k | iq2000 \
| m32r | m68000 | m68k | m88k | mcore \
| k1om \
| le32 | le64 \
| lm32 \
| m32c | m32r | m32rle | m68000 | m68k | m88k \
| maxq | mb | microblaze | microblazeel | mcore | mep | metag \
| mips | mipsbe | mipseb | mipsel | mipsle \
| mips16 \
| mips64 | mips64el \
| mips64vr | mips64vrel \
| mips64octeon | mips64octeonel \
| mips64orion | mips64orionel \
| mips64r5900 | mips64r5900el \
| mips64vr | mips64vrel \
| mips64vr4100 | mips64vr4100el \
| mips64vr4300 | mips64vr4300el \
| mips64vr5000 | mips64vr5000el \
| mips64vr5900 | mips64vr5900el \
| mipsisa32 | mipsisa32el \
| mipsisa32r2 | mipsisa32r2el \
| mipsisa64 | mipsisa64el \
| mipsisa64r2 | mipsisa64r2el \
| mipsisa64sb1 | mipsisa64sb1el \
| mipsisa64sr71k | mipsisa64sr71kel \
| mipsr5900 | mipsr5900el \
| mipstx39 | mipstx39el \
| mn10200 | mn10300 \
| moxie \
| mt \
| msp430 \
| nds32 | nds32le | nds32be \
| nios | nios2 | nios2eb | nios2el \
| ns16k | ns32k \
| openrisc | or32 \
| open8 \
| or1k | or32 \
| pdp10 | pdp11 | pj | pjl \
| powerpc | powerpc64 | powerpc64le | powerpcle | ppcbe \
| powerpc | powerpc64 | powerpc64le | powerpcle \
| pyramid \
| sh | sh[1234] | sh[23]e | sh[34]eb | shbe | shle | sh[1234]le | sh3ele \
| rl78 | rx \
| score \
| sh | sh[1234] | sh[24]a | sh[24]aeb | sh[23]e | sh[34]eb | sheb | shbe | shle | sh[1234]le | sh3ele \
| sh64 | sh64le \
| sparc | sparc64 | sparc86x | sparclet | sparclite | sparcv9 | sparcv9b \
| strongarm \
| tahoe | thumb | tic4x | tic80 | tron \
| v850 | v850e \
| sparc | sparc64 | sparc64b | sparc64v | sparc86x | sparclet | sparclite \
| sparcv8 | sparcv9 | sparcv9b | sparcv9v \
| spu \
| tahoe | tic4x | tic54x | tic55x | tic6x | tic80 | tron \
| ubicom32 \
| v850 | v850e | v850e1 | v850e2 | v850es | v850e2v3 \
| we32k \
| x86 | xscale | xstormy16 | xtensa \
| z8k)
| x86 | xc16x | xstormy16 | xtensa \
| z8k | z80)
basic_machine=$basic_machine-unknown
;;
m6811 | m68hc11 | m6812 | m68hc12)
# Motorola 68HC11/12.
c54x)
basic_machine=tic54x-unknown
;;
c55x)
basic_machine=tic55x-unknown
;;
c6x)
basic_machine=tic6x-unknown
;;
m6811 | m68hc11 | m6812 | m68hc12 | m68hcs12x | nvptx | picochip)
basic_machine=$basic_machine-unknown
os=-none
;;
m88110 | m680[12346]0 | m683?2 | m68360 | m5200 | v70 | w65 | z8k)
;;
ms1)
basic_machine=mt-unknown
;;
strongarm | thumb | xscale)
basic_machine=arm-unknown
;;
xgate)
basic_machine=$basic_machine-unknown
os=-none
;;
xscaleeb)
basic_machine=armeb-unknown
;;
xscaleel)
basic_machine=armel-unknown
;;
# We use `pc' rather than `unknown'
# because (1) that's what they normally are, and
@@ -293,59 +364,83 @@ case $basic_machine in
# Recognize the basic CPU types with company name.
580-* \
| a29k-* \
| aarch64-* | aarch64_be-* \
| alpha-* | alphaev[4-8]-* | alphaev56-* | alphaev6[78]-* \
| alpha64-* | alpha64ev[4-8]-* | alpha64ev56-* | alpha64ev6[78]-* \
| alphapca5[67]-* | alpha64pca5[67]-* | arc-* \
| alphapca5[67]-* | alpha64pca5[67]-* | arc-* | arceb-* \
| arm-* | armbe-* | armle-* | armeb-* | armv*-* \
| avr-* \
| bs2000-* \
| c[123]* | c30-* | [cjt]90-* | c4x-* | c54x-* | c55x-* | c6x-* \
| clipper-* | cydra-* \
| avr-* | avr32-* \
| be32-* | be64-* \
| bfin-* | bs2000-* \
| c[123]* | c30-* | [cjt]90-* | c4x-* \
| c8051-* | clipper-* | craynv-* | cydra-* \
| d10v-* | d30v-* | dlx-* \
| elxsi-* \
| f30[01]-* | f700-* | fr30-* | frv-* | fx80-* \
| f30[01]-* | f700-* | fido-* | fr30-* | frv-* | fx80-* \
| h8300-* | h8500-* \
| hppa-* | hppa1.[01]-* | hppa2.0-* | hppa2.0[nw]-* | hppa64-* \
| hexagon-* \
| i*86-* | i860-* | i960-* | ia64-* \
| ip2k-* | iq2000-* \
| m32r-* \
| k1om-* \
| le32-* | le64-* \
| lm32-* \
| m32c-* | m32r-* | m32rle-* \
| m68000-* | m680[012346]0-* | m68360-* | m683?2-* | m68k-* \
| m88110-* | m88k-* | mcore-* \
| m88110-* | m88k-* | maxq-* | mcore-* | metag-* \
| microblaze-* | microblazeel-* \
| mips-* | mipsbe-* | mipseb-* | mipsel-* | mipsle-* \
| mips16-* \
| mips64-* | mips64el-* \
| mips64vr-* | mips64vrel-* \
| mips64octeon-* | mips64octeonel-* \
| mips64orion-* | mips64orionel-* \
| mips64r5900-* | mips64r5900el-* \
| mips64vr-* | mips64vrel-* \
| mips64vr4100-* | mips64vr4100el-* \
| mips64vr4300-* | mips64vr4300el-* \
| mips64vr5000-* | mips64vr5000el-* \
| mips64vr5900-* | mips64vr5900el-* \
| mipsisa32-* | mipsisa32el-* \
| mipsisa32r2-* | mipsisa32r2el-* \
| mipsisa64-* | mipsisa64el-* \
| mipsisa64r2-* | mipsisa64r2el-* \
| mipsisa64sb1-* | mipsisa64sb1el-* \
| mipsisa64sr71k-* | mipsisa64sr71kel-* \
| mipsr5900-* | mipsr5900el-* \
| mipstx39-* | mipstx39el-* \
| mmix-* \
| mt-* \
| msp430-* \
| none-* | np1-* | nv1-* | ns16k-* | ns32k-* \
| nds32-* | nds32le-* | nds32be-* \
| nios-* | nios2-* | nios2eb-* | nios2el-* \
| none-* | np1-* | ns16k-* | ns32k-* \
| open8-* \
| orion-* \
| pdp10-* | pdp11-* | pj-* | pjl-* | pn-* | power-* \
| powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* | ppcbe-* \
| powerpc-* | powerpc64-* | powerpc64le-* | powerpcle-* \
| pyramid-* \
| romp-* | rs6000-* \
| sh-* | sh[1234]-* | sh[23]e-* | sh[34]eb-* | shbe-* \
| rl78-* | romp-* | rs6000-* | rx-* \
| sh-* | sh[1234]-* | sh[24]a-* | sh[24]aeb-* | sh[23]e-* | sh[34]eb-* | sheb-* | shbe-* \
| shle-* | sh[1234]le-* | sh3ele-* | sh64-* | sh64le-* \
| sparc-* | sparc64-* | sparc86x-* | sparclet-* | sparclite-* \
| sparcv9-* | sparcv9b-* | strongarm-* | sv1-* | sx?-* \
| tahoe-* | thumb-* \
| sparc-* | sparc64-* | sparc64b-* | sparc64v-* | sparc86x-* | sparclet-* \
| sparclite-* \
| sparcv8-* | sparcv9-* | sparcv9b-* | sparcv9v-* | sv1-* | sx?-* \
| tahoe-* \
| tic30-* | tic4x-* | tic54x-* | tic55x-* | tic6x-* | tic80-* \
| tile*-* \
| tron-* \
| v850-* | v850e-* | vax-* \
| ubicom32-* \
| v850-* | v850e-* | v850e1-* | v850es-* | v850e2-* | v850e2v3-* \
| vax-* \
| we32k-* \
| x86-* | x86_64-* | xps100-* | xscale-* | xstormy16-* \
| xtensa-* \
| x86-* | x86_64-* | xc16x-* | xps100-* \
| xstormy16-* | xtensa*-* \
| ymp-* \
| z8k-*)
| z8k-* | z80-*)
;;
# Recognize the basic CPU types without company name, with glob match.
xtensa*)
basic_machine=$basic_machine-unknown
;;
# Recognize the various machine names and aliases which stand
# for a CPU type and a company and sometimes even an OS.
@@ -363,7 +458,7 @@ case $basic_machine in
basic_machine=a29k-amd
os=-udi
;;
abacus)
abacus)
basic_machine=abacus-unknown
;;
adobe68k)
@@ -409,6 +504,10 @@ case $basic_machine in
basic_machine=m68k-apollo
os=-bsd
;;
aros)
basic_machine=i386-pc
os=-aros
;;
aux)
basic_machine=m68k-apple
os=-aux
@@ -417,10 +516,35 @@ case $basic_machine in
basic_machine=ns32k-sequent
os=-dynix
;;
blackfin)
basic_machine=bfin-unknown
os=-linux
;;
blackfin-*)
basic_machine=bfin-`echo $basic_machine | sed 's/^[^-]*-//'`
os=-linux
;;
bluegene*)
basic_machine=powerpc-ibm
os=-cnk
;;
c54x-*)
basic_machine=tic54x-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
c55x-*)
basic_machine=tic55x-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
c6x-*)
basic_machine=tic6x-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
c90)
basic_machine=c90-cray
os=-unicos
;;
cegcc)
basic_machine=arm-unknown
os=-cegcc
;;
convex-c1)
basic_machine=c1-convex
os=-bsd
@@ -445,13 +569,20 @@ case $basic_machine in
basic_machine=j90-cray
os=-unicos
;;
cr16c)
basic_machine=cr16c-unknown
craynv)
basic_machine=craynv-cray
os=-unicosmp
;;
cr16 | cr16-*)
basic_machine=cr16-unknown
os=-elf
;;
crds | unos)
basic_machine=m68k-crds
;;
crisv32 | crisv32-* | etraxfs*)
basic_machine=crisv32-axis
;;
cris | cris-* | etrax*)
basic_machine=cris-axis
;;
@@ -481,6 +612,14 @@ case $basic_machine in
basic_machine=m88k-motorola
os=-sysv3
;;
dicos)
basic_machine=i686-pc
os=-dicos
;;
djgpp)
basic_machine=i586-pc
os=-msdosdjgpp
;;
dpx20 | dpx20-*)
basic_machine=rs6000-bull
os=-bosx
@@ -592,7 +731,6 @@ case $basic_machine in
i370-ibm* | ibm*)
basic_machine=i370-ibm
;;
# I'm not sure what "Sysv32" means. Should this be sysv3.2?
i*86v32)
basic_machine=`echo $1 | sed -e 's/86.*/86-pc/'`
os=-sysv32
@@ -631,6 +769,14 @@ case $basic_machine in
basic_machine=m68k-isi
os=-sysv
;;
m68knommu)
basic_machine=m68k-unknown
os=-linux
;;
m68knommu-*)
basic_machine=m68k-`echo $basic_machine | sed 's/^[^-]*-//'`
os=-linux
;;
m88k-omron*)
basic_machine=m88k-omron
;;
@@ -642,10 +788,21 @@ case $basic_machine in
basic_machine=ns32k-utek
os=-sysv
;;
microblaze*)
basic_machine=microblaze-xilinx
;;
mingw64)
basic_machine=x86_64-pc
os=-mingw64
;;
mingw32)
basic_machine=i386-pc
basic_machine=i686-pc
os=-mingw32
;;
mingw32ce)
basic_machine=arm-unknown
os=-mingw32ce
;;
miniframe)
basic_machine=m68000-convergent
;;
@@ -659,10 +816,6 @@ case $basic_machine in
mips3*)
basic_machine=`echo $basic_machine | sed -e 's/mips3/mips64/'`-unknown
;;
mmix*)
basic_machine=mmix-knuth
os=-mmixware
;;
monitor)
basic_machine=m68k-rom68k
os=-coff
@@ -675,10 +828,21 @@ case $basic_machine in
basic_machine=i386-pc
os=-msdos
;;
ms1-*)
basic_machine=`echo $basic_machine | sed -e 's/ms1-/mt-/'`
;;
msys)
basic_machine=i686-pc
os=-msys
;;
mvs)
basic_machine=i370-ibm
os=-mvs
;;
nacl)
basic_machine=le32-unknown
os=-nacl
;;
ncr3000)
basic_machine=i486-ncr
os=-sysv4
@@ -743,9 +907,11 @@ case $basic_machine in
np1)
basic_machine=np1-gould
;;
nv1)
basic_machine=nv1-cray
os=-unicosmp
neo-tandem)
basic_machine=neo-tandem
;;
nse-tandem)
basic_machine=nse-tandem
;;
nsr-tandem)
basic_machine=nsr-tandem
@@ -754,9 +920,8 @@ case $basic_machine in
basic_machine=hppa1.1-oki
os=-proelf
;;
or32 | or32-*)
openrisc | openrisc-*)
basic_machine=or32-unknown
os=-coff
;;
os400)
basic_machine=powerpc-ibm
@@ -778,6 +943,14 @@ case $basic_machine in
basic_machine=i860-intel
os=-osf
;;
parisc)
basic_machine=hppa-unknown
os=-linux
;;
parisc-*)
basic_machine=hppa-`echo $basic_machine | sed 's/^[^-]*-//'`
os=-linux
;;
pbd)
basic_machine=sparc-tti
;;
@@ -787,6 +960,12 @@ case $basic_machine in
pc532 | pc532-*)
basic_machine=ns32k-pc532
;;
pc98)
basic_machine=i386-pc
;;
pc98-*)
basic_machine=i386-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
pentium | p5 | k5 | k6 | nexgen | viac3)
basic_machine=i586-pc
;;
@@ -816,9 +995,10 @@ case $basic_machine in
;;
power) basic_machine=power-ibm
;;
ppc) basic_machine=powerpc-unknown
ppc | ppcbe) basic_machine=powerpc-unknown
;;
ppc-*) basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'`
ppc-* | ppcbe-*)
basic_machine=powerpc-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
ppcle | powerpclittle | ppc-le | powerpc-little)
basic_machine=powerpcle-unknown
@@ -843,6 +1023,14 @@ case $basic_machine in
basic_machine=i586-unknown
os=-pw32
;;
rdos | rdos64)
basic_machine=x86_64-pc
os=-rdos
;;
rdos32)
basic_machine=i386-pc
os=-rdos
;;
rom68k)
basic_machine=m68k-rom68k
os=-coff
@@ -869,6 +1057,10 @@ case $basic_machine in
sb1el)
basic_machine=mipsisa64sb1el-unknown
;;
sde)
basic_machine=mipsisa32-sde
os=-elf
;;
sei)
basic_machine=mips-sei
os=-seiux
@@ -880,6 +1072,9 @@ case $basic_machine in
basic_machine=sh-hitachi
os=-hms
;;
sh5el)
basic_machine=sh5le-unknown
;;
sh64)
basic_machine=sh64-unknown
;;
@@ -901,6 +1096,9 @@ case $basic_machine in
basic_machine=i860-stratus
os=-sysv4
;;
strongarm-* | thumb-*)
basic_machine=arm-`echo $basic_machine | sed 's/^[^-]*-//'`
;;
sun2)
basic_machine=m68000-sun
;;
@@ -957,17 +1155,9 @@ case $basic_machine in
basic_machine=t90-cray
os=-unicos
;;
tic54x | c54x*)
basic_machine=tic54x-unknown
os=-coff
;;
tic55x | c55x*)
basic_machine=tic55x-unknown
os=-coff
;;
tic6x | c6x*)
basic_machine=tic6x-unknown
os=-coff
tile*)
basic_machine=$basic_machine-unknown
os=-linux-gnu
;;
tx39)
basic_machine=mipstx39-unknown
@@ -1029,9 +1219,16 @@ case $basic_machine in
basic_machine=hppa1.1-winbond
os=-proelf
;;
xbox)
basic_machine=i686-pc
os=-mingw32
;;
xps | xps100)
basic_machine=xps100-honeywell
;;
xscale-* | xscalee[bl]-*)
basic_machine=`echo $basic_machine | sed 's/^xscale/arm/'`
;;
ymp)
basic_machine=ymp-cray
os=-unicos
@@ -1040,6 +1237,10 @@ case $basic_machine in
basic_machine=z8k-unknown
os=-sim
;;
z80-*-coff)
basic_machine=z80-unknown
os=-sim
;;
none)
basic_machine=none-none
os=-none
@@ -1059,6 +1260,9 @@ case $basic_machine in
romp)
basic_machine=romp-ibm
;;
mmix)
basic_machine=mmix-knuth
;;
rs6000)
basic_machine=rs6000-ibm
;;
@@ -1075,13 +1279,10 @@ case $basic_machine in
we32k)
basic_machine=we32k-att
;;
sh3 | sh4 | sh[34]eb | sh[1234]le | sh[23]ele)
sh[1234] | sh[24]a | sh[24]aeb | sh[34]eb | sh[1234]le | sh[23]ele)
basic_machine=sh-unknown
;;
sh64)
basic_machine=sh64-unknown
;;
sparc | sparcv9 | sparcv9b)
sparc | sparcv8 | sparcv9 | sparcv9b | sparcv9v)
basic_machine=sparc-sun
;;
cydra)
@@ -1125,9 +1326,12 @@ esac
if [ x"$os" != x"" ]
then
case $os in
# First match some system type aliases
# that might get confused with valid system types.
# First match some system type aliases
# that might get confused with valid system types.
# -solaris* is a basic system type, with this one exception.
-auroraux)
os=-auroraux
;;
-solaris1 | -solaris1.*)
os=`echo $os | sed -e 's|solaris1|sunos4|'`
;;
@@ -1148,26 +1352,31 @@ case $os in
# Each alternative MUST END IN A *, to match a version number.
# -sysv* is not here because it comes later, after sysvr4.
-gnu* | -bsd* | -mach* | -minix* | -genix* | -ultrix* | -irix* \
| -*vms* | -sco* | -esix* | -isc* | -aix* | -sunos | -sunos[34]*\
| -hpux* | -unos* | -osf* | -luna* | -dgux* | -solaris* | -sym* \
| -*vms* | -sco* | -esix* | -isc* | -aix* | -cnk* | -sunos | -sunos[34]*\
| -hpux* | -unos* | -osf* | -luna* | -dgux* | -auroraux* | -solaris* \
| -sym* | -kopensolaris* | -plan9* \
| -amigaos* | -amigados* | -msdos* | -newsos* | -unicos* | -aof* \
| -aos* \
| -aos* | -aros* \
| -nindy* | -vxsim* | -vxworks* | -ebmon* | -hms* | -mvs* \
| -clix* | -riscos* | -uniplus* | -iris* | -rtu* | -xenix* \
| -hiux* | -386bsd* | -knetbsd* | -mirbsd* | -netbsd* | -openbsd* \
| -hiux* | -386bsd* | -knetbsd* | -mirbsd* | -netbsd* \
| -bitrig* | -openbsd* | -solidbsd* \
| -ekkobsd* | -kfreebsd* | -freebsd* | -riscix* | -lynxos* \
| -bosx* | -nextstep* | -cxux* | -aout* | -elf* | -oabi* \
| -ptx* | -coff* | -ecoff* | -winnt* | -domain* | -vsta* \
| -udi* | -eabi* | -lites* | -ieee* | -go32* | -aux* \
| -chorusos* | -chorusrdb* \
| -cygwin* | -pe* | -psos* | -moss* | -proelf* | -rtems* \
| -mingw32* | -linux-gnu* | -linux-uclibc* | -uxpv* | -beos* | -mpeix* | -udk* \
| -chorusos* | -chorusrdb* | -cegcc* \
| -cygwin* | -msys* | -pe* | -psos* | -moss* | -proelf* | -rtems* \
| -mingw32* | -mingw64* | -linux-gnu* | -linux-android* \
| -linux-newlib* | -linux-musl* | -linux-uclibc* \
| -uxpv* | -beos* | -mpeix* | -udk* \
| -interix* | -uwin* | -mks* | -rhapsody* | -darwin* | -opened* \
| -openstep* | -oskit* | -conix* | -pw32* | -nonstopux* \
| -storm-chaos* | -tops10* | -tenex* | -tops20* | -its* \
| -os2* | -vos* | -palmos* | -uclinux* | -nucleus* \
| -morphos* | -superux* | -rtmk* | -rtmk-nova* | -windiss* \
| -powermax* | -dnix* | -nx6 | -nx7 | -sei* | -dragonfly*)
| -powermax* | -dnix* | -nx6 | -nx7 | -sei* | -dragonfly* \
| -skyos* | -haiku* | -rdos* | -toppers* | -drops* | -es*)
# Remember, each alternative MUST END IN *, to match a version number.
;;
-qnx*)
@@ -1185,7 +1394,7 @@ case $os in
os=`echo $os | sed -e 's|nto|nto-qnx|'`
;;
-sim | -es1800* | -hms* | -xray | -os68k* | -none* | -v88r* \
| -windows* | -osx | -abug | -netware* | -os9* | -beos* \
| -windows* | -osx | -abug | -netware* | -os9* | -beos* | -haiku* \
| -macos* | -mpw* | -magic* | -mmixware* | -mon960* | -lnews*)
;;
-mac*)
@@ -1206,7 +1415,7 @@ case $os in
-opened*)
os=-openedition
;;
-os400*)
-os400*)
os=-os400
;;
-wince*)
@@ -1255,7 +1464,7 @@ case $os in
-sinix*)
os=-sysv4
;;
-tpf*)
-tpf*)
os=-tpf
;;
-triton*)
@@ -1291,8 +1500,13 @@ case $os in
-aros*)
os=-aros
;;
-kaos*)
os=-kaos
-zvmoe)
os=-zvmoe
;;
-dicos*)
os=-dicos
;;
-nacl*)
;;
-none)
;;
@@ -1316,6 +1530,12 @@ else
# system, and we'll never get to this point.
case $basic_machine in
score-*)
os=-elf
;;
spu-*)
os=-elf
;;
*-acorn)
os=-riscix1.2
;;
@@ -1325,9 +1545,24 @@ case $basic_machine in
arm*-semi)
os=-aout
;;
c4x-* | tic4x-*)
os=-coff
;;
c4x-* | tic4x-*)
os=-coff
;;
c8051-*)
os=-elf
;;
hexagon-*)
os=-elf
;;
tic54x-*)
os=-coff
;;
tic55x-*)
os=-coff
;;
tic6x-*)
os=-coff
;;
# This must come before the *-dec entry.
pdp10-*)
os=-tops20
@@ -1346,19 +1581,22 @@ case $basic_machine in
;;
m68000-sun)
os=-sunos3
# This also exists in the configure program, but was not the
# default.
# os=-sunos4
;;
m68*-cisco)
os=-aout
;;
mep-*)
os=-elf
;;
mips*-cisco)
os=-elf
;;
mips*-*)
os=-elf
;;
or1k-*)
os=-elf
;;
or32-*)
os=-coff
;;
@@ -1371,9 +1609,15 @@ case $basic_machine in
*-be)
os=-beos
;;
*-haiku)
os=-haiku
;;
*-ibm)
os=-aix
;;
*-knuth)
os=-mmixware
;;
*-wec)
os=-proelf
;;
@@ -1476,7 +1720,7 @@ case $basic_machine in
-sunos*)
vendor=sun
;;
-aix*)
-cnk*|-aix*)
vendor=ibm
;;
-beos*)
@@ -1539,7 +1783,7 @@ case $basic_machine in
esac
echo $basic_machine$os
exit 0
exit
# Local variables:
# eval: (add-hook 'write-file-hooks 'time-stamp)
deps/jemalloc/configure
Vendored
+2625 -1077
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/configure.ac
Vendored
+786 -237
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/coverage.sh
Vendored Executable
+16
View File
@@ -0,0 +1,16 @@
#!/bin/sh
set -e
objdir=$1
suffix=$2
shift 2
objs=$@
gcov -b -p -f -o "${objdir}" ${objs}
# Move gcov outputs so that subsequent gcov invocations won't clobber results
# for the same sources with different compilation flags.
for f in `find . -maxdepth 1 -type f -name '*.gcov'` ; do
mv "${f}" "${f}.${suffix}"
done
deps/jemalloc/doc/jemalloc.3
Vendored
+465 -310
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/doc/jemalloc.html
Vendored
+489 -507
View File
File diff suppressed because one or more lines are too long
deps/jemalloc/doc/jemalloc.xml.in
Vendored
+657 -592
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/include/jemalloc/internal/arena.h
Vendored
+628 -308
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/include/jemalloc/internal/atomic.h
Vendored
+156 -21
View File
@@ -11,22 +11,8 @@
#define atomic_read_uint64(p) atomic_add_uint64(p, 0)
#define atomic_read_uint32(p) atomic_add_uint32(p, 0)
#if (LG_SIZEOF_PTR == 3)
# define atomic_read_z(p) \
(size_t)atomic_add_uint64((uint64_t *)p, (uint64_t)0)
# define atomic_add_z(p, x) \
(size_t)atomic_add_uint64((uint64_t *)p, (uint64_t)x)
# define atomic_sub_z(p, x) \
(size_t)atomic_sub_uint64((uint64_t *)p, (uint64_t)x)
#elif (LG_SIZEOF_PTR == 2)
# define atomic_read_z(p) \
(size_t)atomic_add_uint32((uint32_t *)p, (uint32_t)0)
# define atomic_add_z(p, x) \
(size_t)atomic_add_uint32((uint32_t *)p, (uint32_t)x)
# define atomic_sub_z(p, x) \
(size_t)atomic_sub_uint32((uint32_t *)p, (uint32_t)x)
#endif
#define atomic_read_z(p) atomic_add_z(p, 0)
#define atomic_read_u(p) atomic_add_u(p, 0)
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
@@ -37,12 +23,17 @@ 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. */
#ifdef __GCC_HAVE_SYNC_COMPARE_AND_SWAP_8
#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)
{
@@ -56,6 +47,20 @@ 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)
@@ -70,7 +75,7 @@ atomic_sub_uint64(uint64_t *p, uint64_t x)
return (OSAtomicAdd64(-((int64_t)x), (int64_t *)p));
}
#elif (defined(__amd64_) || defined(__x86_64__))
# elif (defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE uint64_t
atomic_add_uint64(uint64_t *p, uint64_t x)
{
@@ -97,8 +102,43 @@ atomic_sub_uint64(uint64_t *p, uint64_t x)
return (x);
}
#else
# if (LG_SIZEOF_PTR == 3)
# 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
@@ -119,6 +159,20 @@ 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)
@@ -133,7 +187,7 @@ 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__))
#elif (defined(__i386__) || defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE uint32_t
atomic_add_uint32(uint32_t *p, uint32_t x)
{
@@ -160,9 +214,90 @@ atomic_sub_uint32(uint32_t *p, uint32_t x)
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
+4 -2
View File
@@ -9,12 +9,14 @@
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern malloc_mutex_t base_mtx;
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 */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/chunk.h
Vendored
+7 -9
View File
@@ -28,20 +28,14 @@
#ifdef JEMALLOC_H_EXTERNS
extern size_t opt_lg_chunk;
#ifdef JEMALLOC_SWAP
extern bool opt_overcommit;
#endif
extern const char *opt_dss;
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
/* Protects stats_chunks; currently not used for any other purpose. */
extern malloc_mutex_t chunks_mtx;
/* Chunk statistics. */
extern chunk_stats_t stats_chunks;
#endif
#ifdef JEMALLOC_IVSALLOC
extern rtree_t *chunks_rtree;
#endif
extern size_t chunksize;
extern size_t chunksize_mask; /* (chunksize - 1). */
@@ -49,9 +43,14 @@ 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, bool base, bool *zero);
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 */
/******************************************************************************/
@@ -60,6 +59,5 @@ bool chunk_boot(void);
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#include "jemalloc/internal/chunk_swap.h"
#include "jemalloc/internal/chunk_dss.h"
#include "jemalloc/internal/chunk_mmap.h"
deps/jemalloc/include/jemalloc/internal/chunk_dss.h
Vendored
+18 -10
View File
@@ -1,25 +1,34 @@
#ifdef JEMALLOC_DSS
/******************************************************************************/
#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
/*
* Protects sbrk() calls. This avoids malloc races among threads, though it
* does not protect against races with threads that call sbrk() directly.
*/
extern malloc_mutex_t dss_mtx;
void *chunk_alloc_dss(size_t size, bool *zero);
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_dealloc_dss(void *chunk, size_t size);
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 */
/******************************************************************************/
@@ -27,4 +36,3 @@ bool chunk_dss_boot(void);
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_DSS */
deps/jemalloc/include/jemalloc/internal/chunk_mmap.h
Vendored
+3 -4
View File
@@ -9,11 +9,10 @@
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
void *chunk_alloc_mmap(size_t size);
void *chunk_alloc_mmap_noreserve(size_t size);
void chunk_dealloc_mmap(void *chunk, size_t size);
bool pages_purge(void *addr, size_t length);
bool chunk_mmap_boot(void);
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 */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/chunk_swap.h
Vendored
-34
View File
@@ -1,34 +0,0 @@
#ifdef JEMALLOC_SWAP
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
extern malloc_mutex_t swap_mtx;
extern bool swap_enabled;
extern bool swap_prezeroed;
extern size_t swap_nfds;
extern int *swap_fds;
#ifdef JEMALLOC_STATS
extern size_t swap_avail;
#endif
void *chunk_alloc_swap(size_t size, bool *zero);
bool chunk_in_swap(void *chunk);
bool chunk_dealloc_swap(void *chunk, size_t size);
bool chunk_swap_enable(const int *fds, unsigned nfds, bool prezeroed);
bool chunk_swap_boot(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_SWAP */
deps/jemalloc/include/jemalloc/internal/ckh.h
Vendored
+5 -12
View File
@@ -5,7 +5,7 @@ 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 void ckh_hash_t (const void *, size_t[2]);
typedef bool ckh_keycomp_t (const void *, const void *);
/* Maintain counters used to get an idea of performance. */
@@ -17,7 +17,7 @@ typedef bool ckh_keycomp_t (const void *, const void *);
* 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)
#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
@@ -30,11 +30,6 @@ struct ckhc_s {
};
struct ckh_s {
#ifdef JEMALLOC_DEBUG
#define CKH_MAGIC 0x3af2489d
uint32_t magic;
#endif
#ifdef CKH_COUNT
/* Counters used to get an idea of performance. */
uint64_t ngrows;
@@ -47,7 +42,7 @@ struct ckh_s {
/* Used for pseudo-random number generation. */
#define CKH_A 1103515241
#define CKH_C 12347
uint32_t prn_state;
uint32_t prng_state;
/* Total number of items. */
size_t count;
@@ -80,11 +75,9 @@ 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);
void ckh_string_hash(const void *key, size_t r_hash[2]);
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);
void ckh_pointer_hash(const void *key, size_t r_hash[2]);
bool ckh_pointer_keycomp(const void *k1, const void *k2);
#endif /* JEMALLOC_H_EXTERNS */
deps/jemalloc/include/jemalloc/internal/ctl.h
Vendored
+31 -32
View File
@@ -2,6 +2,8 @@
#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;
@@ -11,28 +13,29 @@ typedef struct ctl_stats_s ctl_stats_t;
struct ctl_node_s {
bool named;
union {
struct {
const char *name;
/* If (nchildren == 0), this is a terminal node. */
unsigned nchildren;
const ctl_node_t *children;
} named;
struct {
const ctl_node_t *(*index)(const size_t *, size_t,
size_t);
} indexed;
} u;
int (*ctl)(const size_t *, size_t, void *, size_t *, void *,
size_t);
};
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;
#ifdef JEMALLOC_STATS
arena_stats_t astats;
/* Aggregate stats for small size classes, based on bin stats. */
@@ -41,13 +44,11 @@ struct ctl_arena_stats_s {
uint64_t ndalloc_small;
uint64_t nrequests_small;
malloc_bin_stats_t *bstats; /* nbins elements. */
malloc_bin_stats_t bstats[NBINS];
malloc_large_stats_t *lstats; /* nlclasses elements. */
#endif
};
struct ctl_stats_s {
#ifdef JEMALLOC_STATS
size_t allocated;
size_t active;
size_t mapped;
@@ -61,11 +62,8 @@ struct ctl_stats_s {
uint64_t nmalloc; /* huge_nmalloc */
uint64_t ndalloc; /* huge_ndalloc */
} huge;
#endif
unsigned narenas;
ctl_arena_stats_t *arenas; /* (narenas + 1) elements. */
#ifdef JEMALLOC_SWAP
size_t swap_avail;
#endif
};
#endif /* JEMALLOC_H_STRUCTS */
@@ -79,29 +77,30 @@ 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 (JEMALLOC_P(mallctl)(name, oldp, oldlenp, newp, newlen) \
if (je_mallctl(name, oldp, oldlenp, newp, newlen) \
!= 0) { \
malloc_write("<jemalloc>: Failure in xmallctl(\""); \
malloc_write(name); \
malloc_write("\", ...)\n"); \
malloc_printf( \
"<jemalloc>: Failure in xmallctl(\"%s\", ...)\n", \
name); \
abort(); \
} \
} while (0)
#define xmallctlnametomib(name, mibp, miblenp) do { \
if (JEMALLOC_P(mallctlnametomib)(name, mibp, miblenp) != 0) { \
malloc_write( \
"<jemalloc>: Failure in xmallctlnametomib(\""); \
malloc_write(name); \
malloc_write("\", ...)\n"); \
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 (JEMALLOC_P(mallctlbymib)(mib, miblen, oldp, oldlenp, newp, \
if (je_mallctlbymib(mib, miblen, oldp, oldlenp, newp, \
newlen) != 0) { \
malloc_write( \
"<jemalloc>: Failure in xmallctlbymib()\n"); \
deps/jemalloc/include/jemalloc/internal/extent.h
Vendored
+3 -6
View File
@@ -9,24 +9,23 @@ typedef struct extent_node_s extent_node_t;
/* Tree of extents. */
struct extent_node_s {
#if (defined(JEMALLOC_SWAP) || defined(JEMALLOC_DSS))
/* Linkage for the size/address-ordered tree. */
rb_node(extent_node_t) link_szad;
#endif
/* Linkage for the address-ordered tree. */
rb_node(extent_node_t) link_ad;
#ifdef JEMALLOC_PROF
/* Profile counters, used for huge objects. */
prof_ctx_t *prof_ctx;
#endif
/* 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;
@@ -34,9 +33,7 @@ typedef rb_tree(extent_node_t) extent_tree_t;
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#if (defined(JEMALLOC_SWAP) || defined(JEMALLOC_DSS))
rb_proto(, extent_tree_szad_, extent_tree_t, extent_node_t)
#endif
rb_proto(, extent_tree_ad_, extent_tree_t, extent_node_t)
deps/jemalloc/include/jemalloc/internal/hash.h
Vendored
+302 -37
View File
@@ -1,3 +1,8 @@
/*
* The following hash function is based on MurmurHash3, placed into the public
* domain by Austin Appleby. See http://code.google.com/p/smhasher/ for
* details.
*/
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
@@ -14,56 +19,316 @@
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
uint64_t hash(const void *key, size_t len, uint64_t seed);
uint32_t hash_x86_32(const void *key, int len, uint32_t seed);
void hash_x86_128(const void *key, const int len, uint32_t seed,
uint64_t r_out[2]);
void hash_x64_128(const void *key, const int len, const uint32_t seed,
uint64_t r_out[2]);
void hash(const void *key, size_t len, const uint32_t seed,
size_t r_hash[2]);
#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)
/******************************************************************************/
/* Internal implementation. */
JEMALLOC_INLINE uint32_t
hash_rotl_32(uint32_t x, int8_t r)
{
const uint64_t m = 0xc6a4a7935bd1e995LLU;
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);
return (x << r) | (x >> (32 - r));
}
while(data != end) {
uint64_t k = *data++;
JEMALLOC_INLINE uint64_t
hash_rotl_64(uint64_t x, int8_t r)
{
return (x << r) | (x >> (64 - r));
}
k *= m;
k ^= k >> r;
k *= m;
JEMALLOC_INLINE uint32_t
hash_get_block_32(const uint32_t *p, int i)
{
h ^= k;
h *= m;
}
return (p[i]);
}
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;
}
JEMALLOC_INLINE uint64_t
hash_get_block_64(const uint64_t *p, int i)
{
h ^= h >> r;
h *= m;
h ^= h >> r;
return (p[i]);
}
JEMALLOC_INLINE uint32_t
hash_fmix_32(uint32_t h)
{
h ^= h >> 16;
h *= 0x85ebca6b;
h ^= h >> 13;
h *= 0xc2b2ae35;
h ^= h >> 16;
return (h);
}
JEMALLOC_INLINE uint64_t
hash_fmix_64(uint64_t k)
{
k ^= k >> 33;
k *= QU(0xff51afd7ed558ccdLLU);
k ^= k >> 33;
k *= QU(0xc4ceb9fe1a85ec53LLU);
k ^= k >> 33;
return (k);
}
JEMALLOC_INLINE uint32_t
hash_x86_32(const void *key, int len, uint32_t seed)
{
const uint8_t *data = (const uint8_t *) key;
const int nblocks = len / 4;
uint32_t h1 = seed;
const uint32_t c1 = 0xcc9e2d51;
const uint32_t c2 = 0x1b873593;
/* body */
{
const uint32_t *blocks = (const uint32_t *) (data + nblocks*4);
int i;
for (i = -nblocks; i; i++) {
uint32_t k1 = hash_get_block_32(blocks, i);
k1 *= c1;
k1 = hash_rotl_32(k1, 15);
k1 *= c2;
h1 ^= k1;
h1 = hash_rotl_32(h1, 13);
h1 = h1*5 + 0xe6546b64;
}
}
/* tail */
{
const uint8_t *tail = (const uint8_t *) (data + nblocks*4);
uint32_t k1 = 0;
switch (len & 3) {
case 3: k1 ^= tail[2] << 16;
case 2: k1 ^= tail[1] << 8;
case 1: k1 ^= tail[0]; k1 *= c1; k1 = hash_rotl_32(k1, 15);
k1 *= c2; h1 ^= k1;
}
}
/* finalization */
h1 ^= len;
h1 = hash_fmix_32(h1);
return (h1);
}
UNUSED JEMALLOC_INLINE void
hash_x86_128(const void *key, const int len, uint32_t seed,
uint64_t r_out[2])
{
const uint8_t * data = (const uint8_t *) key;
const int nblocks = len / 16;
uint32_t h1 = seed;
uint32_t h2 = seed;
uint32_t h3 = seed;
uint32_t h4 = seed;
const uint32_t c1 = 0x239b961b;
const uint32_t c2 = 0xab0e9789;
const uint32_t c3 = 0x38b34ae5;
const uint32_t c4 = 0xa1e38b93;
/* body */
{
const uint32_t *blocks = (const uint32_t *) (data + nblocks*16);
int i;
for (i = -nblocks; i; i++) {
uint32_t k1 = hash_get_block_32(blocks, i*4 + 0);
uint32_t k2 = hash_get_block_32(blocks, i*4 + 1);
uint32_t k3 = hash_get_block_32(blocks, i*4 + 2);
uint32_t k4 = hash_get_block_32(blocks, i*4 + 3);
k1 *= c1; k1 = hash_rotl_32(k1, 15); k1 *= c2; h1 ^= k1;
h1 = hash_rotl_32(h1, 19); h1 += h2;
h1 = h1*5 + 0x561ccd1b;
k2 *= c2; k2 = hash_rotl_32(k2, 16); k2 *= c3; h2 ^= k2;
h2 = hash_rotl_32(h2, 17); h2 += h3;
h2 = h2*5 + 0x0bcaa747;
k3 *= c3; k3 = hash_rotl_32(k3, 17); k3 *= c4; h3 ^= k3;
h3 = hash_rotl_32(h3, 15); h3 += h4;
h3 = h3*5 + 0x96cd1c35;
k4 *= c4; k4 = hash_rotl_32(k4, 18); k4 *= c1; h4 ^= k4;
h4 = hash_rotl_32(h4, 13); h4 += h1;
h4 = h4*5 + 0x32ac3b17;
}
}
/* tail */
{
const uint8_t *tail = (const uint8_t *) (data + nblocks*16);
uint32_t k1 = 0;
uint32_t k2 = 0;
uint32_t k3 = 0;
uint32_t k4 = 0;
switch (len & 15) {
case 15: k4 ^= tail[14] << 16;
case 14: k4 ^= tail[13] << 8;
case 13: k4 ^= tail[12] << 0;
k4 *= c4; k4 = hash_rotl_32(k4, 18); k4 *= c1; h4 ^= k4;
case 12: k3 ^= tail[11] << 24;
case 11: k3 ^= tail[10] << 16;
case 10: k3 ^= tail[ 9] << 8;
case 9: k3 ^= tail[ 8] << 0;
k3 *= c3; k3 = hash_rotl_32(k3, 17); k3 *= c4; h3 ^= k3;
case 8: k2 ^= tail[ 7] << 24;
case 7: k2 ^= tail[ 6] << 16;
case 6: k2 ^= tail[ 5] << 8;
case 5: k2 ^= tail[ 4] << 0;
k2 *= c2; k2 = hash_rotl_32(k2, 16); k2 *= c3; h2 ^= k2;
case 4: k1 ^= tail[ 3] << 24;
case 3: k1 ^= tail[ 2] << 16;
case 2: k1 ^= tail[ 1] << 8;
case 1: k1 ^= tail[ 0] << 0;
k1 *= c1; k1 = hash_rotl_32(k1, 15); k1 *= c2; h1 ^= k1;
}
}
/* finalization */
h1 ^= len; h2 ^= len; h3 ^= len; h4 ^= len;
h1 += h2; h1 += h3; h1 += h4;
h2 += h1; h3 += h1; h4 += h1;
h1 = hash_fmix_32(h1);
h2 = hash_fmix_32(h2);
h3 = hash_fmix_32(h3);
h4 = hash_fmix_32(h4);
h1 += h2; h1 += h3; h1 += h4;
h2 += h1; h3 += h1; h4 += h1;
r_out[0] = (((uint64_t) h2) << 32) | h1;
r_out[1] = (((uint64_t) h4) << 32) | h3;
}
UNUSED JEMALLOC_INLINE void
hash_x64_128(const void *key, const int len, const uint32_t seed,
uint64_t r_out[2])
{
const uint8_t *data = (const uint8_t *) key;
const int nblocks = len / 16;
uint64_t h1 = seed;
uint64_t h2 = seed;
const uint64_t c1 = QU(0x87c37b91114253d5LLU);
const uint64_t c2 = QU(0x4cf5ad432745937fLLU);
/* body */
{
const uint64_t *blocks = (const uint64_t *) (data);
int i;
for (i = 0; i < nblocks; i++) {
uint64_t k1 = hash_get_block_64(blocks, i*2 + 0);
uint64_t k2 = hash_get_block_64(blocks, i*2 + 1);
k1 *= c1; k1 = hash_rotl_64(k1, 31); k1 *= c2; h1 ^= k1;
h1 = hash_rotl_64(h1, 27); h1 += h2;
h1 = h1*5 + 0x52dce729;
k2 *= c2; k2 = hash_rotl_64(k2, 33); k2 *= c1; h2 ^= k2;
h2 = hash_rotl_64(h2, 31); h2 += h1;
h2 = h2*5 + 0x38495ab5;
}
}
/* tail */
{
const uint8_t *tail = (const uint8_t*)(data + nblocks*16);
uint64_t k1 = 0;
uint64_t k2 = 0;
switch (len & 15) {
case 15: k2 ^= ((uint64_t)(tail[14])) << 48;
case 14: k2 ^= ((uint64_t)(tail[13])) << 40;
case 13: k2 ^= ((uint64_t)(tail[12])) << 32;
case 12: k2 ^= ((uint64_t)(tail[11])) << 24;
case 11: k2 ^= ((uint64_t)(tail[10])) << 16;
case 10: k2 ^= ((uint64_t)(tail[ 9])) << 8;
case 9: k2 ^= ((uint64_t)(tail[ 8])) << 0;
k2 *= c2; k2 = hash_rotl_64(k2, 33); k2 *= c1; h2 ^= k2;
case 8: k1 ^= ((uint64_t)(tail[ 7])) << 56;
case 7: k1 ^= ((uint64_t)(tail[ 6])) << 48;
case 6: k1 ^= ((uint64_t)(tail[ 5])) << 40;
case 5: k1 ^= ((uint64_t)(tail[ 4])) << 32;
case 4: k1 ^= ((uint64_t)(tail[ 3])) << 24;
case 3: k1 ^= ((uint64_t)(tail[ 2])) << 16;
case 2: k1 ^= ((uint64_t)(tail[ 1])) << 8;
case 1: k1 ^= ((uint64_t)(tail[ 0])) << 0;
k1 *= c1; k1 = hash_rotl_64(k1, 31); k1 *= c2; h1 ^= k1;
}
}
/* finalization */
h1 ^= len; h2 ^= len;
h1 += h2;
h2 += h1;
h1 = hash_fmix_64(h1);
h2 = hash_fmix_64(h2);
h1 += h2;
h2 += h1;
r_out[0] = h1;
r_out[1] = h2;
}
/******************************************************************************/
/* API. */
JEMALLOC_INLINE void
hash(const void *key, size_t len, const uint32_t seed, size_t r_hash[2])
{
#if (LG_SIZEOF_PTR == 3 && !defined(JEMALLOC_BIG_ENDIAN))
hash_x64_128(key, len, seed, (uint64_t *)r_hash);
#else
uint64_t hashes[2];
hash_x86_128(key, len, seed, hashes);
r_hash[0] = (size_t)hashes[0];
r_hash[1] = (size_t)hashes[1];
#endif
}
#endif
#endif /* JEMALLOC_H_INLINES */
deps/jemalloc/include/jemalloc/internal/huge.h
Vendored
+13 -8
View File
@@ -9,29 +9,34 @@
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
#ifdef JEMALLOC_STATS
/* Huge allocation statistics. */
extern uint64_t huge_nmalloc;
extern uint64_t huge_ndalloc;
extern size_t huge_allocated;
#endif
/* 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,
void *huge_malloc(size_t size, bool zero, dss_prec_t dss_prec);
void *huge_palloc(size_t size, size_t alignment, bool zero,
dss_prec_t dss_prec);
bool 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);
size_t alignment, bool zero, bool try_tcache_dalloc, dss_prec_t dss_prec);
#ifdef JEMALLOC_JET
typedef void (huge_dalloc_junk_t)(void *, size_t);
extern huge_dalloc_junk_t *huge_dalloc_junk;
#endif
void huge_dalloc(void *ptr, bool unmap);
size_t huge_salloc(const void *ptr);
#ifdef JEMALLOC_PROF
dss_prec_t huge_dss_prec_get(arena_t *arena);
prof_ctx_t *huge_prof_ctx_get(const void *ptr);
void huge_prof_ctx_set(const void *ptr, prof_ctx_t *ctx);
#endif
bool huge_boot(void);
void huge_prefork(void);
void huge_postfork_parent(void);
void huge_postfork_child(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/jemalloc_internal.h.in
Vendored
+638 -398
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc/include/jemalloc/internal/jemalloc_internal_defs.h.in
Vendored
+205
View File
@@ -0,0 +1,205 @@
#ifndef JEMALLOC_INTERNAL_DEFS_H_
#define JEMALLOC_INTERNAL_DEFS_H_
/*
* 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
/*
* 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
/*
* 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 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_CODE_COVERAGE enables test code coverage analysis. */
#undef JEMALLOC_CODE_COVERAGE
/*
* 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 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
/*
* 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
/*
* Define if operating system has alloca.h header.
*/
#undef JEMALLOC_HAS_ALLOCA_H
/* C99 restrict keyword supported. */
#undef JEMALLOC_HAS_RESTRICT
/* For use by hash code. */
#undef JEMALLOC_BIG_ENDIAN
/* 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
#endif /* JEMALLOC_INTERNAL_DEFS_H_ */
deps/jemalloc/include/jemalloc/internal/jemalloc_internal_macros.h
Vendored
+51
View File
@@ -0,0 +1,51 @@
/*
* JEMALLOC_ALWAYS_INLINE and JEMALLOC_INLINE are used within header files for
* functions that are static inline functions if inlining is enabled, and
* single-definition library-private functions if inlining is disabled.
*
* JEMALLOC_ALWAYS_INLINE_C and JEMALLOC_INLINE_C are for use in .c files, in
* which case the denoted functions are always static, regardless of whether
* inlining is enabled.
*/
#if defined(JEMALLOC_DEBUG) || defined(JEMALLOC_CODE_COVERAGE)
/* Disable inlining to make debugging/profiling easier. */
# define JEMALLOC_ALWAYS_INLINE
# define JEMALLOC_ALWAYS_INLINE_C static
# define JEMALLOC_INLINE
# define JEMALLOC_INLINE_C static
# define inline
#else
# define JEMALLOC_ENABLE_INLINE
# ifdef JEMALLOC_HAVE_ATTR
# define JEMALLOC_ALWAYS_INLINE \
static inline JEMALLOC_ATTR(unused) JEMALLOC_ATTR(always_inline)
# define JEMALLOC_ALWAYS_INLINE_C \
static inline JEMALLOC_ATTR(always_inline)
# else
# define JEMALLOC_ALWAYS_INLINE static inline
# define JEMALLOC_ALWAYS_INLINE_C static inline
# endif
# define JEMALLOC_INLINE static inline
# define JEMALLOC_INLINE_C static inline
# ifdef _MSC_VER
# define inline _inline
# endif
#endif
#ifdef JEMALLOC_CC_SILENCE
# define UNUSED JEMALLOC_ATTR(unused)
#else
# define UNUSED
#endif
#define ZU(z) ((size_t)z)
#define QU(q) ((uint64_t)q)
#define QI(q) ((int64_t)q)
#ifndef __DECONST
# define __DECONST(type, var) ((type)(uintptr_t)(const void *)(var))
#endif
#ifndef JEMALLOC_HAS_RESTRICT
# define restrict
#endif
deps/jemalloc/include/jemalloc/internal/mb.h
Vendored
+8 -1
View File
@@ -54,7 +54,7 @@ mb_write(void)
);
#endif
}
#elif (defined(__amd64_) || defined(__x86_64__))
#elif (defined(__amd64__) || defined(__x86_64__))
JEMALLOC_INLINE void
mb_write(void)
{
@@ -87,6 +87,13 @@ mb_write(void)
: "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
deps/jemalloc/include/jemalloc/internal/mutex.h
Vendored
+43 -30
View File
@@ -1,22 +1,42 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#ifdef JEMALLOC_OSSPIN
typedef OSSpinLock malloc_mutex_t;
#else
typedef pthread_mutex_t malloc_mutex_t;
#endif
typedef struct malloc_mutex_s malloc_mutex_t;
#ifdef PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP
# define MALLOC_MUTEX_INITIALIZER PTHREAD_ADAPTIVE_MUTEX_INITIALIZER_NP
#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
# define MALLOC_MUTEX_INITIALIZER PTHREAD_MUTEX_INITIALIZER
# 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
@@ -24,11 +44,15 @@ typedef pthread_mutex_t malloc_mutex_t;
#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_destroy(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 */
/******************************************************************************/
@@ -36,7 +60,6 @@ void malloc_mutex_destroy(malloc_mutex_t *mutex);
#ifndef JEMALLOC_ENABLE_INLINE
void malloc_mutex_lock(malloc_mutex_t *mutex);
bool malloc_mutex_trylock(malloc_mutex_t *mutex);
void malloc_mutex_unlock(malloc_mutex_t *mutex);
#endif
@@ -46,37 +69,27 @@ malloc_mutex_lock(malloc_mutex_t *mutex)
{
if (isthreaded) {
#ifdef JEMALLOC_OSSPIN
OSSpinLockLock(mutex);
#ifdef _WIN32
EnterCriticalSection(&mutex->lock);
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLockLock(&mutex->lock);
#else
pthread_mutex_lock(mutex);
pthread_mutex_lock(&mutex->lock);
#endif
}
}
JEMALLOC_INLINE bool
malloc_mutex_trylock(malloc_mutex_t *mutex)
{
if (isthreaded) {
#ifdef JEMALLOC_OSSPIN
return (OSSpinLockTry(mutex) == false);
#else
return (pthread_mutex_trylock(mutex) != 0);
#endif
} else
return (false);
}
JEMALLOC_INLINE void
malloc_mutex_unlock(malloc_mutex_t *mutex)
{
if (isthreaded) {
#ifdef JEMALLOC_OSSPIN
OSSpinLockUnlock(mutex);
#ifdef _WIN32
LeaveCriticalSection(&mutex->lock);
#elif (defined(JEMALLOC_OSSPIN))
OSSpinLockUnlock(&mutex->lock);
#else
pthread_mutex_unlock(mutex);
pthread_mutex_unlock(&mutex->lock);
#endif
}
}
deps/jemalloc/include/jemalloc/internal/private_namespace.h
Vendored
-195
View File
@@ -1,195 +0,0 @@
#define arena_bin_index JEMALLOC_N(arena_bin_index)
#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_large JEMALLOC_N(arena_dalloc_large)
#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_new JEMALLOC_N(arena_new)
#define arena_palloc JEMALLOC_N(arena_palloc)
#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_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_salloc_demote JEMALLOC_N(arena_salloc_demote)
#define arena_stats_merge JEMALLOC_N(arena_stats_merge)
#define arena_tcache_fill_small JEMALLOC_N(arena_tcache_fill_small)
#define arenas_bin_i_index JEMALLOC_N(arenas_bin_i_index)
#define arenas_extend JEMALLOC_N(arenas_extend)
#define arenas_lrun_i_index JEMALLOC_N(arenas_lrun_i_index)
#define atomic_add_uint32 JEMALLOC_N(atomic_add_uint32)
#define atomic_add_uint64 JEMALLOC_N(atomic_add_uint64)
#define atomic_sub_uint32 JEMALLOC_N(atomic_sub_uint32)
#define atomic_sub_uint64 JEMALLOC_N(atomic_sub_uint64)
#define base_alloc JEMALLOC_N(base_alloc)
#define base_boot JEMALLOC_N(base_boot)
#define base_node_alloc JEMALLOC_N(base_node_alloc)
#define base_node_dealloc JEMALLOC_N(base_node_dealloc)
#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_alloc_mmap_noreserve JEMALLOC_N(chunk_alloc_mmap_noreserve)
#define chunk_alloc_swap JEMALLOC_N(chunk_alloc_swap)
#define chunk_boot JEMALLOC_N(chunk_boot)
#define chunk_dealloc JEMALLOC_N(chunk_dealloc)
#define chunk_dealloc_dss JEMALLOC_N(chunk_dealloc_dss)
#define chunk_dealloc_mmap JEMALLOC_N(chunk_dealloc_mmap)
#define chunk_dealloc_swap JEMALLOC_N(chunk_dealloc_swap)
#define chunk_dss_boot JEMALLOC_N(chunk_dss_boot)
#define chunk_in_dss JEMALLOC_N(chunk_in_dss)
#define chunk_in_swap JEMALLOC_N(chunk_in_swap)
#define chunk_mmap_boot JEMALLOC_N(chunk_mmap_boot)
#define chunk_swap_boot JEMALLOC_N(chunk_swap_boot)
#define chunk_swap_enable JEMALLOC_N(chunk_swap_enable)
#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 create_zone JEMALLOC_N(create_zone)
#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 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 hash JEMALLOC_N(hash)
#define huge_boot JEMALLOC_N(huge_boot)
#define huge_dalloc JEMALLOC_N(huge_dalloc)
#define huge_malloc JEMALLOC_N(huge_malloc)
#define huge_palloc JEMALLOC_N(huge_palloc)
#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 idalloc JEMALLOC_N(idalloc)
#define imalloc JEMALLOC_N(imalloc)
#define ipalloc JEMALLOC_N(ipalloc)
#define iralloc JEMALLOC_N(iralloc)
#define isalloc JEMALLOC_N(isalloc)
#define ivsalloc JEMALLOC_N(ivsalloc)
#define jemalloc_darwin_init JEMALLOC_N(jemalloc_darwin_init)
#define jemalloc_postfork JEMALLOC_N(jemalloc_postfork)
#define jemalloc_prefork JEMALLOC_N(jemalloc_prefork)
#define malloc_cprintf JEMALLOC_N(malloc_cprintf)
#define malloc_mutex_destroy JEMALLOC_N(malloc_mutex_destroy)
#define malloc_mutex_init JEMALLOC_N(malloc_mutex_init)
#define malloc_mutex_lock JEMALLOC_N(malloc_mutex_lock)
#define malloc_mutex_trylock JEMALLOC_N(malloc_mutex_trylock)
#define malloc_mutex_unlock JEMALLOC_N(malloc_mutex_unlock)
#define malloc_printf JEMALLOC_N(malloc_printf)
#define malloc_write JEMALLOC_N(malloc_write)
#define mb_write JEMALLOC_N(mb_write)
#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_lookup JEMALLOC_N(prof_lookup)
#define prof_malloc JEMALLOC_N(prof_malloc)
#define prof_mdump JEMALLOC_N(prof_mdump)
#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_init JEMALLOC_N(prof_tdata_init)
#define pthread_create JEMALLOC_N(pthread_create)
#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_set JEMALLOC_N(rtree_set)
#define s2u JEMALLOC_N(s2u)
#define sa2u JEMALLOC_N(sa2u)
#define stats_arenas_i_bins_j_index JEMALLOC_N(stats_arenas_i_bins_j_index)
#define stats_arenas_i_index JEMALLOC_N(stats_arenas_i_index)
#define stats_arenas_i_lruns_j_index JEMALLOC_N(stats_arenas_i_lruns_j_index)
#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_print JEMALLOC_N(stats_print)
#define szone2ozone JEMALLOC_N(szone2ozone)
#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_bin_flush_large JEMALLOC_N(tcache_bin_flush_large)
#define tcache_bin_flush_small JEMALLOC_N(tcache_bin_flush_small)
#define tcache_boot JEMALLOC_N(tcache_boot)
#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_event JEMALLOC_N(tcache_event)
#define tcache_get JEMALLOC_N(tcache_get)
#define tcache_stats_merge JEMALLOC_N(tcache_stats_merge)
#define thread_allocated_get JEMALLOC_N(thread_allocated_get)
#define thread_allocated_get_hard JEMALLOC_N(thread_allocated_get_hard)
#define u2s JEMALLOC_N(u2s)
deps/jemalloc/include/jemalloc/internal/private_namespace.sh
Vendored Executable
+5
View File
@@ -0,0 +1,5 @@
#!/bin/sh
for symbol in `cat $1` ; do
echo "#define ${symbol} JEMALLOC_N(${symbol})"
done
deps/jemalloc/include/jemalloc/internal/private_symbols.txt
Vendored
+413
View File
@@ -0,0 +1,413 @@
a0calloc
a0free
a0malloc
arena_alloc_junk_small
arena_bin_index
arena_bin_info
arena_boot
arena_dalloc
arena_dalloc_bin
arena_dalloc_bin_locked
arena_dalloc_junk_large
arena_dalloc_junk_small
arena_dalloc_large
arena_dalloc_large_locked
arena_dalloc_small
arena_dss_prec_get
arena_dss_prec_set
arena_malloc
arena_malloc_large
arena_malloc_small
arena_mapbits_allocated_get
arena_mapbits_binind_get
arena_mapbits_dirty_get
arena_mapbits_get
arena_mapbits_large_binind_set
arena_mapbits_large_get
arena_mapbits_large_set
arena_mapbits_large_size_get
arena_mapbits_small_runind_get
arena_mapbits_small_set
arena_mapbits_unallocated_set
arena_mapbits_unallocated_size_get
arena_mapbits_unallocated_size_set
arena_mapbits_unzeroed_get
arena_mapbits_unzeroed_set
arena_mapbitsp_get
arena_mapbitsp_read
arena_mapbitsp_write
arena_mapp_get
arena_maxclass
arena_new
arena_palloc
arena_postfork_child
arena_postfork_parent
arena_prefork
arena_prof_accum
arena_prof_accum_impl
arena_prof_accum_locked
arena_prof_ctx_get
arena_prof_ctx_set
arena_prof_promoted
arena_ptr_small_binind_get
arena_purge_all
arena_quarantine_junk_small
arena_ralloc
arena_ralloc_junk_large
arena_ralloc_no_move
arena_redzone_corruption
arena_run_regind
arena_salloc
arena_stats_merge
arena_tcache_fill_small
arenas
arenas_booted
arenas_cleanup
arenas_extend
arenas_initialized
arenas_lock
arenas_tls
arenas_tsd
arenas_tsd_boot
arenas_tsd_cleanup_wrapper
arenas_tsd_get
arenas_tsd_get_wrapper
arenas_tsd_init_head
arenas_tsd_set
atomic_add_u
atomic_add_uint32
atomic_add_uint64
atomic_add_z
atomic_sub_u
atomic_sub_uint32
atomic_sub_uint64
atomic_sub_z
base_alloc
base_boot
base_calloc
base_node_alloc
base_node_dealloc
base_postfork_child
base_postfork_parent
base_prefork
bitmap_full
bitmap_get
bitmap_info_init
bitmap_info_ngroups
bitmap_init
bitmap_set
bitmap_sfu
bitmap_size
bitmap_unset
bt_init
buferror
choose_arena
choose_arena_hard
chunk_alloc
chunk_alloc_dss
chunk_alloc_mmap
chunk_boot
chunk_dealloc
chunk_dealloc_mmap
chunk_dss_boot
chunk_dss_postfork_child
chunk_dss_postfork_parent
chunk_dss_prec_get
chunk_dss_prec_set
chunk_dss_prefork
chunk_in_dss
chunk_npages
chunk_postfork_child
chunk_postfork_parent
chunk_prefork
chunk_unmap
chunks_mtx
chunks_rtree
chunksize
chunksize_mask
ckh_bucket_search
ckh_count
ckh_delete
ckh_evict_reloc_insert
ckh_insert
ckh_isearch
ckh_iter
ckh_new
ckh_pointer_hash
ckh_pointer_keycomp
ckh_rebuild
ckh_remove
ckh_search
ckh_string_hash
ckh_string_keycomp
ckh_try_bucket_insert
ckh_try_insert
ctl_boot
ctl_bymib
ctl_byname
ctl_nametomib
ctl_postfork_child
ctl_postfork_parent
ctl_prefork
dss_prec_names
extent_tree_ad_first
extent_tree_ad_insert
extent_tree_ad_iter
extent_tree_ad_iter_recurse
extent_tree_ad_iter_start
extent_tree_ad_last
extent_tree_ad_new
extent_tree_ad_next
extent_tree_ad_nsearch
extent_tree_ad_prev
extent_tree_ad_psearch
extent_tree_ad_remove
extent_tree_ad_reverse_iter
extent_tree_ad_reverse_iter_recurse
extent_tree_ad_reverse_iter_start
extent_tree_ad_search
extent_tree_szad_first
extent_tree_szad_insert
extent_tree_szad_iter
extent_tree_szad_iter_recurse
extent_tree_szad_iter_start
extent_tree_szad_last
extent_tree_szad_new
extent_tree_szad_next
extent_tree_szad_nsearch
extent_tree_szad_prev
extent_tree_szad_psearch
extent_tree_szad_remove
extent_tree_szad_reverse_iter
extent_tree_szad_reverse_iter_recurse
extent_tree_szad_reverse_iter_start
extent_tree_szad_search
get_errno
hash
hash_fmix_32
hash_fmix_64
hash_get_block_32
hash_get_block_64
hash_rotl_32
hash_rotl_64
hash_x64_128
hash_x86_128
hash_x86_32
huge_allocated
huge_boot
huge_dalloc
huge_dalloc_junk
huge_dss_prec_get
huge_malloc
huge_mtx
huge_ndalloc
huge_nmalloc
huge_palloc
huge_postfork_child
huge_postfork_parent
huge_prefork
huge_prof_ctx_get
huge_prof_ctx_set
huge_ralloc
huge_ralloc_no_move
huge_salloc
iallocm
icalloc
icalloct
idalloc
idalloct
imalloc
imalloct
ipalloc
ipalloct
iqalloc
iqalloct
iralloc
iralloct
iralloct_realign
isalloc
isthreaded
ivsalloc
ixalloc
jemalloc_postfork_child
jemalloc_postfork_parent
jemalloc_prefork
malloc_cprintf
malloc_mutex_init
malloc_mutex_lock
malloc_mutex_postfork_child
malloc_mutex_postfork_parent
malloc_mutex_prefork
malloc_mutex_unlock
malloc_printf
malloc_snprintf
malloc_strtoumax
malloc_tsd_boot
malloc_tsd_cleanup_register
malloc_tsd_dalloc
malloc_tsd_malloc
malloc_tsd_no_cleanup
malloc_vcprintf
malloc_vsnprintf
malloc_write
map_bias
mb_write
mutex_boot
narenas_auto
narenas_total
narenas_total_get
ncpus
nhbins
opt_abort
opt_dss
opt_junk
opt_lg_chunk
opt_lg_dirty_mult
opt_lg_prof_interval
opt_lg_prof_sample
opt_lg_tcache_max
opt_narenas
opt_prof
opt_prof_accum
opt_prof_active
opt_prof_final
opt_prof_gdump
opt_prof_leak
opt_prof_prefix
opt_quarantine
opt_redzone
opt_stats_print
opt_tcache
opt_utrace
opt_valgrind
opt_xmalloc
opt_zero
p2rz
pages_purge
pow2_ceil
prof_backtrace
prof_boot0
prof_boot1
prof_boot2
prof_bt_count
prof_ctx_get
prof_ctx_set
prof_dump_open
prof_free
prof_gdump
prof_idump
prof_interval
prof_lookup
prof_malloc
prof_mdump
prof_postfork_child
prof_postfork_parent
prof_prefork
prof_promote
prof_realloc
prof_sample_accum_update
prof_sample_threshold_update
prof_tdata_booted
prof_tdata_cleanup
prof_tdata_get
prof_tdata_init
prof_tdata_initialized
prof_tdata_tls
prof_tdata_tsd
prof_tdata_tsd_boot
prof_tdata_tsd_cleanup_wrapper
prof_tdata_tsd_get
prof_tdata_tsd_get_wrapper
prof_tdata_tsd_init_head
prof_tdata_tsd_set
quarantine
quarantine_alloc_hook
quarantine_boot
quarantine_booted
quarantine_cleanup
quarantine_init
quarantine_tls
quarantine_tsd
quarantine_tsd_boot
quarantine_tsd_cleanup_wrapper
quarantine_tsd_get
quarantine_tsd_get_wrapper
quarantine_tsd_init_head
quarantine_tsd_set
register_zone
rtree_delete
rtree_get
rtree_get_locked
rtree_new
rtree_postfork_child
rtree_postfork_parent
rtree_prefork
rtree_set
s2u
sa2u
set_errno
small_size2bin
stats_cactive
stats_cactive_add
stats_cactive_get
stats_cactive_sub
stats_chunks
stats_print
tcache_alloc_easy
tcache_alloc_large
tcache_alloc_small
tcache_alloc_small_hard
tcache_arena_associate
tcache_arena_dissociate
tcache_bin_flush_large
tcache_bin_flush_small
tcache_bin_info
tcache_boot0
tcache_boot1
tcache_booted
tcache_create
tcache_dalloc_large
tcache_dalloc_small
tcache_destroy
tcache_enabled_booted
tcache_enabled_get
tcache_enabled_initialized
tcache_enabled_set
tcache_enabled_tls
tcache_enabled_tsd
tcache_enabled_tsd_boot
tcache_enabled_tsd_cleanup_wrapper
tcache_enabled_tsd_get
tcache_enabled_tsd_get_wrapper
tcache_enabled_tsd_init_head
tcache_enabled_tsd_set
tcache_event
tcache_event_hard
tcache_flush
tcache_get
tcache_initialized
tcache_maxclass
tcache_salloc
tcache_stats_merge
tcache_thread_cleanup
tcache_tls
tcache_tsd
tcache_tsd_boot
tcache_tsd_cleanup_wrapper
tcache_tsd_get
tcache_tsd_get_wrapper
tcache_tsd_init_head
tcache_tsd_set
thread_allocated_booted
thread_allocated_initialized
thread_allocated_tls
thread_allocated_tsd
thread_allocated_tsd_boot
thread_allocated_tsd_cleanup_wrapper
thread_allocated_tsd_get
thread_allocated_tsd_get_wrapper
thread_allocated_tsd_init_head
thread_allocated_tsd_set
tsd_init_check_recursion
tsd_init_finish
u2rz
deps/jemalloc/include/jemalloc/internal/private_unnamespace.sh
Vendored Executable
+5
View File
@@ -0,0 +1,5 @@
#!/bin/sh
for symbol in `cat $1` ; do
echo "#undef ${symbol}"
done
deps/jemalloc/include/jemalloc/internal/prn.h → deps/jemalloc/include/jemalloc/internal/prng.h
Vendored
+4 -4
View File
@@ -4,7 +4,7 @@
/*
* Simple linear congruential pseudo-random number generator:
*
* prn(y) = (a*x + c) % m
* prng(y) = (a*x + c) % m
*
* where the following constants ensure maximal period:
*
@@ -25,7 +25,7 @@
* uint32_t state : Seed value.
* const uint32_t a, c : See above discussion.
*/
#define prn32(r, lg_range, state, a, c) do { \
#define prng32(r, lg_range, state, a, c) do { \
assert(lg_range > 0); \
assert(lg_range <= 32); \
\
@@ -34,8 +34,8 @@
r >>= (32 - lg_range); \
} while (false)
/* Same as prn32(), but 64 bits of pseudo-randomness, using uint64_t. */
#define prn64(r, lg_range, state, a, c) do { \
/* 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); \
\
deps/jemalloc/include/jemalloc/internal/prof.h
Vendored
+167 -101
View File
@@ -1,4 +1,3 @@
#ifdef JEMALLOC_PROF
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
@@ -9,29 +8,46 @@ 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_BT_MAX_DEFAULT 7
#define LG_PROF_SAMPLE_DEFAULT 0
#ifdef JEMALLOC_PROF
# define PROF_PREFIX_DEFAULT "jeprof"
#else
# define PROF_PREFIX_DEFAULT ""
#endif
#define LG_PROF_SAMPLE_DEFAULT 19
#define LG_PROF_INTERVAL_DEFAULT -1
#define LG_PROF_TCMAX_DEFAULT -1
/*
* Hard limit on stack backtrace depth. Note that the version of
* prof_backtrace() that is based on __builtin_return_address() necessarily has
* a hard-coded number of backtrace frame handlers.
* 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.
*/
#if (defined(JEMALLOC_PROF_LIBGCC) || defined(JEMALLOC_PROF_LIBUNWIND))
# define LG_PROF_BT_MAX ((ZU(1) << (LG_SIZEOF_PTR+3)) - 1)
#else
# define LG_PROF_BT_MAX 7 /* >= LG_PROF_BT_MAX_DEFAULT */
#endif
#define PROF_BT_MAX (1U << LG_PROF_BT_MAX)
#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
#define PROF_CKH_MINITEMS 64
/* Size of memory buffer to use when writing dump files. */
#define PROF_DUMP_BUF_SIZE 65536
#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 */
/******************************************************************************/
@@ -109,8 +125,19 @@ struct prof_ctx_s {
/* Associated backtrace. */
prof_bt_t *bt;
/* Protects cnt_merged and cnts_ql. */
malloc_mutex_t lock;
/* 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.
* - Dumping a heap profile that includes 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;
@@ -123,7 +150,11 @@ struct prof_ctx_s {
* this context.
*/
ql_head(prof_thr_cnt_t) cnts_ql;
/* Linkage for list of contexts to be dumped. */
ql_elm(prof_ctx_t) dump_link;
};
typedef ql_head(prof_ctx_t) prof_ctx_list_t;
struct prof_tdata_s {
/*
@@ -145,9 +176,14 @@ struct prof_tdata_s {
void **vec;
/* Sampling state. */
uint64_t prn_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 */
@@ -162,14 +198,18 @@ extern bool opt_prof;
* to notice state changes.
*/
extern bool opt_prof_active;
extern size_t opt_lg_prof_bt_max; /* Maximum backtrace depth. */
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 ssize_t opt_lg_prof_tcmax; /* lg(max per thread bactrace cache) */
extern char opt_prof_prefix[PATH_MAX + 1];
extern char opt_prof_prefix[
/* Minimize memory bloat for non-prof builds. */
#ifdef JEMALLOC_PROF
PATH_MAX +
#endif
1];
/*
* Profile dump interval, measured in bytes allocated. Each arena triggers a
@@ -186,42 +226,25 @@ extern uint64_t prof_interval;
*/
extern bool prof_promote;
/* (1U << opt_lg_prof_bt_max). */
extern unsigned prof_bt_max;
/* Thread-specific backtrace cache, used to reduce bt2ctx contention. */
#ifndef NO_TLS
extern __thread prof_tdata_t *prof_tdata_tls
JEMALLOC_ATTR(tls_model("initial-exec"));
# define PROF_TCACHE_GET() prof_tdata_tls
# define PROF_TCACHE_SET(v) do { \
prof_tdata_tls = (v); \
pthread_setspecific(prof_tdata_tsd, (void *)(v)); \
} while (0)
#else
# define PROF_TCACHE_GET() \
((prof_tdata_t *)pthread_getspecific(prof_tdata_tsd))
# define PROF_TCACHE_SET(v) do { \
pthread_setspecific(prof_tdata_tsd, (void *)(v)); \
} while (0)
#endif
/*
* Same contents as b2cnt_tls, but initialized such that the TSD destructor is
* called when a thread exits, so that prof_tdata_tls contents can be merged,
* unlinked, and deallocated.
*/
extern pthread_key_t prof_tdata_tsd;
void bt_init(prof_bt_t *bt, void **vec);
void prof_backtrace(prof_bt_t *bt, unsigned nignore, unsigned max);
void prof_backtrace(prof_bt_t *bt, unsigned nignore);
prof_thr_cnt_t *prof_lookup(prof_bt_t *bt);
#ifdef JEMALLOC_JET
size_t prof_bt_count(void);
typedef int (prof_dump_open_t)(bool, const char *);
extern prof_dump_open_t *prof_dump_open;
#endif
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 */
/******************************************************************************/
@@ -233,13 +256,13 @@ bool prof_boot2(void);
\
assert(size == s2u(size)); \
\
prof_tdata = PROF_TCACHE_GET(); \
if (prof_tdata == NULL) { \
prof_tdata = prof_tdata_init(); \
if (prof_tdata == NULL) { \
prof_tdata = prof_tdata_get(true); \
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; \
} \
break; \
} \
\
if (opt_prof_active == false) { \
@@ -249,13 +272,13 @@ bool prof_boot2(void);
/* Don't bother with sampling logic, since sampling */\
/* interval is 1. */\
bt_init(&bt, prof_tdata->vec); \
prof_backtrace(&bt, nignore, prof_bt_max); \
prof_backtrace(&bt, nignore); \
ret = prof_lookup(&bt); \
} else { \
if (prof_tdata->threshold == 0) { \
/* Initialize. Seed the prng differently for */\
/* each thread. */\
prof_tdata->prn_state = \
prof_tdata->prng_state = \
(uint64_t)(uintptr_t)&size; \
prof_sample_threshold_update(prof_tdata); \
} \
@@ -272,7 +295,7 @@ bool prof_boot2(void);
if (size >= prof_tdata->threshold - \
prof_tdata->accum) { \
bt_init(&bt, prof_tdata->vec); \
prof_backtrace(&bt, nignore, prof_bt_max); \
prof_backtrace(&bt, nignore); \
ret = prof_lookup(&bt); \
} else \
ret = (prof_thr_cnt_t *)(uintptr_t)1U; \
@@ -280,23 +303,61 @@ bool prof_boot2(void);
} while (0)
#ifndef JEMALLOC_ENABLE_INLINE
malloc_tsd_protos(JEMALLOC_ATTR(unused), prof_tdata, prof_tdata_t *)
prof_tdata_t *prof_tdata_get(bool create);
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);
void prof_ctx_set(const void *ptr, size_t usize, 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_malloc(const void *ptr, size_t usize, prof_thr_cnt_t *cnt);
void prof_realloc(const void *ptr, size_t usize, prof_thr_cnt_t *cnt,
size_t old_usize, 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(bool create)
{
prof_tdata_t *prof_tdata;
cassert(config_prof);
prof_tdata = *prof_tdata_tsd_get();
if (create && prof_tdata == NULL)
prof_tdata = prof_tdata_init();
return (prof_tdata);
}
JEMALLOC_INLINE void
prof_sample_threshold_update(prof_tdata_t *prof_tdata)
{
/*
* The body of this function is compiled out unless heap profiling is
* enabled, so that it is possible to compile jemalloc with floating
* point support completely disabled. Avoiding floating point code is
* important on memory-constrained systems, but it also enables a
* workaround for versions of glibc that don't properly save/restore
* floating point registers during dynamic lazy symbol loading (which
* internally calls into whatever malloc implementation happens to be
* integrated into the application). Note that some compilers (e.g.
* gcc 4.8) may use floating point registers for fast memory moves, so
* jemalloc must be compiled with such optimizations disabled (e.g.
* -mno-sse) in order for the workaround to be complete.
*/
#ifdef JEMALLOC_PROF
uint64_t r;
double u;
cassert(config_prof);
/*
* Compute sample threshold as a geometrically distributed random
* variable with mean (2^opt_lg_prof_sample).
@@ -313,14 +374,15 @@ prof_sample_threshold_update(prof_tdata_t *prof_tdata)
* Luc Devroye
* Springer-Verlag, New York, 1986
* pp 500
* (http://cg.scs.carleton.ca/~luc/rnbookindex.html)
* (http://luc.devroye.org/rnbookindex.html)
*/
prn64(r, 53, prof_tdata->prn_state,
(uint64_t)6364136223846793005LLU, (uint64_t)1442695040888963407LLU);
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;
#endif
}
JEMALLOC_INLINE prof_ctx_t *
@@ -329,13 +391,12 @@ 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. */
dassert(chunk->arena->magic == ARENA_MAGIC);
ret = arena_prof_ctx_get(ptr);
} else
ret = huge_prof_ctx_get(ptr);
@@ -344,18 +405,17 @@ prof_ctx_get(const void *ptr)
}
JEMALLOC_INLINE void
prof_ctx_set(const void *ptr, prof_ctx_t *ctx)
prof_ctx_set(const void *ptr, size_t usize, 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. */
dassert(chunk->arena->magic == ARENA_MAGIC);
arena_prof_ctx_set(ptr, ctx);
arena_prof_ctx_set(ptr, usize, ctx);
} else
huge_prof_ctx_set(ptr, ctx);
}
@@ -365,11 +425,13 @@ 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_TCACHE_GET();
assert(prof_tdata != NULL);
prof_tdata = prof_tdata_get(false);
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) {
@@ -388,19 +450,20 @@ prof_sample_accum_update(size_t size)
}
JEMALLOC_INLINE void
prof_malloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt)
prof_malloc(const void *ptr, size_t usize, prof_thr_cnt_t *cnt)
{
cassert(config_prof);
assert(ptr != NULL);
assert(size == isalloc(ptr));
assert(usize == isalloc(ptr, true));
if (opt_lg_prof_sample != 0) {
if (prof_sample_accum_update(size)) {
if (prof_sample_accum_update(usize)) {
/*
* 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
* be a difference between the usize passed to
* PROF_ALLOC_PREP() and prof_malloc().
*/
assert((uintptr_t)cnt == (uintptr_t)1U);
@@ -408,17 +471,17 @@ prof_malloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt)
}
if ((uintptr_t)cnt > (uintptr_t)1U) {
prof_ctx_set(ptr, cnt->ctx);
prof_ctx_set(ptr, usize, cnt->ctx);
cnt->epoch++;
/*********/
mb_write();
/*********/
cnt->cnts.curobjs++;
cnt->cnts.curbytes += size;
cnt->cnts.curbytes += usize;
if (opt_prof_accum) {
cnt->cnts.accumobjs++;
cnt->cnts.accumbytes += size;
cnt->cnts.accumbytes += usize;
}
/*********/
mb_write();
@@ -428,27 +491,28 @@ prof_malloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt)
mb_write();
/*********/
} else
prof_ctx_set(ptr, (prof_ctx_t *)(uintptr_t)1U);
prof_ctx_set(ptr, usize, (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_realloc(const void *ptr, size_t usize, prof_thr_cnt_t *cnt,
size_t old_usize, 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));
assert(usize == isalloc(ptr, true));
if (opt_lg_prof_sample != 0) {
if (prof_sample_accum_update(size)) {
if (prof_sample_accum_update(usize)) {
/*
* Don't sample. The size passed to
* Don't sample. The usize 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
* its actual usize was insufficient to cross
* the sample threshold.
*/
cnt = (prof_thr_cnt_t *)(uintptr_t)1U;
@@ -463,10 +527,10 @@ prof_realloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt,
* 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);
malloc_mutex_lock(old_ctx->lock);
old_ctx->cnt_merged.curobjs--;
old_ctx->cnt_merged.curbytes -= old_size;
malloc_mutex_unlock(&old_ctx->lock);
old_ctx->cnt_merged.curbytes -= old_usize;
malloc_mutex_unlock(old_ctx->lock);
told_cnt = (prof_thr_cnt_t *)(uintptr_t)1U;
}
} else
@@ -475,23 +539,23 @@ prof_realloc(const void *ptr, size_t size, prof_thr_cnt_t *cnt,
if ((uintptr_t)told_cnt > (uintptr_t)1U)
told_cnt->epoch++;
if ((uintptr_t)cnt > (uintptr_t)1U) {
prof_ctx_set(ptr, cnt->ctx);
prof_ctx_set(ptr, usize, cnt->ctx);
cnt->epoch++;
} else
prof_ctx_set(ptr, (prof_ctx_t *)(uintptr_t)1U);
} else if (ptr != NULL)
prof_ctx_set(ptr, usize, (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;
told_cnt->cnts.curbytes -= old_usize;
}
if ((uintptr_t)cnt > (uintptr_t)1U) {
cnt->cnts.curobjs++;
cnt->cnts.curbytes += size;
cnt->cnts.curbytes += usize;
if (opt_prof_accum) {
cnt->cnts.accumobjs++;
cnt->cnts.accumbytes += size;
cnt->cnts.accumbytes += usize;
}
}
/*********/
@@ -510,9 +574,12 @@ 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) {
assert(size == isalloc(ptr));
prof_thr_cnt_t *tcnt = prof_lookup(ctx->bt);
prof_thr_cnt_t *tcnt;
assert(size == isalloc(ptr, true));
tcnt = prof_lookup(ctx->bt);
if (tcnt != NULL) {
tcnt->epoch++;
@@ -533,10 +600,10 @@ prof_free(const void *ptr, size_t size)
* OOM during free() cannot be propagated, so operate
* directly on cnt->ctx->cnt_merged.
*/
malloc_mutex_lock(&ctx->lock);
malloc_mutex_lock(ctx->lock);
ctx->cnt_merged.curobjs--;
ctx->cnt_merged.curbytes -= size;
malloc_mutex_unlock(&ctx->lock);
malloc_mutex_unlock(ctx->lock);
}
}
}
@@ -544,4 +611,3 @@ prof_free(const void *ptr, size_t size)
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_PROF */
deps/jemalloc/include/jemalloc/internal/public_namespace.sh
Vendored Executable
+6
View File
@@ -0,0 +1,6 @@
#!/bin/sh
for nm in `cat $1` ; do
n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'`
echo "#define je_${n} JEMALLOC_N(${n})"
done
deps/jemalloc/include/jemalloc/internal/public_unnamespace.sh
Vendored Executable
+6
View File
@@ -0,0 +1,6 @@
#!/bin/sh
for nm in `cat $1` ; do
n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'`
echo "#undef je_${n}"
done
deps/jemalloc/include/jemalloc/internal/ql.h
Vendored
+18 -18
View File
@@ -1,61 +1,61 @@
/*
* List definitions.
*/
#define ql_head(a_type) \
#define ql_head(a_type) \
struct { \
a_type *qlh_first; \
}
#define ql_head_initializer(a_head) {NULL}
#define ql_head_initializer(a_head) {NULL}
#define ql_elm(a_type) qr(a_type)
#define ql_elm(a_type) qr(a_type)
/* List functions. */
#define ql_new(a_head) do { \
#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_elm_new(a_elm, a_field) qr_new((a_elm), a_field)
#define ql_first(a_head) ((a_head)->qlh_first)
#define ql_first(a_head) ((a_head)->qlh_first)
#define ql_last(a_head, a_field) \
#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) \
#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) \
#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 { \
#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) \
#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 { \
#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 { \
#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 { \
#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); \
} \
@@ -66,18 +66,18 @@ struct { \
} \
} while (0)
#define ql_head_remove(a_head, a_type, a_field) do { \
#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 { \
#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) \
#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) \
#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
+11 -11
View File
@@ -1,28 +1,28 @@
/* Ring definitions. */
#define qr(a_type) \
#define qr(a_type) \
struct { \
a_type *qre_next; \
a_type *qre_prev; \
}
/* Ring functions. */
#define qr_new(a_qr, a_field) do { \
#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_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_prev(a_qr, a_field) ((a_qr)->a_field.qre_prev)
#define qr_before_insert(a_qrelm, a_qr, a_field) do { \
#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) \
#define qr_after_insert(a_qrelm, a_qr, a_field) \
do \
{ \
(a_qr)->a_field.qre_next = (a_qrelm)->a_field.qre_next; \
@@ -31,7 +31,7 @@ struct { \
(a_qrelm)->a_field.qre_next = (a_qr); \
} while (0)
#define qr_meld(a_qr_a, a_qr_b, a_field) do { \
#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); \
@@ -42,10 +42,10 @@ struct { \
/* 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) \
#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 { \
#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 \
@@ -54,13 +54,13 @@ struct { \
(a_qr)->a_field.qre_prev = (a_qr); \
} while (0)
#define qr_foreach(var, a_qr, a_field) \
#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) \
#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)) \
deps/jemalloc/include/jemalloc/internal/quarantine.h
Vendored
+67
View File
@@ -0,0 +1,67 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
typedef struct quarantine_obj_s quarantine_obj_t;
typedef struct quarantine_s quarantine_t;
/* 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
struct quarantine_obj_s {
void *ptr;
size_t usize;
};
struct quarantine_s {
size_t curbytes;
size_t curobjs;
size_t first;
#define LG_MAXOBJS_INIT 10
size_t lg_maxobjs;
quarantine_obj_t objs[1]; /* Dynamically sized ring buffer. */
};
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
quarantine_t *quarantine_init(size_t lg_maxobjs);
void quarantine(void *ptr);
void quarantine_cleanup(void *arg);
bool quarantine_boot(void);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifndef JEMALLOC_ENABLE_INLINE
malloc_tsd_protos(JEMALLOC_ATTR(unused), quarantine, quarantine_t *)
void quarantine_alloc_hook(void);
#endif
#if (defined(JEMALLOC_ENABLE_INLINE) || defined(JEMALLOC_QUARANTINE_C_))
malloc_tsd_externs(quarantine, quarantine_t *)
malloc_tsd_funcs(JEMALLOC_ALWAYS_INLINE, quarantine, quarantine_t *, NULL,
quarantine_cleanup)
JEMALLOC_ALWAYS_INLINE void
quarantine_alloc_hook(void)
{
quarantine_t *quarantine;
assert(config_fill && opt_quarantine);
quarantine = *quarantine_tsd_get();
if (quarantine == NULL)
quarantine_init(LG_MAXOBJS_INIT);
}
#endif
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/rb.h
Vendored
+46 -50
View File
@@ -22,10 +22,6 @@
#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) \
@@ -223,88 +219,88 @@ a_prefix##reverse_iter(a_rbt_type *rbtree, a_type *start, \
* The following API is generated:
*
* static void
* ex_new(ex_t *extree);
* ex_new(ex_t *tree);
* Description: Initialize a red-black tree structure.
* Args:
* extree: Pointer to an uninitialized red-black tree object.
* tree: Pointer to an uninitialized red-black tree object.
*
* static ex_node_t *
* ex_first(ex_t *extree);
* ex_first(ex_t *tree);
* static ex_node_t *
* ex_last(ex_t *extree);
* Description: Get the first/last node in extree.
* ex_last(ex_t *tree);
* Description: Get the first/last node in tree.
* Args:
* extree: Pointer to an initialized red-black tree object.
* Ret: First/last node in extree, or NULL if extree is empty.
* 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 *extree, ex_node_t *node);
* ex_next(ex_t *tree, ex_node_t *node);
* static ex_node_t *
* ex_prev(ex_t *extree, ex_node_t *node);
* ex_prev(ex_t *tree, ex_node_t *node);
* Description: Get node's successor/predecessor.
* Args:
* extree: Pointer to an initialized red-black tree object.
* node : A node in extree.
* Ret: node's successor/predecessor in extree, or NULL if node is
* 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 *extree, ex_node_t *key);
* ex_search(ex_t *tree, ex_node_t *key);
* Description: Search for node that matches key.
* Args:
* extree: Pointer to an initialized red-black tree object.
* key : Search key.
* Ret: Node in extree that matches key, or NULL if no match.
* 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 *extree, ex_node_t *key);
* ex_nsearch(ex_t *tree, ex_node_t *key);
* static ex_node_t *
* ex_psearch(ex_t *extree, ex_node_t *key);
* 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 extree.
* key in tree.
* Args:
* extree: Pointer to an initialized red-black tree object.
* key : Search key.
* Ret: Node in extree that matches key, or if no match, hypothetical
* node's successor/predecessor (NULL if no successor/predecessor).
* 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 *extree, ex_node_t *node);
* Description: Insert node into extree.
* ex_insert(ex_t *tree, ex_node_t *node);
* Description: Insert node into tree.
* Args:
* extree: Pointer to an initialized red-black tree object.
* node : Node to be inserted into extree.
* tree: Pointer to an initialized red-black tree object.
* node: Node to be inserted into tree.
*
* static void
* ex_remove(ex_t *extree, ex_node_t *node);
* Description: Remove node from extree.
* ex_remove(ex_t *tree, ex_node_t *node);
* Description: Remove node from tree.
* Args:
* extree: Pointer to an initialized red-black tree object.
* node : Node in extree to be removed.
* tree: Pointer to an initialized red-black tree object.
* node: Node in tree to be removed.
*
* static ex_node_t *
* ex_iter(ex_t *extree, ex_node_t *start, ex_node_t *(*cb)(ex_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 *extree, ex_node_t *start, ex_node *(*cb)(ex_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 extree, starting at node.
* If extree is modified, iteration must be immediately
* 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:
* extree: 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 extree.
* arg : Opaque pointer passed to cb().
* 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.
*/
deps/jemalloc/include/jemalloc/internal/rtree.h
Vendored
+32 -21
View File
@@ -14,17 +14,18 @@ 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
#define RTREE_NODESIZE (1U << 16)
typedef void *(rtree_alloc_t)(size_t);
typedef void (rtree_dalloc_t)(void *);
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
struct rtree_s {
rtree_alloc_t *alloc;
rtree_dalloc_t *dalloc;
malloc_mutex_t mutex;
void **root;
unsigned height;
@@ -35,27 +36,31 @@ struct rtree_s {
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
rtree_t *rtree_new(unsigned bits);
rtree_t *rtree_new(unsigned bits, rtree_alloc_t *alloc, rtree_dalloc_t *dalloc);
void rtree_delete(rtree_t *rtree);
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);
#ifdef JEMALLOC_DEBUG
uint8_t 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);
uint8_t rtree_get(rtree_t *rtree, uintptr_t key);
bool rtree_set(rtree_t *rtree, uintptr_t key, uint8_t 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 * \
JEMALLOC_INLINE uint8_t \
f(rtree_t *rtree, uintptr_t key) \
{ \
void *ret; \
uint8_t ret; \
uintptr_t subkey; \
unsigned i, lshift, height, bits; \
void **node, **child; \
@@ -65,12 +70,12 @@ f(rtree_t *rtree, uintptr_t key) \
i < height - 1; \
i++, lshift += bits, node = child) { \
bits = rtree->level2bits[i]; \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR + \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR + \
3)) - bits); \
child = (void**)node[subkey]; \
if (child == NULL) { \
RTREE_UNLOCK(&rtree->mutex); \
return (NULL); \
return (0); \
} \
} \
\
@@ -81,7 +86,10 @@ f(rtree_t *rtree, uintptr_t key) \
bits = rtree->level2bits[i]; \
subkey = (key << lshift) >> ((ZU(1) << (LG_SIZEOF_PTR+3)) - \
bits); \
ret = node[subkey]; \
{ \
uint8_t *leaf = (uint8_t *)node; \
ret = leaf[subkey]; \
} \
RTREE_UNLOCK(&rtree->mutex); \
\
RTREE_GET_VALIDATE \
@@ -120,7 +128,7 @@ RTREE_GET_GENERATE(rtree_get)
#undef RTREE_GET_VALIDATE
JEMALLOC_INLINE bool
rtree_set(rtree_t *rtree, uintptr_t key, void *val)
rtree_set(rtree_t *rtree, uintptr_t key, uint8_t val)
{
uintptr_t subkey;
unsigned i, lshift, height, bits;
@@ -135,14 +143,14 @@ rtree_set(rtree_t *rtree, uintptr_t key, void *val)
bits);
child = (void**)node[subkey];
if (child == NULL) {
child = (void**)base_alloc(sizeof(void *) <<
rtree->level2bits[i+1]);
size_t size = ((i + 1 < height - 1) ? sizeof(void *)
: (sizeof(uint8_t))) << rtree->level2bits[i+1];
child = (void**)rtree->alloc(size);
if (child == NULL) {
malloc_mutex_unlock(&rtree->mutex);
return (true);
}
memset(child, 0, sizeof(void *) <<
rtree->level2bits[i+1]);
memset(child, 0, size);
node[subkey] = child;
}
}
@@ -150,7 +158,10 @@ rtree_set(rtree_t *rtree, uintptr_t key, void *val)
/* 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;
{
uint8_t *leaf = (uint8_t *)node;
leaf[subkey] = val;
}
malloc_mutex_unlock(&rtree->mutex);
return (false);
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
-34
View File
@@ -1,25 +1,16 @@
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#define UMAX2S_BUFSIZE 65
#ifdef JEMALLOC_STATS
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;
#endif
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
typedef struct chunk_stats_s chunk_stats_t;
#endif
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#ifdef JEMALLOC_STATS
#ifdef JEMALLOC_TCACHE
struct tcache_bin_stats_s {
/*
* Number of allocation requests that corresponded to the size of this
@@ -27,7 +18,6 @@ struct tcache_bin_stats_s {
*/
uint64_t nrequests;
};
#endif
struct malloc_bin_stats_s {
/*
@@ -52,13 +42,11 @@ struct malloc_bin_stats_s {
*/
uint64_t nrequests;
#ifdef JEMALLOC_TCACHE
/* Number of tcache fills from this bin. */
uint64_t nfills;
/* Number of tcache flushes to this bin. */
uint64_t nflushes;
#endif
/* Total number of runs created for this bin's size class. */
uint64_t nruns;
@@ -69,9 +57,6 @@ struct malloc_bin_stats_s {
*/
uint64_t reruns;
/* High-water mark for this bin. */
size_t highruns;
/* Current number of runs in this bin. */
size_t curruns;
};
@@ -93,9 +78,6 @@ struct malloc_large_stats_s {
*/
uint64_t nrequests;
/* High-water mark for this size class. */
size_t highruns;
/* Current number of runs of this size class. */
size_t curruns;
};
@@ -127,14 +109,10 @@ struct arena_stats_s {
*/
malloc_large_stats_t *lstats;
};
#endif /* JEMALLOC_STATS */
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
struct chunk_stats_s {
# ifdef JEMALLOC_STATS
/* Number of chunks that were allocated. */
uint64_t nchunks;
# endif
/* High-water mark for number of chunks allocated. */
size_t highchunks;
@@ -146,7 +124,6 @@ struct chunk_stats_s {
*/
size_t curchunks;
};
#endif /* JEMALLOC_STATS */
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
@@ -154,24 +131,14 @@ struct chunk_stats_s {
extern bool opt_stats_print;
#ifdef JEMALLOC_STATS
extern size_t stats_cactive;
#endif
char *u2s(uint64_t x, unsigned base, char *s);
#ifdef JEMALLOC_STATS
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
void stats_print(void (*write)(void *, const char *), void *cbopaque,
const char *opts);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#ifdef JEMALLOC_STATS
#ifndef JEMALLOC_ENABLE_INLINE
size_t stats_cactive_get(void);
@@ -202,6 +169,5 @@ stats_cactive_sub(size_t size)
}
#endif
#endif /* JEMALLOC_STATS */
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/tcache.h
Vendored
+215 -203
View File
@@ -1,4 +1,3 @@
#ifdef JEMALLOC_TCACHE
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
@@ -6,6 +5,16 @@ 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
@@ -22,17 +31,26 @@ typedef struct tcache_s tcache_t;
#define LG_TCACHE_MAXCLASS_DEFAULT 15
/*
* (1U << opt_lg_tcache_gc_sweep) is the approximate number of allocation
* events between full GC sweeps (-1: disabled). Integer rounding may cause
* the actual number to be slightly higher, since GC is performed
* incrementally.
* 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 LG_TCACHE_GC_SWEEP_DEFAULT 13
#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.
@@ -42,9 +60,7 @@ struct tcache_bin_info_s {
};
struct tcache_bin_s {
# ifdef JEMALLOC_STATS
tcache_bin_stats_t tstats;
# endif
int low_water; /* Min # cached since last GC. */
unsigned lg_fill_div; /* Fill (ncached_max >> lg_fill_div). */
unsigned ncached; /* # of cached objects. */
@@ -52,12 +68,8 @@ struct tcache_bin_s {
};
struct tcache_s {
# ifdef JEMALLOC_STATS
ql_elm(tcache_t) link; /* Used for aggregating stats. */
# endif
# ifdef JEMALLOC_PROF
uint64_t prof_accumbytes;/* Cleared after arena_prof_accum() */
# endif
arena_t *arena; /* This thread's arena. */
unsigned ev_cnt; /* Event count since incremental GC. */
unsigned next_gc_bin; /* Next bin to GC. */
@@ -76,29 +88,11 @@ struct tcache_s {
extern bool opt_tcache;
extern ssize_t opt_lg_tcache_max;
extern ssize_t opt_lg_tcache_gc_sweep;
extern tcache_bin_info_t *tcache_bin_info;
/* Map of thread-specific caches. */
#ifndef NO_TLS
extern __thread tcache_t *tcache_tls
JEMALLOC_ATTR(tls_model("initial-exec"));
# define TCACHE_GET() tcache_tls
# define TCACHE_SET(v) do { \
tcache_tls = (tcache_t *)(v); \
pthread_setspecific(tcache_tsd, (void *)(v)); \
} while (0)
#else
# define TCACHE_GET() ((tcache_t *)pthread_getspecific(tcache_tsd))
# define TCACHE_SET(v) do { \
pthread_setspecific(tcache_tsd, (void *)(v)); \
} while (0)
#endif
extern pthread_key_t tcache_tsd;
/*
* Number of tcache bins. There are nbins small-object bins, plus 0 or more
* Number of tcache bins. There are NBINS small-object bins, plus 0 or more
* large-object bins.
*/
extern size_t nhbins;
@@ -106,134 +100,178 @@ extern size_t nhbins;
/* Maximum cached size class. */
extern size_t tcache_maxclass;
/* Number of tcache allocation/deallocation events between incremental GCs. */
extern unsigned tcache_gc_incr;
void tcache_bin_flush_small(tcache_bin_t *tbin, size_t binind, unsigned rem
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
, tcache_t *tcache
#endif
);
void tcache_bin_flush_large(tcache_bin_t *tbin, size_t binind, unsigned rem
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
, tcache_t *tcache
#endif
);
tcache_t *tcache_create(arena_t *arena);
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);
#ifdef JEMALLOC_STATS
void tcache_thread_cleanup(void *arg);
void tcache_stats_merge(tcache_t *tcache, arena_t *arena);
#endif
bool tcache_boot(void);
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);
tcache_t *tcache_get(void);
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);
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_))
JEMALLOC_INLINE tcache_t *
tcache_get(void)
/* Map of thread-specific caches. */
malloc_tsd_externs(tcache, tcache_t *)
malloc_tsd_funcs(JEMALLOC_ALWAYS_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_ALWAYS_INLINE, tcache_enabled, tcache_enabled_t,
tcache_enabled_default, malloc_tsd_no_cleanup)
JEMALLOC_INLINE void
tcache_flush(void)
{
tcache_t *tcache;
if ((isthreaded & opt_tcache) == false)
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_ALWAYS_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_GET();
if ((uintptr_t)tcache <= (uintptr_t)2) {
tcache = *tcache_tsd_get();
if ((uintptr_t)tcache <= (uintptr_t)TCACHE_STATE_MAX) {
if (tcache == TCACHE_STATE_DISABLED)
return (NULL);
if (tcache == NULL) {
tcache = tcache_create(choose_arena());
if (tcache == NULL)
return (NULL);
} else {
if (tcache == (void *)(uintptr_t)1) {
if (create == false) {
/*
* Make a note that an allocator function was
* called after the tcache_thread_cleanup() was
* called.
* 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.
*/
TCACHE_SET((uintptr_t)2);
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
JEMALLOC_ALWAYS_INLINE void
tcache_event(tcache_t *tcache)
{
if (tcache_gc_incr == 0)
if (TCACHE_GC_INCR == 0)
return;
tcache->ev_cnt++;
assert(tcache->ev_cnt <= tcache_gc_incr);
if (tcache->ev_cnt == tcache_gc_incr) {
size_t binind = tcache->next_gc_bin;
tcache_bin_t *tbin = &tcache->tbins[binind];
tcache_bin_info_t *tbin_info = &tcache_bin_info[binind];
if (tbin->low_water > 0) {
/*
* Flush (ceiling) 3/4 of the objects below the low
* water mark.
*/
if (binind < nbins) {
tcache_bin_flush_small(tbin, binind,
tbin->ncached - tbin->low_water +
(tbin->low_water >> 2)
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
, tcache
#endif
);
} else {
tcache_bin_flush_large(tbin, binind,
tbin->ncached - tbin->low_water +
(tbin->low_water >> 2)
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
, tcache
#endif
);
}
/*
* Reduce fill count by 2X. Limit lg_fill_div such that
* the fill count is always at least 1.
*/
if ((tbin_info->ncached_max >> (tbin->lg_fill_div+1))
>= 1)
tbin->lg_fill_div++;
} else if (tbin->low_water < 0) {
/*
* Increase fill count by 2X. Make sure lg_fill_div
* stays greater than 0.
*/
if (tbin->lg_fill_div > 1)
tbin->lg_fill_div--;
}
tbin->low_water = tbin->ncached;
tcache->next_gc_bin++;
if (tcache->next_gc_bin == nhbins)
tcache->next_gc_bin = 0;
tcache->ev_cnt = 0;
}
assert(tcache->ev_cnt <= TCACHE_GC_INCR);
if (tcache->ev_cnt == TCACHE_GC_INCR)
tcache_event_hard(tcache);
}
JEMALLOC_INLINE void *
JEMALLOC_ALWAYS_INLINE void *
tcache_alloc_easy(tcache_bin_t *tbin)
{
void *ret;
@@ -249,7 +287,7 @@ tcache_alloc_easy(tcache_bin_t *tbin)
return (ret);
}
JEMALLOC_INLINE void *
JEMALLOC_ALWAYS_INLINE void *
tcache_alloc_small(tcache_t *tcache, size_t size, bool zero)
{
void *ret;
@@ -257,37 +295,44 @@ tcache_alloc_small(tcache_t *tcache, size_t size, bool zero)
tcache_bin_t *tbin;
binind = SMALL_SIZE2BIN(size);
assert(binind < nbins);
assert(binind < NBINS);
tbin = &tcache->tbins[binind];
size = arena_bin_info[binind].reg_size;
ret = tcache_alloc_easy(tbin);
if (ret == NULL) {
ret = tcache_alloc_small_hard(tcache, tbin, binind);
if (ret == NULL)
return (NULL);
}
assert(arena_salloc(ret) == arena_bin_info[binind].reg_size);
assert(tcache_salloc(ret) == arena_bin_info[binind].reg_size);
if (zero == false) {
#ifdef JEMALLOC_FILL
if (opt_junk)
memset(ret, 0xa5, size);
else if (opt_zero)
memset(ret, 0, size);
#endif
} else
if (config_fill) {
if (opt_junk) {
arena_alloc_junk_small(ret,
&arena_bin_info[binind], false);
} else if (opt_zero)
memset(ret, 0, size);
}
VALGRIND_MAKE_MEM_UNDEFINED(ret, 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);
}
#ifdef JEMALLOC_STATS
tbin->tstats.nrequests++;
#endif
#ifdef JEMALLOC_PROF
tcache->prof_accumbytes += arena_bin_info[binind].reg_size;
#endif
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 *
JEMALLOC_ALWAYS_INLINE void *
tcache_alloc_large(tcache_t *tcache, size_t size, bool zero)
{
void *ret;
@@ -296,7 +341,7 @@ tcache_alloc_large(tcache_t *tcache, size_t size, bool zero)
size = PAGE_CEILING(size);
assert(size <= tcache_maxclass);
binind = nbins + (size >> PAGE_SHIFT) - 1;
binind = NBINS + (size >> LG_PAGE) - 1;
assert(binind < nhbins);
tbin = &tcache->tbins[binind];
ret = tcache_alloc_easy(tbin);
@@ -309,74 +354,53 @@ tcache_alloc_large(tcache_t *tcache, size_t size, bool zero)
if (ret == NULL)
return (NULL);
} else {
#ifdef JEMALLOC_PROF
arena_chunk_t *chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ret);
size_t pageind = (((uintptr_t)ret - (uintptr_t)chunk) >>
PAGE_SHIFT);
chunk->map[pageind-map_bias].bits &= ~CHUNK_MAP_CLASS_MASK;
#endif
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) {
#ifdef JEMALLOC_FILL
if (opt_junk)
memset(ret, 0xa5, size);
else if (opt_zero)
memset(ret, 0, size);
#endif
} else
if (config_fill) {
if (opt_junk)
memset(ret, 0xa5, size);
else if (opt_zero)
memset(ret, 0, size);
}
VALGRIND_MAKE_MEM_UNDEFINED(ret, size);
} else {
VALGRIND_MAKE_MEM_UNDEFINED(ret, size);
memset(ret, 0, size);
}
#ifdef JEMALLOC_STATS
tbin->tstats.nrequests++;
#endif
#ifdef JEMALLOC_PROF
tcache->prof_accumbytes += size;
#endif
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)
JEMALLOC_ALWAYS_INLINE void
tcache_dalloc_small(tcache_t *tcache, void *ptr, size_t binind)
{
arena_t *arena;
arena_chunk_t *chunk;
arena_run_t *run;
arena_bin_t *bin;
tcache_bin_t *tbin;
tcache_bin_info_t *tbin_info;
size_t pageind, binind;
arena_chunk_map_t *mapelm;
assert(arena_salloc(ptr) <= small_maxclass);
assert(tcache_salloc(ptr) <= SMALL_MAXCLASS);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
arena = chunk->arena;
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> PAGE_SHIFT;
mapelm = &chunk->map[pageind-map_bias];
run = (arena_run_t *)((uintptr_t)chunk + (uintptr_t)((pageind -
(mapelm->bits >> PAGE_SHIFT)) << PAGE_SHIFT));
dassert(run->magic == ARENA_RUN_MAGIC);
bin = run->bin;
binind = ((uintptr_t)bin - (uintptr_t)&arena->bins) /
sizeof(arena_bin_t);
assert(binind < nbins);
#ifdef JEMALLOC_FILL
if (opt_junk)
memset(ptr, 0x5a, arena_bin_info[binind].reg_size);
#endif
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)
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
, tcache
#endif
);
1), tcache);
}
assert(tbin->ncached < tbin_info->ncached_max);
tbin->avail[tbin->ncached] = ptr;
@@ -385,38 +409,27 @@ tcache_dalloc_small(tcache_t *tcache, void *ptr)
tcache_event(tcache);
}
JEMALLOC_INLINE void
JEMALLOC_ALWAYS_INLINE void
tcache_dalloc_large(tcache_t *tcache, void *ptr, size_t size)
{
arena_t *arena;
arena_chunk_t *chunk;
size_t pageind, binind;
size_t binind;
tcache_bin_t *tbin;
tcache_bin_info_t *tbin_info;
assert((size & PAGE_MASK) == 0);
assert(arena_salloc(ptr) > small_maxclass);
assert(arena_salloc(ptr) <= tcache_maxclass);
assert(tcache_salloc(ptr) > SMALL_MAXCLASS);
assert(tcache_salloc(ptr) <= tcache_maxclass);
chunk = (arena_chunk_t *)CHUNK_ADDR2BASE(ptr);
arena = chunk->arena;
pageind = ((uintptr_t)ptr - (uintptr_t)chunk) >> PAGE_SHIFT;
binind = nbins + (size >> PAGE_SHIFT) - 1;
binind = NBINS + (size >> LG_PAGE) - 1;
#ifdef JEMALLOC_FILL
if (opt_junk)
if (config_fill && opt_junk)
memset(ptr, 0x5a, size);
#endif
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)
#if (defined(JEMALLOC_STATS) || defined(JEMALLOC_PROF))
, tcache
#endif
);
1), tcache);
}
assert(tbin->ncached < tbin_info->ncached_max);
tbin->avail[tbin->ncached] = ptr;
@@ -428,4 +441,3 @@ tcache_dalloc_large(tcache_t *tcache, void *ptr, size_t size)
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
#endif /* JEMALLOC_TCACHE */
deps/jemalloc/include/jemalloc/internal/tsd.h
Vendored
+434
View File
@@ -0,0 +1,434 @@
/******************************************************************************/
#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);
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
typedef struct tsd_init_block_s tsd_init_block_t;
typedef struct tsd_init_head_s tsd_init_head_t;
#endif
/*
* 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 tsd_init_head_t a_name##_tsd_init_head; \
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 tsd_init_head_t a_name##_tsd_init_head = { \
ql_head_initializer(blocks), \
MALLOC_MUTEX_INITIALIZER \
}; \
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) { \
tsd_init_block_t block; \
wrapper = tsd_init_check_recursion( \
&a_name##_tsd_init_head, &block); \
if (wrapper) \
return (wrapper); \
wrapper = (a_name##_tsd_wrapper_t *) \
malloc_tsd_malloc(sizeof(a_name##_tsd_wrapper_t)); \
block.data = wrapper; \
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(); \
} \
tsd_init_finish(&a_name##_tsd_init_head, &block); \
} \
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
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
struct tsd_init_block_s {
ql_elm(tsd_init_block_t) link;
pthread_t thread;
void *data;
};
struct tsd_init_head_s {
ql_head(tsd_init_block_t) blocks;
malloc_mutex_t lock;
};
#endif
#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);
#if (!defined(JEMALLOC_MALLOC_THREAD_CLEANUP) && !defined(JEMALLOC_TLS) && \
!defined(_WIN32))
void *tsd_init_check_recursion(tsd_init_head_t *head,
tsd_init_block_t *block);
void tsd_init_finish(tsd_init_head_t *head, tsd_init_block_t *block);
#endif
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/internal/util.h
Vendored
+162
View File
@@ -0,0 +1,162 @@
/******************************************************************************/
#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_ARG_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
#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
#ifndef assert_not_implemented
#define assert_not_implemented(e) do { \
if (config_debug && !(e)) \
not_implemented(); \
} while (0)
#endif
/* Use to assert a particular configuration, e.g., cassert(config_debug). */
#define cassert(c) do { \
if ((c) == false) \
not_reached(); \
} while (0)
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
int buferror(int err, char *buf, size_t buflen);
uintmax_t malloc_strtoumax(const char *restrict nptr,
char **restrict 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 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/internal/zone.h
Vendored
-23
View File
@@ -1,23 +0,0 @@
#ifndef JEMALLOC_ZONE
# error "This source file is for zones on Darwin (OS X)."
#endif
/******************************************************************************/
#ifdef JEMALLOC_H_TYPES
#endif /* JEMALLOC_H_TYPES */
/******************************************************************************/
#ifdef JEMALLOC_H_STRUCTS
#endif /* JEMALLOC_H_STRUCTS */
/******************************************************************************/
#ifdef JEMALLOC_H_EXTERNS
malloc_zone_t *create_zone(void);
void szone2ozone(malloc_zone_t *zone);
#endif /* JEMALLOC_H_EXTERNS */
/******************************************************************************/
#ifdef JEMALLOC_H_INLINES
#endif /* JEMALLOC_H_INLINES */
/******************************************************************************/
deps/jemalloc/include/jemalloc/jemalloc.h.in
Vendored
-66
View File
@@ -1,66 +0,0 @@
#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"
#ifndef JEMALLOC_P
# define JEMALLOC_P(s) s
#endif
#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)
#define ALLOCM_SUCCESS 0
#define ALLOCM_ERR_OOM 1
#define ALLOCM_ERR_NOT_MOVED 2
extern const char *JEMALLOC_P(malloc_conf);
extern void (*JEMALLOC_P(malloc_message))(void *, const char *);
void *JEMALLOC_P(malloc)(size_t size) JEMALLOC_ATTR(malloc);
void *JEMALLOC_P(calloc)(size_t num, size_t size) JEMALLOC_ATTR(malloc);
int JEMALLOC_P(posix_memalign)(void **memptr, size_t alignment, size_t size)
JEMALLOC_ATTR(nonnull(1));
void *JEMALLOC_P(realloc)(void *ptr, size_t size);
void JEMALLOC_P(free)(void *ptr);
size_t JEMALLOC_P(malloc_usable_size)(const void *ptr);
void JEMALLOC_P(malloc_stats_print)(void (*write_cb)(void *, const char *),
void *cbopaque, const char *opts);
int JEMALLOC_P(mallctl)(const char *name, void *oldp, size_t *oldlenp,
void *newp, size_t newlen);
int JEMALLOC_P(mallctlnametomib)(const char *name, size_t *mibp,
size_t *miblenp);
int JEMALLOC_P(mallctlbymib)(const size_t *mib, size_t miblen, void *oldp,
size_t *oldlenp, void *newp, size_t newlen);
int JEMALLOC_P(allocm)(void **ptr, size_t *rsize, size_t size, int flags)
JEMALLOC_ATTR(nonnull(1));
int JEMALLOC_P(rallocm)(void **ptr, size_t *rsize, size_t size,
size_t extra, int flags) JEMALLOC_ATTR(nonnull(1));
int JEMALLOC_P(sallocm)(const void *ptr, size_t *rsize, int flags)
JEMALLOC_ATTR(nonnull(1));
int JEMALLOC_P(dallocm)(void *ptr, int flags) JEMALLOC_ATTR(nonnull(1));
#ifdef __cplusplus
};
#endif
#endif /* JEMALLOC_H_ */
deps/jemalloc/include/jemalloc/jemalloc.sh
Vendored Executable
+28
View File
@@ -0,0 +1,28 @@
#!/bin/sh
objroot=$1
cat <<EOF
#ifndef JEMALLOC_H_
#define JEMALLOC_H_
#ifdef __cplusplus
extern "C" {
#endif
EOF
for hdr in jemalloc_defs.h jemalloc_rename.h jemalloc_macros.h \
jemalloc_protos.h jemalloc_mangle.h ; do
cat "${objroot}include/jemalloc/${hdr}" \
| grep -v 'Generated from .* by configure\.' \
| sed -e 's/^#define /#define /g' \
| sed -e 's/ $//g'
echo
done
cat <<EOF
#ifdef __cplusplus
};
#endif
#endif /* JEMALLOC_H_ */
EOF
deps/jemalloc/include/jemalloc/jemalloc_defs.h.in
Vendored
+9 -152
View File
@@ -1,167 +1,24 @@
#ifndef JEMALLOC_DEFS_H_
#define JEMALLOC_DEFS_H_
/*
* If JEMALLOC_PREFIX is defined, it will cause all public APIs to be prefixed.
* This makes it possible, with some care, to use multiple allocators
* simultaneously.
*
* In many cases it is more convenient to manually prefix allocator function
* calls than to let macros do it automatically, particularly when using
* multiple allocators simultaneously. Define JEMALLOC_MANGLE before
* #include'ing jemalloc.h in order to cause name mangling that corresponds to
* the API prefixing.
*/
#undef JEMALLOC_PREFIX
#undef JEMALLOC_CPREFIX
#if (defined(JEMALLOC_PREFIX) && defined(JEMALLOC_MANGLE))
#undef JEMALLOC_P
#endif
/*
* 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 OSAtomic*() functions are available, as provided by Darwin, and
* documented in the atomic(3) manual page.
*/
#undef JEMALLOC_OSATOMIC
/*
* Defined if OSSpin*() functions are available, as provided by Darwin, and
* documented in the spinlock(3) manual page.
*/
#undef JEMALLOC_OSSPIN
/* Defined if __attribute__((...)) syntax is supported. */
#undef JEMALLOC_HAVE_ATTR
#ifdef JEMALLOC_HAVE_ATTR
# define JEMALLOC_ATTR(s) __attribute__((s))
#else
# define JEMALLOC_ATTR(s)
#endif
/* JEMALLOC_CC_SILENCE enables code that silences unuseful compiler warnings. */
#undef JEMALLOC_CC_SILENCE
/* Support the experimental API. */
#undef JEMALLOC_EXPERIMENTAL
/*
* 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_TINY enables support for tiny objects, which are smaller than one
* quantum.
*/
#undef JEMALLOC_TINY
/*
* 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
/* JEMALLOC_SWAP enables mmap()ed swap file support. */
#undef JEMALLOC_SWAP
/* Support memory filling (junk/zero). */
#undef JEMALLOC_FILL
/* Support optional abort() on OOM. */
#undef JEMALLOC_XMALLOC
/* Support SYSV semantics. */
#undef JEMALLOC_SYSV
/* Support lazy locking (avoid locking unless a second thread is launched). */
#undef JEMALLOC_LAZY_LOCK
/* Determine page size at run time if defined. */
#undef DYNAMIC_PAGE_SHIFT
/* One page is 2^STATIC_PAGE_SHIFT bytes. */
#undef STATIC_PAGE_SHIFT
/* TLS is used to map arenas and magazine caches to threads. */
#undef NO_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.
* Define overrides for non-standard allocator-related functions if they are
* present on the system.
*/
#undef JEMALLOC_OVERRIDE_MEMALIGN
#undef JEMALLOC_OVERRIDE_VALLOC
/*
* Darwin (OS X) uses zones to work around Mach-O symbol override shortcomings.
*/
#undef JEMALLOC_ZONE
#undef JEMALLOC_ZONE_VERSION
/* If defined, use mremap(...MREMAP_FIXED...) for huge realloc(). */
#undef JEMALLOC_MREMAP_FIXED
/*
* Methods for purging unused pages differ between operating systems.
* At least Linux omits the "const" in:
*
* 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.
* size_t malloc_usable_size(const void *ptr);
*
* Match the operating system's prototype.
*/
#undef JEMALLOC_PURGE_MADVISE_DONTNEED
#undef JEMALLOC_PURGE_MADVISE_FREE
#undef JEMALLOC_USABLE_SIZE_CONST
/* 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
#endif /* JEMALLOC_DEFS_H_ */
deps/jemalloc/include/jemalloc/jemalloc_macros.h.in
Vendored
+61
View File
@@ -0,0 +1,61 @@
#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@"
# define MALLOCX_LG_ALIGN(la) (la)
# if LG_SIZEOF_PTR == 2
# define MALLOCX_ALIGN(a) (ffs(a)-1)
# else
# define MALLOCX_ALIGN(a) \
((a < (size_t)INT_MAX) ? ffs(a)-1 : ffs(a>>32)+31)
# endif
# define MALLOCX_ZERO ((int)0x40)
/* Bias arena index bits so that 0 encodes "MALLOCX_ARENA() unspecified". */
# define MALLOCX_ARENA(a) ((int)(((a)+1) << 8))
#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
#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
deps/jemalloc/include/jemalloc/jemalloc_mangle.sh
Vendored Executable
+45
View File
@@ -0,0 +1,45 @@
#!/bin/sh
public_symbols_txt=$1
symbol_prefix=$2
cat <<EOF
/*
* 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
EOF
for nm in `cat ${public_symbols_txt}` ; do
n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'`
echo "# define ${n} ${symbol_prefix}${n}"
done
cat <<EOF
#endif
/*
* The ${symbol_prefix}* 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
EOF
for nm in `cat ${public_symbols_txt}` ; do
n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'`
echo "# undef ${symbol_prefix}${n}"
done
cat <<EOF
#endif
EOF
deps/jemalloc/include/jemalloc/jemalloc_protos.h.in
Vendored
+58
View File
@@ -0,0 +1,58 @@
/*
* 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 jemalloc_mangle@install_suffix@.h).
*/
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);
JEMALLOC_EXPORT void *@je_@mallocx(size_t size, int flags);
JEMALLOC_EXPORT void *@je_@rallocx(void *ptr, size_t size, int flags);
JEMALLOC_EXPORT size_t @je_@xallocx(void *ptr, size_t size, size_t extra,
int flags);
JEMALLOC_EXPORT size_t @je_@sallocx(const void *ptr, int flags);
JEMALLOC_EXPORT void @je_@dallocx(void *ptr, int flags);
JEMALLOC_EXPORT size_t @je_@nallocx(size_t size, int flags);
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);
JEMALLOC_EXPORT void @je_@malloc_stats_print(void (*write_cb)(void *,
const char *), void *@je_@cbopaque, const char *opts);
JEMALLOC_EXPORT size_t @je_@malloc_usable_size(
JEMALLOC_USABLE_SIZE_CONST 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
#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
deps/jemalloc/include/jemalloc/jemalloc_rename.sh
Vendored Executable
+22
View File
@@ -0,0 +1,22 @@
#!/bin/sh
public_symbols_txt=$1
cat <<EOF
/*
* 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.
*/
#ifndef JEMALLOC_NO_RENAME
EOF
for nm in `cat ${public_symbols_txt}` ; do
n=`echo ${nm} |tr ':' ' ' |awk '{print $1}'`
m=`echo ${nm} |tr ':' ' ' |awk '{print $2}'`
echo "# define je_${n} ${m}"
done
cat <<EOF
#endif
EOF
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/src/arena.c
Vendored
+1558 -1685
View File
File diff suppressed because it is too large Load Diff

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