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:v5.0.10
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:win-2.8.19-rc1
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
2093 Commits
v5.0.10 .. win-2.8.19-rc1
Author SHA1 Message Date
Alexis Campailla 0f79d72396 Workaround for getpeername() issue 
getpeername() is returning invalid addresses on ipv6 sockets accepted
with AcceptEx, filling only 16 bytes of the address structure.
Providing a workaround by saving the remote address returned by
GetAcceptExSockaddrs, which is valid.
 2015-02-25 12:27:07 +01:00
Alexis Campailla 322d708f21 getpeername returns 0 on error 2015-02-24 18:11:23 +01:00
Alexis Campailla c3f758c923 Update redis.windows.conf 2015-02-24 18:09:07 +01:00
Alexis Campailla fe6fff14d1 redlis-cli crash on non-printable chars 
The debug version of isprint() raises an assert when the input value
is not EOF or in the range 0 through 0xFF, inclusive.
The trick is that the char value needs to be passed as unsigned,
before being converted to int.
 2015-02-24 17:16:52 +01:00
Alexis Campailla a39e4ae65d Handle absolute paths for config files 2015-02-24 16:28:33 +01:00
Alexis Campailla b02a6d191d Minor documenation bug (confusing dir and heapdir) 
See https://github.com/MSOpenTech/redis/issues/218
 2015-02-24 15:31:11 +01:00
Scott Ganyo 1fe315d6d9 Update README.md 2015-02-13 14:24:44 +01:00
Robert Hahn b167e972d2 issue 180: fixed handling of the alt gr key on international keyboards 2015-02-13 14:09:50 +01:00
Alexis Campailla 768e061172 Abort on OOM 
It looks like https://github.com/MSOpenTech/redis/commit/62cc0e320d9afe28cc44a642d9e8f5e1c0e5d775#diff-07ce465381b20e496c6fb51bedd3491fR3105
inadvertently made the Windows side not abort when an OOM occurs.
 2015-02-13 03:03:58 +01:00
Alexis Campailla b93dc1c727 Port diskless replication to Windows 
During diskless replication the master forks a child, which on posix
simply inherits the socket file descriptors for the connections to
the slaves.
A unix pipe is also used for the child to report the results back
to the master.

The bulk of the porting work is in making sure that the socket
file descriptors and pipe file descriptor are propagated correctly
from the master to its child.
 2015-02-06 16:08:50 +01:00
Alexis Campailla b713fda58f Remove useless qfork synchronization event 
There is no apparent reason for the child to wait for the
startOperation event to be signaled.
 2015-02-06 16:06:53 +01:00
Alexis Campailla 5677f532f8 Refactor BeginForkOperation 
Refactor BeginForkOperation in preparation for diskless replication:
- Separate copying of operation data and child process creation
- Provide specific entry points for each operation type
 2015-02-06 15:42:47 +01:00
orangemocha 3942d52957 Merge pull request #216 from MSOpenTech/orangemocha-CrashOnLargeInsert 
Bug fixes
 2015-02-02 16:40:01 +01:00
Alexis Campailla 9fd03a982d Fix AV in AllocHeapBlock 2015-02-02 16:10:05 +01:00
Alexis Campailla bc8d23c64a Merge branch release-2.8.17 into 2.8 2015-01-06 14:40:16 -05:00
Alexis Campailla 7796c039cb Fix redis-cli pipe mode 
Pipe mode was broken due to CRT's translation of CRLF sequences.
 2015-01-02 17:48:09 -05:00
Alexis Campailla 69ff2feb70 Merge version 2.8.19 from antirez into 2.8 2014-12-26 16:14:57 +01:00
Alexis Campailla bc2ce9d6e0 Merge version 2.8.18 from antirez into 2.8 
Conflicts:
	deps/linenoise/linenoise.c
	deps/lua/src/lua_cjson.c
	redis.conf
	src/anet.c
	src/config.c
	src/networking.c
	src/object.c
	src/rdb.c
	src/redis.c
	src/redis.h
	src/replication.c
	tests/support/server.tcl
 2014-12-26 15:20:36 +01:00
Alexis Campailla 8d5716c03e Merge branch 'release-2.8.17' into 2.8 
Conflicts:
	bin/release/redis-2.8.17.zip
 2014-12-24 00:20:37 +01:00
Alexis Campailla a87bca3e7b replaced // comments #2150 2014-12-24 00:18:58 +01:00
Alexis Campailla a659c7c13a Merge commit 'c94db2ba3849ffe2ae1b3da04bd2860a601afcd5' into release-2.8.17 
Conflicts:
	bin/release/redis-2.8.17.zip
 2014-12-23 15:01:22 +01:00
Alexis Campailla 1ba3989480 replaced // comments #2150 2014-12-23 01:46:44 +01:00
Alexis Campailla 5c3f6b9d1f redis-benchmark AUTH command to be discarded after the first send #2150 
Cherry pick 560ec60ed4
 2014-12-22 20:54:17 +01:00
Alexis Campailla c94db2ba38 Fix missing time functions initialization in redis-benchmark 
Commit 08e4226e59 broke redis-benchmark.

Made time function initialization automatic on first execution.
Added explicit initialization of time functions in redis-benchmark, to
avoid additional cost on first benchmark execution.
 2014-12-20 18:03:34 +01:00
Alexis Campailla 3b1d4ec032 Move release binaries to release page 
Adopting the suggestions at https://github.com/MSOpenTech/redis/issues/172
We no longer commit the binaries into the repo.
Instead, we create periodic releases on the release page.
 2014-12-16 16:36:10 +01:00
antirez 902b877dc0 Redis 2.8.19. 2014-12-16 09:18:20 +01:00
antirez 02d465c8fe Don't log admin commands in MONITOR. 
Otherwise there are security risks, especially when providing Redis as a
service, the user may "sniff" for admin commands renamed to an
unguessable string via rename-command in redis.conf.
 2014-12-13 09:00:03 +01:00
antirez 4d8f4262a4 List of commands flagged as admin commands modified. 
The old list did not made much sense... and the flag is currently not
used at all, so no side effects.
 2014-12-13 08:59:43 +01:00
Salvatore Sanfilippo d4fc65309d Merge pull request #2092 from rhoml/rhoml-install-sentinel 
Adds a symlink for redis-sentinel when Make install
 2014-12-12 22:58:56 +01:00
antirez e47e460f1f Lua cmsgpack lib updated to latest version. 
It fixes a bad bug that crashes the server in certain conditions
as shown in issue #2210.
 2014-12-12 22:57:09 +01:00
Rhommel Lamas 5509c141f8 Add symlink to redis-sentinel during make install 2014-12-12 20:00:08 +01:00
antirez 7de1ef773e SORT: Don't sort Set elements if not needed. 
Related to #2094.
 2014-12-11 16:01:21 +01:00
Matt Stancliff e945a546af Fix zero-ordering SORT when called against lists 
People mostly use SORT against lists, but our prior
behavior was pretending lists were an unordered bag
requiring a forced-sort when no sort was requested.

We can just use the native list ordering to ensure
consistency across replicaion and scripting calls.

Closes #2079
Closes #545 (again)
 2014-12-11 16:01:09 +01:00
Ben Dowling d81c38316b Update redis_init_script.tpl 
status command currently reports success when redis has crashed and the pid file still exists. Changing to check the actual process is running.
 2014-12-11 15:21:53 +01:00
zhanghailei dba57ea910 FIXED redis-benchmark's idle mode.With idle mode shouldn't create write event 2014-12-11 15:21:53 +01:00
Sun He 888ea17548 zipmap.c: update comments above 2014-12-09 16:01:13 +01:00
Deepak Verma 86ebc139c2 replaced // comments #2150 2014-12-09 15:26:26 +01:00
azure provisioned user 3d73f088e1 redis-benchmark AUTH command to be discarded after the first send #2150 2014-12-09 15:26:26 +01:00
Sun He 76d53a6770 sds.c: Correct two spelling mistakes in comments 2014-12-09 15:14:14 +01:00
Sun He 4848cf9335 sds.c/sdscatvprintf: set va_end to finish va_list cpy 2014-12-09 15:14:14 +01:00
Sun He d2f584fa53 sds.c: Correct some comments 2014-12-09 15:14:13 +01:00
Serghei Iakovlev 2ed3f09cc3 Update whatisdoing.sh 
Improved getting pid
 2014-12-09 13:01:31 +01:00
antirez 77b997d429 Include stropts only if __sun is defined. 2014-12-09 12:59:43 +01:00
Jan-Erik Rediger d409371193 Fix implicit declaration of ioctl on Solaris 2014-12-09 12:57:59 +01:00
Johan Bergström 23b96c02a5 Silence _BSD_SOURCE warnings in glibc 2.20 and forward 
See https://sourceware.org/glibc/wiki/Release/2.20#Packaging_Changes
 2014-12-09 12:24:30 +01:00
antirez a47a042ea6 Mark whatisdoing.sh as deprecated in top-comment. 2014-12-09 12:18:47 +01:00
Serghei Iakovlev b5737d2d19 getting pid fixes 
```sh
$ ~ pidof redis-server
# nothing

$ ~ ps aux | grep [r]edis
redis      593  0.0  0.0  36900  5564 ?        Ssl  Dec02   1:37 /usr/bin/redis-server 127.0.0.1:6379
klay     15927  0.0  0.0  16772  6068 pts/6    S+   13:58   0:00 redis-cli

$ ~ uname -a
Linux edge 3.17.4-1-ARCH #1 SMP PREEMPT Fri Nov 21 21:14:42 CET 2014 x86_64 GNU/Linux
```
 2014-12-09 12:09:31 +01:00
Sun He a598e0894b sparkline.c: AddSample skip Empty label 2014-12-09 11:28:53 +01:00
Sun He 7d480abab4 sparkline.c: mov label-ini into the AddSample Function 2014-12-09 11:28:53 +01:00
Jan-Erik Rediger 2f3c860979 Only ignore sigpipe in interactive mode 
This allows shell pipes to correctly end redis-cli.

Ref #2066
 2014-12-09 09:01:10 +01:00
antirez 0c211a1953 Simplify lua_cmsgpack macro and fix build on old Linux distros. 
Thanks to @badboy for the help in checking the build after the fix.
 2014-12-05 10:52:56 +01:00
Alexis Campailla 08e4226e59 Revisit time functions 
Addressing https://github.com/MSOpenTech/redis/issues/188

- Made gettimeofday use GetSystemTimePreciseAsFileTime when available
- Introduced definitions for gettimeofday_fast and gettimeofday_precise,
  so that calls can be specialized based on specific usage requirements
- Introduced GetHighResRelativeTime, for high-resolution relative time
  measurements on all Windows versions.
 2014-12-04 18:26:53 +01:00
Alexis Campailla 7827f41357 Issue 173: add child process log messages to main log file 
Slave processes were not using the master process log file.
On Unix this is relying on the server.logfile variable being available
to the slave processes through fork(), and reopening the logfile
in the slaves (on every log event).
On Windows we don't use server.logfile and require an explicity call
to setLogFile.
I resorted to explicitly passing the logfile to the slaves as a
command line argument, so the logfile argument (and logging) can be
available to the slave before qfork and globals setup have completed.

Writing to the same file atomically from multiple processes requires
using CreateFile with FILE_APPEND_DATA, instead of fopen, which provides
atomicity on Unix but not on Windows.

Also changed the implementation to not reopen the logfile on every
log event, and not flushing the file on every write. Performance is
dramaticaly improved this way.
 2014-12-04 16:02:45 +01:00
antirez d6ada4d1a8 Redis 2.8.18. 2014-12-04 11:36:52 +01:00
antirez 7c671c09a1 Check that tcp-backlog is matched by /proc/sys/net/core/somaxconn. 2014-12-04 11:07:55 +01:00
antirez d56ef62983 Network bandwidth tracking + refactoring. 
Track bandwidth used by clients and replication (but diskless
replication is not tracked since the actual transfer happens in the
child process).

This includes a refactoring that makes tracking new instantaneous
metrics simpler.
 2014-12-03 12:17:18 +01:00
antirez f60c8c0999 Don't show the ASCII logo if syslog is enabled. 
Closes issue #1935.
 2014-12-03 10:52:09 +01:00
antirez 04231cb7d7 Regression test for issue #2175. 2014-12-03 10:41:06 +01:00
antirez 6e0277769d Handle infinite explicitly in createStringObjectFromLongLong(). 2014-12-03 10:41:06 +01:00
antirez 42695fdfe6 Use exp format and more precision output for ZSCAN. 
Ref: issue #2175
 2014-12-03 10:41:06 +01:00
antirez 164ad2ead1 Over 80 chars comment trimmed in pfcountCommand(). 2014-12-02 17:03:15 +01:00
antirez 0e40124d0f Mark PFCOUNT as read-only, even if not true. 
PFCOUNT is technically speaking a write command, since the cached value
of the HLL is exposed in the data structure (design error, mea culpa), and
can be modified by PFCOUNT.

However if we flag PFCOUNT as "w", read only slaves can't execute the
command, which is a problem since there are environments where slaves
are used to scale PFCOUNT reads.

Nor it is possible to just prevent PFCOUNT to modify the data structure
in slaves, since without the cache we lose too much efficiency.

So while this commit allows slaves to create a temporary inconsistency
(the strings representing the HLLs in the master and slave can be
different in certain moments) it is actually harmless.

In the long run this should be probably fixed by turning the HLL into a
more opaque representation, for example by storing the cached value in
the part of the string which is not exposed (this should be possible
with SDS strings).
 2014-12-02 17:01:55 +01:00
antirez d0cc96f053 Mark diskless replication as experimental in redis.conf. 2014-12-02 14:13:44 +01:00
antirez a81043959a Test: wait for actual startup in start_server. 
start_server now uses return value from Tcl exec to get the server pid,
however this introduces errors that depend from timing: a lot of the
testing code base assumed the server to be actually up and running when
server_start returns.

So the old code that waits to see the pid in the log file was restored.
 2014-11-28 12:57:40 +01:00
antirez f0c00a14ed Test: try to cleanup still running Redis instances on exit. 
It's hard to run the Redis test continuously if it leaks processes on
exceptions / errors.
 2014-11-28 12:57:30 +01:00
antirez 3ac62dff23 Test framework: exit on timeout with report. 
There was no sane way to detect tests that may never end because of
Redis bugs or tests bugs.
 2014-11-28 12:54:17 +01:00
antirez 60c09dcf09 redis-benchmark: default num of requests is now 100000. 
10000 completes in a too short time and may easily provide unreliable
figures because of tiny duration.
 2014-11-28 10:38:58 +01:00
Matthias Petschick e689e93663 fix benchmark memleak in loop mode 2014-11-28 10:38:54 +01:00
antirez 01854f87e8 Document redis-cli --stat in --help output. 2014-11-25 18:25:58 +01:00
Matt Stancliff e305df099a Fix lua-cmsgpack 64 bit integer on 32 bit platform 
This syncs lua-cmsgpack with the mattsta/lua-cmsgpack upstream.

Fixes #2161
 2014-11-25 15:56:57 +01:00
antirez 9854d03fad Avoid valgrind memory leak false positive in processInlineBuffer(). 
zmalloc(0) cauesd to actually trigger a non-zero allocation since with
standard libc malloc we have our own zmalloc header for memory tracking,
but at the same time the returned pointer is at the end of the block and
not in the middle. This triggers a false positive when testing with
valgrind.

When the inline protocol args count is 0, we now avoid reallocating
c->argv, preventing the issue to happen.
 2014-11-25 14:51:04 +01:00
antirez 5ddf56f9b1 Attempt to prevent false positives in replication test. 2014-11-24 11:55:16 +01:00
antirez b19e745780 scripting.tcl tests order fixed to match unstable branch. 2014-11-14 17:28:58 +01:00
antirez ce96069676 lua_cjson.c Lua includes: angled -> quoted. 2014-11-14 17:28:54 +01:00
antirez ead8e31b28 Fix non-linux builds error introduced with THP checks. 2014-11-14 17:28:49 +01:00
Matt Stancliff 5d8c88fdc1 Lua: add cmsgpack scripting tests 
Basically: test to make sure we can load cmsgpack
and do some sanity checks to make sure pack/unpack works
properly.  We also have a bonus test for circular encoding
and decoding because I was curious how it worked.
 2014-11-14 17:28:45 +01:00
Matt Stancliff 381863be5c Lua: upgrade cmsgpack to 0.4.0 
Main reasons for upgrade:
  - Remove a warning when building Redis
  - Add multi pack/unpack
  - Improve memory usage and use Lua allocator properly
  - Fix some edge case encoding/decoding bugs
 2014-11-14 17:28:40 +01:00
Matt Stancliff a7b58f1569 Lua: remove new warning added by cjson header 
clang doesn't like "extern inline" when no definition
is given right away.
 2014-11-14 17:28:35 +01:00
Matt Stancliff 56e1861398 Lua: Use Redis solaris compatability for cjson too 
cjson calls isinf, but some Solaris versions don't have isinf
even with the attempted fix we have in deps/Makefile.

We can harmlessly include the Redis solarisfixes.h header to
give cjson proper isinf.

Note: cjson has a compile-time setting for using their own defined
isinf, but the Redis definition in solarisfixes.h is more complete.

Fixes antirez#1620
 2014-11-14 17:28:32 +01:00
Matt Stancliff e2983dd03f Lua: Upgrade cjson to 2.1.0 (2012-03-01) 
The new cjson has some improvements over our current version including
increased platform compatability, a new resource limit to restrict
decode depth, and better invalid number handling.

One minor change was required to deps/Makefile because this version
of cjson doesn't export itself globally, so we added a quick little
define of -DENABLE_CJSON_GLOBAL.

cjson now has an optional higher performing float parsing interface,
but we are not including it (g_fmt.c, dtoa.c) because it requires
endianness declaration during compile time.

This commit is exactly lua_cjson.c from 2.1.0 with one minor
change of altering the two Lua includes for local search
instead of system-wide importing.
 2014-11-14 17:28:27 +01:00
Matt Stancliff 2828bbd511 Lua: add cjson scripting test 
Two simple decode tests added mainly to check that
the 'cjson' global gets registered and is usable.
 2014-11-14 17:28:23 +01:00
Matt Stancliff 1e501d9fd6 Lua: Add bitop 
A few people have written custom C commands because bit
manipulation isn't exposed through Lua.  Let's give
them Mike Pall's bitop.

This adds bitop 1.0.2 (2012-05-08) from http://bitop.luajit.org/

bitop is imported as "bit" into the global namespace.

New Lua commands: bit.tobit, bit.tohex, bit.bnot, bit.band, bit.bor, bit.bxor,
bit.lshift, bit.rshift, bit.arshift, bit.rol, bit.ror, bit.bswap

Verification of working (the asserts would abort on error, so (nil) is correct):
127.0.0.1:6379> eval "assert(bit.tobit(1) == 1); assert(bit.band(1) == 1); assert(bit.bxor(1,2) == 3); assert(bit.bor(1,2,4,8,16,32,64,128) == 255)" 0
(nil)
127.0.0.1:6379> eval 'assert(0x7fffffff == 2147483647, "broken hex literals"); assert(0xffffffff == -1 or 0xffffffff == 2^32-1, "broken hex literals"); assert(tostring(-1) == "-1", "broken tostring()"); assert(tostring(0xffffffff) == "-1" or tostring(0xffffffff) == "4294967295", "broken tostring()")' 0
(nil)

Tests also integrated into the scripting tests and can be run with:
./runtest --single unit/scripting

Tests are excerpted from `bittest.lua` included in the bitop distribution.
 2014-11-14 17:28:19 +01:00
unknown eb88474985 issue 181: config file ignored when --service-name argument passed 2014-11-13 16:41:51 -08:00
antirez d9326cb83d THP detection for LATENCY DOCTOR. 2014-11-12 11:17:31 +01:00
antirez 011cfdfb6b Check THP support at startup and warn about it. 2014-11-12 11:17:27 +01:00
antirez c0266d87ae THP detection / reporting functions added. 2014-11-12 11:17:24 +01:00
antirez a5fcf44fcc Diskless SYNC: fix RDB EOF detection. 
RDB EOF detection was relying on the final part of the RDB transfer to
be a magic 40 bytes EOF marker. However as the slave is put online
immediately, and because of sockets timeouts, the replication stream is
actually contiguous with the RDB file.

This means that to detect the EOF correctly we should either:

1) Scan all the stream searching for the mark. Sucks CPU-wise.
2) Start to send the replication stream only after an acknowledge.
3) Implement a proper chunked encoding.

For now solution "2" was picked, so the master does not start to send
ASAP the stream of commands in the case of diskless replication. We wait
for the first REPLCONF ACK command from the slave, that certifies us
that the slave correctly loaded the RDB file and is ready to get more
data.
 2014-11-11 17:27:22 +01:00
antirez ec1ab0f18b Disconnect timedout slave: regression introduced with diskless repl. 2014-11-11 17:25:58 +01:00
unknown 9acbc2cc68 issue 182: provide docx file content as github markdown for easier reading for those without a .docx file viewer. 2014-11-08 23:02:51 -08:00
unknown 0287e7da15 binaries for issue 170 2014-11-05 15:39:56 -08:00
unknown 47406f9591 issue 170: Patching part of the LP64/LLP64 ABI issue in order to ease immediate replication pain. 2014-11-05 15:39:34 -08:00
Matt Stancliff eb67ca0f00 redis-cli: ignore SIGPIPE network errors 
Closes #2066
 2014-10-30 12:02:32 +01:00
antirez bb9ad32cb0 Fix DEBUG POPULATE warning for lack of casting. 2014-10-29 14:36:55 +01:00
antirez a37f8ec8b4 DEBUG POPULATE two args form implemented. 
The old DEBUG POPULATE form for automatic creation of test keys is:

    DEBUG POPULATE <count>

Now an additional form is available:

    DEBUG POPULATE <count> <prefix>

When prefix is not specified, it defaults to "key", so the keys are
named incrementally from key:0 to key:<count-1>. Otherwise the specified
prefix is used instead of "key".

The command is useful in order to populate different Redis instances
with key names guaranteed to don't collide. There are other debugging
uses, for example it is possible to add additional N keys using a count
of N and a random prefix at every call.
 2014-10-29 14:36:55 +01:00
antirez b2509d7759 redis-cli: add missing newline in error message. 2014-10-29 14:35:48 +01:00
antirez 9b7a190c82 Diskless replication: missing listRewind() added. 
This caused BGSAVE to be triggered a second time without any need when
we switch from socket to disk target via the command

    CONFIG SET repl-diskless-sync no

and there is already a slave waiting for the BGSAVE to start.
Also comments clarified about what is happening.
 2014-10-29 14:34:58 +01:00
antirez 49629a2e34 Log slave ip:port in more log messages. 2014-10-29 14:34:58 +01:00
antirez 081eeb9f6b Use new slave name function for diskless repl reporting. 2014-10-29 14:34:58 +01:00
antirez 72ea77af0b Added a function to get slave name for logs. 2014-10-29 14:34:51 +01:00
antirez 16319c69b1 Diskless replication: log BGSAVE delay only when it is non-zero. 2014-10-29 14:34:25 +01:00
antirez bafd3f646f Document repl-diskless-sync-delay in redis.conf. 2014-10-29 14:34:25 +01:00
antirez f71c4d54d1 Diskless sync delay is now configurable. 2014-10-29 14:34:18 +01:00
antirez 4b520177b3 Remove duplicated log message about starting BGSAVE. 2014-10-29 14:33:50 +01:00
antirez 7e8a471baf Diskless replication documented inside example redis.conf. 2014-10-29 14:33:50 +01:00
antirez d57ed95222 Diskless replication tested with the multiple slaves consistency test. 2014-10-29 14:33:50 +01:00
antirez 7d32dfc69f Diskless replication: child -> parent communication improved. 
Child now reports full info to the parent including IDs of slaves in
failure state and exit code.
 2014-10-29 14:33:50 +01:00
antirez 7728ee809d Translate rio fdset target EWOULDBLOCK error into ETIMEDOUT. 
EWOULDBLOCK with the fdset rio target is returned when we try to write
but the send timeout socket option triggered an error. Better to
translate the error in something the user can actually recognize as a
timeout.
 2014-10-29 14:33:50 +01:00
antirez 7cffd74b38 Diskless replication: set / reset socket send timeout. 
We need to avoid that a child -> slaves transfer can continue forever.
We use the same timeout used as global replication timeout, which is
documented to also affect I/O operations during bulk transfers.
 2014-10-29 14:33:50 +01:00
antirez eff8fb6f39 anet.c: new API anetSendTimeout(). 2014-10-29 14:33:50 +01:00
antirez c38bb5b2ea Diskless replication: less debugging printfs around. 2014-10-29 14:33:50 +01:00
antirez 348c4f6ce3 rio.c fdset write() method fixed: wrong type for return value. 2014-10-29 14:33:50 +01:00
antirez a7d0137075 rio fdset target: handle short writes. 
While the socket is set in blocking mode, we still can get short writes
writing to a socket.
 2014-10-29 14:33:50 +01:00
antirez 23bf435f54 anet.c: API to set sockets back to blocking mode. 2014-10-29 14:33:50 +01:00
antirez 9a521e6c74 Diskless replication: rio fdset target new supports buffering. 
To perform a socket write() for each RDB rio API write call was
extremely unefficient, so now rio has minimal buffering capabilities.
Writes are accumulated into a buffer and only when a given limit is
reacehd are actually wrote to the N slaves FDs.

Trivia: rio lacked support for buffering since our targets were:

1) Memory buffers.
2) C standard I/O.

Both were buffered already.
 2014-10-29 14:33:50 +01:00
antirez 9588973fda Diskless replication: Various fixes to backgroundSaveDoneHandlerSocket() 2014-10-29 14:33:50 +01:00
antirez 8a618971bd Diskless replication: don't send "\n" pings to slaves. 
This is useful for normal replication in order to refresh the slave
when we are persisting on disk, but for diskless replication the
child is already receiving data while in WAIT_BGSAVE_END state.
 2014-10-29 14:33:50 +01:00
antirez b794b94797 Diskless replication: remove 40 bytes EOF mark from end of RDB file. 2014-10-29 14:33:50 +01:00
antirez 3f1f29fdf3 Diskless replication: swap inverted branches to compute read len. 2014-10-29 14:33:50 +01:00
antirez dbc6e9a1ef Diskless replication: don't enter the read-payload branch forever. 2014-10-29 14:33:49 +01:00
antirez adf55b2337 syncReadLine(): actually enforce buffer size limits. 2014-10-29 14:33:49 +01:00
antirez 89f82a25e1 Diskless replication: EOF:<mark> streaming support slave side. 2014-10-29 14:33:49 +01:00
antirez 583a762cd9 Diskless replication: redis.conf and CONFIG SET/GET support. 2014-10-29 14:33:49 +01:00
antirez 31b0e13268 Diskless replication: trigger a BGSAVE after a config change. 
If we turn from diskless to disk-based replication via CONFIG SET, we
need a way to start a BGSAVE if there are slaves alerady waiting for a
BGSAVE to start. Normally with disk-based replication we do it as soon
as the previous child exits, but when there is a configuration change
via CONFIG SET, we may have slaves in WAIT_BGSAVE_START state without
an RDB background process currently active.
 2014-10-29 14:33:49 +01:00
antirez d403e4ae63 Diskless replication flag renamed repl_diskless -> repl_diskless_sync. 2014-10-29 14:33:49 +01:00
antirez 2f0b58e4ed Diskless replication: trigger diskless RDB transfer if needed. 2014-10-29 14:33:49 +01:00
antirez 82bfae5b70 Diskless replication: handle putting the slave online. 2014-10-29 14:33:49 +01:00
antirez 132c108f17 Diskless replication: read report from child. 2014-10-29 14:33:49 +01:00
antirez e82ec2c5d8 Diskless replication: child writes report to parent. 2014-10-29 14:33:49 +01:00
antirez 2ed67e4a9e rio.c fdset target: tolerate (and report) a subset of FDs in error. 
Fdset target is used when we want to write an RDB file directly to
slave's sockets. In this setup as long as there is a single slave that
is still receiving our payload, we want to continue sennding instead of
aborting. However rio calls should abort of no FD is ok.

Also we want the errors reported so that we can signal the parent who is
ok and who is broken, so there is a new set integers with the state of
each fd. Zero is ok, non-zero is the errno of the failure, if avaialble,
or a generic EIO.
 2014-10-29 14:33:49 +01:00
antirez 51864fcdd7 Diskless replication: parent-child pipe and a few TODOs. 2014-10-29 14:33:49 +01:00
antirez 1b4cadb664 Diskless replication: RDB -> slaves transfer draft implementation. 2014-10-29 14:33:49 +01:00
antirez 38dee9dc51 rio.c: draft implementation of fdset target implemented. 2014-10-29 14:33:49 +01:00
antirez 5d295fea12 rio.c refactoring before adding a new target. 2014-10-29 14:33:49 +01:00
antirez aef4c60c78 Add some comments in syncCommand() to clarify RDB target. 2014-10-29 14:33:49 +01:00
antirez 81abd1e178 Define different types of RDB childs. 
We need to remember what is the saving strategy of the current RDB child
process, since the configuration may be modified at runtime via CONFIG
SET and still we'll need to understand, when the child exists, what to
do and for what goal the process was initiated: to create an RDB file
on disk or to write stuff directly to slave's sockets.
 2014-10-29 14:33:49 +01:00
antirez 1e633f21c5 RDB file creation refactored to target non-disk target. 2014-10-29 14:33:49 +01:00
antirez ff8a3baaf7 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.
 2014-10-29 14:33:30 +01:00
Jonathan Pickett 29fa2f09a4 Replaced WIN32_IOCP guard around QueryPerformnaceCounter code with _WIN32. Updating binary package. 2014-10-28 22:01:00 -07:00
Jonathan Pickett 063be1b6be Merge pull request #186 from deepakverma/2.8 
Using QueryPerformancecounter to measure latency in redis-benchmark.exe
 2014-10-28 21:03:53 -07:00
deepakv 7c6644442b Using QueryPerformancecounter to measure latency in redis-benchmark.exe 
Redis-benchmark on windows seems to be not able to calculate the time granularity in micro seconds.
     Came across this article http://msdn.microsoft.com/en-us/library/windows/desktop/dn553408(v=vs.85).aspx and updated the benchmark for latency measurements accordingly.

     Windows:
     Following is output from windows for micro seconds taken for each request. Where for any request taking less than 1 ms is being calculated as 0
     redis-benchmark.exe -h <azurerediscache> -a pwd -c 1 -d 1024 -t get -n 10
     0,0,0,0,0,0,0,0

     Linux:
     src/redis-benchmark -h <azurerediscache> -a pwd -c 1 -d 1024 -t get -n 10
     792,800,808,820,837,845,888,891,971,2498

     After the change on windows now it shows following:
     796,809,887,890,893,943,944,974
 2014-10-23 12:10:18 -07:00
antirez b6203138d3 Linenoise README updated to match source code. 2014-10-06 10:02:08 +02:00
Miguel Parramon 85b15acb44 redis.conf small grammar fix 
😄

Closes #2034
 2014-10-06 10:02:08 +02:00
Ezequiel Lovelle 09fc0fbbe4 Fix typo 
Closes #2029
 2014-10-06 10:02:08 +02:00
Gregory Petrosyan 03b36eef72 Fix typos in comments 
Closes #2002
 2014-10-06 10:01:36 +02:00
Matt Stancliff e1ccd7243f Add missing 'by' 2014-10-06 10:01:36 +02:00
Juarez Bochi 364aeed96d Fix typo in scripting.c comment 
Closes #1960
 2014-10-06 10:01:36 +02:00
T.J. Schuck a7a2f69bdb Fix typo radis -> redis 
Closes #1938
 2014-10-06 10:01:36 +02:00
Matt Stancliff 1adc4a7888 Cleanup wording of dictScan() comment 
Some language in the comment was difficult
to understand, so this commit: clarifies wording, removes
unnecessary words, and relocates some dependent clauses
closer to what they actually describe.

I also tried to break up longer chains of thought
(if X, then Y, and Q, and also F, so obviously M)
into more manageable chunks for ease of understanding.
 2014-10-06 10:01:36 +02:00
Matt Stancliff c3375fbdff Remove trailing spaces from tests 2014-10-06 10:01:27 +02:00
Matt Stancliff c4d32c03d0 Fix spelling in some test cases 2014-10-06 10:01:05 +02:00
Jan-Erik Rediger 919a4c3126 Fix sentinel.conf typo 
Closes #1739
 2014-10-06 10:00:48 +02:00
Xiaojie Zhang 8a1d16fe5c Fix comment inconsistencies in ziplist.c 
Closes #1523
 2014-10-06 10:00:48 +02:00
Matt Stancliff e0ffc863c6 Clean up text throughout project 
- Remove trailing newlines from redis.conf
  - Fix comment misspelling
  - Clarifies zipEncodeLength usage and a C API mention (#1243, #1242)
  - Fix cluster typos (inspired by @papanikge #1507)
  - Fix rewite -> rewrite in a few places (inspired by #682)

Closes #1243, #1242, #1507
 2014-10-06 10:00:43 +02:00
Matt Stancliff 22e6f434f8 Spell software correctly 2014-10-06 09:59:52 +02:00
Aaron Rutkovsky 3101760937 Fix typos 
Closes #1513
 2014-10-06 09:59:02 +02:00
Ben d3005539c5 Grammar and typo fixes in redis.conf 
Closes #1441
 2014-10-06 09:58:56 +02:00
xuxiang 3c99e429cf Fix typo in redis.h 
Closes #1386
 2014-10-06 09:58:18 +02:00
Ted Nyman 55597bbd67 Fix on-place -> in-place 
Closes #1373
 2014-10-06 09:58:18 +02:00
Michael Parker a5044c254d Fix hash table size in comment for dictScan 
Closes #1351
 2014-10-06 09:58:18 +02:00
Jan-Erik Rediger 33daf8e9a9 Fix many small typos 
Closes #1871
 2014-10-06 09:58:11 +02:00
zionwu b72422e100 Fix incorrect comments 
error != success; and 0 != number of bytes written

Closes #1806
 2014-10-06 09:57:01 +02:00
Jan-Erik Rediger 6cdd38f947 Fix typo in redis.conf 
Closes #1713
 2014-10-06 09:57:01 +02:00
Juarez Bochi 6b64fa49f8 Fix typo 
Closes #1682
 2014-10-06 09:57:01 +02:00
Agis Anastasopoulos fc9bce45c5 Fix typo 
Closes #1544
 2014-10-06 09:57:01 +02:00
Jan-Erik Rediger 575e293ddd Fix typo: ad -> and 
Closes #1537
 2014-10-06 09:57:01 +02:00
Matt Stancliff d05fcaffa0 Update linenoise accessory files 
The core linenoise code was being backported, but not
the README or example.  It's less confusing for users
if everything matches across directories.

Fix inspired by @thrig

Closes #1872
 2014-10-06 09:57:01 +02:00
Manuel Meurer a99d187bc4 Improve disable save comment in redis.conf 
Closes #1897
 2014-10-06 09:57:01 +02:00
Lynn 904bd0400c Remove surplus double quotes 
Closes #1877
 2014-10-06 09:57:01 +02:00
jonathan pickett 18d1138e75 redis-server not being built in debug because of project configuration manager setting 2014-10-01 15:46:58 -07:00
jonathan pickett 450fd83a9a I found a case where the temp-rewriteaof-old-xxxx.aof files were not being deleted. Added the _O_TEMPORARY flag to the open() call to guarantee that the underlying CreateFile call is being called with the FILE_FLAG_DELETE_ON_CLOSE. 2014-10-01 15:43:31 -07:00
jonathan pickett bd27d3bd9b Restoring the ability to compile without IOCP. To do so remove the WIN32_IOCP definition in the C/C++\Preporcessor\Preprocessor Definitions in all projects. Performance is 30-40% better with IOCP. 2014-09-23 15:28:38 -07:00
antirez 04b6945ebc Linenoise lib updated. 2014-09-22 14:56:01 +02:00
jonathan pickett a9e3429538 issue 169: missed adding app.config 2014-09-21 13:04:33 -07:00
jonathan pickett 51cac40c73 Added a project to take current redis version number in src\version.h and propagate this number into the release documentation and the name of the release zip file. This replaces the powershell script used previously to build the binary release package zip. 2014-09-19 17:12:12 -07:00
jonathan pickett 96898fa51c Merge branch '2.8antirez' into 2.8 2014-09-19 12:09:05 -07:00
antirez ade6c5e782 Redis 2.8.17 2014-09-19 17:49:16 +02:00
Matt Stancliff cf143b571e Fix hiredis getaddrinfo leak 
Fixed in Redis by 1a5e5b6, but since that part of code
is largely copy/paste from Redis, the fix needs to be
ported over too.

Closes #2012
 2014-09-19 14:21:36 +02:00
jonathan pickett c7b6d311df finishing 2.8.15-->2.8.16 version update 2014-09-16 18:53:49 -07:00
jonathan pickett 18534064c9 merge antirez changes 2.8.15-->2.8.16 2014-09-16 17:05:36 -07:00
antirez a69534967b Redis 2.8.16. 2014-09-16 15:15:56 +02:00
antirez 5e38bc389c Better truncated AOF loading tests. 
Now there are tests to write more data after loading a truncated AOF,
testing that the loaded data is correct, appending more, and testing
again.
 2014-09-16 11:05:50 +02:00
antirez 01f7db608d Seek at the end of AOF after truncate call. 
It is not clear if files open in append only mode will automatically fix
their offset after a truncate(2) operation. This commit makes sure that
we reposition the AOF file descriptor offset at the end of the file
after a truncated AOF is loaded and trimmed to the last valid command.
 2014-09-16 11:05:50 +02:00
antirez 9faeff0215 On AOF end of file, truncate the AOF to last valid command. 
Recently we introduced the ability to load truncated AOFs, but
unfortuantely the support was broken since the server, after loading the
truncated AOF, continues appending to the file that is corrupted at the
end. The problem is fixed only in the next AOF rewrite.

This commit fixes the issue by truncating the AOF to the last valid
opcode, and aborting if it is not possible to truncate the file
correctly.
 2014-09-16 10:41:15 +02:00
antirez fbd9dc6039 Don't propagate SAVE. 
This is a general fix (check that dirty delta is positive) but actually
should have as the only effect fixing the SAVE propagation to
AOF and slaves.
 2014-09-16 10:41:15 +02:00
jonathan pickett 915c0dd656 Merge changes from Posix Redis through 2.8.15 into MSOpenTech branch 2014-09-15 18:10:10 -07:00
antirez 551ffc486c version.h updated to 2.8.15. 2014-09-12 16:16:58 +02:00
antirez 78dca8977f Redis 2.8.15. 2014-09-12 16:15:48 +02:00
antirez a6695e341b Sentinel sentinelGetLeader() top comment improved. 2014-09-11 19:28:10 +02:00
antirez 9fe4429a09 Sentinel: fix computation of total number of votes. 
The code to check the number of voters was never updated to follow the new
Sentinel specification, so the number of voters was computed using only
the set of Sentinels that provided a vote.

This means that there is a changing majority on partitions, even if
usually the issue is not triggered because of the configured quorum
check (what was broken was the other implicit check that requires anyway
half of the known sentinels to agree in order to start a failover).
 2014-09-11 19:28:09 +02:00
antirez 6fbcb19275 luaRedisGenericCommand() cached argv handling simplified. 
As discussed in issue #1945.
 2014-09-10 15:47:02 +02:00
xiaost 8375a6d0aa Limit the *SCAN command dictScan iterations 
*SCAN will cause redis server to hang for seconds
after millions of keys was deleted by SCAN/DEL pairs
 2014-09-10 09:58:20 +02:00
antirez 55024e9a9a Sentinel: don't set announce-ip if is empty. 2014-09-10 09:26:29 +02:00
antirez 3cc31da349 Sentinel: clarify announce-ip/port options in sentinel.conf. 2014-09-10 09:26:29 +02:00
antirez 8c3507aa8f Sentinel: announce ip/port changes + rewrite. 
The original implementation was modified in order to allow to
selectively announce a different IP or port, and to rewrite the two
options in the config file after a rewrite.
 2014-09-10 09:26:29 +02:00
Dara Kong 9f3abf436b sentinel: Decouple bind address from address sent to other sentinels 
There are instances such as EC2 where the bind address is private
(behind a NAT) and cannot be accessible from WAN.

https://groups.google.com/d/msg/redis-db/PVVvjO4nMd0/P3oWC036v3cJ
 2014-09-10 09:26:29 +02:00
jonathan pickett 874c21d801 issue 166: master/slave sync with RDB > 2G. cause: off_t defined as long. longs are 4-bytes on x64 Windows. Redefined impacted vars as int64_t. 2014-09-09 19:30:36 -07:00
antirez ff47d5ebf6 Tests for aof-load-truncated = yes. 2014-09-08 10:58:00 +02:00
antirez b3fdddfd1f AOF tests fixed turning aof-load-truncated to no. 
When aof-load-truncated option was introduced, with a default of "yes",
the past behavior of the server to abort with trunncated AOF changed, so
we need to explicitly configure the tests to abort with truncated AOF
by setting the option to no.
 2014-09-08 10:58:00 +02:00
antirez 670084b724 Make aof-load-truncated option actually configurable. 2014-09-08 10:58:00 +02:00
antirez 03a5c0ecb5 Correct cleanup when aborting AOF loading. 
Because of the new ability to start with a truncated AOF, we need
to correctly release all the memory on EOF error. Otherwise there is a
small leak, that is not really a problem, but causes a false positive in
the tests that detect memory leaks.
 2014-09-08 10:58:00 +02:00
antirez 683cf5b872 AOF ability to load truncated files. 2014-09-08 10:57:57 +02:00
antirez 94f7979414 Test AOF format error detection. 2014-09-08 10:57:11 +02:00
antirez 537f3053fa AOF loading: split handling of format errors from unexpected EOF. 2014-09-08 10:57:10 +02:00
jonathan pickett 763d7b7c6a updated release notes for persistence-available flag 2014-09-06 14:46:26 -07:00
jonathan pickett 69a5ea090a 2.8.14 binaries with persistence-available flag 2014-09-06 14:38:11 -07:00
jonathan pickett b001508333 Merge branch '2.8' of https://github.com/MSOpenTech/redis into 2.8 2014-09-06 14:23:46 -07:00
jonathan pickett 0c3a93b8a2 adding persistence-available flag 2014-09-06 14:23:36 -07:00
jonathan pickett c9c6ac2056 updated release package, nuspecs, and docs 2014-09-06 14:14:04 -07:00
jonathan pickett 1688851b13 Merge antirez 2.8.14 into MSOPENTech 2.8 branch 
Conflicts:
	src/ae.c
	src/bitops.c
	src/dict.c
	src/fmacros.h
	src/hyperloglog.c
	src/networking.c
	src/redis-check-dump.c
	src/redis.c
	src/sds.c
	src/t_string.c
	src/t_zset.c
	src/util.c
 2014-09-06 13:44:19 -07:00
Matt Stancliff 667e2991dc Return empty string if GETRANGE of empty string 
Previously, GETRANGE of a key containing nothing ("")
would allocate a large (size_t)-1 return value causing
crashes on 32bit builds when it tried to allocate the
4 GB return string.
 2014-09-04 11:47:21 +02:00
Matt Stancliff 4284b07220 Increase size of range request in getrange 
32 bit builds don't have a big enough long to capture
the same range as a 64 bit build.  If we use "long long"
we get proper size limits everywhere.

Also updates size of unsigned comparison to fit new size of `end`.

Fixes #1981
 2014-09-04 11:46:57 +02:00
Kirk Shoop a2119ce3a5 appveyor config 2014-09-04 00:22:25 -07:00
jonathan pickett 8bea36f67b forgot to add the 2.8.13 binary to the last commit 2014-09-03 16:06:00 -07:00
Kirk Shoop e5328c06b4 add nuget and chocolatey badges 2014-09-03 07:55:46 -07:00
jonathan pickett 0e03891633 updated release notes for 2.8.13 and updated name of pre-compiled binaries zip package 2014-09-02 20:06:06 -07:00
jonathan pickett 92e96ad66b Merge branch '2.8' of https://github.com/MSOpenTech/Redis into 2.8 2014-09-02 19:54:05 -07:00
antirez db6e874a25 Redis 2.8.14. 2014-09-01 17:00:38 +02:00
Matt Stancliff dd36f18c76 Sentinel: Abort Hello quicker if not connected 
We can save a little work by aborting when we enter the function
if we're disconnected.
 2014-09-01 16:35:04 +02:00
Paddy Byers d1f89bdc2b Add regression test for issue #1939 2014-09-01 10:41:15 +02:00
Paddy Byers 59ad68ca4f Store the length of the static argv when first allocated. 2014-09-01 10:40:04 +02:00
Kirk Shoop (MS OPEN TECH) 95f845b691 add script to push nupkgs 2014-08-29 10:45:54 -07:00
antirez 8fb8a474fd Fix dictRehash assert casting type. 
Also related to #1929.
 2014-08-27 10:31:52 +02:00
antirez 85b3db4065 Cast to right type in dictNext(). 
This closes issue #1929, the other part was fixed in the context of issue
 2014-08-27 10:31:48 +02:00
antirez 3a706ba792 Use long for rehash and iterator index in dict.h. 
This allows to support datasets with more than 2 billion of keys
(possible in very large memory instances, this bug was actually
reported).

Closes issue #1814.
 2014-08-27 10:31:44 +02:00
antirez 97fdcd8746 Missing va_end also added in hiredis copy of sds.c. 2014-08-27 10:31:41 +02:00
Mariano Pérez Rodríguez 45fac9481a Missing assert removal from sdsIncrLen() 
Companion for 8eeb1802ec, but dealing with hiredis.
 2014-08-27 10:31:35 +02:00
antirez c3eac3ddff No longer useful assert removed from sdsIncrLen(). 2014-08-27 10:31:32 +02:00
Matt Stancliff fdcadd16c6 Fix redis-benchmark abort output 2014-08-27 10:31:29 +02:00
Matt Stancliff ac48aada30 Deny CLIENT command in scripts 
We don't want scripts doing CLIENT SETNAME
or CLIENT KILL or CLIENT LIST or CLIENT PAUSE.

Originally reported by Chris Wj then proper
action inspired by Itamar Haber.

Reference: https://groups.google.com/forum/#!topic/redis-db/09B2EYwyVgk
 2014-08-27 10:31:20 +02:00
Matt Stancliff abe29c83e6 redis-benchmark: Remove unused buffer 
This just deletes old code that didn't get removed when
logic changed.  We were setting offsets that never
got read anywhere.

Since clients are now just cloned, we don't need to track
per-client buffer offsets anywhere because they are all
the same from the original client.
 2014-08-27 10:30:49 +02:00
CharSyam 8fcce90b25 redis-benchmark: add auth option 
Closes #1097
 2014-08-27 10:30:49 +02:00
Matt Stancliff 4ac8472cac Rename two 'buf' vars to 'ip' for better clarity 
Clearly ip[32] is wrong, but it's less clear that buf[32] was wrong
without further reading.
 2014-08-27 10:30:42 +02:00
Eiichi Sato f7b5e2d11e Sentinel: fix bufsize to support IPv6 address 
Closes #1914
 2014-08-27 10:30:37 +02:00
Jan-Erik Rediger 32a94beca4 Use correct github url to find commands.json 
Once this is merged:
  - merge the latest changes to commands.json in antirez/redis-doc
  - re-run: utils/generate-command-help.rb > src/help.h

Then we'll have nice and easy tab-completed help in redis-cli again.

Closes #1909
 2014-08-27 10:30:34 +02:00
Mariano Pérez Rodríguez f862e5ddfa Stop tests from leaving a black background 
Uses ANSI "default background" color code after closing tests
so any non-black terminals don't remain polluted.

Fixes #1649
Closes #1912
 2014-08-27 10:30:29 +02:00
Matt Stancliff 32168a5e8a Sentinel: Reject config from STDIN 
Sentinel needs to die with a more accurate error message
when attempted to open a config from STDIN.

See: https://groups.google.com/forum/#!topic/redis-db/sYx7VNMWaNM
 2014-08-27 10:30:26 +02:00
Matt Stancliff 4bf7c5d0f8 Remove unused global variable 
It has an important name, but nothing uses it...
 2014-08-27 10:30:22 +02:00
antirez 33b37b1400 Fix invalid expire error for SET family commands. 2014-08-27 10:30:19 +02:00
Cong Ding ae2cd2e16f Remove unused function 
Closes #878
 2014-08-27 10:30:11 +02:00
David Palm 8c0743d80e Extend range of bytesToHuman to TB and PB 
Also adds a fallthrough case for when given
large values (like overflow numbers of 2^64 by mistake).

Closes #858
 2014-08-27 10:30:08 +02:00
yoav ef12b4c3eb Add error check for writing RDB checksum 
Closes #857
 2014-08-27 10:30:04 +02:00
Dov Murik 1580a378b8 redis-cli: fix prompt after shutdown command 
Fix redis-cli prompt to state "not connected" after a SHUTDOWN command
is sent.
 2014-08-27 10:30:01 +02:00
xiaoyu 467848460a Clarify argument to dict macro 
d is more clear because the type of argument is dict not dictht

Closes #513
 2014-08-27 10:29:57 +02:00
antirez 5e707c66dd Fix undefined behavior in ll2string(). 
The bug was found by @CAFxX, thanks!
See issue #1940.
 2014-08-27 10:29:53 +02:00
antirez 65d47452f8 Remove warnings and improve integer sign correctness. 2014-08-27 10:29:26 +02:00
antirez 09757420a4 Add -W to compilation flags. 2014-08-27 10:25:57 +02:00
antirez 134345e332 Use unsigned integers in SDS header. 
This raises the max string to 4GB without any downside.
 2014-08-27 10:25:54 +02:00
Pieter Noordhuis f89a7bbe11 va_copy must be matched by va_end 
Hat tip to @rfuchs. See: https://github.com/redis/hiredis/pull/178.

Fixes #1187
 2014-08-27 10:25:44 +02:00
charsyam dbcf381099 Remove duplicate prototypes in redis.h 
Also moves acceptHandler() to be near the other accept...() functions.

Closes #1105
 2014-08-27 10:25:40 +02:00
NanXiao 6bc74b4393 Use resolv library in Solaris 
For some Solaris flavours, the inet_aton in in resolv library.
Not linking this library will introduce link error.

Improves compatability with older Solaris and still
works on new Solaris.

Closes #1092
 2014-08-27 10:25:37 +02:00
Daniel Price 1aff711d90 Define AR to help with lua cross-compilation 
Closes #997
 2014-08-27 10:25:34 +02:00
Matt Stancliff 866f6bcc80 Cleanup double semicolons 
Closes #1161
 2014-08-27 10:25:31 +02:00
Chris Lamb 29ed617ac4 redis-check-dump: Prevent segfault if can't malloc 
Found by The Mayhem Team (Alexandre Rebert, Thanassis Avgerinos,
Sang Kil Cha, David Brumley, Manuel Egele) Cylab, Carnegie Mellon
University. See http://bugs.debian.org/716259 for more.

Signed-off-by: Chris Lamb <lamby@debian.org>

Fixes #1191
 2014-08-27 10:25:28 +02:00
charsyam a43e45e391 Remove unused LINE_BUFLEN definition 
Closes #1129
 2014-08-27 10:25:25 +02:00
Matt Stancliff 25a4699a9e Reject MOVE to non-integer DBs 
Previously, "MOVE key somestring" would move the key to
DB 0 which is just unexpected and wrong.
String as DB == error.

Test added too.

Modified by @antirez in order to use the getLongLongFromObject() API
instead of strtol().

Fixes #1428
 2014-08-27 10:25:22 +02:00
Matt Stancliff 09a3976478 pubsub: Return integers for NUMSUB, not strings 
Also adds test for numsub — due to tcl being tcl,
it doesn't capture the "numberness" of the fix,
but now we at least have one test case for numsub.

Closes #1561
 2014-08-27 10:25:15 +02:00
Xiaojie Zhang 60c748628b Avoid unnecessary decoding in ziplist.c 
Closes #1519
 2014-08-27 10:25:12 +02:00
Samuel Chou 96614f5866 Improve accuracy of HAVE_ATOMIC dependency check 
[I had to split out the clang check due to
 clang *really* not liking the __GLIBC_PREREQ macro; -matt]

Closes #1456
 2014-08-27 10:25:07 +02:00
Jim Howard d1df50ad59 Remove redundant event loop fd processing 
Reason:
https://github.com/antirez/redis/pull/1450#issuecomment-30432925

Closes #1450
 2014-08-27 10:25:04 +02:00
Xian Li 282c6106ce memtest: Add missing free() 
Closes #1614
 2014-08-27 10:25:00 +02:00
Joshua Anderson 25b5abe8cf src/Makefile: Reword "to run make test" message 
Closes #1610
 2014-08-27 10:24:56 +02:00
cubicdaiya a7c46df612 Use 'void' for zero-argument functions 
According to the C standard,
it is desirable to give the type 'void'
to functions have no argument.

Closes #1631
 2014-08-27 10:24:52 +02:00
Matt Stancliff 8dbd0134e5 Fix assert technical correctness 
dictAdd returns DICT_OK, not REDIS_OK. They both
have the same underlying values, so it works even though
the code is technically wrong.

Fixes #1512
 2014-08-27 10:24:46 +02:00
Jan-Erik Rediger 0c8999dfeb install_server.sh: add missing bang 
This was discovered by _bodya and reported in the IRC channel.
Everything worked fine as these scripts are always executed as shell
scripts.

Closes #1728
 2014-08-27 10:24:42 +02:00
Edgars Irmejs 0f9ab49d58 Change unixsocketperm comment to 700 from 755 
According to unix manuals, "Connecting to the socket object requires
read/write permission." -- mode 755 is useless for anybody
other than the owner.

Fixes #1696
 2014-08-27 10:24:39 +02:00
ripcurld00d 4be50f7fbb redis-check-dump: use names instead of magic nums 
Use constants to avoid magic numbers in `types`, which is an array
that stores the names of `REDIS` types.

Closes #1681
 2014-08-27 10:24:36 +02:00
Kashif Rasul 8fc2247c9b Fix issues raised by clang analyzer 
Modified by @antirez since the original fix to genInfoString() looked
weak. Probably the clang analyzer complained about `section` being
possibly NULL, and strcasecmp() called with a NULL pointer. In the
practice this can never happen, still for the sake of correctness
the right fix is not to modify only the first call, but to set `section`
to the value of "default" if it happens to be NULL.

Closes #1660
 2014-08-27 10:24:33 +02:00
antirez ebd25710dd Force quit when receiving a second SIGINT. 
Also quit ASAP when we are still loading a DB, since care is not needed
in this special condition, especially for a SIGINT.
 2014-08-27 10:24:29 +02:00
Matt Stancliff 524b2626f7 Add graceful exit when Ctrl-C is received 2014-08-27 10:24:26 +02:00
Matt Stancliff 78bde472c0 Fix intset midpoint selection 
The classic (min+max)/2 is provably unsafe.  Fixed
as recommended in research:
http://googleresearch.blogspot.com/2006/06/extra-extra-read-all-about-it-nearly.html

Fix inspired by @wjin, but I used a different approach.
(later, I found @kuebler fixed the same issue too).

Fixes #1741, #1602
 2014-08-27 10:24:22 +02:00
Jan-Erik Rediger 07b2779bff Handle large getrange requests 
Previously the end was casted to a smaller type
which resulted in a wrong check and failed
with values larger than handled by unsigned.

Closes #1847, #1844
 2014-08-27 10:24:19 +02:00
Matt Stancliff 4024220fe6 scripting: no eval with negative key count 
Negative key count causes segfault in Lua functions.

Fixes #1842
Closes #1843
 2014-08-27 10:24:15 +02:00
Jan-Erik Rediger 50669f7b66 redis-cli: fix latency result output 
(Cleaned up a little by @mattsta)

Closes #1774
 2014-08-27 10:24:11 +02:00
Wei Jin 9b53f79e00 Remove redundant else/return block 
Fixes #1741
 2014-08-27 10:24:08 +02:00
antirez 9840df7bad Clarify AIX "#undef hz". 2014-08-27 10:22:59 +02:00
siahl b5dc2c5b91 Add support for compiling on AIX 
Closes #1900
 2014-08-27 10:22:28 +02:00
Matt Stancliff cbd1607c6a redis-cli: Add --no-raw option 
Some people need formatted output even when they have no
interactive tty.

Fixes #760
 2014-08-27 10:21:19 +02:00
Matt Stancliff 9a039f76c2 redis-cli: stop showing incorrectly selected DB 
Previously redis-cli would happily show "-1" or "99999"
as valid DB choices.

Now, if the SELECT call returned an error, we don't update
the DB number in the CLI.

Inspired by @anupshendkar in #1313

Fixes #566, #1313
 2014-08-27 10:21:19 +02:00
Matt Stancliff 28b54b5d94 redis-cli: Re-attach selected DB after auth 
Previously, if you did SELECT then AUTH, redis-cli
would show your SELECT'd db even though it didn't
happen.

Note: running into this situation is a (hopefully) very limited
used case of people using multiple DBs and AUTH all at the same
time.

Fixes antirez#1639
 2014-08-27 10:21:19 +02:00
jonathan pickett ff6aee6fa3 merge for 2.8.13 2014-08-19 17:46:27 -07:00
jonathan pickett 79afb8e15e Merge branch '2.8' of https://github.com/MSOpenTech/Redis into 2.8 
Conflicts:
	src/aof.c
 2014-08-19 17:45:29 -07:00
jonathan pickett 6f4ce3f83e documenting problem discovered in issue 157 2014-08-19 13:27:16 -07:00
jonathan pickett c3358cc869 Merge branch '2.8' of https://github.com/MSOpenTech/Redis into 2.8 2014-08-19 12:38:13 -07:00
jonathan pickett 36f30412b5 issue 161: Memory allocator not compiled to be thread safe. Background AOF fsync code (bio.c) was allocating memory. This caused heap corruption. 2014-08-14 19:40:56 -07:00
jonathan pickett 16cefcf0b5 reworked COW join for Win8+ 2014-08-12 12:20:35 -07:00
Jonathan Pickett 6d364f41ce Merge pull request #156 from kenny-evitt/service-doc-minor-fixes 
Minor fixes to 'service' document
 2014-08-04 15:58:37 -07:00
jonathan pickett 31e9621f06 code review feedback on 432a38e53d7844a7bcf8131ddaa4c86d: moved heap memory mapped file into Redis sub directory under local app data folder 2014-07-31 17:23:13 -07:00
jonathan pickett 8d28cf7542 issue 155: temp-rewriteaof-old-xxxxx.aof files not being cleaned up 2014-07-31 15:53:00 -07:00
Kenny Evitt 33018d61a5 Fix text for "Starting the Service" section. 2014-07-30 15:26:03 -04:00
Kenny Evitt 96f3fc015a Remove blank heading lines. 2014-07-30 15:24:53 -04:00
jonathan pickett f52ce3c2c7 added heapdir directive to allow for configuration of where QFork memory mapped file lives 2014-07-29 15:41:41 -07:00
jonathan pickett 94e40d1034 Merge branch '2.8' of https://github.com/MSOpenTech/Redis into 2.8 2014-07-28 18:11:09 -07:00
jonathan pickett 8cb1d65616 Merge branch '2.8antirez' into 2.8: raise version from 2.8.12 to 2.8.13 
Conflicts:
	src/Makefile.dep
	src/aof.c
	src/redis-cli.c
	tests/instances.tcl
	tests/integration/replication.tcl
 2014-07-28 18:10:57 -07:00
jonathan pickett c6c93699cc Service install with a full path to the conf file was failing. Part of this was due to not passing quotes along with arguments with spaces. The other part was not adding access allowed ACEs on all accessed paths. 2014-07-28 15:15:09 -07:00
Yossi Gottlieb ff58d5a380 Fail SYNC if background save child aborted due to a signal. 2014-07-28 14:46:03 +02:00
Salvatore Sanfilippo eb7b3427e7 Merge pull request #1894 from veselov/2.8 
* fixed doc URL for keyspace events
 2014-07-28 09:38:33 +02:00
vps 4d7b3815d5 * fixed doc URL for keyspace events 2014-07-27 20:14:39 -07:00
antirez 9fa51dd4d1 Example redis.conf: improve slaveof description. 2014-07-25 14:20:31 +02:00
jonathan pickett fb12b5738d issue 154: [save ""] not working after fix for issue 153 2014-07-23 10:34:32 -07:00
antirez d558af8e6f Fix util.c compilation by including stdint.h. 2014-07-23 18:01:51 +02:00
antirez e426e77896 Faster ll2string() implementation. 
Based on ideas documented in this blog post:

https://www.facebook.com/notes/facebook-engineering/three-optimization-tips-for-c/10151361643253920

The original code was modified to handle signed integers, reformetted to
fit inside the Redis code base, and was stress-tested with a program
in order to validate the implementation against snprintf().

Redis was measured to be measurably faster from the point of view of
clients in real-world operations because of this change, since sometimes
number to string conversion is used extensively (for example every time
a GET results into an integer encoded object to be returned to the
user).
 2014-07-23 14:58:26 +02:00
antirez 8a44e10848 PFSELFTEST: less false positives. 
This is just a quickfix, for the nature of the test the right way to fix
it is to average the error of N runs, since otherwise it is always
possible to get a false positive with a bad run, or to minimize too much
this possibility we may end testing with too much "large" error ranges.
 2014-07-23 11:45:03 +02:00
jonathan pickett 9e191b4b58 setup changes in preparation for official Windows 2.8.12 release 2014-07-22 17:07:42 -07:00
jonathan pickett dbc7669fff issue 152: master DNS alias not resolving 2014-07-22 16:54:51 -07:00
jonathan pickett 5e05dad83a issue 153: command line parsing dying on imbalanced quotes in comments 2014-07-22 15:58:40 -07:00
antirez 0bfdd4bb49 Test: check sorted set elements order after union. 
This is not a regression but issue #1786 showed the need for this test.
 2014-07-22 17:58:41 +02:00
antirez 098bb97703 Remove useless var and check in zunionInterGenericCommand(). 2014-07-22 17:58:26 +02:00
antirez 316e1db970 ZUNIONSTORE reimplemented for speed. 
The user @kjmph provided excellent ideas to improve speed of ZUNIONSTORE
(in certain cases by many order of magnitude), together with an
implementation of the ideas.

While the ideas were sounding, the implementation could be improved both
in terms of speed and clearness, so that's my attempt at reimplementing
the speedup proposed, trying to improve by directly using just a
dictionary with an embedded score inside, and reusing the single-pass
aggregate + order-later approach.

Note that you can't apply this commit without applying the previous
commit in this branch that adds a double in the dictEntry value union.

Issue #1786.
 2014-07-22 17:58:05 +02:00
antirez c8a6cd94b3 Add double field in dict.c entry value union. 2014-07-22 17:53:06 +02:00
jonathan pickett 192b285b14 corrected the TCP keepalive code for WinSock 2014-07-21 17:35:00 -07:00
jonathan pickett d47adb19ac improved path parsing for included conf files 2014-07-21 17:33:12 -07:00
antirez 7b11dfe05e Test: Pub/Sub PING. 2014-07-18 12:03:38 +02:00
antirez 156029b336 Test: small integer sharing depends on maxmemory policy. 2014-07-18 10:55:19 +02:00
antirez 06e9b3cad8 PING: backward compatible error for wrong number of args. 2014-07-18 10:46:21 +02:00
antirez ab5f20d19c tryObjectEncoding(): use shared objects with maxmemory and non-LRU policy. 
In order to make sure every object has its own private LRU counter, when
maxmemory is enabled tryObjectEncoding() does not use the pool of shared
integers. However when the policy is not LRU-based, it does not make
sense to do so, and it is much better to save memory using shared
integers.
 2014-07-18 10:46:05 +02:00
antirez 70e39481ab Variadic PING with support for Pub/Sub. 
PING can now be called with an additional arugment, behaving exactly
like the ECHO command. PING can now also be called in Pub/Sub mode (with
one more more subscriptions to channels / patterns) in order to trigger
the delivery of an asynchronous pong message with the optional payload.

This fixes issue #420.
 2014-07-18 10:41:47 +02:00
antirez 294bcfc4d2 PubSub clients refactoring and new PUBSUB flag. 
The code tested many times if a client had active Pub/Sub subscriptions
by checking the length of a list and dictionary where the patterns and
channels are stored. This was substituted with a client flag called
REDIS_PUBSUB that is simpler to test for. Moreover in order to manage
this flag some code was refactored.

This commit is believed to have no effects in the behavior of the
server.
 2014-07-18 10:41:22 +02:00
jonathan pickett b7a045d747 Creation of QFork memory mapped file now respects the 'dir' .conf file directive. This allows for the QFork file to live in separate directory from the main executable. 2014-07-16 15:48:40 -07:00
jonathan pickett 4e71fc1c6f issue 145: there were a few places where our fwrite implementation was not used because of include file ordering 2014-07-15 13:13:04 -07:00
jonathan pickett 4df3af4e6e issue 146: changes to network layer between 2.8.9 and 2.8.12 broke IOCP async connect code 2014-07-14 19:57:21 -07:00
antirez 0c29cadd46 Redis 2.8.13. 2014-07-14 17:48:14 +02:00
michael-grunder 3df2ab673c Fix OBJECT arity 
Previously, the command definition for the OBJECT command specified
a minimum of two args (and that it was variadic), which meant that
if you sent this:

OBJECT foo

When cluster was enabled, it would result in an assertion/SEGFAULT
when Redis was attempting to extract keys.

It appears that OBJECT is not variadic, and only ever takes 3 args.

https://gist.github.com/michael-grunder/25960ce1508396d0d36a
 2014-07-14 15:38:32 +02:00
antirez c970507478 Fix typo in LATENCY DOCTOR output. 2014-07-11 10:57:34 +02:00
antirez bb10ad9c13 Test: more reliable AOF rewrite test under write load. 2014-07-10 16:42:52 +02:00
antirez c785d4cccc LATENCY command / monitor basic tests. 2014-07-10 16:24:24 +02:00
antirez e1fd73f634 LATENCY HISTORY returns empty array if no data is available. 
Previously we returned an error.
 2014-07-10 16:24:24 +02:00
antirez 315536d012 Test: enable latency monitor in the default config. 
This way as a side effect of running the test we also stress the latency
monitor data collection.
 2014-07-10 15:04:59 +02:00
antirez f0e67afd94 Test: AOF rewrite during write load. 2014-07-10 11:24:59 +02:00
jonathan pickett 128f5b53de issue 148: post build step fails when current directory is not in path 2014-07-09 20:54:10 -07:00
antirez 1f3c097b1c Fixed a few missing newline in createLatencyReport(). 2014-07-09 20:02:01 +02:00
antirez e4db34e97c createLatencyReport(): compile before commit avoids commits. 2014-07-09 20:01:57 +02:00
antirez f9765990b2 createLatencyReport(), fix mount option name. 2014-07-09 20:01:53 +02:00
antirez 54ca682887 createLatencyReport(): initialize all advices to zero. 2014-07-09 20:01:48 +02:00
antirez 95d880a7bf LATENCY DOCTOR first implementation complete. 2014-07-09 20:01:44 +02:00
antirez 2e581217d9 Fix analyzeLatencyForEvent() MAD computation. 2014-07-09 20:01:41 +02:00
antirez 121a9a3579 LATENCY DOCTOR: initial draft and events summary output. 2014-07-09 20:01:37 +02:00
antirez 3dd113bf33 Latency: low level time series analysis implemented. 2014-07-09 20:01:33 +02:00
antirez 3e8f5113fa LATENCY SAMPLES renamed LATENCY HISTORY. 2014-07-09 20:01:30 +02:00
antirez c145854bac LATENCY RESET implemented. 2014-07-09 20:01:27 +02:00
antirez 65dbb3546a redis-benchmark: abort when all clients are disconnected. 2014-07-09 20:01:18 +02:00
antirez 4c68319370 Use fsync instead of aof_fsync in final AOF sync. 
This happens in the child process so we don't care about latency:
better to sync metadata as well.
 2014-07-09 20:01:01 +02:00
antirez f3eab7a9b4 Fixed conditional for aof-write-pending-fsync latency event selection. 2014-07-09 20:00:57 +02:00
antirez f3c4111edd Dependencies updated. 2014-07-09 20:00:50 +02:00
antirez 07b4d33734 Cast void* to char* to avoid waring in latencyCommand(). 2014-07-09 19:58:06 +02:00
antirez 58fb90fdac Properly initialize min/max in latency.c. 2014-07-09 19:57:59 +02:00
antirez c005315a64 latencyStartMonitor() modified to avoid warnings. 2014-07-09 19:57:53 +02:00
antirez e81b51ad8c Latency monitor: specialize delayed aof writes events. 2014-07-09 19:57:48 +02:00
antirez 8cc912db43 LATENCY GRAPH: filling under the curve is more readable. 2014-07-09 19:57:42 +02:00
antirez e905588062 LATENCY GRAPH implemented. 2014-07-09 19:57:37 +02:00
antirez b38996aab1 latencyTimeSeries structure max field type fixed. 2014-07-09 19:57:34 +02:00
antirez 062564c261 Free labels in freeSparklineSequence(). 2014-07-09 19:57:28 +02:00
antirez d20d89e630 LATENCY LATEST: add the max field. 2014-07-09 19:57:22 +02:00
antirez dc1c91a8bf Latency monitor trheshold value is now configurable. 
This commit adds both support for redis.conf and CONFIG SET/GET.
 2014-07-09 19:57:17 +02:00
antirez 350792bd72 ASCII sparklines generation API. 2014-07-09 19:57:08 +02:00
antirez ed95915d33 License added to latency.h. 2014-07-09 19:56:47 +02:00
antirez 34afc91030 Latency monitor turned off by default. 
It is not a good idea to bloat the code with gettimeofday() calls if the
instance is working well, and turning monitoring on at runtime is a
joke.
 2014-07-09 19:56:43 +02:00
antirez 35a62f7408 Latency monitor: more hooks around the code. 2014-07-09 19:56:43 +02:00
antirez 855c97ce1b Latency monitor: don't add new samples in the same second. 
Instead we update the old sample with the new latency if it is greater.
 2014-07-09 19:56:35 +02:00
antirez 3cb03b68a4 LATENCY LATEST implemented. 2014-07-09 19:56:30 +02:00
antirez e400ae60f5 Latency monitor: command duration is in useconds. Convert. 2014-07-09 19:04:02 +02:00
antirez a57eb3272d LATENCY SAMPLES implemented. 2014-07-09 19:03:59 +02:00
antirez 72c84acc99 Latency monitor: collect slow commands. 
We introduce the distinction between slow and fast commands since those
are two different sources of latency. An O(1) or O(log N) command without
side effects (can't trigger deletion of large objects as a side effect of
its execution) if delayed is a symptom of inherent latency of the system.

A non-fast command (commands that may run large O(N) computations) if
delayed may just mean that the user is executing slow operations.

The advices LATENCY should provide in this two different cases are
different, so we log the two classes of commands in a separated way.
 2014-07-09 19:03:26 +02:00
antirez 58ecc0bd68 Latency monitor: basic samples collection. 2014-07-09 18:57:57 +02:00
jonathan pickett 88a62862d3 issue 144: spaces in path breaking QFork child launch 2014-07-07 15:22:26 -07:00
jonathan pickett deb9d5cb66 formatting changes 2014-07-03 20:01:08 -07:00
jonathan pickett 2f2985275b Updating readme.md for the latest build. Also tweaking the heap size requirement in the unit test configuration so that I can run more instances without running out of disk space. 2014-07-03 19:53:24 -07:00
jonathan pickett 52c98f5cb8 issue 126: fwrite() failure with buffers straddling page boundaries during background AOF save 2014-07-03 18:47:10 -07:00
jonathan pickett 9613f9a2a6 issue 143: Added more detailed error messages for ERROR_COMMITMENT_LIMIT and ERROR_DISK_FULL errors in QFork code. 2014-07-03 15:30:56 -07:00
antirez 0baab66d16 Fixed 2.8 build breakage introduced back porting COMMAND. 2014-07-02 22:42:00 +02:00
jonathan pickett 96e0be50bb Merge branch '2.8antirez' into 2.8. Updates version from 2.8.9 to 2.8.12. 
Conflicts:
	deps/hiredis/async.c
	deps/hiredis/examples/example-ae.c
	deps/hiredis/hiredis.c
	deps/hiredis/hiredis.h
	deps/hiredis/net.c
	deps/hiredis/net.h
	deps/jemalloc/COPYING
	deps/jemalloc/INSTALL
	deps/jemalloc/README
	deps/jemalloc/bin/pprof
	deps/jemalloc/config.guess
	deps/jemalloc/config.sub
	deps/jemalloc/include/jemalloc/internal/ckh.h
	deps/jemalloc/include/jemalloc/internal/hash.h
	deps/jemalloc/include/jemalloc/internal/prng.h
	deps/jemalloc/include/jemalloc/internal/ql.h
	deps/jemalloc/include/jemalloc/internal/qr.h
	deps/jemalloc/include/jemalloc/internal/quarantine.h
	deps/jemalloc/include/jemalloc/internal/rb.h
	deps/jemalloc/include/jemalloc/internal/tcache.h
	deps/jemalloc/include/jemalloc/internal/tsd.h
	deps/jemalloc/include/jemalloc/internal/util.h
	deps/jemalloc/src/bitmap.c
	deps/jemalloc/src/ckh.c
	deps/jemalloc/src/quarantine.c
	deps/jemalloc/src/tsd.c
	deps/jemalloc/test/integration/aligned_alloc.c
	deps/jemalloc/test/unit/bitmap.c
	src/aof.c
	src/redis-cli.c
	src/redis.h
	src/util.c
 2014-07-01 22:54:48 -07:00
antirez 670cf9598e Fix Solaris compilation due to ctime_r() call. 
Introduced in Redis 2.8.10 because of a change in Sentinel.
This closes issue #1837.
 2014-06-30 16:29:14 +02:00
antirez 41d7c90417 Test: fix instances.tcl restart_instance abort error. 2014-06-30 12:06:24 +02:00
antirez 63feb93a9f COMMAND COUNT subcommand added. 2014-06-27 18:38:08 +02:00
antirez 0a6649a7c0 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 18:38:08 +02:00
antirez 53377f8c6d COMMANDS command renamed COMMAND. 2014-06-27 18:38:03 +02:00
antirez f3efd529df 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 18:37:34 +02:00
Matt Stancliff fdc5dbd5b1 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 18:37:29 +02:00
antirez afe949efcb Remove infinite loop from PSYNC test. 
Added for debugging and forgot there.
 2014-06-27 18:36:59 +02:00
antirez 1066fed769 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.
 2014-06-27 18:36:59 +02:00
Matt Stancliff cd4f0dc775 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.
 2014-06-27 18:36:59 +02:00
Matt Stancliff 58fea46976 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.
 2014-06-27 18:36:59 +02:00
Matt Stancliff ff3ca17bac Use predefined macro for used_memory() update 2014-06-27 18:36:59 +02:00
antirez 04749ee015 Fixed assert conditional in ROLE command test. 2014-06-26 22:13:57 +02:00
antirez 60ff8095d6 No more trailing spaces in Redis source code. 2014-06-26 18:52:16 +02:00
antirez e9cd75dcd1 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 18:43:42 +02:00
Kirk Shoop (MS OPEN TECH) a4d22bfa8c updated license urls 2014-06-25 15:37:44 -07:00
Kirk Shoop (MS OPEN TECH) 8b9aa92123 added nuget release notes 2014-06-25 15:16:27 -07:00
Kirk Shoop (MS OPEN TECH) 084073fced Update package script and release notes 2014-06-25 12:09:56 -07:00
jonathan pickett 4600a06960 issue 136: --serviceName argument not recognized in config.c. 2014-06-24 17:39:47 -07:00
antirez ac6b656fd6 Old form of CLIENT KILL should still allow suicide. 2014-06-24 12:49:18 +02:00
jonathan pickett b6e7f2d45e updated release notes for 2.8.9 nuget and chocolatey packages 2014-06-23 20:08:16 -07:00
jonathan pickett a14400a858 issue 134: command line parsing of [save ""] broken 2014-06-23 18:55:42 -07:00
antirez 32ebb3e651 Redis 2.8.12. 2014-06-23 16:56:18 +02:00
antirez 5d16a838db 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 14:04:09 +02:00
antirez d3c2918434 Sentinel implementation of ROLE. 2014-06-23 12:07:55 +02:00
antirez 95f492ee57 Silence different signs comparison warning in sds.c. 2014-06-23 11:53:50 +02:00
antirez 364464e790 Use Redis updated sds.c for deps/hiredis. 2014-06-23 11:53:45 +02:00
Matt Stancliff 8066334986 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 11:53:40 +02:00
Matt Stancliff e59eef246b 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 11:53:25 +02:00
antirez a8ecd3f040 Dummy zmalloc.h restored into deps/hiredis. 2014-06-23 11:53:15 +02:00
Matt Stancliff 6ee25dc439 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 11:53:12 +02:00
Matt Stancliff ed705f4789 Cancel SHUTDOWN if initial AOF is being written 
Fixes #1826 (and many other reports of the same problem)
 2014-06-23 10:09:25 +02:00
antirez 9a5408f23e Basic tests for the ROLE command. 2014-06-23 09:09:02 +02:00
antirez 061fd99767 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.
 2014-06-21 16:04:20 +02:00
antirez 2b805ce119 Allow to call ROLE in LOADING state. 2014-06-21 15:29:12 +02:00
antirez 8c460a2872 ROLE command: array len fixed for slave output. 2014-06-21 15:29:11 +02:00
antirez 7d0992dad6 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 15:27:57 +02:00
antirez 674194ad42 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 15:27:57 +02:00
antirez 61d9a73d87 CLIENT KILL: fix closing link of the current client. 2014-06-21 15:27:57 +02:00
antirez 09dc6dadc1 New features for CLIENT KILL. 2014-06-21 15:27:57 +02:00
antirez cad13223f0 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 15:27:53 +02:00
antirez b6a26b52bf 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 15:27:12 +02:00
antirez 8060de98f8 ROLE output improved for slaves. 
Info about the replication state with the master added.
 2014-06-21 15:27:12 +02:00
antirez 41a15205b6 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 15:27:12 +02:00
antirez d8d415e717 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.
 2014-06-21 15:27:08 +02:00
antirez 52189cb9c3 Use sdscatfmt() in getClientInfoString() to make it faster. 2014-06-21 15:27:02 +02:00
antirez 4acc3daa92 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.
 2014-06-21 15:26:58 +02:00
antirez 3a915ace44 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.
 2014-06-21 15:26:54 +02:00
antirez ebc62a6f35 Jemalloc updated to 3.6.0. 
Not a single bug in about 3 months, and our previous version was
too old (3.2.0).
 2014-06-20 15:01:36 +02:00
antirez a0e0fba7fa Jemalloc update script added. 2014-06-20 14:53:47 +02:00
jonathan pickett 411b6e9119 issue 129: correct name of file 2014-06-19 17:59:44 -07:00
antirez 2355ae9bc6 Sentinel test: tolerate larger delays in init tests. 2014-06-19 16:10:20 +02:00
antirez fd2faee5de Sentinel test: unit 02, avoid some time related false positives. 2014-06-19 16:10:17 +02:00
antirez 93ee0f2666 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.
 2014-06-19 15:25:57 +02:00
antirez 9b883974af 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.
 2014-06-19 10:35:03 +02:00
antirez 7665cc67b3 Sentinel test: add manual failover test. 2014-06-19 10:32:38 +02:00
antirez 1267e28edc Test: use higher level redis.tcl proc to read replies. 2014-06-18 16:25:45 +02:00
antirez 2cff0c041f Test: Tcl client initial support for automatic reconnection. 2014-06-18 16:19:22 +02:00
antirez 9a39db3bb6 Test: tcl client, unset deferred var on close. 2014-06-18 16:19:19 +02:00
antirez b0964b7dd9 redis.tcl: return I/O error message when peer closes connection. 2014-06-18 16:18:45 +02:00
antirez eef369b66a Sentinel framework new layout back ported 2.8. 
Functionally the same, but makes cherry picking simpler.
 2014-06-18 16:14:33 +02:00
Matt Stancliff f392c7737c Add correct exit value to failed tests 
(Note from @antirez: modified to apply to 2.8)
 2014-06-18 15:03:06 +02:00
Alex Suraci d178cd03ed add missing signal.h include 2014-06-18 09:34:23 +02:00
Matt Stancliff 1b044af402 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.
 2014-06-17 23:40:05 +02:00
jonathan pickett 56f504cb48 if "redis-server --service-install config.conf" and the .conf file had an include statement, then "redis-server --service-start" would fail because the current directory was set to %systemdir% on startup. Changed the current directory to the executable directory when --service-run is encountered 2014-06-16 16:17:37 -07:00
jonathan pickett e5de315940 fix for issue 130: recursive .conf file includes not working 2014-06-16 12:43:44 -07:00
antirez 258f3cd360 Scripting: regression test for issue #1811. 2014-06-16 10:30:42 +02:00
antirez ca6b95dfdd 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.
 2014-06-16 10:30:37 +02:00
jonathan pickett 721a9a2467 issue 125: sentinel command line argument not being recognized 2014-06-15 08:18:38 -07:00
jonathan pickett 0713fce727 issue 123: changed error output in Win32_service and Win32_QFork to use redisLog. Also fixed a few compiler warnings. 2014-06-11 19:22:19 -07:00
antirez b0f3922383 Redis 2.8.11. 2014-06-11 10:30:05 +02:00
antirez a3b0dbccf5 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.
 2014-06-11 10:14:26 +02:00
Matt Stancliff ba76daa42e 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.
 2014-06-11 10:14:26 +02:00
jonathan pickett c83b639b45 issue 119: error reported in sentinel.tcl testing. Error expected. Moved output to loglevel verbose. 2014-06-09 13:51:14 -07:00
jonathan pickett 0df1ab8c74 issue 121: restore redis-cli command history 2014-06-09 13:34:16 -07:00
Matt Stancliff 8f774ee7ca 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 11:43:19 +02:00
Matt Stancliff 2af2414344 Fix lack of SA_ONSTACK under Cygwin 
Fixes #232
 2014-06-09 11:43:19 +02:00
Matt Stancliff b4f9761d85 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 11:39:44 +02:00
Andy Grunwald 6ab4775a2a Fixed typo in word avarege in result message of --intrinsic-latency analyzer 2014-06-06 11:20:04 +02:00
Jan-Erik Rediger b98ee12f61 Small typo fixed 2014-06-06 10:45:46 +02:00
yoav 3783d90e53 Fix eval usage in tests to conform with eval semantics 2014-06-06 10:44:08 +02:00
zionwu e3112bed8d fix issue 1787 2014-06-06 10:38:04 +02:00
jonathan pickett ca941f6c0b chose wrong option for binaries in last merge 2014-06-05 15:36:18 -07:00
jonathan pickett 15a236015f Merge branch '2.8' of https://github.com/MSOpenTech/Redis into 2.8 
Conflicts:
	bin/release/redis-2.8.9.zip
 2014-06-05 15:20:08 -07:00
jonathan pickett d0c274bbf3 Issue 120: sentinel reporting error writing to config file. open(O_RDONLY) followed by fsync() would result in access denied error in windows. In the context of the code changed, ifdefing out fsync on Windows was the safest thing to do. 2014-06-05 15:18:25 -07:00
antirez 987a47b823 Redis 2.8.10. 2014-06-05 11:02:23 +02:00
antirez 430bec3461 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 10:50:31 +02:00
antirez 3a30be51aa Tests for min-slaves-* feature. 2014-06-05 10:50:31 +02:00
antirez 42504169c7 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 18:58:58 +02:00
antirez 26359d570b Regression test for issue #1118. 2014-06-04 18:51:31 +02:00
antirez 768994b683 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 18:45:49 +02:00
jonathan pickett d623ca7e56 code review changes for commit 3ee3c4bdaf 2014-06-01 18:12:34 -07:00
jonathan pickett 228bf0350e resolving merge conflict from another local branch 2014-06-01 14:32:32 -07:00
jonathan pickett 803462c2c9 Merge branch '2.8' of https://github.com/MSOpenTech/Redis into 2.8 2014-06-01 14:31:33 -07:00
jonathan pickett 21fafdc620 issue 111: ANSI escape sequences not rendering on Windows terminal 2014-06-01 14:30:13 -07:00
jonathan pickett d4051d4c38 issue 105: 32-bit version not running on server 2003. Adjusted project settings and selected APIs in use via runtime version checks and windows version macro. 2014-05-29 22:33:19 -07:00
antirez 1c0c42e668 More trailing spaces in sentinel.c removed. 2014-05-28 15:46:14 +02:00
jonathan pickett 88606c1277 Merge branch '2.8' of https://github.com/MSOpenTech/redis into 2.8 2014-05-27 18:14:17 -07:00
jonathan pickett 4f0e17fb35 issue #93: cleaning up memory mapped files left over in cases where FILE_FLAG_DELETE_ON_CLOSE will not clean up file 2014-05-27 17:37:10 -07:00
Matt Stancliff 75266018ee 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 17:53:40 +02:00
antirez edc41baca5 Sentinel example config: explain you don't need to specify slaves. 2014-05-26 10:18:07 +02:00
jonathan pickett a4dd5cba88 issue #108: removing post build script from 32-bit release target. Script is 64-bit only. 2014-05-23 12:54:15 -07:00
jonathan pickett e97604ce63 issue #112: Adding permission for NETWORK_SERVICE on the folder redis-server.exe is in during --service-install. This allows for RDB files to be written and for .conf files to open in RW mode. Fixed a documentation typo. 2014-05-22 14:36:35 -07:00
antirez 1a14eba2d5 Test: fixed scripting.tcl test false positive. 2014-05-22 15:46:06 +02:00
antirez f482349707 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).
 2014-05-22 15:46:02 +02:00
antirez f3d3c606f1 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.
 2014-05-22 15:45:58 +02:00
antirez 4113473997 While ANET_ERR is -1, check syscall retval for -1 itself. 2014-05-22 15:45:54 +02:00
antirez e8669ca8a7 Regression test for issue #1764. 2014-05-20 16:20:45 +02:00
michael-grunder d491a479bc 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 16:20:38 +02:00
antirez d575f7a147 Remove trailing spaces from scripting.c 2014-05-20 16:11:28 +02:00
antirez 6d2fddd2a3 Remove trailing spaces from sentinel.c. 2014-05-20 14:22:34 +02:00
antirez ad935769ae HyperLogLog regression test for issue #1762. 2014-05-19 15:45:24 +02:00
Mike Trinkala 43278af63e Correct the HyperLogLog stale cache flag to prevent unnecessary computations. 
Set the MSB as documented.
 2014-05-19 15:45:24 +02:00
jonathan pickett 1c3b70c971 Finishing merge of branch '2.8.4_msopen' of https://github.com/MSOpenTech/redis into 2.8.4_msopen 
Conflicts:
	bin/release/redis-2.8.4.zip
	src/Win32_Interop/Win32_QFork.cpp
	src/sds.c
 2014-05-16 12:49:22 -07:00
jonathan pickett 3ee3c4bdaf Merge branch 'antirez_2.8' into 2.8.4_msopen. Bumps version from 2.8.4 to 2.8.9 
Conflicts:
	00-RELEASENOTES
	deps/linenoise/linenoise.c
	src/anet.c
	src/debug.c
	src/redis-cli.c
	src/redis.c
	src/sentinel.c
	src/t_zset.c
 2014-05-15 19:18:26 -07:00
jonathan pickett f441b8d0f2 Issue 108: Added 32-bit support back in. QFork code requires a contiguous memory allocation for heap. Due to this requirement and 32-bit memory model limitations, maxheap is limited to about 512MB in 32-bit executable. 32-bit version appears to be about 30% slower than 64-bit version on 64-bit OS. 2014-05-15 18:24:59 -07:00
antirez 3203e2dcdd Fix 2.8 backport of fastscript branch. 
No REDIS_ENCODING_EMBSTR in 2.8 internals.
 2014-05-14 18:39:42 +02:00
Akos Vandra 6e478ec890 Fixed possible buffer overflow bug if RDB file is corrupted. 
(Note: commit message modified by @antirez for clarity).
 2014-05-12 11:49:16 +02:00
Akos Vandra c0eaac9572 fixed possible buffer overflow error 2014-05-12 11:19:36 +02:00
antirez 23011dea1f Sentinel: Add "dir /tmp" directive in example sentinel.conf. 2014-05-12 10:46:56 +02:00
antirez 6ae4c1fcbc DEBUG POPULATE: call dictExpand() to avoid useless rehashing. 2014-05-09 15:05:31 +02:00
jonathan pickett ed1e8ac956 directory rename 2014-05-08 20:56:17 -07:00
jonathan pickett 991653211b directory rename 2014-05-08 20:55:08 -07:00
jonathan pickett 1fd9047397 issue 99: adding sample to demonstrate config of master + 3 slaves + 3 sentinels on one box. Discovered along the way that maxheap would not propagate through .conf include statements. It does now. 2014-05-08 20:07:39 -07:00
antirez 13d8b2b096 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 16:49:45 +02:00
antirez 909d1883e7 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 16:49:45 +02:00
jonathan pickett 01eaf183b3 fix for issue #106: K32QueryWorkingSetEx not in kernel32.dll on Vista/Server 2008 2014-05-07 16:44:06 -07:00
jonathan pickett 176d1cd04b added comment for code changed in commit feb6c2884a 2014-05-07 16:41:49 -07:00
antirez 80b96bc4a7 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 16:17:11 +02:00
antirez 093eab56de 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 16:17:11 +02:00
antirez 32971f6251 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 16:17:11 +02:00
antirez 87516f1a4d 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 16:17:11 +02:00
antirez 1b5524f697 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 16:17:11 +02:00
antirez 8237d88f48 Scripting: cache argv in luaRedisGenericCommand(). 
~ 4% consistently measured speed improvement.
 2014-05-07 16:17:11 +02:00
antirez 6938a3d58f 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 16:17:11 +02:00
antirez 08ea98c9e9 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 16:17:11 +02:00
antirez 3a1a0c3240 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 16:17:11 +02:00
antirez d70281575c Define HAVE_ATOMIC for clang. 2014-05-07 16:17:11 +02:00
antirez 67e350367d 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 16:17:11 +02:00
jonathan pickett feb6c2884a fix for issue #97: Added a modified sample to demonstrate large async GET operation. Also discovered that reply reading in async.c would stop if the reply data exceeded the buffer allocated for the read. redisAeAddRead() would not reschedule a read if already in read mode. To fix this is added redisAeForceAddRead(). 2014-05-01 15:19:18 -07:00
jonathan pickett 9f871c4ec5 Issue #95: Adding ability to specify custom service name/have multiple Redis service instances running.(Forgive the whitespace changes in Win32_Service.cpp. My editor was set to keep tabs when I wrote the code originally. Tabs are gone now.) 2014-04-30 14:34:01 -07:00
jonathan pickett de0a37aa76 Forgot to add custom build step for EventLog.mc to release target. Fixes issue 101. 2014-04-24 19:05:00 -07:00
jonathan pickett a2ef6a4fdc Added support for SIO_LOOPBACK_FAST_PATH (http://msdn.microsoft.com/en-us/library/windows/desktop/jj841212%28v=vs.85%29.aspx) . This speeds up localhost performance on Win8/Server2012+ significantly. In order to take advantage of this clients will have to enable socket fast loopback before connect(). 2014-04-23 11:53:13 -07:00
Matt Stancliff b06de30dd7 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 16:14:40 +02:00
Matt Stancliff e9d0d86dbf Check key expiration before deleting 
Deleting an expired key should return 0, not success.

Fixes #1648
 2014-04-23 16:14:40 +02:00
Glauber Costa 96b5dc0b98 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 15:35:12 +02:00
Salvatore Sanfilippo 3d2af57974 Merge pull request #1700 from nirvdrum/patch-1 
Fixed typos.
 2014-04-23 11:55:50 +02:00
antirez 35252037fd 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 10:21:26 +02:00
Kevin Menard 662f505727 Fixed typos. 2014-04-22 10:51:19 -04:00
antirez de7ebf83e1 redis-cli help.h updated. 2014-04-22 16:15:01 +02:00
antirez 91dcc5b12f generate-command-help.rb updated with new hyperloglog group. 2014-04-22 16:15:01 +02:00
antirez 26375709c5 Redis 2.8.9. 2014-04-22 10:14:57 +02:00
jonathan pickett cbd76db6a3 fixes for issue #94: service install will now self elevate 2014-04-18 18:55:19 -07:00
antirez c5ee48a9d9 Fuzzy test for ZREMRANGEBYLEX added. 2014-04-18 16:16:20 +02:00
antirez be145453c6 ZREMRANGEBYLEX memory leak removed calling zslFreeLexRange(). 2014-04-18 16:16:20 +02:00
antirez 115782b1d7 PFCOUNT multi-key test added. 2014-04-18 16:16:20 +02:00
antirez 2cd20d3c4c 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 16:16:20 +02:00
antirez 3f5cd0c660 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 16:16:20 +02:00
antirez f0eca148ff PFCOUNT support for multi-key union. 2014-04-18 16:16:20 +02:00
antirez d25fd8be64 HyperLogLog low level merge extracted from PFMERGE. 2014-04-18 16:16:20 +02:00
antirez 4a26648a22 ZREMRANGEBYLEX implemented. 2014-04-18 16:16:20 +02:00
antirez 3fc22561e7 Always pass sorted set range objects by reference. 2014-04-18 16:16:04 +02:00
antirez ea0af9ace3 ZREMRANGE* commands refactored into a single generic function. 2014-04-18 16:14:38 +02:00
antirez 019e6f1889 Sorted set lex ranges stress tester. 2014-04-18 16:14:38 +02:00
jonathan pickett e674609fcb adding working hiredis async example to solution based off of example-ae.c source. For issues 18 + 86 2014-04-16 17:16:27 -07:00
jonathan pickett ff367b50e5 fdapi not properly mapping fstat in config.c, so we ended up invoking CRT version without the rfd->fd translation resulting in the CRT calling exit(). Fixes issue 43. 2014-04-16 15:25:21 -07:00
antirez 9c9aef24a2 Basic ZRANGEBYLEX / ZLEXCOUNT tests. 2014-04-17 00:08:38 +02:00
antirez 11f89497b4 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 00:08:37 +02:00
antirez d975bb587f ZLEXCOUNT implemented. 
Like ZCOUNT for lexicographical ranges.
 2014-04-16 15:26:28 +02:00
antirez 71dfb87f76 More HyperLogLog tests. 2014-04-16 15:26:28 +02:00
antirez f2e59eae44 HyperLogLog invalid representation error code set to INVALIDOBJ. 2014-04-16 15:26:28 +02:00
antirez 469b7c5168 PFDEBUG TODENSE added. 
Converts HyperLogLogs from sparse to dense. Used for testing.
 2014-04-16 15:26:28 +02:00
antirez 614fcd491e User-defined switch point between sparse-dense HLL encodings. 2014-04-16 15:26:28 +02:00
antirez 08da2b79fc PFSELFTEST improved with sparse encoding checks. 2014-04-16 15:26:27 +02:00
antirez 5a0bab2a8f PFDEBUG ENCODING added. 2014-04-16 15:26:27 +02:00
antirez 05563310f4 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 15:26:27 +02:00
antirez e77a3e40f2 Error message for invalid HLL objects unified. 2014-04-16 15:26:27 +02:00
antirez c4b2a7a7f4 PFMERGE fixed to work with sparse encoding. 2014-04-16 15:26:27 +02:00
antirez ee682c4c48 Mark PFDEBUG as write command in the commands table. 
It is safer since it is able to have side effects.
 2014-04-16 15:26:27 +02:00
antirez 9ef8ef410b 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 15:26:27 +02:00
antirez 91e05e1618 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 15:26:27 +02:00
antirez 68a4bf70e0 hllSparseAdd(): speed optimization. 
Mostly by reordering opcodes check conditional by frequency of opcodes
in larger sparse-encoded HLLs.
 2014-04-16 15:26:27 +02:00
antirez 21e328a6d7 Detect corrupted sparse HLLs in hllSparseSum(). 2014-04-16 15:26:27 +02:00
antirez 4244aa29f9 hllSparseAdd(): faster code removing conditional. 
Bottleneck found profiling. Big run time improvement found when testing
after the change.
 2014-04-16 15:26:27 +02:00
antirez 4cd11077d2 Comment typo in hllSparseAdd(). first -> fits. 2014-04-16 15:26:27 +02:00
antirez 08183cfef5 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 15:26:27 +02:00
antirez d428569d13 More robust HLL_SPARSE macros protecting 'p' with parens. 
Now the macros will work with arguments such as "ptr+1".
 2014-04-16 15:26:27 +02:00
antirez 2d1e35a58a hllSparseAdd() opcode seek stop condition fixed. 2014-04-16 15:26:27 +02:00
antirez 29c7d1974b 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 15:26:27 +02:00
antirez 023f2f7a10 Fixed memmove() count in hllSparseAdd(). 2014-04-16 15:26:27 +02:00
antirez 91391760ca 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 15:26:27 +02:00
antirez 5d87160f8a 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 15:26:27 +02:00
antirez a18e7afbe8 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 15:26:27 +02:00
antirez 5c86d9e7b0 PFDEBUG added, PFGETREG removed. 
PFDEBUG will be the interface to do debugging tasks with a key
containing an HLL object.
 2014-04-16 15:26:27 +02:00
antirez 0e0ddcc4fa 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 15:26:27 +02:00
antirez 533235f7e3 hllSparseAdd() sanity check for span != 0 added. 2014-04-16 15:26:27 +02:00
antirez f12fb49bb2 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 15:26:27 +02:00
antirez a3e2adc0ae Fix seqlen computation in hllSparseAdd(). 2014-04-16 15:26:27 +02:00
antirez eaa73ef73a Abstract hllSparseAdd() / hllDenseAdd() via hllAdd(). 2014-04-16 15:26:27 +02:00
antirez 95de3eaea0 hllSparseSum(): multiply 1 * runlen for zero entries. 2014-04-16 15:26:27 +02:00
antirez c277169ed7 Macro HLL_SPARSE_XZERO_LEN fixed. 2014-04-16 15:26:27 +02:00
antirez b6ca71a40a Fix HLL sparse object creation #2. 
Two vars initialized to wrong values in createHLLObject().
 2014-04-16 15:26:27 +02:00
antirez 8b39fee4c8 Increment pointer while iterating sparse HLL object. 2014-04-16 15:26:27 +02:00
antirez c1ddead779 Fix HLL sparse object creation. 
The function didn't considered the fact that each XZERO opcode is
two bytes.
 2014-04-16 15:26:27 +02:00
antirez 0a7ba6faef Create HyperLogLog objects with sparse encoding. 2014-04-16 15:26:27 +02:00
antirez 8e889728bf HyperLogLog sparse to dense conversion function. 2014-04-16 15:26:27 +02:00
antirez aadd513160 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 15:26:27 +02:00
antirez e5e5facba1 hllCount() refactored to support multiple representations. 2014-04-16 15:26:27 +02:00
antirez 88847a216f hllAdd() refactored into two functions. 
Also dense representation access macro renamed accordingly.
 2014-04-16 15:26:27 +02:00
antirez 21c97e71cd 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 15:26:27 +02:00
antirez ed96e115d0 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 15:26:27 +02:00
antirez 93c75f1bf4 HyperLogLog sparse representation description and macros. 2014-04-16 15:26:26 +02:00
antirez 2b1385b9ec 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 15:26:22 +02:00
antirez 12d1d18b59 ZRANGEBYLEX and ZREVRANGEBYLEX implementation. 2014-04-16 15:26:09 +02:00
antirez a10bfded15 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 15:26:09 +02:00
antirez ee764b0f8d 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 15:26:09 +02:00
antirez 091f5677a0 Initial HyperLogLog tests. 2014-04-16 15:26:09 +02:00
antirez a97675923d Return "WRONGTYPE" error on PF* type mismatch. 2014-04-16 15:26:09 +02:00
antirez faa7f259cc 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 15:26:09 +02:00
antirez 8ae6887091 Make hll-gnuplot-graph.rb callable from cli. 2014-04-16 15:26:09 +02:00
antirez bdd8701c60 Remove HyperLogLog type checking duplicated code. 2014-04-16 15:26:09 +02:00
antirez b3baa51403 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 15:26:09 +02:00
antirez 348d5ea246 PFCOUNT: unshare the object when cached cardinality is modified. 2014-04-16 15:26:09 +02:00
antirez ba0dc0be2c PFSELFTEST improved to test the approximation error. 2014-04-16 15:26:09 +02:00
antirez c2cd48718d 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 15:26:09 +02:00
antirez ff551fad24 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 15:26:09 +02:00
Raymond Myers e08100a9ba Fixed pfadd/pfcount commands emitting hll* events instead of pf* events 2014-04-16 15:26:09 +02:00
Raymond Myers 23acae6e1c Change HLL* to PF* in error messages 2014-04-16 15:26:09 +02:00
antirez e3773b0a0e 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 15:26:09 +02:00
antirez badf23f57b HyperLogLog API prefix modified from "P" to "PF". 
Using both the initials of Philippe Flajolet instead of just "P".
 2014-04-16 15:26:03 +02:00
antirez 6e71459c73 Makefile.dep updated with hyperloglog.o deps. 2014-04-16 15:25:28 +02:00
antirez 2ee2bec2d6 HyperLogLog: make API use the P prefix in honor of Philippe Flajolet. 2014-04-16 15:24:15 +02:00
antirez 1e90c0066c HLLMERGE fixed by adding a... missing loop! 2014-04-16 15:23:21 +02:00
antirez be7eef911b hll-gnuplot-graph.rb: added new filter "all". 2014-04-16 15:23:21 +02:00
antirez f2e707b679 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 15:23:21 +02:00
antirez be7fe2b92b 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 15:23:21 +02:00
antirez 39dd1f648d 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 15:23:21 +02:00
antirez f07b1514bb 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 15:23:21 +02:00
antirez 32e9996846 hll-gnuplot-graph.rb: Use |error| when filter is :max 2014-04-16 15:23:21 +02:00
antirez 9faa80ee8c Ignore txt files inside utils/hyperloglog. 
Those are generated to trace graphs using gnuplot.
 2014-04-16 15:23:21 +02:00
antirez aad5959ddc 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 15:23:21 +02:00
antirez d6ce659ffd hll-gnuplot-graph.rb added to plot HyperLogLog error graphs. 2014-04-16 15:23:21 +02:00
antirez c3df5965f3 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 15:23:21 +02:00
antirez 9e178afae9 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 15:22:56 +02:00
antirez 1c795db9d0 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 15:19:40 +02:00
antirez 7f30998432 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 15:19:40 +02:00
antirez 430d0ade75 Remove a few useless operations from hllCount() fast path. 2014-04-16 15:19:40 +02:00
antirez 83650a72a1 HLLCOUNT 3x faster taking fast path for default params. 2014-04-16 15:19:40 +02:00
antirez 3afd25ecc7 Use processor base types in HLL_(GET|SET)_REGISTER. 
This speedups the macros by a noticeable factor.
 2014-04-16 15:19:40 +02:00
antirez 0ffd5e4a30 HyperLogLog: use precomputed table for 2^(-M[i]). 2014-04-16 15:19:40 +02:00
antirez ca6973d3b6 hll-err.rb: speedup using pipelining. 2014-04-16 15:19:40 +02:00
antirez 2500f2a334 hll-err.rb added to test error rate of Redis HyperLogLog. 2014-04-16 15:19:40 +02:00
antirez 42592a7faa 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 15:19:40 +02:00
antirez 3035075b2c HLLCOUNT implemented. 2014-04-16 15:19:40 +02:00
antirez 959d0f012a HLLADD implemented. 2014-04-16 15:19:33 +02:00
antirez ac29bb2ae4 hllAdd() low level HyperLogLog "add" implemented. 2014-04-16 15:19:05 +02:00
antirez 2a91f548c4 HyperLogLog: redefine constants using "P". 2014-04-16 15:19:05 +02:00
antirez 3f159bbd44 HLL_SET_REGISTER fixed. 
There was an error in the first version of the macro.
Now the HLLSELFTEST test reports success.
 2014-04-16 15:19:05 +02:00
antirez a18e880bf1 Use REDIS_HLL_REGISTER_MAX when possible. 2014-04-16 15:19:05 +02:00
antirez 00409d8742 HLL_(SET|GET)_REGISTER types fixed. 2014-04-16 15:19:05 +02:00
antirez de3f821af6 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 15:18:57 +02:00
antirez 02d88fb201 HyperLogLog: initial sketch of registers access. 2014-04-16 15:17:26 +02:00
jonathan pickett a7f2df706e Removing install project since we have NuGet and Chocolatey installs. Removing RedisWAInst since Azure fork will handle their own installation. 2014-04-15 19:01:55 -07:00
jonathan pickett 884dd54bbc since redis now has built-in service support, and Sentinel should be used for monitoring, this project is no longer needed 2014-04-15 18:05:51 -07:00
jonathan pickett 3cd77a5cb5 fox for issue 84. now recognizing "stdout" as as stdout. Also added docs to binary zip generation script and modified redis.windows.conf to remove references to unsupported daemonize flag 2014-04-15 16:21:39 -07:00
jonathan pickett bcfdbf7bcc fox for issue #90. info.rdb_changes_since_last_save not being properly updated. Also dealt with the exit code from a failed fork not making it back to the parent. 2014-04-15 14:21:25 -07:00
jonathan pickett 765cdbe670 fix for issue #39 2014-04-14 19:10:33 -07:00
jonathan pickett 631e8e1afb Bringing forward a fix from 2.6.14 (commit 99e01c7e6b). Replaces the fork/execv construction with CreateProcess. 2014-04-11 20:03:15 -07:00
jonathan pickett 959c9d3469 Added preshutdown notification support to service to deal with Issue 87. Also improved message filter support for eventlog. 2014-04-11 19:31:05 -07:00
jonathan pickett 2d29a8f207 Added support for redirecting Redis logging to the Windows Application log. Repurposed --syslog flags for this. See redis.windows.conf for information about the repurposed flags. When redis is running as a service it will use the Windows EventLog by default. 2014-04-08 18:16:09 -07:00
jonathan pickett 34250336c1 Release build still had link dependency on DLL versions of CRT. Also updated release documentation 2014-04-07 19:00:35 -07:00
jonathan pickett 9651d90354 sprintf argument change to fix a compiler warning caused forked process to fail to map control structure unexpectedly. Fixing in a different manner. 2014-04-07 18:37:25 -07:00
jonathan pickett 6eb1ec2217 added Windows Service support to Redis 2014-04-03 16:39:02 -07:00
jonathan pickett e2a48cd715 PS scripts to automate NuGet/Chocolatey package generation. Also eliminated VC redist dependency in Chocolatey spec 2014-04-02 16:48:17 -07:00
jonathan pickett 0df3c6b4b4 Updated the README.md for the current build. Removed APIBridge.dll from the powershell script that builds the zip file containing the binaries. Updated the zip file. 2014-04-01 17:11:17 -07:00
jonathan pickett 8b9a8a0a41 In linenoise.c in linenoiseRaw() isatty() is called on stdin in order to determine where to send the redis-cli prompt. The FDAPI implementation of isatty was not handling this case, so the prompt was not showing up. 2014-04-01 16:04:14 -07:00
jonathan pickett c3c91f5ab0 Moved projects to VS2013. Removed dependency on CRT runtime DLLs. Removed APIBridge.DLL. Added MSOpenTech copyright notices. 2014-03-31 17:01:21 -07:00
antirez ddbdd3b0df Document TTL changes in the 2.6 -> 2.8 migration doc. 2014-03-27 15:53:14 +01:00
antirez d2e59c2715 Redis 2.8.8. 2014-03-25 11:30:42 +01:00
antirez 99fc582f8c Test: do not complain when "leaks" can't run because process died. 2014-03-25 09:34:11 +01:00
antirez 3580bb485a 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 09:07:21 +01:00
Matt Stancliff c3510af1c0 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 09:07:21 +01:00
Matt Stancliff 771f8ad0e7 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 09:07:21 +01:00
Matt Stancliff 3ce742d1d5 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 09:07:21 +01:00
Matt Stancliff b59585396b 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 09:07:21 +01:00
Matt Stancliff 6826af1b50 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 09:07:21 +01:00
Matt Stancliff 611372fa97 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 09:07:21 +01:00
Matt Stancliff 3bd3240660 Sentinel: 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;
 2014-03-25 08:06:04 +01:00
antirez 79349affd5 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.
 2014-03-25 08:03:11 +01:00
Matt Stancliff 80dec5e4df Sentinel: Notify user when config can't be saved 2014-03-25 08:03:07 +01:00
Jan-Erik Rediger a2ec9a9068 Small typo fixed 2014-03-25 08:02:56 +01:00
Matt Stancliff 88c6c669c3 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 21:56:22 +01:00
Jan-Erik Rediger 6182dad37c 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 18:12:58 +01:00
antirez 4ebc7e3738 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 12:03:44 +01:00
antirez 72257f4b87 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 10:23:58 +01:00
antirez 7054f2a029 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 10:02:58 +01:00
antirez 7a20f09647 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 09:45:18 +01:00
jonathan pickett f5b410a52e package description update 2014-03-21 20:41:15 -07:00
jonathan pickett 4c0e75ed8b updating binaries for last change 2014-03-21 18:55:19 -07:00
jonathan pickett 8d0309eebf Bringing over fix from azure fork that fixes partial synchronization (https://github.com/WindowsAzure/redis/commit/f5be20b0201d9c492506e20254d38053223201c5 ). I added '#ifdef win32' blocks around the changes so we could rebase this forward. 2014-03-21 18:45:33 -07:00
antirez 001775f5fb 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 11:19:28 +01:00
antirez c68189a19f Specify lruclock in redisServer structure via REDIS_LRU_BITS. 
The padding field was totally useless: removed.
 2014-03-21 11:17:52 +01:00
antirez ff8c81873a Set LRU parameters via REDIS_LRU_BITS define. 2014-03-21 11:17:36 +01:00
antirez e3b71a1c02 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 11:16:12 +01:00
antirez 0937377aa2 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 11:16:12 +01:00
antirez 686839b477 Sentinel test: 02 unit better coverage + refactoring. 2014-03-21 11:16:12 +01:00
antirez 6d0e408a27 Sentinel test: foreach_instance_id implements 'break'. 2014-03-21 11:16:12 +01:00
antirez ba2edc4191 Sentinel: instance_is_killed proc added to sentinel.tcl. 2014-03-21 11:16:12 +01:00
antirez 9c2063fb8f Sentinel: propagate down-after-ms changes to slaves and sentinels. 2014-03-21 11:16:12 +01:00
antirez ffa8f479a5 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 11:16:12 +01:00
antirez 42091a79bb 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 11:16:11 +01:00
antirez 38241c4b1e 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 11:16:11 +01:00
jonathan pickett 16ca1634ae adding nuget and chocolatey packaging specifications 2014-03-20 17:23:38 -07:00
jonathan pickett decf224e85 With sufficiently large allocations dlcalloc was not clearing allocated memory because of a compilation flag. This was causing a unit test to fail. 2014-03-14 15:46:49 -07:00
Matt Stancliff 9de0755869 Fix segfault from accessing array out of bounds 
argc == 2; argv[2] == crash
 2014-03-14 22:57:10 +01:00
antirez a31a0b4336 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 22:57:06 +01:00
antirez 6b0e36ffc6 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 11:04:54 +01:00
antirez bd48ff69b0 Sentinel: HELLO processing refactored into sentinelProcessHelloMessage(). 2014-03-14 10:56:56 +01:00
antirez 3703112671 Linenoise updated, multiline mode enabled in redis-cli. 2014-03-13 15:12:04 +01:00
zhanghailei ccd2c18c1c According to context,the size should be 16 rather than 64 2014-03-11 10:12:19 +01:00
zhanghailei daa5d2a6c9 FIXED a typo more thank should be more than 2014-03-11 10:12:13 +01:00
zhanghailei ca0720b694 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 10:12:00 +01:00
Matt Stancliff ffe742f92a 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 10:10:54 +01:00
antirez 6fb9e2d717 Typo in sentinel.conf, exists -> exits. 2014-03-11 10:10:30 +01:00
jonathan pickett 53f0c8f6e2 using memtoull instead of _atoi64 while parsing maxmemory and maxheap flags. 'redis-server --maxmem 500m --maxheap 2g' works 2014-03-10 16:19:49 -07:00
antirez eb9e1526e6 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.
 2014-03-10 23:04:37 +01:00
Matt Stancliff a6970570e3 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.
 2014-03-10 15:47:20 +01:00
antirez fe0ab7d234 Fixed memory leak in SORT LIMIT option argument parsing on error. 2014-03-10 15:45:29 +01:00
jonathan pickett 0080e0fc70 Added --maxheap flag to control size of memory mapped file created. Heap can now exceed size of physical memory. See long comment in Win32_qfork.cpp about implications of using --maxheap flag. 2014-03-07 19:51:02 -08:00
antirez 464fef9bf8 Redis 2.8.7. 2014-03-05 14:42:50 +01:00
antirez aab16ead92 Cast saveparams[].seconds to long for %ld format specifier. 2014-03-05 11:26:50 +01:00
antirez 86afc0706d Merge branch '2.8' of github.com:/antirez/redis into 2.8 2014-03-05 10:22:56 +01:00
antirez 55b8f6ec1c Document why we update peak memory in INFO. 2014-03-05 10:16:20 +01:00
Matt Stancliff 647a261465 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 10:16:16 +01:00
antirez 9c66cd91be Sentinel test: Makefile target added. 2014-03-05 10:16:12 +01:00
michael-grunder 6991792943 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 10:16:06 +01:00
antirez 7d65b7199a BITPOS fuzzy testing. 2014-03-05 10:16:02 +01:00
antirez c19cfde65d Basic BITPOS tests. 2014-03-05 10:15:55 +01:00
antirez 55f4b20f31 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 10:15:32 +01:00
antirez 1606978a0b 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 10:15:32 +01:00
antirez 6200191943 CONFIG REWRITE should be logged at WARNING level. 2014-03-05 10:15:32 +01:00
antirez 2af2173a60 Sentinel test: debugging console improved. 2014-03-05 10:15:32 +01:00
antirez 1dc1e31c2b Sentinel test: initial debugging console. 2014-03-05 10:15:32 +01:00
antirez b132121a71 Sentinel test: be more patient in create_redis_master_slave_cluster. 2014-03-05 10:15:32 +01:00
antirez d1d706c923 Sentiel test: add test start time in output. 2014-03-05 10:15:32 +01:00
antirez c99cd2fd40 Sentinel test: use 1000 as retry in initial 00 unit test. 2014-03-05 10:15:32 +01:00
antirez f33d4a6c24 Sentinel test: initial tests in 03 unit. 2014-03-05 10:15:32 +01:00
antirez c2d99d49b5 Sentinel test: foreach_instance_id now supports 'continue'. 2014-03-05 10:15:32 +01:00
antirez e40f8378bd Sentienl test: fixed typo in unit 03 top comment. 2014-03-05 10:15:32 +01:00
antirez a0335e18ba Sentinel test: Makefile target added. 2014-02-28 16:00:17 +01:00
antirez 950cb76e7b BITPOS fuzzy testing. 2014-02-27 15:58:34 +01:00
antirez 42e3630d77 Basic BITPOS tests. 2014-02-27 15:58:30 +01:00
antirez 1892b56224 warnigns -> warnings in redisBitpos(). 2014-02-27 15:58:27 +01:00
antirez 2c8036f7b2 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 15:58:21 +01:00
antirez 3294f74fef 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 15:58:14 +01:00
antirez 82d2e295b8 Added two more BITCOUNT tests stressing misaligned access. 2014-02-27 15:57:56 +01:00
antirez a3eb3f9c3b 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 15:57:53 +01:00
antirez 30a92b6c76 Fix misaligned word access in redisPopcount(). 2014-02-27 15:57:49 +01:00
antirez c1cc28f230 warnigns -> warnings in redisBitpos(). 2014-02-27 15:55:56 +01:00
antirez d79f9ebdb5 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 15:55:47 +01:00
antirez 25e2791ec3 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 15:54:31 +01:00
antirez c955b47d62 Added two more BITCOUNT tests stressing misaligned access. 2014-02-27 15:53:40 +01:00
antirez bfe4c560e6 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 15:53:40 +01:00
antirez 580ed0768f Fix misaligned word access in redisPopcount(). 2014-02-27 15:53:40 +01:00
jonathan pickett 411ef3f609 Cleaned up duplicate WIN32_IOCP flag in projects. Also added compressOutputToReleaseFolder.ps1 script to release build of Redis. 2014-02-25 18:03:23 -08:00
jonathan pickett 30a26d36ab added --bypass-system-reserve flag for Azure team to test maximum heap allocation 2014-02-25 18:01:26 -08:00
antirez d90e3e3a70 Sentinel test: add stub for unit 04. 2014-02-25 15:55:37 +01:00
antirez 89d95ec2e7 Sentinel test: added TODO items in 02 unit. 2014-02-25 15:55:37 +01:00
antirez 65194d11c3 redis-cli: also remove useless uint8_t. 2014-02-25 15:09:15 +01:00
antirez 16c2189d4e 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 15:09:15 +01:00
antirez 1b2bcd4210 redis-cli: check argument existence for --pattern. 2014-02-25 15:09:15 +01:00
antirez 6ec3dc2c42 redis-cli: --intrinsic-latency run mode added. 2014-02-25 15:09:15 +01:00
antirez 02ae6cb01a redis-cli: added comments to split program in parts. 2014-02-25 15:09:15 +01:00
antirez fe05112029 Sentinel test: check role at end of unit 01. 2014-02-25 15:08:41 +01:00
michael-grunder cf76fdd04f 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 15:08:38 +01:00
antirez 9e34e296cf Sentinel test: kill masters instead of using DEBUG SLEEP in all tests. 2014-02-25 15:08:34 +01:00
antirez dfacfafc2a 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 10:30:35 +01:00
antirez 2fab983c69 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 10:30:35 +01:00
antirez a6d560db64 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 10:30:35 +01:00
antirez 3f16fee39a 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 10:30:35 +01:00
antirez e8e2ac072c 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 10:30:35 +01:00
antirez 85fa77e02e Sentinel: log quorum with +monitor event. 2014-02-25 10:30:35 +01:00
antirez 2d48d7c2fc 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 10:30:35 +01:00
antirez 6e6106797e Sentinel: generate +monitor events at startup. 2014-02-25 10:30:35 +01:00
antirez 96162c0c58 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 10:30:35 +01:00
antirez 39eacde136 Sentinel: added missing exit(1) after checking for config file. 2014-02-25 10:30:35 +01:00
antirez 28aa15a6ea Sentinel test: tmp dir and gitignore added. 2014-02-25 10:30:35 +01:00
antirez 3917224f7c Sentinel test: minor fixes to --pause-on-error. 2014-02-25 10:30:35 +01:00
antirez 001ab29915 Sentinel test: --pause-on-error option added. 
Pause the test with running instances available for state inspection on
error.
 2014-02-25 10:30:35 +01:00
antirez d31a394148 Sentinel test: added empty units to fill later. 2014-02-25 10:30:35 +01:00
antirez d83ab8f9d4 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 10:30:35 +01:00
antirez 1dfab3d107 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 10:30:35 +01:00
antirez aee165e084 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 10:30:35 +01:00
antirez 73b290cc5b Sentinel test: ability to run just a subset of test files. 2014-02-25 10:30:35 +01:00
Matt Stancliff 3ced78dc6c 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 10:30:35 +01:00
jonathan pickett f5245a44e4 sdshdr definition in deps/hiredis/sds.h must match the definition in src/sds.h. Specifically, sizeof (C struct with a char* member) calculates the char* as a pointer. sizeof(C-struct with char[] member) calculates char[] as empty character array. BTW, the parent struct from member pointer dereference is an abomination. 2014-02-24 15:59:52 -08:00
antirez 6441a41f94 Sentinel: report instances role switch events. 
This is useful mostly for debugging of issues.
 2014-02-20 12:28:03 +01:00
antirez 05dde71919 Sentinel test: some reliability fixes to 00-base tests. 2014-02-20 12:28:03 +01:00
antirez c499d82f28 Sentinel test: check that role matches at end of 00-base. 2014-02-20 12:28:03 +01:00
antirez c01aff1234 Sentinel test: ODOWN and agreement. 2014-02-20 12:28:03 +01:00
antirez 56cb037a86 Sentinel test: check reconfig of slaves and old master. 2014-02-20 12:28:03 +01:00
antirez d4edf63bc4 Sentinel test: basic failover tested. Framework improvements. 2014-02-20 12:28:02 +01:00
antirez 722d4f0c41 Sentinel test: basic tests for MONITOR and auto-discovery. 2014-02-20 12:28:02 +01:00
antirez b26660e130 Sentinel test: info fields, master-slave setup, fixes. 2014-02-20 12:28:02 +01:00
antirez ac5fd5a519 Prefix test file names with numbers to force exec order. 2014-02-20 12:28:02 +01:00
antirez 890559ea1c Sentinel test: provide basic commands to access instances. 2014-02-20 12:28:02 +01:00
jonathan pickett 208eb29003 Fixed 1-off bug allocating last block of MMF. Added additional heap allocation with maxmemory flag to allow for fragmentation. 2014-02-19 18:10:05 -08:00
antirez 905c55d5ae Sentinel: SENTINEL_SLAVE_RECONF_RETRY_PERIOD -> RECONF_TIMEOUT 
Rename define to match the new meaning.
 2014-02-18 10:30:29 +01:00
antirez 1b345ec3f0 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 10:30:28 +01:00
jonathan pickett bdd9804c42 modified binaries after maxmemory change 2014-02-17 18:21:13 -08:00
jonathan pickett 92258f87ba Removing maxheapgb flag in favor of utilizing existing maxmemory flag. Now honoring maxmemory flag in .conf files. Slightly modified default MMF startup requirements to [phys - min(0.2*phys, 3GB)] 2014-02-17 18:11:17 -08:00
antirez 351e790032 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 17:38:58 +01:00
antirez f71c7de88e Test: colorstr moved to util.tcl. 2014-02-17 17:38:54 +01:00
antirez 0886c98fef 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 12:29:54 +01:00
antirez 4237f14a8a 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 12:14:19 +01:00
antirez 5efee4f0bb 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 12:10:12 +01:00
jonathan pickett 8d9e238a20 /tmp is not guaranteed to be available on Windows. Substituting directory in TEMP environment variable. Unit test still fails for other reasons. 2014-02-14 10:56:20 -08:00
jonathan pickett a695b4e4a4 updating binaries 2014-02-14 10:53:13 -08:00
antirez 1d4d9e7b14 Redis 2.8.6. 2014-02-13 15:43:40 +01:00
antirez 85492dcfef 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 15:13:35 +01:00
antirez eb10bb31ac Log when CONFIG REWRITE goes bad. 2014-02-13 14:33:50 +01:00
antirez ebdb37cea7 Test: regression for issue #1549. 
It was verified that reverting the commit that fixes the bug, the test
no longer passes.
 2014-02-13 12:27:08 +01:00
antirez 9b73a274ac 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 12:16:34 +01:00
antirez 96973a7c33 AOF write error: retry with a frequency of 1 hz. 2014-02-12 16:57:17 +01:00
antirez fadbbdd3f4 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 16:57:13 +01:00
jonathan pickett 8c87102147 Not updating slave acknowledgement time when IOCP operation completes. breaks integration/replication-4 unit test. 2014-02-11 18:51:21 -08:00
jonathan pickett 4e45ebf5a1 unit tests not working in 2.6.14 on windows still not working in 2.8.4 2014-02-11 18:49:40 -08:00
jonathan pickett e4ac644fc1 Changing heap size calculation to be 80% of physical ram. This allow the server to start on Azure S and XS machines. 2014-02-11 15:01:23 -08:00
antirez 688d32e16b Don't count time to feed MONITORs in SLOWLOG. 2014-02-07 18:29:26 +01:00
antirez 3e4968339b Sentinel: allow SHUTDOWN command in Sentinel mode. 2014-02-07 11:22:30 +01:00
Jonathan Pickett 4bbb8e6bce fixed rebase error that caused redis to panic with BLPOP 2014-02-05 17:28:05 -08:00
Jonathan Pickett 2959371b3f adding --maxheapgb flag to startup to control size of memory mapped file created 2014-02-05 11:05:54 -08:00
antirez 301a0cfc69 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 16:41:04 +01:00
antirez 4e809a9a19 Redis 2.8.5. 2014-02-04 11:17:21 +01:00
Jonathan Pickett 2277ed0ea8 Merge branch '2_6_14_rebase' into 2.8.4 2014-02-03 17:10:36 -08:00
Jonathan Pickett 62cc0e320d Merge branch '2_6_14_rebase' into 2.8.4 
Conflicts:
	00-RELEASENOTES
	deps/hiredis/hiredis.c
	deps/hiredis/sds.c
	deps/jemalloc/ChangeLog
	deps/jemalloc/Makefile.in
	deps/jemalloc/VERSION
	deps/jemalloc/configure
	deps/jemalloc/configure.ac
	deps/jemalloc/doc/jemalloc.3
	deps/jemalloc/doc/jemalloc.html
	deps/jemalloc/doc/jemalloc.xml.in
	deps/jemalloc/include/jemalloc/internal/arena.h
	deps/jemalloc/include/jemalloc/internal/chunk.h
	deps/jemalloc/include/jemalloc/internal/chunk_dss.h
	deps/jemalloc/include/jemalloc/internal/chunk_mmap.h
	deps/jemalloc/include/jemalloc/internal/ctl.h
	deps/jemalloc/include/jemalloc/internal/extent.h
	deps/jemalloc/include/jemalloc/internal/huge.h
	deps/jemalloc/include/jemalloc/internal/jemalloc_internal.h.in
	deps/jemalloc/include/jemalloc/internal/private_namespace.h
	deps/jemalloc/include/jemalloc/internal/prof.h
	deps/jemalloc/include/jemalloc/internal/rtree.h
	deps/jemalloc/include/jemalloc/jemalloc.h.in
	deps/jemalloc/include/jemalloc/jemalloc_defs.h.in
	deps/jemalloc/src/arena.c
	deps/jemalloc/src/base.c
	deps/jemalloc/src/chunk.c
	deps/jemalloc/src/chunk_dss.c
	deps/jemalloc/src/chunk_mmap.c
	deps/jemalloc/src/ctl.c
	deps/jemalloc/src/huge.c
	deps/jemalloc/src/jemalloc.c
	deps/jemalloc/src/mutex.c
	deps/jemalloc/src/prof.c
	deps/jemalloc/src/rtree.c
	deps/jemalloc/src/stats.c
	deps/jemalloc/src/tcache.c
	deps/jemalloc/src/util.c
	deps/jemalloc/src/zone.c
	deps/jemalloc/test/thread_arena.c
	deps/linenoise/linenoise.c
	redis.conf
	src/Makefile.dep
	src/ae.c
	src/anet.c
	src/anet.h
	src/aof.c
	src/bitops.c
	src/config.c
	src/config.h
	src/db.c
	src/debug.c
	src/dict.c
	src/memtest.c
	src/mkreleasehdr.sh
	src/networking.c
	src/object.c
	src/rdb.c
	src/redis-benchmark.c
	src/redis-cli.c
	src/redis.c
	src/redis.h
	src/release.c
	src/replication.c
	src/rio.c
	src/rio.h
	src/scripting.c
	src/sds.c
	src/sentinel.c
	src/t_list.c
	src/t_string.c
	src/t_zset.c
	src/version.h
	src/zmalloc.c
	tests/assets/default.conf
	tests/integration/rdb.tcl
	tests/test_helper.tcl
	tests/unit/basic.tcl
	tests/unit/expire.tcl
	tests/unit/obuf-limits.tcl
	tests/unit/pubsub.tcl
	tests/unit/scripting.tcl

Still having problems with BLPOP
 2014-02-03 17:06:54 -08:00
antirez ddcf160309 Move mstime_t define outside sentinel.c. 
The define is now used in other parts of Redis 2.8 tree instead of long
long.

A nice side effect is that now 2.8 and unstable sentinel.c files are
identical as it should be.
 2014-02-03 16:34:46 +01:00
antirez c5bc592650 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 16:29:25 +01:00
antirez 5201ca0ca1 Allow CONFIG and SHUTDOWN while in stale-slave state. 2014-02-03 15:51:07 +01:00
antirez 3da5cbe5bb 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 15:46:47 +01:00
PatrickJS 0be31e2d22 update copyright year 2014-02-03 11:19:25 +01:00
antirez 66304fb122 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 16:25:13 +01:00
antirez 1406112ec4 Redis.conf comment about tcp-backlog option improved. 2014-01-31 15:03:25 +01:00
antirez 917b851491 Option "backlog" renamed "tcp-backlog". 
This is especially important since we already have a concept of backlog
(the replication backlog).
 2014-01-31 15:03:22 +01:00
Nenad Merdanovic 8dda9dbef0 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 15:03:19 +01:00
antirez f0652c37a5 Sentinel: check arity for SENTINEL MASTER command. 
This fixes issue #1530.
 2014-01-31 10:14:07 +01:00
Jonathan Pickett 4440406bb5 Using project reference rather than linker library input. This fixes the annoying 'do you want to rebuild the universe just because you pressed F5' problem. Built is now built. 2014-01-28 18:08:38 -08:00
Jonathan Pickett 8812fe61bd added lua and hiredis libs as project references 2014-01-28 11:58:14 -08:00
Jonathan Pickett 66b7f8a044 working through issues regarding sentinal not working 2014-01-28 11:49:20 -08:00
antirez a2c9d38a9f SENTINEL SET master quorum implemented. 2014-01-28 11:23:51 +01:00
antirez 89a731b3aa Fixed inverted if condition in MISCONF error code path. 2014-01-28 10:10:56 +01:00
Jonathan Pickett e1f312e547 redis log not working with anything ofther than stdout 2014-01-27 21:12:31 -08:00
Jonathan Pickett e35841ce76 --loglevel command line argument was not working 2014-01-27 19:26:47 -08:00
antirez ecfefde760 Don't log MONITOR clients as disconnecting slaves. 2014-01-25 11:54:02 +01:00
Jonathan Pickett 6daa29da46 added failure error messages back in to QFork MasterInit 2014-01-22 17:03:44 -08:00
antirez 43503ae5d6 redis-cli --help output improved with --scan and periods. 2014-01-22 12:08:08 +01:00
antirez b9c84d4aef redis-cli: support for --scan option. 2014-01-22 12:08:04 +01:00
antirez 248e916550 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 09:56:35 +01:00
Jonathan Pickett ce6c1f08b0 continuing code cleanup by moving win32 specific code into Win32_Interop directory 2014-01-17 18:24:31 -08:00
Jonathan Pickett f00034a438 fixed _MSC_VER version check 2014-01-17 14:41:53 -08:00
Jonathan Pickett 60a82ae058 Header file changes for compiling under vs2013. Solution files will remain at VS2012 for awhile. To build under VS2013 all that is required is to converting the solution files. 2014-01-17 12:40:51 -08:00
Jonathan Pickett 8b3daf2bb8 compile/link cleanup 2014-01-16 14:13:51 -08:00
Jonathan Pickett 93581da537 race condition for resetting completed event causing hard to reproduce AOF load errors 2014-01-16 14:12:36 -08:00
Jonathan Pickett d30d0c6a31 moving win32 specific code into a static lib so that it compiles only once, also some minor code tidying 2014-01-15 16:32:38 -08:00
antirez d401022d14 Fix typo in aofRewriteBufferAppend() comment. 2014-01-14 15:38:26 +01:00
antirez 4ea91aadfb 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 11:28:18 +01:00
antirez 71f7eb2590 Redis 2.8.4. 2014-01-13 17:09:58 +01:00
antirez df00b6ac05 SENTINEL SET: error on bad option name + flush config on error. 2014-01-13 16:39:56 +01:00
antirez b8db8a0c48 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 16:39:51 +01:00
antirez 8ed0b49525 Sentinel: fix wrong arity error message. 2014-01-13 16:39:47 +01:00
antirez 560e548dc6 Sentinel: SENTINEL REMOVE command added. 
The command totally removes a monitored master.
 2014-01-13 16:39:44 +01:00
antirez 0ca750d94a 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.
 2014-01-13 16:39:40 +01:00
antirez 1f9580c125 Sentinel: flush config on disk when new master is added. 2014-01-13 16:39:36 +01:00
antirez bb207007ca anetResolveIP() prototype added to anet.h. 2014-01-13 16:39:31 +01:00
antirez b7dc3204a6 Sentinel: SENTINEL MONITOR command implemented. 
It allows to add new masters to monitor at runtime.
 2014-01-13 16:39:27 +01:00
antirez 2e0ba7bb5b 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.
 2014-01-13 16:39:24 +01:00
antirez 52cf0975c7 Sentinel: added SENTINEL MASTER <name> command. 
With SENTINEL MASTERS it was already possible to list all the configured
masters, but not a specific one.
 2014-01-13 16:39:18 +01:00
antirez 6bcc370c9d Add all the configurable fields to addReplySentinelRedisInstance(). 
Note: the auth password with the master is voluntarily not exposed.
 2014-01-13 16:39:15 +01:00
antirez ea2bffa030 Trip comment to 80 cols in SentinelCommand(). 2014-01-13 16:39:11 +01:00
Jonathan Pickett 0d7366c329 workaround for deadlocking on RDB file copy during multiple master/slave synchronizations 2014-01-09 19:42:44 -08:00
antirez bca65866ff Test: regression for issues #1483. 2014-01-09 11:19:11 +01:00
antirez e4c51759f5 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 11:12:53 +01:00
Jonathan Pickett 8a47dc3ca5 more useful error message 2014-01-08 18:08:38 -08:00
Jonathan Pickett e12c4cdf15 Sync socket write would return WSAEWOULDBLOCK when send buffer full. Detecting this and adding a small delay then retrying so that migrate tests can finish. 2014-01-08 17:07:58 -08:00
Jonathan Pickett c32ccaa868 added early break to test if protocol error detected. went back to provious windows version of test. 2014-01-08 15:39:47 -08:00
antirez 51f71e2d8e Fix keyspace events flags-to-string conversion. 
Fixes issue #1491 on Github.
 2014-01-08 17:18:00 +01:00
antirez 05d219a6fa Test: stress events flags to/from string conversion. 2014-01-08 17:16:09 +01:00
antirez 2a1a31ca9d 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 14:27:49 +01:00
antirez 418d3d358a Clarify a comment in slaveTryPartialResynchronization(). 2014-01-08 14:11:02 +01:00
Jonathan Pickett 4d88ec9886 improving error reporting when fork fails for testing 2014-01-07 18:04:03 -08:00
Jonathan Pickett 663850ea73 Added dictionary master hash to global state being copied into qforked process. This fixes the fork/expire problems. 2014-01-07 17:04:52 -08:00
Jonathan Pickett f5a1b485c3 getting unit tests back to antirez verbatim 2014-01-06 17:31:47 -08:00
Jonathan Pickett 78b051a8be cleaning up unit tests. leaving in win32 checks around failing tests in aof, bgsave and scripting. 2014-01-03 19:45:45 -08:00
Jonathan Pickett 1849e30c7a adding delay to test because debug loadaof is taking too long for test to pass 2014-01-03 18:43:18 -08:00
Jonathan Pickett bddf4ac4da with recent changes these tests are working on windows 2014-01-03 18:22:49 -08:00
Jonathan Pickett 5decabc2c7 failing to erase from socket/fd backmaps caused confusion in multi.tcl 2014-01-03 18:00:59 -08:00
Jonathan Pickett c53e73573c FILE*->fd not handled well in ftruncate. Fix allows broken tests in aof.tcl to work. 2014-01-03 17:20:11 -08:00
Jonathan Pickett 449d6623ab a fix to get the integration tests working again, and added warnings about tests bypassed for Win32 2014-01-03 15:58:28 -08:00
Jonathan Pickett bf22d5646d replacing win32_cow implementation with win32_qfork. replication currently broken, but all other unit tests are working. 2014-01-02 12:06:06 -08:00
antirez 0a1a236e3e Log disconnection with slave only when ip:port is available. 2013-12-25 18:41:10 +01:00
antirez 27d06111db anetPeerToString / SockName: port can be NULL on errors too. 2013-12-25 18:39:49 +01:00
antirez 5b7c16137d 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 18:38:33 +01:00
antirez d07d4a876c Remove useless goto from anetTcpGenericConnect(). 2013-12-25 18:24:04 +01:00
antirez 9d1997706c 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 18:16:46 +01:00
Jonathan Pickett 5e2bb3e678 Removed x86 and non-IOCP targets in preparation for new qfork implementation. Only supported build henceforth will be x64 with IOCP. 2013-12-23 19:29:38 -08:00
antirez 4ad219adc8 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 12:50:52 +01:00
antirez c6db326d1d 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 11:34:15 +01:00
antirez 4456ee1173 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 11:54:10 +01:00
antirez 5fa937d956 Slave disconnection is an event worth logging. 2013-12-22 10:16:02 +01:00
antirez 9ba9d78ada Log when a slave lose the connection with its master. 2013-12-21 00:24:29 +01:00
antirez d33f3689d7 Clarify include directive behavior in example redis.conf. 2013-12-19 16:03:52 +01:00
antirez fb8a480f54 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 16:03:49 +01:00
Yubao Liu d62bbfda16 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 16:03:37 +01:00
antirez 090bcc946f CONFIG REWRITE: old development comments removed. 2013-12-19 16:02:50 +01:00
antirez ddd529bc3b 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 16:02:46 +01:00
antirez 673c42bb4e Example redis.conf formatted to better show appendfilename option. 2013-12-19 10:19:13 +01:00
antirez 33f6f35f34 Makefile.dep updated. 2013-12-13 13:14:47 +01:00
antirez 8eb1cb3b52 SDIFF iterator misuse bug regression test added. 
See commit c00453d for more info about the bug.
 2013-12-13 11:37:35 +01:00
antirez 993e0ede76 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 11:29:59 +01:00
antirez 2507e366b2 Sentinel: dead code removed. 2013-12-13 11:01:20 +01:00
antirez f9e9448a5b Makefile: remove odd syntax not compatible with some make versions. 
See issue #1448.
 2013-12-12 15:19:29 +01:00
Yubao Liu 45c60a903a fix typo in redis.conf and sentinel.conf 2013-12-12 11:32:05 +01:00
codeeply 5e75e681fe comment mistake fixed 2013-12-12 11:25:48 +01:00
antirez 8995634014 Redis 2.8.3. 2013-12-11 15:31:57 +01:00
antirez 9a8ae5a553 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 15:24:50 +01:00
Yossi Gottlieb 0ff078d8d0 Return proper error on requests with an unbalanced number of quotes. 2013-12-11 13:21:55 +01:00
Yossi Gottlieb 25ba2e9607 Fix wrong repldboff type which causes dropped replication in rare cases. 2013-12-11 11:37:58 +01:00
antirez 563d6b3f98 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 18:42:22 +01:00
antirez b6610a569d 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 18:18:36 +01:00
antirez 26cf5c8ac6 Log empty DB + Loading data into two separated messages. 2013-12-10 17:51:16 +01:00
antirez 303cc97ff9 Don't send more than 1 newline/sec while loading RDB. 2013-12-10 17:50:26 +01:00
antirez 75bf5a4a4a 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).
 2013-12-10 15:40:36 +01:00
antirez 8d0083ba25 Handle inline requested terminated with just \n. 2013-12-10 15:40:33 +01:00
antirez fba0b23e72 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 12:49:27 +01:00
antirez dceaca1f69 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 12:49:23 +01:00
antirez bf307cfb4d Fixed typo in redis.conf. 2013-12-06 10:48:40 +01:00
Anurag Ramdasan 4f9d30b33b Grammar fix. 2013-12-05 18:54:01 +01:00
Anurag Ramdasan b67f39da09 fixed typo 2013-12-05 17:18:37 +01:00
Anurag Ramdasan 4c53178c6c Fixed grammar: 'usually' to 'usual' 2013-12-05 16:42:28 +01:00
antirez 7f6743a581 Fixed grammar: before H the article is a, not an. 2013-12-05 16:37:21 +01:00
antirez edae78999c Fixed typos in redis.conf file. 2013-12-05 16:30:15 +01:00
antirez 3d7263aa41 Removed old comments and dead code from freeClient(). 2013-12-03 13:54:15 +01:00
antirez 333453646c Grammar fix in freeClient(). 2013-12-03 13:40:51 +01:00
antirez 4d650a3baa Redis 2.8.2. 2013-12-02 16:07:46 +01:00
antirez dd0ac4ac72 Sentinel: don't write HZ when flushing config. 
See issue #1419.
 2013-12-02 15:56:42 +01:00
antirez 75347ada7f 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 15:56:39 +01:00
antirez 83333b08d0 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 16:25:49 +01:00
antirez 50d140e90b SLAVEOF command refactored into a proper API. 
We now have replicationSetMaster() and replicationUnsetMaster() that can
be called in other contexts (for instance Redis Cluster).
 2013-11-28 16:19:16 +01:00
antirez 2eb8a46061 Reply to PING with error when there is a MISCONF state. 2013-11-28 16:16:58 +01:00
antirez a46f841df3 Sentinel: log vote received from other Sentinels. 2013-11-28 15:23:51 +01:00
Jonathan Pickett 7da645073b removing if{} block used for testing 2013-11-26 17:43:29 -08:00
Jonathan Pickett 926e5283c6 Isolating the socket differences beween Linux and Windows in Win32_FDAPI and APIBridge code. 2013-11-26 17:19:22 -08:00
huangz1990 e5c577e679 fix a bug in sentinel.c about pub/sub link 2013-11-26 15:11:58 +01:00
antirez dd239c37db Redis 2.8.1. 2013-11-25 11:21:49 +01:00
Salvatore Sanfilippo c61cc761d5 Merge pull request #1370 from yaauie/deprecate-redis-copy-script-2.8 
Deprecate redis copy script 2.8
 2013-11-25 02:10:32 -08:00
antirez d13635b2a9 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 10:57:20 +01:00
antirez d240202261 Sentinel: fix type specifier for Hello msg generation. 
This fixes issue #1395.
 2013-11-25 10:25:10 +01:00
antirez e83741746c Fix false positive in memory efficiency test. 
Fixes issue #1298.
 2013-11-25 10:21:18 +01:00
antirez 4e6d34bd2c Redis 2.8.0. 2013-11-22 10:25:29 +01:00
antirez f75fccafe7 Sentinel: example sentinel.conf updated. 2013-11-21 17:07:10 +01:00
antirez 312ca4dacc Sentinel: different comments updated to new implementation. 2013-11-21 16:23:11 +01:00
antirez c84275ee60 Sentinel: cleanup around SENTINEL_INFO_VALIDITY_TIME. 2013-11-21 16:06:07 +01:00
antirez 0b2639123b Sentinel: removed mem leak and useless code. 2013-11-21 15:44:10 +01:00
antirez 750b007d7c Sentinel: manual failover works again. 2013-11-21 15:24:27 +01:00
antirez 4bd1dd1c53 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 15:24:22 +01:00
antirez 46e2f3468c 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 15:24:15 +01:00
antirez 87098a39f0 Sentinel: Hello message sending code refactored. 2013-11-21 15:24:11 +01:00
antirez 9ecd819ecb Sentinel: select slave with best (greater) replication offset. 2013-11-21 15:24:08 +01:00
antirez eeb8cb3305 Sentinel: take the replication offset in slaves state. 2013-11-21 15:24:03 +01:00
antirez c517f72343 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.
 2013-11-21 15:23:47 +01:00
antirez 812f76a850 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.
 2013-11-21 15:22:55 +01:00
antirez fc93198ff9 Sentinel: various fixes to leader election implementation. 2013-11-21 15:22:52 +01:00
antirez 44b3684633 Sentinel: failover script execution fixed. 2013-11-21 15:22:48 +01:00
antirez 6d0400f569 Sentinel: no longer used defines removed. 2013-11-21 15:22:44 +01:00
antirez 1bbce5bf01 Sentinel: when writing config on disk, remember sentinels runid. 2013-11-21 15:22:27 +01:00
antirez 679dc8b09d Sentinel: arity of known-sentinel/slave is 4 not 3. 2013-11-21 15:22:21 +01:00
antirez 3e1fc6278d Sentinel: rewriteConfigSentinelOption() sub-iterators var typo fixed. 2013-11-21 15:22:18 +01:00
antirez 9a9c0cfaa6 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.
 2013-11-21 15:22:14 +01:00
antirez 93d924ff1c Sentinel: sentinelFlushConfig() to CONFIG REWRITE + fsync. 2013-11-21 15:22:11 +01:00
antirez a52909c5f2 Sentinel: CONFIG REWRITE support for Sentinel config. 2013-11-21 15:22:07 +01:00
antirez 8c3e197040 Sentinel: can-failover option removed, many comments fixed. 2013-11-21 15:22:02 +01:00
antirez 04b1fb0b1a Fix typo 'configuraiton' in rewriteConfigRewriteLine() comment. 2013-11-21 15:21:58 +01:00
antirez 0b9853ecdc 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.
 2013-11-21 15:21:55 +01:00
antirez 737062745d Sentinel: failover abort function simplified. 2013-11-21 15:21:50 +01:00
antirez 66b03c1a40 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.
 2013-11-21 15:21:46 +01:00
antirez 8ba31c218b 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).
 2013-11-21 15:21:41 +01:00
antirez ccaba966bc Sentinel: state machine and timeouts simplified. 2013-11-21 15:21:37 +01:00
antirez 3c4497e83c Sentinel: election timeout define. 2013-11-21 15:21:34 +01:00
antirez e15ba6a697 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).
 2013-11-21 15:21:30 +01:00
antirez 1a6abe7d79 Sentinel: master address selection in get-master-address refactored. 2013-11-21 15:21:26 +01:00
antirez 0eeb0a0782 Sentinel: fix conditional to only affect slaves with wrong master. 2013-11-21 15:21:21 +01:00
antirez 64c8de8657 Sentinel: simplify and refactor slave reconfig code. 2013-11-21 15:20:56 +01:00
antirez 782f9cacaf Sentinel: reconfigure slaves to right master. 2013-11-21 15:20:44 +01:00
antirez 7dbc0a63f5 Sentinel: remember last time slave changed master. 2013-11-21 15:20:40 +01:00
antirez 612dbb2a91 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.
 2013-11-21 15:20:36 +01:00
antirez 4ccf807abc Sentinel: safer slave reconfig, master reported role should match. 2013-11-21 15:20:31 +01:00
antirez e98d82c639 Sentinel: role reporting fixed and added in SENTINEL output. 2013-11-21 15:20:25 +01:00
antirez be19e5450c Sentinel: being a master and reporting as slave is considered SDOWN. 2013-11-21 15:20:20 +01:00
antirez 8c551d65a1 Sentinel: make sure role_reported is always updated. 2013-11-21 15:20:15 +01:00
antirez 9577fed8a3 Sentinel: track role change time. Wait before reconfigurations. 2013-11-21 15:20:11 +01:00
antirez fc10fb17da Sentinel: fix no-down check in master->slave conversion code. 2013-11-21 15:20:07 +01:00
antirez b02ef3d59a Sentinel: readd slaves back after a master reset. 2013-11-21 15:19:56 +01:00
antirez 1a1dc3de38 Sentinel: sentinelResetMaster() new flag to avoid removing set of sentinels. 
This commit also removes some dead code and cleanup generic flags.
 2013-11-21 15:19:52 +01:00
antirez 9f0e52a13d Sentinel: receive Pub/Sub messages from slaves. 2013-11-21 15:19:49 +01:00
antirez f7604b4c07 Sentinel: change event name when converting master to slave. 2013-11-21 15:19:45 +01:00
antirez 1569326277 Sentinel: added config-epoch to SENTINEL masters output. 2013-11-21 15:19:41 +01:00
antirez 9f20780de6 Sentinel: new failover algo, desync slaves and update config epoch. 2013-11-21 15:19:34 +01:00
antirez 2488257ad8 Sentinel: when starting failover seek for votes ASAP. 2013-11-21 15:19:30 +01:00
antirez 6593f33222 Sentinel: +new-epoch events. 2013-11-21 15:19:26 +01:00
antirez 48acc675dd Sentinel: wait some time between failover attempts. 2013-11-21 15:19:22 +01:00
antirez 447c2787a0 Sentinel: allow to vote for myself. 2013-11-21 15:19:16 +01:00
antirez eba4775b5d Sentinel: fix PUBLISH to masters and slaves. 2013-11-21 15:19:11 +01:00
antirez b95c6ed7b7 Sentinel: epoch introduced in leader vote. 2013-11-21 15:18:52 +01:00
antirez 663d79c0d5 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.
 2013-11-21 15:18:45 +01:00
antirez b72985d7f7 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.
 2013-11-21 15:18:41 +01:00
antirez fe7f96f18c Sentinel: remove code not useful in the new design. 2013-11-21 15:18:36 +01:00
antirez be2ef1b59f 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.
 2013-11-21 15:18:31 +01:00
Jonathan Pickett 14e809d0f7 adjusting unit test timing and forcing double evaluation rather than string evaluation 2013-11-12 18:36:17 -08:00
Jonathan Pickett e0f78cd084 eliminating COW dependent tests to verify that IOCP is not responsible for crashes and other strange behavior being seen 2013-11-11 16:17:23 -08:00
antirez 98682cd178 Log to what master a slave is going to connect to. 2013-11-11 09:25:41 +01:00
antirez 68b911478f Fixed typo in release candidate notes. 2013-11-10 10:25:09 +01:00
Ryan Biesemeyer 7d30b3121d Deprecate utils/redis-copy.rb in favor of redis-copy gem 2013-11-08 06:33:04 +00:00
antirez 875bc51909 Fix broken rdbWriteRaw() return value check in rdb.c. 
Thanks to @PhoneLi for reporting.
 2013-11-07 23:54:46 +01:00
Jonathan Pickett 59cdeb4e89 Disabling the COW replacement for fork() temporarily. This appears to fix crashing issues with redis. Also commenting out related unit tests. Replication is broken until COW (or a better fork replacement) is put back in. 2013-11-06 11:46:27 -08:00
antirez c874e39c45 Sentinel: sentinelSendSlaveOf() was missing a var and the prototype. 2013-11-06 11:29:57 +01:00
antirez 1767998751 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 11:29:53 +01:00
antirez 97810c45e8 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 11:29:49 +01:00
antirez f7f97bf730 Redis 2.7.106 (2.8 Release Candidate 6). 2013-11-06 09:24:26 +01:00
antirez 3a66e0c157 Fixed typo in parseScanCursorOrReply(): ad -> at. 
Thanks to @badboy for reporting it.
 2013-11-05 17:25:34 +01:00
antirez 060d56e7eb 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 17:25:29 +01:00
antirez 162acd8ac2 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 17:24:33 +01:00
antirez 406be0ea28 Use strtoul() instead of sscanf() in SCAN implementation. 2013-11-05 17:24:30 +01:00
antirez 4b615eea8f Added tests for [SHZ]SCAN with MATCH. 2013-11-05 17:24:26 +01:00
antirez 54a5a7dff8 HSCAN/ZSCAN: skip value when matching. 
This fixes issue #1360 and #1362.
 2013-11-05 17:24:22 +01:00
antirez bebbc7f9f7 Pass int64_t to intsetGet() instead of long long. 2013-11-05 17:24:17 +01:00
antirez 736e343509 removed not used vars in dictScan(). 2013-11-05 17:24:12 +01:00
Jonathan Pickett 6710555bc8 moved logging for hiredis and rediccli usage 2013-10-31 18:16:24 -07:00
antirez 59b30b327d SCAN: no sdsEncodedObject() API in Redis 2.8. 2013-10-31 18:22:29 +01:00
antirez 3d3e350b12 SSCAN with integer encoded object test improved. 2013-10-31 10:38:56 +01:00
antirez 3cc7d486bc Inverted variable boolean value and name after scanGenericCommand() refactoring. 2013-10-31 10:38:52 +01:00
antirez 06455014aa scanGenericCommand() refactoring and handling of integer encoded elements. 
This commit fixes issue #1354.
 2013-10-31 10:38:48 +01:00
antirez 141f30f421 Regression test added for [SHZ]SCAN issue #1354. 2013-10-31 10:38:45 +01:00
Jonathan Pickett 3032831099 2nd unit test needs to be investigated because it is frequently hanging unit test pass 2013-10-30 19:15:50 -07:00
antirez c53cab5dde Test: added a SCAN test trying to trigger HT resize. 2013-10-30 16:51:48 +01:00
antirez ab6f4195dd Test: added ZSCAN test. 2013-10-30 16:27:43 +01:00
antirez 13e879c9cf Test: added HSCAN test. 2013-10-30 16:27:39 +01:00
antirez 9ea69a58ac Test: added SSCAN test. 2013-10-30 16:27:36 +01:00
antirez bd1962acc4 SCAN test keys sorting turned into more idiomatic Tcl. 2013-10-30 16:27:31 +01:00
antirez 02617b6e92 SCAN: tests moved to unit/scan.tcl. 2013-10-30 16:27:28 +01:00
Jonathan Pickett e2162fc107 "r client list" returning information on 3 connections. Fixed loop logic to check for omem in all connections. 2013-10-29 19:21:02 -07:00
antirez bb2df795e0 Aesthetic fix (missing space) into HSCAN and ZSCAN implementations. 
Thanks to @badboy for reporting.
 2013-10-29 16:21:21 +01:00
antirez 35816a2e4d ZSCAN implemented. 2013-10-29 16:21:18 +01:00
antirez 1cc7e646af HSCAN implemented. 2013-10-29 16:21:13 +01:00
antirez 24ecabc995 SSCAN implemented. 2013-10-29 16:21:01 +01:00
antirez 17785f40b5 dictScan(): empty hash table requires special handling. 2013-10-29 16:20:13 +01:00
antirez 7f063b1c8c 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.
 2013-10-29 16:20:10 +01:00
antirez 9e62d9e79c 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.
 2013-10-29 16:20:06 +01:00
antirez 1a06bbe430 Fixed typos in dictScan() comment. 2013-10-29 16:20:02 +01:00
antirez 0880bec4e2 dictScan() algorithm documented. 2013-10-29 16:19:57 +01:00
antirez 1d659784f4 SCAN: stay inside 80 cols. 2013-10-29 16:19:54 +01:00
antirez 854b8be616 Revert "Fixed typo in SCAN comment. iff -> if." 
Probably here Pieter means "if and only if".

This reverts commit 43fdf3b404.
 2013-10-29 16:19:49 +01:00
antirez 5fa5153bf5 SCAN: simplify keys list cleanup using listSetFreeMethod(). 2013-10-29 16:19:45 +01:00
antirez 108c6e6734 SCAN: Fix test after option renamed from PATTERN to MATCH. 2013-10-29 16:19:41 +01:00
antirez 21da9b06a4 SCAN: improve variable names for readability. 2013-10-29 16:19:37 +01:00
antirez 907e06cd61 SCAN: remove additional newlines to conform to Redis code base. 2013-10-29 16:19:33 +01:00
antirez b44f1589a4 SCAN: remove useless assertion, already enforced by command table. 2013-10-29 16:19:28 +01:00
antirez d1c1f62ed5 SCAN: use define REDIS_LONGSTR_SIZE instead of fixed len. 2013-10-29 16:19:25 +01:00
antirez 8cdaf604ce Fixed typo in SCAN comment. iff -> if. 2013-10-29 16:19:21 +01:00
antirez 33409b4c6c SCAN option name changed: pattern -> match. 2013-10-29 16:19:17 +01:00
Pieter Noordhuis fd22106a98 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.
 2013-10-29 16:19:12 +01:00
Pieter Noordhuis 8c4ff679fe SCAN requires at least 1 argument 2013-10-29 16:19:08 +01:00
Pieter Noordhuis 0dd95e23ab Add SCAN command 2013-10-29 16:18:24 +01:00
Jonathan Pickett 07f2edf5e3 upping timeout for AOF rewrite on windows 2013-10-28 17:07:39 -07:00
antirez 9546f7846b redis-benchmark: update help for new __rand_int__ form. 2013-10-28 18:14:11 +01:00
Jonathan Pickett 37e9d78774 for non-IOCP build: fixes breakage in integration/replicationxx, unit/dump, and unit/instrospection unit tests 2013-10-25 16:02:00 -07:00
Jonathan Pickett 8dc2d3096d Fixing non-IOCP networking issues. Remapping WSAEWOULDBLOCK==>EAGAIN, and non-blocking send() can return 0 without having client disconnect. 2013-10-23 19:11:47 -07:00
Jonathan Pickett 5fbc129b4e Work in progress. Temporary move towards separate IOCP/non-IOCP builds because of thread state issues. Neither build set is particularily stable at the moment. 2013-10-18 13:34:16 -07:00
Jonathan Pickett e1f6d21cdf stability fixes 2013-10-09 15:12:57 -07:00
antirez 508d3c1bb7 Redis 2.7.105 (2.8 Release Candidate 5). 2013-10-09 13:08:42 +02:00
antirez df0c96002d 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 16:14:57 +02:00
antirez d7fa6d9aba 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 13:01:49 +02:00
antirez c8c1006cf4 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 12:27:40 +02:00
antirez 81f614ef6f Add REWRITE to CONFIG subcommands help message. 2013-10-04 12:27:36 +02:00
Jonathan Pickett 7071e9d599 Fix for issue #59. Generated FDs are used everywhere until a SOCKET handle is used by the Winsock APIs. The win32_socketmap code generates the simulated FDs and handles the mapping from these to socket handles. 2013-10-03 16:12:54 -07:00
Michel Martens 20c506f432 Document the redis-cli --csv option. 2013-09-26 10:12:51 +02:00
Jonathan Pickett de48871570 converting main solution to vs2012 2013-09-20 14:16:30 -07:00
Jonathan Pickett 3561ada2db upgraded tool chain to vs2012 2013-09-20 13:05:30 -07:00
antirez af967ff9cb Allow AUTH / PING when disconnected from slave and serve-stale-data is no. 2013-09-17 09:46:40 +02:00
antirez 2d55f1ca7f 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 10:37:55 +02:00
Maxim Zakharov 9a0f27a5d0 mistype fixed 2013-09-03 15:15:42 +02:00
Maxim Zakharov c713b1ebdd A mistype fixed 2013-09-03 15:15:34 +02:00
antirez 56ce068664 Redis 2.7.104 (2.8 Release Candidate 4). 2013-08-30 09:52:18 +02:00
antirez 23acf93f97 Test: Lua stack leak regression test added. 2013-08-30 08:59:16 +02:00
antirez 0ea9a20d47 Test: added a memory efficiency test. 2013-08-30 08:48:07 +02:00
antirez 752b1fca90 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 11:49:28 +02:00
Salvatore Sanfilippo 3e62d7c308 Merge pull request #1244 from yihuang/2.8 
fix lua_cmsgpack pack map as array
 2013-08-27 06:15:32 -07:00
antirez a57294b339 Fix an hypothetical issue in processMultibulkBuffer(). 2013-08-27 13:00:14 +02:00
antirez facc5c0661 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 12:36:09 +02:00
antirez 79a1d335c8 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 12:30:42 +02:00
antirez 0f32c37fc9 Fix DEBUG SDSLEN after 2.8 back port. 2013-08-27 12:26:02 +02:00
antirez df21771c69 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 12:22:12 +02:00
antirez f05b1f6889 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 12:22:07 +02:00
antirez 9c3e6c3aa0 Update server.lastbgsave_status when fork() fails. 2013-08-27 10:16:25 +02:00
antirez 9df63fcabf Only run the fast active expire cycle if master & enabled. 2013-08-27 09:31:43 +02:00
antirez f5903846bc Opening TCP listening ports refactored into a function. 2013-08-22 14:06:54 +02:00
antirez bd6327855b Print error message when can't bind * on any address. 2013-08-22 13:03:07 +02:00
antirez 2c94d80f58 Fix for issue #1214 simplified. 2013-08-21 11:41:35 +02:00
Allan 778f753deb fixed initServer fail while having no IPv6 nor IPv4 2013-08-21 11:41:30 +02:00
Allan 6c2b34a465 fixed initServer failed if no IPV4 or no IPV6 2013-08-21 11:41:25 +02:00
Allan 0a4656d63f fixed bug issue of #1213 2013-08-21 11:41:21 +02:00
antirez 7c861340b1 Use printf %zu specifier to print private_dirty. 2013-08-20 12:09:36 +02:00
antirez 2c018b14fb dictFingerprint(): cast pointers to integer of same size. 2013-08-20 12:09:31 +02:00
antirez 6a88a84ec7 Redis 2.7.103 (2.8 Release Candidate 3). 2013-08-19 18:04:32 +02:00
antirez 5621272575 Revert "Fixed type in dict.c comment: 265 -> 256." 
This reverts commit 009515f914.
 2013-08-19 17:25:57 +02:00
antirez 009515f914 Fixed type in dict.c comment: 265 -> 256. 2013-08-19 15:11:07 +02:00
antirez 837817b2b7 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 15:02:16 +02:00
antirez bd353d9a3c 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 15:02:13 +02:00
antirez c7da5fc6c1 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 15:02:10 +02:00
antirez 8368550733 Fix comments for correctness in zunionInterGenericCommand(). 
Related to issue #1240.
 2013-08-19 15:02:06 +02:00
antirez 3ad87c652b 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 15:02:02 +02:00
antirez 05379f49ee 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 15:01:58 +02:00
yihuang 706056ca16 fix lua_cmsgpack pack map as array 2013-08-13 17:48:40 +08:00
antirez 14a1cba3e3 Use precomptued objects for bulk and mbulk prefixes. 2013-08-12 12:43:26 +02:00
antirez 004f00bfa4 replicationFeedSlaves() func name typo: feedReplicationBacklogWithObject -> feedReplicationBacklog. 2013-08-12 12:38:52 +02:00
antirez 45d4e06e93 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 12:31:20 +02:00
antirez 3eab283b65 Fix a PSYNC bug caused by a variable name typo. 2013-08-12 11:51:03 +02:00
antirez 0e72cc601f Fix sdsempty() prototype in sds.h. 2013-08-12 11:38:28 +02:00
antirez 689829f2e1 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 16:47:18 +02:00
antirez 4fe67cc103 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 14:31:54 +02:00
Jan-Erik Rediger a256b83448 Little typo 2013-08-07 16:09:33 +02:00
antirez a79862de2a redis-benchmark: fix memory leak introduced by 346256f 2013-08-07 16:09:13 +02:00
antirez bcc7965931 redis-benchmark: max pipeline length hardcoded limit removed. 2013-08-07 16:09:09 +02:00
antirez 46c1bafeab redis-benchmark: fix db selection when :rand: feature is used. 2013-08-07 11:22:33 +02:00
antirez 151ad68540 redis-benchmark: ability to SELECT a specifid db number. 2013-08-07 11:22:28 +02:00
antirez fa48b1fa32 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 15:36:43 +02:00
antirez 31d0f34100 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 15:36:39 +02:00
antirez 00c8cfef74 Some activeExpireCycle() refactoring. 2013-08-06 15:36:35 +02:00
antirez 8686a70303 Remove dead code and fix comments for new expire code. 2013-08-06 15:36:31 +02:00
antirez d54e373b84 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.
 2013-08-06 15:36:27 +02:00
antirez 500155b91b 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.
 2013-08-06 15:36:23 +02:00
antirez cc94697284 Redis 2.7.102 (2.8 Release Candidate 2). 2013-07-30 20:20:02 +02:00
antirez f8e43aba78 Test: regression test for issue #1221. 2013-07-29 17:40:18 +02:00
antirez 13f7ade551 Fix replicationFeedSlaves() off-by-one bug. 
This fixes issue #1221.
 2013-07-28 12:50:35 +02:00
antirez 78644f5e3e 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 11:02:20 +02:00
antirez aea8020554 Use latest sds.c in the hiredis library under deps. 2013-07-25 10:34:02 +02:00
antirez c3e1724a57 Ignore sdsrange return value. 2013-07-24 18:59:36 +02:00
antirez f899ab55ca 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 11:22:52 +02:00
antirez a1d37ba469 Inline protocol improved to accept quoted strings. 2013-07-24 10:38:05 +02:00
antirez 57c8e026b9 Every function inside sds.c is now commented. 2013-07-23 16:36:14 +02:00
Henry Rawas 3339a22399 Install MSI for Redis 2013-07-21 12:45:59 -07:00
Henry Rawas aa392e3fb7 Issue #51. Path to zip files changed 2013-07-21 12:45:58 -07:00
Henry Rawas 2d92308f38 Issue #43 - Fix async connect and write for sentinel. 2013-07-21 12:45:58 -07:00
Henry Rawas eff7cdfad0 Issue #46 - file length calculation overflow. Switch to 64bit values. 2013-07-21 12:45:35 -07:00
Henry Rawas b4554da5fd #41 aeWinReceiveDone not initialized 2013-07-21 12:45:16 -07:00
Henry Rawas aedf3bfcd1 #43 Redis-sentinel async communication fixes for windows 2013-07-21 12:45:16 -07:00
Henry Rawas fe7c28cc50 remove app.config from proj 2013-07-21 12:44:24 -07:00
Henry Rawas e4b13e5eb0 more small typos 2013-07-21 12:44:24 -07:00
Henry Rawas 17dd06cc97 Fixed typo in executable name 2013-07-21 12:44:23 -07:00
Henry Rawas 0929eec2f7 Fix some bad hyphen and quote chras in readme 2013-07-21 12:44:22 -07:00
Henry Rawas 1e6147fada Add missing exe files, quote for subscription, x64 switch, name conflict 
Conflicts:

	.gitignore
 2013-07-21 12:44:22 -07:00
Henry Rawas 5180055384 Fix Azure Installer README formatting 2013-07-21 12:44:21 -07:00
Henry Rawas b1f0b92f2d adding Windows Azure Installer 2013-07-21 12:44:20 -07:00
Henry Rawas c84e5ba84d update readme, disable permission test 2013-07-21 12:42:45 -07:00
Henry Rawas 3a5ee522b5 move release bin directory 2013-07-21 12:42:44 -07:00
Henry Rawas 8f77c71751 Avoid using closed handle and double free memory after replication failure 2013-07-21 12:42:40 -07:00
Henry Rawas 22c2b323da Issue #33 inconsistency for float rounding 2013-07-21 12:42:36 -07:00
Henry Rawas 96418fd48e Issue #32 - showing 32/64 bit in logo and version 2013-07-21 12:42:32 -07:00
Henry Rawas 1f42527c58 Modify and enable protocol desync test for windows 2013-07-21 12:42:29 -07:00
Henry Rawas 89cae3d62e Improve non blocking opening and closing of sockets 2013-07-21 12:42:09 -07:00
Henry Rawas cb85f493cb 2.6 windows binaries (not signed) 2013-07-21 12:41:14 -07:00
Henry Rawas 896c1e659f Redis 2.6 for Windows 2013-07-21 12:39:55 -07:00
Henry Rawas 19c476f4ce Updated Readme for 2.6.8 2013-07-21 12:26:08 -07:00
Henry Rawas 239f5233d5 move release bin directory 2013-07-21 12:26:07 -07:00
Henry Rawas 097dd94bb4 add license 2013-07-21 12:26:05 -07:00
Henry Rawas 07b031839a Avoid using closed handle and double free memory after replication failure 2013-07-21 12:26:05 -07:00
Henry Rawas c179bcfd66 Issue #33 inconsistency for float rounding 2013-07-21 12:26:02 -07:00
Henry Rawas a6a8aebec9 Issue #32 - showing 32/64 bit in logo and version 2013-07-21 12:25:58 -07:00
Henry Rawas 81eed0db38 Modify and enable protocol desync test for windows 2013-07-21 12:25:55 -07:00
Henry Rawas e955536015 Improve non blocking opening and closing of sockets 2013-07-21 12:25:55 -07:00
Henry Rawas 47bbccf9be 2.6 windows binaries (not signed) 2013-07-21 11:55:26 -07:00
Henry Rawas 11d870bd55 Redis 2.6 for Windows 2013-07-21 11:54:30 -07:00
antirez f3b44f291d Merge remote-tracking branch 'origin/2.8' into 2.8 2013-07-19 14:55:49 +02:00
antirez 5d0ad2f983 Fixed typo in 2.8 release notes. 2013-07-18 16:10:31 +02:00
Salvatore Sanfilippo 349f6349be Merge pull request #1203 from ronnix/patch-1 
Fix a few typos in release notes
 2013-07-18 03:27:41 -07:00
Ronan Amicel 5b19e9d433 Fix a few typos in release notes 2013-07-18 12:23:47 +02:00
antirez 3cb3714e99 Redis 2.7.101 (2.8 Release Candidate 1). 2013-07-18 11:26:53 +02:00
antirez a50635bb6c 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 15:04:22 +02:00
antirez 2e449d42e0 Fixed typo in rio.h, simgle -> single. 2013-07-16 15:43:41 +02:00
yoav 9f6f436a51 Chunked loading of RDB to prevent redis from stalling reading very large keys. 2013-07-16 15:41:59 +02:00
antirez 112e763618 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 15:08:57 +02:00
Ted Nyman fe04710908 Make sure the log standardizes on 'timeout' 2013-07-12 23:12:27 +02:00
antirez 98757b4a6e Use the environment locale for strcoll() collation. 2013-07-12 13:39:48 +02:00
antirez 18fabeb264 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 13:39:44 +02:00
antirez d8fcbb6645 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 13:39:40 +02:00
Jan-Erik Rediger 1bcbb7a90c Wrap IPv6 in brackets in the prompt. 2013-07-12 10:58:47 +02:00
antirez 1e23848ed3 Sentinel: embed IPv6 address into [] when naming slave/sentinel instance. 2013-07-11 17:10:09 +02:00
antirez 076f6395b9 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 17:09:44 +02:00
antirez 4cfd70e903 hiredis: minimal IPv6 support. 2013-07-11 17:09:35 +02:00
antirez 98d0abcecd 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.
 2013-07-11 17:09:30 +02:00
antirez 8669e70921 anet.c: save some vertical space. 2013-07-11 17:09:25 +02:00
antirez 9e089e7c5d anet.c: use SO_REUSEADDR when creating listening sockets. 
It used to be ok, but the socket option was removed when adding IPv6
support.
 2013-07-11 17:09:22 +02:00
antirez cdf801d92e Use getClientPeerId() for MONITOR implementation. 2013-07-11 17:09:18 +02:00
antirez 3472d045d8 getClientPeerId() refactored into two functions. 2013-07-11 17:09:14 +02:00
antirez da18366609 getClientPeerId() now reports errors. 
We now also use it in CLIENT KILL implementation.
 2013-07-11 17:09:10 +02:00
antirez 4fa68b2815 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.
 2013-07-11 17:09:04 +02:00
antirez a7451c1b6d All IP string repr buffers are now REDIS_IP_STR_LEN bytes. 2013-07-11 17:07:52 +02:00
antirez 2e75d3947c IPv6: bind IPv4 and IPv6 interfaces by default. 2013-07-11 17:07:41 +02:00
antirez b5423b099c Fix old anetPeerToString() API call in replication.c 2013-07-11 17:07:38 +02:00
antirez 1e7949227c 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.
 2013-07-11 17:07:34 +02:00
Geoff Garside 0d8f254359 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.
 2013-07-11 17:07:31 +02:00
Geoff Garside 5938126c98 Cleanup main() and BACKTRACE mistaken pulled while rebasing. 2013-07-11 17:07:27 +02:00
Geoff Garside d3bf85423a Fix calls to anetPeerToString() missing buffer size. 2013-07-11 17:07:22 +02:00
Geoff Garside 47620a3186 Document port6 and bind6 config options. 
Add commented port6 and bind6 options to default redis.conf file.
 2013-07-11 17:07:18 +02:00
Geoff Garside fd8a6ae7a6 Add anetTcp6Server() function. 
Refactor the common code from anetTcpServer into internal function which
can be used by both anetTcpServer and anetTcp6Server.
 2013-07-11 17:07:13 +02:00
Geoff Garside f638236902 Add static anetV6Only() function. 
This function sets the IPV6_V6ONLY option to 1 to use separate stack
IPv6 sockets.
 2013-07-11 17:07:07 +02:00
Geoff Garside 71795d4e7e 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.
 2013-07-11 17:07:03 +02:00
Geoff Garside d37d006cd2 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.
 2013-07-11 17:06:55 +02:00
Geoff Garside 57e2b23dfb 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.
 2013-07-11 17:06:43 +02:00
Geoff Garside 39d57c057f 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.
 2013-07-11 17:06:35 +02:00
Geoff Garside f1151f869e 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.
 2013-07-11 17:05:21 +02:00
Geoff Garside fefa14914e 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.
 2013-07-11 17:05:16 +02:00
Geoff Garside 77ddec8582 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.
 2013-07-11 17:05:12 +02:00
Geoff Garside 4b2e374e4a Update calls to anetResolve to include buffer size 2013-07-11 17:05:08 +02:00
Geoff Garside 3570411def Update calls to anetPeerToString to include ip_len. 2013-07-11 17:05:03 +02:00
Geoff Garside 68d72aa5b1 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.
 2013-07-11 17:04:32 +02:00
Geoff Garside 30836721f2 Update anetTcpAccept & anetPeerToString calls. 
Add the additional ip buffer length argument to function calls of
anetTcpAccept and anetPeerToString in network.c and cluster.c
 2013-07-11 17:02:29 +02:00
Geoff Garside 920ab4c926 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.
 2013-07-11 17:02:12 +02:00
Geoff Garside aee5ee70fa 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.
 2013-07-11 17:02:08 +02:00
Geoff Garside e50f2c93bc 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.
 2013-07-11 17:02:04 +02:00
Geoff Garside d36dc1fc82 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.
 2013-07-11 17:01:59 +02:00
Geoff Garside 855ff8df9f 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.
 2013-07-11 17:01:55 +02:00
antirez 33c5f0a243 Example redis.conf: bind to multiple interfaces documented. 2013-07-08 10:42:37 +02:00
antirez 323e5e2058 Remove cluster stuff from closeListeningSockets() for 2.8. 2013-07-08 10:31:49 +02:00
antirez fc022ca300 Binding multiple IPs done properly with multiple sockets. 2013-07-08 10:30:54 +02:00
antirez 00063bea33 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.
 2013-07-08 10:28:02 +02:00
antirez 6dabd34ad0 Ability to bind multiple addresses. 2013-07-08 10:27:21 +02:00
antirez 14857decfa anet.c: Allow creation of TCP listening sockets bound to N addresses. 2013-07-08 10:25:51 +02:00
antirez 2150cfebb7 sds.c: new function sdsjoin() to join strings. 2013-07-08 10:25:47 +02:00
antirez 3e3501262b 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 12:18:55 +02:00
antirez 24b3799238 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 12:02:09 +02:00
antirez 2f6560f351 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 17:48:49 +02:00
charsyam 847a40805f fix randstring bug 2013-07-02 12:25:30 +02:00
antirez aaf5623535 Only allow basenames for dbfilename and appendfilename. 
This fixes issue #1094.
 2013-07-02 12:24:12 +02:00
antirez 25a3ad7f74 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 12:24:08 +02:00
antirez 028fdbb70b getAbsolutePath() moved into utils.c 2013-07-02 12:24:05 +02:00
antirez 8090c37f71 CONFIG SET maxclients. 2013-07-01 10:44:08 +02:00
antirez 75dc48f04a ae.c event loop: API to resize the fd set size on the run. 2013-07-01 10:44:04 +02:00
antirez 0a0fd37e5b Allow SHUTDOWN in loading state. 2013-06-27 12:14:23 +02:00
antirez 7bfbc38068 Redis 2.7.3 2013-06-26 15:26:11 +02:00
antirez bb0d0fd479 function renamed: popcount_binary -> redisPopcount. 2013-06-26 15:24:58 +02:00
antirez cdf79c063f 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 15:24:30 +02:00
antirez 7cca200603 Test: add some AOF testing to EVALSHA replication test. 2013-06-26 15:24:26 +02:00
antirez 29308077a8 Flush the replication script cache after SCRIPT FLUSH. 2013-06-26 15:24:22 +02:00
antirez 9faa6b9dc3 Test: EVALSHA replication. 2013-06-26 15:24:19 +02:00
antirez b0e0f3c5b2 Test: randomInt() behavior commented. 2013-06-26 15:24:16 +02:00
antirez 4e2b97bb89 Test: replication-3 test speedup in master-slave setup. 2013-06-26 15:24:12 +02:00
antirez d34c1ac208 Force propagation of SCRIPT LOAD to AOF. 2013-06-26 15:24:09 +02:00
antirez b7233f40fb SCRIPT FLUSH comment minor pedantic improvement. 2013-06-26 15:24:04 +02:00
antirez 8450d6e24d 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.
 2013-06-26 15:23:52 +02:00
antirez 545fe0c318 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.
 2013-06-26 15:23:29 +02:00
antirez 9d894b1b8c 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.
 2013-06-26 15:23:15 +02:00
antirez 8328d993e1 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).
 2013-06-26 15:21:55 +02:00
antirez aeab473dfb Allow PUBSUB NUMSUB without channels. 
The result is an empty list but it is handy to call it programmatically.
 2013-06-26 15:21:12 +02:00
antirez a8f1474dfd 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.
 2013-06-26 15:21:08 +02:00
antirez 72d3fbea65 Fix comment typo in integration/aof.tcl. 2013-06-26 15:21:05 +02:00
YAMAMOTO Takashi 3642557ae6 don't assume time_t == long 
time_t is always 64bit on recent versions of NetBSD.
 2013-06-26 15:20:59 +02:00
YAMAMOTO Takashi 30f3ae03a5 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.
 2013-06-26 15:20:56 +02:00
YAMAMOTO Takashi f27896a17d 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
 2013-06-26 15:20:32 +02:00
YAMAMOTO Takashi 9a7b09ed35 don't define _XOPEN_SOURCE for NetBSD 
on NetBSD, defining _XOPEN_SOURCE hides extensions
like inet_aton, strcasecmp, etc.
 2013-06-26 15:20:03 +02:00
Jan-Erik Rediger 90c3a14ac0 Initialize char* to NULL to remove compiler warning 2013-06-20 16:59:04 +02:00
antirez f2f2b4eb9d Redis 2.6.14 2013-06-20 10:36:40 +02:00
antirez 252f0b6556 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 10:24:43 +02:00
antirez 9af8125c7d 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 10:24:31 +02:00
antirez ee92c043fc Test: regression test for #1163. 2013-06-19 18:53:17 +02:00
antirez be8d5fd954 Test: regression test for #1163. 2013-06-19 18:53:14 +02:00
antirez 8e63fd0734 Allow writes from scripts called by AOF loading in read-only slaves. 
This fixes issue #1163
 2013-06-19 18:29:10 +02:00
antirez 86147afd08 Allow writes from scripts called by AOF loading in read-only slaves. 
This fixes issue #1163
 2013-06-19 18:29:05 +02:00
antirez 449ec38fe4 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 14:45:38 +02:00
antirez bf8e078aa2 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 19:38:23 +02:00
antirez b3fbc84cc1 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 19:38:19 +02:00
Marc-Antoine Perennou 1cf839a216 test-server: only listen to 127.0.0.1 
Signed-off-by: Marc-Antoine Perennou <Marc-Antoine@Perennou.com>
 2013-06-18 19:38:16 +02:00
ioddly e34f00842f Try to report source of bad Lua API calls 2013-06-18 19:38:10 +02:00
antirez a73da27853 Binary safe dump of object content in redisLogObjectDebugInfo(). 2013-06-04 15:56:26 +02:00
antirez b435817b17 Binary safe dump of object content in redisLogObjectDebugInfo(). 2013-06-04 15:55:35 +02:00
antirez 054632f551 CONFIG SET: accept slave-priority zero, it is valid. 2013-05-31 19:31:51 +02:00
antirez d34c623e10 Test: avoid a false positive in min-slaves test. 2013-05-31 11:43:23 +02:00
antirez fe4b62fcaf 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.
 2013-05-31 07:20:19 +02:00
antirez 971217c7b4 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).
 2013-05-31 07:19:45 +02:00
antirez f5275da6e9 Refresh good slaves count when setting slave state as online. 2013-05-31 07:19:42 +02:00
antirez 9a7ea3dae9 Tests added for min-slaves feature. 2013-05-30 18:54:39 +02:00
antirez 4333e6ce20 Make tests compatible with new INFO replication output. 2013-05-30 11:43:53 +02:00
antirez 9d1bc3efcf min-replicas-to-write: only deny write commands. 
I guess I needed another coffee...
 2013-05-30 11:32:00 +02:00
antirez d0d67f8d42 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 11:31:46 +02:00
antirez fc2587ec6e min-slaves-to-write: initial description of the feature in redis.conf 2013-05-30 11:30:34 +02:00
antirez 75fcf5c936 repl_offset field in INFO replication is now just offset. 2013-05-29 19:56:40 +02:00
antirez c64338aa8f 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 19:55:11 +02:00
antirez 889d39b5ba Accept REPLCONF in any state. 2013-05-28 15:26:55 +02:00
antirez 1b87b3ef38 A comment about BLPOP timeout did not reflected actual behavior. 2013-05-27 19:34:12 +02:00
antirez 2d497014ea Version bumped to 2.7.2. 2013-05-27 11:43:47 +02:00
antirez 75bf91abea redis.conf updated: repl-timeout now uesd by masters as well. 2013-05-27 11:43:21 +02:00
antirez 308940aa2c 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 11:43:17 +02:00
antirez 45e6a4023a 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 11:43:14 +02:00
antirez a74f8fe1ad Don't ACK the master after every command. 
Sending an ACK is now moved into the replicationSendAck() function.
 2013-05-27 11:43:10 +02:00
antirez 0000d5334d Make sure that REPLCONF ACK really has no return value. 2013-05-27 11:43:06 +02:00
antirez 146f1d7d86 Replication: send REPLCONF ACK to master. 2013-05-27 11:43:03 +02:00
antirez 1e77b77de4 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 11:43:00 +02:00
antirez 0387f86947 Fixed a bug in no queueing replies to master. 2013-05-25 01:00:16 +02:00
antirez 7ae29d5d9e 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 19:48:03 +02:00
antirez 302b0a5c7b 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 18:42:18 +02:00
antirez 0b398b6ffa Don't stop reading redis.conf if line has no args. 
Should be "continue" and was "return".

This fixes issue #1110
 2013-05-18 16:22:24 +02:00
antirez 059018c4cd CONFIG REWRITE: remove cluster stuff for 2.8 back port. 2013-05-15 12:02:44 +02:00
antirez 414d5db85c Use memtoll() when parsing the backlog-size option. 2013-05-15 12:01:32 +02:00
antirez 604efb02ab CONFIG REWRITE: backlog size is a bytes option. 2013-05-15 12:01:27 +02:00
antirez 7d338fd982 CONFIG REWRITE: bindaddr -> bind. 2013-05-15 12:01:23 +02:00
antirez 2173f30836 CONFIG REWRITE: when rewriting amount of bytes use GB, MB, KB if possible. 2013-05-15 12:01:14 +02:00
antirez 901ceab20b CONFIG REWRITE: correctly escape the notify-keyspace-events option. 2013-05-15 12:01:06 +02:00
antirez b4ce02988c CONFIG REWRITE: "active-rehashing" -> "activerehashing". 2013-05-15 12:01:02 +02:00
antirez 2a29cf430a CONFIG REWRITE: fixed typo in AOF fsync policy. 2013-05-15 12:00:57 +02:00
antirez c95a60960b CONFIG REWRITE: repl-disable-tcp-nodelay is a boolean option. 2013-05-15 12:00:52 +02:00
antirez e8cc959300 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 12:00:47 +02:00
antirez 180cfaae8e 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 12:00:43 +02:00
antirez 417253afc3 CONFIG REWRITE: Use sane perms when creating redis.conf from scratch. 2013-05-15 12:00:38 +02:00
antirez 8feb3d261a CONFIG REWRITE: actually rewrite the config file, atomically. 2013-05-15 12:00:32 +02:00
antirez a461376ca9 CONFIG REWRITE: remove orphaned lines. 2013-05-15 12:00:28 +02:00
antirez 8cf0b66606 CONFIG REWRITE: strip multiple empty lines. 2013-05-15 12:00:24 +02:00
antirez cb8433f313 CONFIG REWRITE: support for client-output-buffer-limit. 2013-05-15 12:00:21 +02:00
antirez f62851aea9 CONFIG REWRITE: support for dir and slaveof. 2013-05-15 12:00:17 +02:00
antirez 91ef10ef45 CONFIG REWRITE: support for save and syslog-facility. 2013-05-15 12:00:13 +02:00
antirez 973f793b04 CONFIG REWRITE: Initial support code and design. 2013-05-15 12:00:03 +02:00
antirez 7a5d3d91a1 Obtain absoute path of configuration file, expose it in INFO. 2013-05-15 11:59:26 +02:00
antirez d8dd70c2dd redis-cli: help.h updated. 2013-05-14 11:22:58 +02:00
antirez feffd5635b redis-cli: help.h updated. 2013-05-14 11:22:50 +02:00
antirez 41babac68d Test: more PSYNC tests (backlog TTL). 2013-05-09 12:52:16 +02:00
antirez cb55ad991b Version bumped to 2.7.1 2013-05-08 15:24:54 +02:00
antirez c847c73e78 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 13:02:53 +02:00
antirez 2c93d34bde 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 11:59:47 +02:00
antirez 2865bb7f1f 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 11:59:42 +02:00
Jiahao Huang 6251c3e66e 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
 2013-05-08 09:59:35 +02:00
Jiahao Huang 3d93dffd5d 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
 2013-05-08 09:56:10 +02:00
antirez 71b442276f Added more info about 32 bit builds into README. 2013-05-08 09:48:49 +02:00
antirez 85671d0e3b Added more info about 32 bit builds into README. 2013-05-08 09:48:46 +02:00
antirez 6a947b1d54 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 17:22:42 +02:00
Michel Martens 736220b86f Reset masterauth if an empty string is configured. 2013-05-02 17:22:38 +02:00
antirez b05d2516fe 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 17:22:30 +02:00
Michel Martens b60465eb78 Reset masterauth if an empty string is configured. 2013-05-02 17:21:55 +02:00
charsyam b37512a28f 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 15:46:59 +02:00
charsyam a94db7908a 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 15:46:56 +02:00
antirez d96497961c Redis 2.6.13 2013-04-30 15:15:56 +02:00
antirez 4f38d03254 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 15:09:34 +02:00
antirez 59ff2fe9a4 Sentinel: more sensible delay in master demote after tilt. 2013-04-30 15:09:31 +02:00
antirez fb0d08e35f 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.
 2013-04-30 15:09:27 +02:00
antirez e7bcec829c 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 15:09:14 +02:00
antirez 4028a777b6 Sentinel: more sensible delay in master demote after tilt. 2013-04-30 15:09:10 +02:00
antirez 70845320cc 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.
 2013-04-30 15:09:06 +02:00
antirez 395e6a5872 Lua updated to version 5.1.5. 2013-04-29 12:02:04 +02:00
antirez 2b427a89fb Lua updated to version 5.1.5. 2013-04-29 12:02:01 +02:00
antirez 34a57a5b52 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 11:34:05 +02:00
antirez d2ff5ed603 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 11:34:02 +02:00
antirez 967ae8ca67 Config option to turn AOF rewrite incremental fsync on/off. 2013-04-24 10:57:38 +02:00
antirez b06f13e7b7 Config option to turn AOF rewrite incremental fsync on/off. 2013-04-24 10:57:35 +02:00
antirez c735116c87 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 10:28:43 +02:00
antirez 8b41e19e47 rio.c: added ability to fdatasync() from time to time while writing. 2013-04-24 10:27:55 +02:00
antirez 9ca306d874 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 10:27:02 +02:00
antirez 5276996cf1 rio.c: added ability to fdatasync() from time to time while writing. 2013-04-24 10:26:58 +02:00
antirez cf7b2c329b 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 14:16:58 +02:00
antirez e426ea5245 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 14:16:47 +02:00
antirez bac29c8b7c Sentinel: turn old master into a slave when it comes back. 2013-04-22 11:26:33 +02:00
antirez d0c9a2a767 Sentinel: turn old master into a slave when it comes back. 2013-04-22 11:26:29 +02:00
antirez a500c120ba Test: split conceptually unrelated comments in RDB test. 2013-04-22 11:26:02 +02:00
antirez 4848af1b9e Test: split conceptually unrelated comments in RDB test. 2013-04-22 11:25:59 +02:00
antirez 172eac504a More explicit panic message on out of memory. 2013-04-19 15:11:59 +02:00
antirez 3a865829a7 More explicit panic message on out of memory. 2013-04-19 15:11:55 +02:00
antirez 0e3baa5241 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 16:23:12 +02:00
antirez 5bf9b53a83 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 16:21:32 +02:00
antirez 42ba5d3788 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 13:19:52 +02:00
antirez 0d7b10ee6e redis-cli: --latency-history mode implemented. 2013-04-11 13:19:48 +02:00
antirez 40861516ac 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 13:19:39 +02:00
antirez 92a7b0102c redis-cli: --latency-history mode implemented. 2013-04-11 13:19:33 +02:00
antirez 9f2e6ac0ed Test: remove useless statements and comments from test default config. 2013-04-04 14:32:17 +02:00
antirez 8d81e68d18 Test: remove useless statements and comments from test default config. 2013-04-04 14:31:30 +02:00
antirez 63978d338d Make rio.c comment 80-columns friendly. 2013-04-03 12:41:31 +02:00
antirez d83d72f380 Make rio.c comment 80-columns friendly. 2013-04-03 12:41:27 +02:00
antirez ed2d988192 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 14:13:03 +02:00
antirez d6b0c18c51 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 14:12:28 +02:00
Johan Bergström 0bc7dbda1c use install as default installer (except on SunOS) 2013-04-02 11:54:20 +02:00
Johan Bergström 034d94193d Silence mkdir output 2013-04-02 11:30:16 +02:00
Johan Bergström 2094eb36e7 Only pass -rdynamic as linker option 2013-04-02 11:30:13 +02:00
Johan Bergström 31844fe138 Remove extra spaces 2013-04-02 11:29:56 +02:00
Johan Bergström c72459bf18 make check is a common naming convention for tests 2013-04-02 11:27:38 +02:00
Johan Bergström ca0b1c2436 Inherit CC for Lua 2013-04-02 11:26:50 +02:00
Johan Bergström 49eef7738e Spaces to tabs 2013-04-02 11:26:41 +02:00
Johan Bergström 0c8690d46e Slightly refactor CFLAGS/LDFLAGS/LIBS 
This way, we can avoid -rdynamic and -pthread warnings on darwin.
 2013-04-02 11:26:37 +02:00
antirez 7da5980848 Redis 2.6.12 2013-03-29 17:42:39 +01:00
charsyam 062f60a36e Support for case unsensitive SET options. 2013-03-29 10:37:33 +01:00
charsyam cf8484b723 Support for case unsensitive SET options. 2013-03-29 10:37:29 +01:00
antirez 4243483d3e Test: Extended SET tests. 2013-03-28 16:45:45 +01:00
antirez d785413d86 Extended SET command implemented (issue #931). 2013-03-28 16:45:37 +01:00
antirez 0c447d3506 Test: Extended SET tests. 2013-03-28 16:39:39 +01:00
antirez 24aa61785d Extended SET command implemented (issue #931). 2013-03-28 16:39:34 +01:00
antirez e8a1a169dd EXPIRE should not resurrect keys. Issue #1026. 2013-03-28 12:49:03 +01:00
antirez dc0f13774b Test: regression test for issue #1026. 2013-03-28 12:48:59 +01:00
antirez 47911b8630 Better DEBUG error message when num of arguments is wrong. 2013-03-28 12:48:55 +01:00
antirez 7ff729e192 Test: verify that lazy-expire works. 2013-03-28 12:48:51 +01:00
antirez 02b9a72bae DEBUG set-active-expire added. 
We need the ability to disable the activeExpireCycle() (active
expired key collection) call for testing purposes.
 2013-03-28 12:48:47 +01:00
antirez e6e07042a1 EXPIRE should not resurrect keys. Issue #1026. 2013-03-28 12:48:38 +01:00
antirez be832929c8 Test: regression test for issue #1026. 2013-03-28 12:48:35 +01:00
antirez 4de4080ea5 Better DEBUG error message when num of arguments is wrong. 2013-03-28 12:48:30 +01:00
antirez 978185bf67 Test: verify that lazy-expire works. 2013-03-28 12:48:26 +01:00
antirez 10d8e6a712 DEBUG set-active-expire added. 
We need the ability to disable the activeExpireCycle() (active
expired key collection) call for testing purposes.
 2013-03-28 12:48:19 +01:00
antirez c6a9a20d3b Test: test replication of MULTI/EXEC. 2013-03-27 12:05:22 +01:00
antirez 61b1d6da18 Test: Restore DB back to 9 after testing MULTI/EXEC with DB 5. 2013-03-27 12:05:19 +01:00
antirez 7009ff98b8 Test: new functions to capture and analyze the replication stream. 2013-03-27 12:05:15 +01:00
antirez 1862239136 Test: read_from_replication_stream, workaround for 2.6 2013-03-27 12:03:46 +01:00
antirez f8791a158b Test: test replication of MULTI/EXEC. 2013-03-27 11:57:32 +01:00
antirez 26319d08b5 Test: Restore DB back to 9 after testing MULTI/EXEC with DB 5. 2013-03-27 11:57:28 +01:00
antirez fc78d978de Test: new functions to capture and analyze the replication stream. 2013-03-27 11:57:23 +01:00
antirez 6cb7860658 Flag PUBLISH as read-only in the command table. 2013-03-27 09:07:43 +01:00
antirez 611dcb56ee 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.
 2013-03-27 09:07:23 +01:00
antirez 889b017403 Transactions: use discardTransaction() in EXEC implementation. 2013-03-27 09:07:18 +01:00
antirez d1369c3d9f 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.
 2013-03-27 09:06:59 +01:00
antirez a452b54849 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 13:59:33 +01:00
antirez 10c6195e3f Flag PUBLISH as read-only in the command table. 2013-03-26 13:59:26 +01:00
antirez 2618101ad5 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 13:59:21 +01:00
antirez 82516e8168 Transactions: use discardTransaction() in EXEC implementation. 2013-03-26 13:59:18 +01:00
antirez d91dfaf6dc 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 13:59:14 +01:00
antirez 150f7e43ca Allow SELECT while loading the DB. 
Fixes issue #1024.
 2013-03-26 13:59:05 +01:00
antirez 6818905406 Allow SELECT while loading the DB. 
Fixes issue #1024.
 2013-03-26 13:59:00 +01:00
NanXiao ef1cf15c14 Update config.c 
Fix bug in configGetCommand function: get correct masterauth value.
 2013-03-25 19:32:24 +01:00
NanXiao 382420ec6e Update config.c 
Fix bug in configGetCommand function: get correct masterauth value.
 2013-03-25 19:32:20 +01:00
antirez 6b7562e204 Test: obuf-limits test false positive removed. 
Fixes #621.
 2013-03-25 11:56:31 +01:00
antirez dee8d84270 Test: obuf-limits test false positive removed. 
Fixes #621.
 2013-03-25 11:56:08 +01:00
antirez 5576a28977 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.
 2013-03-25 11:52:32 +01:00
antirez ec18d4bfba 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.
 2013-03-25 11:52:25 +01:00
antirez 4f8b18f3dd 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 12:55:06 +01:00
antirez 58708fa65a 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 12:55:02 +01:00
antirez c64255990e Test: make sure broken RDB checksum is detected. 2013-03-13 11:15:33 +01:00
antirez ba8367ee49 Test: make sure broken RDB checksum is detected. 2013-03-13 11:15:31 +01:00
antirez ba3c5494d8 Test: more RDB loading checks. 
A test for issue #1001 is included.
 2013-03-13 10:09:21 +01:00
antirez 684b61505a Test: check that Redis starts empty without an RDB file. 2013-03-13 10:09:16 +01:00
antirez 189e865c51 rdbLoad(): rework code to save vertical space. 2013-03-13 10:09:09 +01:00
Damian Janowski b9f8c2a5b0 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.
 2013-03-13 10:08:47 +01:00
antirez 67a745515d Test: more RDB loading checks. 
A test for issue #1001 is included.
 2013-03-13 10:08:40 +01:00
antirez 7b6fb960cc Test: check that Redis starts empty without an RDB file. 2013-03-13 10:08:36 +01:00
antirez 64b91c1493 rdbLoad(): rework code to save vertical space. 2013-03-13 10:08:30 +01:00
Damian Janowski edb3c10361 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.
 2013-03-13 10:08:26 +01:00
antirez 18d16f8592 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 18:36:07 +01:00
antirez b2e42a90d3 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 18:36:02 +01:00
antirez 3a00520e5f Add a missing bugfix entry in 2.6.11 release notes. 2013-03-12 17:55:23 +01:00
antirez 378fcaad55 redis-cli --bigkeys: don't crash with empty DBs. 2013-03-12 09:57:57 +01:00
antirez f93d9929d8 redis-cli --bigkeys: don't crash with empty DBs. 2013-03-12 09:57:49 +01:00
antirez 6589821035 Redis 2.6.11 2013-03-11 19:51:10 +01:00
antirez 48f4f77189 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 11:34:49 +01:00
antirez b3281e93c3 In databasesCron() never test more DBs than we have. 2013-03-11 11:34:45 +01:00
antirez 2677c97d39 Make comment name match var name in activeExpireCycle(). 2013-03-11 11:34:34 +01:00
antirez 1d426bf53b Optimize inner loop of activeExpireCycle() for no-expires case. 2013-03-11 11:34:18 +01:00
antirez da2dd8991b REDIS_DBCRON_DBS_PER_SEC -> REDIS_DBCRON_DBS_PER_CALL 2013-03-11 11:34:14 +01:00
antirez 13f84841b5 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.
 2013-03-11 11:34:10 +01:00
antirez ef3a95fa6f Use unsigned integers for DB ids, for defined wrap-to-zero. 2013-03-11 11:34:05 +01:00
antirez 665b819eb4 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.
 2013-03-11 11:33:59 +01:00
antirez bb562a6414 Actually call databasesCron() inside serverCron(). 2013-03-11 11:30:26 +01:00
antirez e166abad10 Move Redis databases background processing to databasesCron(). 2013-03-11 11:30:21 +01:00
antirez 8e7d962291 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 11:30:13 +01:00
antirez c5afbb6e2a In databasesCron() never test more DBs than we have. 2013-03-11 11:30:09 +01:00
antirez 7e7ee815b0 Make comment name match var name in activeExpireCycle(). 2013-03-11 11:30:04 +01:00
antirez 6a8288766c Optimize inner loop of activeExpireCycle() for no-expires case. 2013-03-11 11:30:01 +01:00
antirez 9afb7789b3 REDIS_DBCRON_DBS_PER_SEC -> REDIS_DBCRON_DBS_PER_CALL 2013-03-11 11:29:57 +01:00
antirez 1a73260dce 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.
 2013-03-11 11:29:53 +01:00
antirez 8abfe5fe8b Use unsigned integers for DB ids, for defined wrap-to-zero. 2013-03-11 11:29:49 +01:00
antirez a4bb4b29fb 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.
 2013-03-11 11:29:45 +01:00
antirez aa7f74c7e9 Actually call databasesCron() inside serverCron(). 2013-03-11 11:29:41 +01:00
antirez 21822fd0e9 Move Redis databases background processing to databasesCron(). 2013-03-11 11:29:34 +01:00
antirez aec5ea5d07 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.
 2013-03-11 11:28:55 +01:00
Gengliang Wang 2d24bf2f94 Removed useless "return" statements in pubsub.c 
(original commit message edited)
 2013-03-06 16:50:56 +01:00
Gengliang Wang 6ae8fb477a Removed useless "return" statements in pubsub.c 
(original commit message edited)
 2013-03-06 16:50:53 +01:00
antirez bc1b2e8f96 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 16:36:57 +01:00
antirez deccd104d8 Add a warning about command renaming in redis.conf. 2013-03-06 16:36:57 +01:00
antirez e7a61d287e 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 16:36:52 +01:00
antirez 78bae8b078 Add a warning about command renaming in redis.conf. 2013-03-06 16:36:52 +01:00
antirez bf6c5d960d 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 12:44:28 +01:00
antirez 95fc9cc0a6 redis-cli: use sdsfreesplitres() instead of hand-coding it. 2013-03-06 12:44:23 +01:00
antirez 5de5efa364 sds.c: sdssplitargs_free() removed as it was a duplicate. 2013-03-06 12:44:19 +01:00
antirez 1f2d5941a1 More specific error message in loadServerConfigFromString(). 2013-03-06 12:44:16 +01:00
antirez 60aa3a67e8 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 12:44:10 +01:00
antirez 921aec17b6 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 12:44:06 +01:00
charsyam e174290307 Don't segfault on unbalanced quotes. 2013-03-06 12:44:02 +01:00
antirez 5bcfa2ebbc 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 12:43:49 +01:00
antirez 00a5e86e44 redis-cli: use sdsfreesplitres() instead of hand-coding it. 2013-03-06 12:43:45 +01:00
antirez f72a3406a7 sds.c: sdssplitargs_free() removed as it was a duplicate. 2013-03-06 12:43:41 +01:00
antirez 68532cf01e More specific error message in loadServerConfigFromString(). 2013-03-06 12:43:37 +01:00
antirez a70e3ffb16 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 12:43:32 +01:00
antirez a6f557657c 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 12:43:27 +01:00
charsyam 5aeb1f71a3 Don't segfault on unbalanced quotes. 2013-03-06 12:43:22 +01:00
antirez a3273dbfa6 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 11:51:26 +01:00
antirez aadb8aaa08 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 11:51:23 +01:00
antirez e9318d92db 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 11:16:24 +01:00
antirez df3b492045 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 11:16:21 +01:00
0x20h 803c5b140d suppress external diff program when using git diff. 2013-03-04 10:59:22 +01:00
0x20h e618eb4fb3 suppress external diff program when using git diff. 2013-03-04 10:58:02 +01:00
Stam He 74373cd930 add a check for aeCreateTimeEvent 
1) Add a check for aeCreateTimeEvent in function initServer.
 2013-03-04 10:52:53 +01:00
Stam He b5047d9b2f point 2 of slave-serve-stale-data miss '-' between 'stale' and 'data' 2013-03-04 10:49:30 +01:00
Stam He a74056f1e0 point 2 of slave-serve-stale-data miss '-' between 'stale' and 'data' 2013-03-04 10:49:19 +01:00
antirez 02cd545f7d A comment in main() clarified. 2013-02-27 12:47:51 +01:00
antirez b148513fec Remove warning when printing redisBuildId(). 2013-02-27 12:47:38 +01:00
antirez 5e9e96999f Remove too agressive/spamming log in rdb.c. 2013-02-27 12:47:33 +01:00
antirez d2a37badc2 Use GCC printf format attribute for redisLog(). 
This commit also fixes redisLog() statements producing warnings.
 2013-02-27 12:47:16 +01:00
antirez 0d8d7f106a Better panic message for failed time event creation. 2013-02-27 12:00:51 +01:00
Stam He 438dfdd13d add a check for aeCreateTimeEvent 
1) Add a check for aeCreateTimeEvent in function initServer.
 2013-02-27 12:00:48 +01:00
Stam He 6f96ac1c1c 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 12:00:43 +01:00
antirez f69b0a0db8 setproctitle.c: declar tmp as static so valgrind will not detect a leak. 2013-02-26 15:51:15 +01:00
antirez ac3100bc3b 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 12:03:48 +01:00
Arnaud Granal fedcd51185 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 18:38:48 +01:00
antirez ff56772115 PSYNC: another change to unexpected reply from PSYNC. 2013-02-13 18:43:45 +01:00
antirez 964ee00de0 PSYNC: More robust handling of unexpected reply to PSYNC. 2013-02-13 18:34:18 +01:00
antirez 3b4c0d80d5 Avoid compiler warning by casting to match printf() specifier. 2013-02-13 13:38:32 +01:00
antirez 34b420dbf3 Replication: more strict error checking for master PING reply. 2013-02-12 16:59:38 +01:00
antirez d96a66f531 Replication: more strict error checking for master PING reply. 2013-02-12 16:59:27 +01:00
antirez 31f0a6ec50 Replication: added new stats counting full and partial resynchronizations. 2013-02-12 16:26:42 +01:00
antirez 5fe2577a19 Return a specific NOAUTH error if authentication is required. 2013-02-12 16:25:47 +01:00
antirez 02ac84ddbd Test: avoid false positives in CLIENT SETNAME closed connection test. 2013-02-12 13:27:32 +01:00
antirez 9854252798 Test: avoid false positives in CLIENT SETNAME closed connection test. 2013-02-12 13:27:30 +01:00
Steven Penny 45bfab8596 Format to fit 80 columns 
This makes it readable on GitHub and editors without auto wrapping.
 2013-02-12 13:11:47 +01:00
Steven Penny 5a8cf340a4 Format to fit 80 columns 
This makes it readable on GitHub and editors without auto wrapping.
 2013-02-12 13:11:44 +01:00
antirez 0ae1a6b1c3 Add missing bracket removed for error after rebase of PSYNC. 2013-02-12 12:58:07 +01:00
antirez 2b1398e870 PSYNC: debugging printf() calls are now logs at DEBUG level. 2013-02-12 12:58:03 +01:00
antirez 5a4ef3e5b6 Remove harmless warning in slaveTryPartialResynchronization(). 2013-02-12 12:57:59 +01:00
antirez 4fe7672dfc 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 12:57:56 +01:00
antirez 8529dd218f SYNC not allowed with pending data on the static output buffer. 2013-02-12 12:57:52 +01:00
antirez 6726ea5b53 Log the unexpected string received in place of the SYNC payload length. 2013-02-12 12:57:48 +01:00
antirez f5edd535d1 After SLAVEOF <newslave> don't allow chained slaves to PSYNC. 2013-02-12 12:57:44 +01:00
antirez 700e5eb4fc PSYNC: work in progress, preview #2, rebased to unstable. 2013-02-12 12:57:40 +01:00
antirez 4d8655cfd3 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 12:57:36 +01:00
antirez 01c21f9943 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 12:57:31 +01:00
antirez 5a35e485f9 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 12:57:26 +01:00
antirez 948fdb5f7d Redis 2.6.10 2013-02-11 13:24:39 +01:00
antirez c970816e4d Makefile: valgrind target added (forces -O0 and libc malloc). 2013-02-11 12:11:26 +01:00
antirez fe97d7b1db Makefile: valgrind target added (forces -O0 and libc malloc). 2013-02-11 12:11:21 +01:00
antirez e88517ac30 TCP keep-alive. Better documentation in redis.conf. 2013-02-11 11:47:35 +01:00
antirez 2d89c53d0d 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.
 2013-02-11 11:47:31 +01:00
antirez 0fe052efaa tcp-keepalive option documented in redis.conf. 2013-02-11 11:47:28 +01:00
antirez e1e8b1cd29 Set SO_KEEPALIVE on client sockets if configured to do so. 2013-02-11 11:47:23 +01:00
antirez 609a3ec1be Add SO_KEEPALIVE support to anet.c. 2013-02-11 11:47:20 +01:00
antirez 0b50d1ce9a Sentinel: advertise the promoted slave address only after successful setup. 2013-02-11 11:47:16 +01:00
Pierre Chapuis c7b9a57fb2 fix comments forgotten in #285 (zipmap -> ziplist) 2013-02-11 11:47:10 +01:00
antirez 8a22934c68 TCP keep-alive. Better documentation in redis.conf. 2013-02-11 11:44:35 +01:00
antirez b06b90b5d8 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.
 2013-02-11 11:44:31 +01:00
antirez 0a14a6542c tcp-keepalive option documented in redis.conf. 2013-02-11 11:44:27 +01:00
antirez 2ed3fc1502 Set SO_KEEPALIVE on client sockets if configured to do so. 2013-02-11 11:44:23 +01:00
antirez f2817cbd9e Add SO_KEEPALIVE support to anet.c. 2013-02-11 11:44:18 +01:00
antirez fcfdbda104 Sentinel: advertise the promoted slave address only after successful setup. 2013-02-11 11:44:14 +01:00
Pierre Chapuis 0d470b4c43 fix comments forgotten in #285 (zipmap -> ziplist) 2013-02-11 11:44:06 +01:00
antirez 63ad0e8867 Make all WATCHers dirty when the slave reloads the DB. 2013-02-08 10:27:26 +01:00
antirez cc55a4525a Make all WATCHers dirty when the slave reloads the DB. 2013-02-08 10:27:21 +01:00
antirez d5d0b467db LASTSAVE is a "random" command. 2013-02-07 19:13:09 +01:00
antirez 5eb0ec142d LASTSAVE is a "random" command. 2013-02-07 19:13:06 +01:00
antirez 5f7dff4d16 TCP_NODELAY after SYNC: changes to the implementation. 2013-02-05 12:05:39 +01:00
charsyam 1d80acae54 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 12:05:24 +01:00
antirez cf0191dc3c TCP_NODELAY after SYNC: changes to the implementation. 2013-02-05 12:05:04 +01:00
charsyam 45b1b2f7ad 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 12:05:01 +01:00
antirez 14daa01c52 Test: No clients timeout while testing. 2013-02-05 12:02:41 +01:00
antirez eb07793b59 Test: No clients timeout while testing. 2013-02-05 12:02:36 +01:00
Johan Bergström 22ccc6db0f Use info nameofexectuable to find current executable 2013-02-05 11:45:11 +01:00
Johan Bergström f9a8b0eced Enforce tcl 8.5 or newer 2013-02-05 11:45:08 +01:00
Johan Bergström 6da090e916 Check available tcl versions 2013-02-05 11:45:05 +01:00
Johan Bergström d2a8bca82b Use info nameofexectuable to find current executable 2013-02-05 11:44:55 +01:00
Johan Bergström 1124cda37a Enforce tcl 8.5 or newer 2013-02-05 11:44:51 +01:00
Johan Bergström e92edd3cee Check available tcl versions 2013-02-05 11:44:44 +01:00
Rock Li 2543fa6476 retval doesn't initalized 
If each if conditions are all fail, variable retval will under uninitlized
 2013-02-05 10:49:24 +01:00
Rock Li 8c0b9f778e retval doesn't initalized 
If each if conditions are all fail, variable retval will under uninitlized
 2013-02-05 10:49:20 +01:00
David Celis 63dc34bd01 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>
 2013-02-04 10:36:24 +01:00
David Celis 44d0eac568 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>
 2013-02-04 10:34:05 +01:00
Gengliang Wang 9ddd0f7796 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 10:33:24 +01:00
Gengliang Wang 57e411c687 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 10:33:21 +01:00
antirez 8024de7eb0 Z*STORE event fixed: generate del only if resulting sorted set is empty. 2013-01-29 13:50:09 +01:00
antirez ca8e7d4f8a Generate del events when S*STORE commands delete the destination key. 2013-01-29 13:43:24 +01:00
antirez 664de412af Tests for keyspace notifications. 2013-01-28 13:19:04 +01:00
antirez d3cf778d82 Send 'expired' events when a key expires by lookup. 2013-01-28 13:18:57 +01:00
antirez 9db8fbcfdd Keyspace notifications: fixed a leak and a bug introduced in the latest commit. 2013-01-28 13:18:53 +01:00
antirez b9bc4f9132 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 13:18:36 +01:00
antirez 6de052cd77 decrRefCount -> decrRefCountVoid in list constructor. 2013-01-28 13:17:37 +01:00
antirez 81af92b023 Enable keyspace events notification when testing. 2013-01-28 13:17:33 +01:00
antirez 4ff138fbbe Keyspace events added for more commands. 2013-01-28 13:17:30 +01:00
antirez 2825f21fd8 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 13:17:26 +01:00
antirez a64383c861 notifyKeyspaceEvent(): release channel names using the right pointers. 2013-01-28 13:17:22 +01:00
antirez fdfb59beae Initial test events for the new keyspace notification API. 2013-01-28 13:17:18 +01:00
antirez dd2ce743ca Fixed over-80-cols comment in db.c 2013-01-28 13:17:15 +01:00
antirez 6e64525cc2 Two fixes to initial keyspace notifications API. 2013-01-28 13:17:11 +01:00
antirez 212edbc409 Keyspace events notification API. 2013-01-28 13:17:00 +01:00
antirez 153766e137 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 10:53:31 +01:00
antirez c17a7f6fbc 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 10:53:28 +01:00
antirez bfad381f57 redis-cli --bigkeys output is now simpler to understand. 2013-01-21 19:16:14 +01:00
antirez 767a53aa84 redis-cli --bigkeys output is now simpler to understand. 2013-01-21 19:16:11 +01:00
antirez 3ff75e58e8 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 19:02:24 +01:00
antirez 742e580f40 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 18:59:52 +01:00
antirez 850117a8a5 Fixed a bug in memtest progress bar, that had no actual effects. 
This closes issue #859, thanks to @erbenmo.
 2013-01-21 12:38:46 +01:00
antirez 1b0f244509 Fixed a bug in memtest progress bar, that had no actual effects. 
This closes issue #859, thanks to @erbenmo.
 2013-01-21 12:38:43 +01:00
Bilal Husain 2f1318ab29 s/adiacent/adjacent/ 
fixed typo in a comment (step 2 memcheck)
 2013-01-21 12:20:07 +01:00
antirez f6f43d7da0 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 12:07:48 +01:00
antirez 580f65a50b 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 12:07:44 +01:00
antirez a154908a8f Additionally two typos fixed thanks to @jodal 2013-01-19 13:46:24 +01:00
antirez 635c532c89 Additionally two typos fixed thanks to @jodal 2013-01-19 13:46:22 +01:00
antirez 39f0a33f78 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 13:30:46 +01:00
antirez da540228d8 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 13:30:41 +01:00
antirez 9e5b70162c 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 13:20:58 +01:00
antirez 1e20c939fe 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 13:20:56 +01:00
antirez d766907cfb 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 12:55:15 +01:00
antirez dbde1d85ea 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 12:55:12 +01:00
guiquanz 560e049947 Fixed many typos. 2013-01-19 11:08:43 +01:00
guiquanz 1caf09399e Fixed many typos. 
Conflicts fixed, mainly because 2.8 has no cluster support / files:
	00-RELEASENOTES
	src/cluster.c
	src/crc16.c
	src/redis-trib.rb
	src/redis.h
 2013-01-19 11:03:19 +01:00
charsyam c2b43eac4f redis-cli prompt bug fix 2013-01-19 10:34:25 +01:00
charsyam ff1e4d22a6 redis-cli prompt bug fix 2013-01-19 10:34:22 +01:00
Jan-Erik Rediger 5e2aa2cafb Always exit if connection fails. 
This avoids unnecessary core dumps. Fixes antirez/redis#894
 2013-01-19 10:29:51 +01:00
Jan-Erik Rediger a8f9cec16a Always exit if connection fails. 
This avoids unnecessary core dumps. Fixes antirez/redis#894
 2013-01-19 10:29:42 +01:00
bitterb aadcda9977 Fix an error reply for CLIENT command 2013-01-19 10:26:32 +01:00
bitterb a732d86fe0 Fix an error reply for CLIENT command 2013-01-19 10:26:25 +01:00
Nathan Parry f2bc198d40 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 11:50:06 +01:00
Nathan Parry 0f4dbd9a11 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 11:49:58 +01:00
antirez 21159d53f5 Redis 2.6.9 2013-01-16 19:54:56 +01:00
antirez c48584b85d redis-cli: save an RDB dump from remote server to local file. 2013-01-16 19:48:30 +01:00
antirez 52f6a1dc16 redis-cli: save an RDB dump from remote server to local file. 2013-01-16 19:48:26 +01:00
antirez f9c1263f31 Tests for CLIENT GETNAME/SETNAME. 2013-01-15 13:34:37 +01:00
antirez 1b2ac3b206 Typo fixed, ASCI -> ASCII. 2013-01-15 13:34:35 +01:00
antirez c5f23ca79b 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 13:34:34 +01:00
antirez 7897e0a0d6 Tests for CLIENT GETNAME/SETNAME. 2013-01-15 13:34:26 +01:00
antirez f31d10b9a6 Typo fixed, ASCI -> ASCII. 2013-01-15 13:34:24 +01:00
antirez 786bd3938e 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 13:34:22 +01:00
antirez 27abaa238f 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 13:33:30 +01:00
antirez aa9497fdb3 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 13:33:27 +01:00
antirez 64ee9c861f Makefile.dep updated. 2013-01-11 23:51:12 +01:00
antirez 78b175a241 Makefile.dep updated. 2013-01-11 23:51:00 +01:00
antirez 14cc0045dc Redis 2.6.8 2013-01-10 17:15:16 +01:00
antirez 023cbc3787 Comment in the call() function clarified a bit. 2013-01-10 12:04:52 +01:00
antirez b70b9c6027 Test: added regression for issue #872. 2013-01-10 12:04:48 +01:00
antirez 46dc3fbb53 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 12:04:45 +01:00
antirez 8a70007f9e Comment in the call() function clarified a bit. 2013-01-10 12:04:24 +01:00
antirez e2e36cf19b Test: added regression for issue #872. 2013-01-10 12:04:19 +01:00
antirez 414e06653c 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 12:04:15 +01:00
antirez cefff3a1b3 Better error reporting when fd event creation fails. 2013-01-03 14:32:32 +01:00
antirez 6570d2f89e 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 14:32:27 +01:00
antirez c1c3e9b756 Better error reporting when fd event creation fails. 2013-01-03 14:32:11 +01:00
antirez 1f8ad7ae55 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 14:32:06 +01:00
antirez 858b5576ac 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 15:21:40 +01:00
antirez 4468ba2317 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 15:21:37 +01:00
antirez 61fb0aacf5 CONTRIBUTING updated with request to add BSD license. 2012-12-17 11:17:55 +01:00
antirez d8a0fddd51 CONTRIBUTING updated with request to add BSD license. 2012-12-17 11:17:52 +01:00
antirez 6fa2610ac9 Added missing license and copyright in deps/hiredis. 2012-12-17 11:11:36 +01:00
antirez 886c9ecb7c Added missing license and copyright in deps/hiredis. 2012-12-17 11:11:27 +01:00
antirez a6d117b6c0 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 17:20:21 +01:00
Patrick TJ McPhee 5207c03c49 Define _XOPEN_SOURCE appropriately on NetBSD. 2012-12-12 18:42:41 +01:00
Patrick TJ McPhee 8e0a9f6027 Define _XOPEN_SOURCE appropriately on NetBSD. 2012-12-12 18:42:36 +01:00
antirez d64a9cf514 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 17:19:58 +01:00
antirez bbee226ecd 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 17:19:54 +01:00
antirez d9301f05e2 Redis 2.6.7 2012-12-03 12:53:09 +01:00
Brian J. McManus 17e31eb03a Issue 804 Add Default-Start and Default-Stop LSB tags for RedHat startup and update-rc.d compatability. 2012-12-03 12:31:12 +01:00
Brian J. McManus 6c9897f6cf Issue 804 Add Default-Start and Default-Stop LSB tags for RedHat startup and update-rc.d compatability. 2012-12-03 12:31:10 +01:00
antirez 984f6edf0a 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 12:17:56 +01:00
antirez a2b3fff235 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 12:17:52 +01:00
antirez 124cb6dda1 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 12:17:33 +01:00
antirez e39735214d 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 12:17:29 +01:00
antirez 07a9f85405 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 20:42:59 +01:00
antirez fbf3e33d18 Test: regression for issue #801. 2012-12-02 20:42:56 +01:00
antirez 64e69c0d5f 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 20:42:51 +01:00
antirez 54b08c86a6 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 20:36:18 +01:00
antirez 6bdcd50b56 Test: regression for issue #801. 2012-12-01 23:07:09 +01:00
antirez cac49a9031 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.
 2012-12-01 12:26:07 +01:00
antirez c1eda786d6 SDIFF fuzz test added. 2012-11-30 17:14:08 +01:00
antirez ccc974d90b 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 17:13:34 +01:00
antirez 2572bb13c6 redis-benchmark: seed the PRNG with time() at startup. 2012-11-30 17:13:25 +01:00
antirez ed70955510 Make an EXEC test more latency proof. 2012-11-29 16:12:26 +01:00
antirez e34c14df1a Make an EXEC test more latency proof. 2012-11-29 16:12:23 +01:00
antirez 3b71404d70 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 14:22:15 +01:00
antirez cd99c14e38 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 10:24:41 +01:00
antirez b41f4e5a48 Redis 2.6.6 2012-11-28 18:53:09 +01:00
antirez 21b26915ce Jemalloc updated to version 3.2.0. 2012-11-28 18:42:41 +01:00
antirez 053f1f357d Jemalloc updated to version 3.2.0. 2012-11-28 18:42:35 +01:00
charsyam 1f0d0b094a Remove unnecessary condition in _dictExpandIfNeeded (dict.c) 2012-11-28 11:45:52 +01:00
Matt Arsenault ccb500cf36 It's a watchdog, not a watchdong. 2012-11-28 11:42:24 +01:00
charsyam ed48550fe6 remove compile warning bioKillThreads 2012-11-28 11:42:20 +01:00
antirez 3ccfb5a4d6 Redis 2.6.5 2012-11-22 17:03:27 +01:00
antirez d0570c9693 EVALSHA is now case insensitive. 
EVALSHA used to crash if the SHA1 was not lowercase (Issue #783).
Fixed using a case insensitive dictionary type for the sha -> script
map used for replication of scripts.
 2012-11-22 15:51:03 +01:00
antirez 9120275dc9 EVALSHA is now case insensitive. 
EVALSHA used to crash if the SHA1 was not lowercase (Issue #783).
Fixed using a case insensitive dictionary type for the sha -> script
map used for replication of scripts.
 2012-11-22 15:51:01 +01:00
antirez cc85ed41df Fix integer overflow in zunionInterGenericCommand(). 
This fixes issue #761.
 2012-11-22 15:28:48 +01:00
antirez de00a5a092 Fix integer overflow in zunionInterGenericCommand(). 
This fixes issue #761.
 2012-11-22 15:28:45 +01:00
antirez 2b45ef020c Test: MULTI state is cleared after EXECABORT error. 2012-11-22 10:36:36 +01:00
antirez 52bd3d8a39 Test: make sure EXEC fails after previous transaction errors. 2012-11-22 10:36:32 +01:00
antirez 37d978d9b4 Test: MULTI/EXEC tests moved into multi.tcl. 2012-11-22 10:36:28 +01:00
antirez 41f0f927c9 Safer handling of MULTI/EXEC on errors. 
After the transcation starts with a MULIT, the previous behavior was to
return an error on problems such as maxmemory limit reached. But still
to execute the transaction with the subset of queued commands on EXEC.

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

With this change:

1) EXEC fails if at least one command was not queued because of an
error. The EXECABORT error is used.
2) A generic error is always reported on EXEC.
3) The client DISCARDs the MULTI state after a failed EXEC, otherwise
pipelining multiple transactions would be basically impossible:
After a failed EXEC the next transaction would be simply queued as
the tail of the previous transaction.
 2012-11-22 10:36:20 +01:00
antirez ca9321afe2 Test: MULTI state is cleared after EXECABORT error. 2012-11-22 10:36:06 +01:00
antirez a527235f33 Test: make sure EXEC fails after previous transaction errors. 2012-11-22 10:36:01 +01:00
antirez 53600e34db Test: MULTI/EXEC tests moved into multi.tcl. 2012-11-22 10:35:50 +01:00
antirez 60f9dac672 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 10:35:47 +01:00
antirez dff1ec4a36 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 10:28:53 +01:00
antirez d4b5c4803b Fast memory test on Redis crash. 2012-11-22 10:28:48 +01:00
antirez dd905417ec Use more fine grained HAVE macros instead of HAVE_PROCFS. 2012-11-22 10:28:43 +01:00
antirez 5ab4151d7f Children creating AOF or RDB files now report memory used by COW. 
Finally Redis is able to report the amount of memory used by
copy-on-write while saving an RDB or writing an AOF file in background.

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

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

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

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

This function is useful to estimate copy-on-write because of childs
saving stuff on disk.
 2012-11-20 12:32:14 +01:00
antirez c342b075df 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 15:26:02 +01:00
antirez 1d8af6074c 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 15:25:57 +01:00
antirez 1a706c729a zmalloc: kill unused __size parameter in update_zmalloc_stat_alloc() macro. 2012-11-14 13:07:54 +01:00
antirez 681c191006 zmalloc: kill unused __size parameter in update_zmalloc_stat_alloc() macro. 2012-11-14 13:07:48 +01:00
antirez 50c41de752 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 23:05:56 +01:00
antirez d85a09dfc6 MIGRATE: fix default timeout to 1000 milliseconds. 
When a timeout <= 0 is provided we set a default timeout of 1 second.
It was set to 1 millisecond for an error resulting from a recent change.
 2012-11-12 18:58:23 +01:00
antirez 26a48504cb MIGRATE timeout should be in milliseconds. 
While it is documented that the MIGRATE timeout is in milliseconds, it
was in seconds instead. This commit fixes the problem.
 2012-11-12 18:57:55 +01:00
antirez ae3aeca828 MIGRATE: fix fd leak due to missing close on error. 2012-11-12 18:56:29 +01:00
antirez 388cc20418 Copyright date fixed in COPYING file. 2012-11-08 19:14:37 +01:00
antirez 5a8234d2c9 Copyright date fixed in COPYING file. 2012-11-08 19:14:29 +01:00
antirez 2861cd84f9 Redis 2.6.4 2012-11-08 18:47:58 +01:00
antirez cda5c0ba06 Make clear that contributing code to the Redis project means to release it under the terms of the BSD license. 2012-11-08 18:43:41 +01:00
antirez 8e30092f06 Make clear that contributing code to the Redis project means to release it under the terms of the BSD license. 2012-11-08 18:43:37 +01:00
antirez 8ddb23b90c BSD license added to every C source and header file. 2012-11-08 18:34:04 +01:00
antirez d288ee655f BSD license added to every C source and header file. 2012-11-08 18:33:13 +01:00
antirez 89514f80e9 Redis 2.6.3 2012-11-06 22:07:51 +01:00
antirez 46c5d39660 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 20:28:15 +01:00
Runzhen Wang 2b28ef0e61 fix a typo in redis.h line 595 comment 2012-11-02 12:12:08 +01:00
Runzhen Wang 3f006679b1 fix a typo in redis.h line 595 comment 2012-11-02 12:12:01 +01:00
antirez 36f026a3a0 More robust handling of AOF rewrite child. 
After the wait3() syscall we used to do something like that:

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

So the AOF rewrite was handled in the else branch without actually
checking if the pid really matches. This commit makes the check explicit
and logs at WARNING level if the pid returned by wait3() does not match
neither the RDB or AOF rewrite child.
 2012-11-01 22:41:57 +01:00
antirez 64be5e365a More robust handling of AOF rewrite child. 
After the wait3() syscall we used to do something like that:

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

So the AOF rewrite was handled in the else branch without actually
checking if the pid really matches. This commit makes the check explicit
and logs at WARNING level if the pid returned by wait3() does not match
neither the RDB or AOF rewrite child.
 2012-11-01 22:41:54 +01:00
Yecheng Fu fecc8797c2 fix typo in comments (redis.c, networking.c) 2012-11-01 22:26:49 +01:00
Salvatore Sanfilippo cc017c5839 Merge pull request #743 from Cofyc/fixtypo 
fix typo in comments (redis.c, networking.c)
 2012-11-01 14:26:01 -07:00
antirez d775b35715 Unix socket clients properly displayed in MONITOR and CLIENT LIST. 
This also fixes issue #745.
 2012-11-01 22:12:50 +01:00
antirez b6ffa85f11 Unix socket clients properly displayed in MONITOR and CLIENT LIST. 
This also fixes issue #745.
 2012-11-01 22:12:45 +01:00
antirez 655e6838fb 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 15:40:46 +01:00
antirez 4917a6a8b3 32 bit build fixed on Linux. 
It failed because of the way jemalloc was compiled (without passing the
right flags to make, but just to configure). Now the same set of flags
are also passed to the make command, fixing the issue.

This fixes issue #744
 2012-11-01 15:37:38 +01:00
Yecheng Fu ed44a74e72 fix typo in comments (redis.c, networking.c) 2012-11-01 18:14:55 +08:00
YAMAMOTO Takashi 428b706308 fix a typo in a comment 2012-10-31 09:30:00 +01:00
YAMAMOTO Takashi 47ff443b53 fix a typo in a comment 2012-10-31 09:29:56 +01:00
antirez 4d9bd53530 Invert two sides of if expression in SET to avoid a lookup. 
Because of the short circuit behavior of && inverting the two sides of
the if expression avoids an hash table lookup if the non-EX variant of
SET is called.

Thanks to Weibin Yao (@yaoweibin on github) for spotting this.
 2012-10-31 09:27:48 +01:00
antirez 9fab63b334 Invert two sides of if expression in SET to avoid a lookup. 
Because of the short circuit behavior of && inverting the two sides of
the if expression avoids an hash table lookup if the non-EX variant of
SET is called.

Thanks to Weibin Yao (@yaoweibin on github) for spotting this.
 2012-10-31 09:27:46 +01:00
antirez 537235bbbb No longer used macro rdbIsOpcode() removed. 2012-10-30 19:12:01 +01:00
antirez 38272f961a help.h update (adds bitop, bitcount, evalsha...) 2012-10-30 19:11:57 +01:00
antirez 6c7e957d17 Ctrl+w support in linenoise. 2012-10-30 19:11:46 +01:00
antirez 6bd105032c Marginally more robust glibc version test for sync_file_range detection. 2012-10-30 19:11:41 +01:00
charsyam 72f42cf40c patch config.h for sync_file_range 2012-10-30 19:11:34 +01:00
antirez 2a49aa1da2 Fix compilation on Linux kernels or glibc versions lacking sync_file_range(). 
This fixes issue #667.

Many thanks to Didier Spezia for the fix.
 2012-10-30 19:11:29 +01:00
antirez 24ee20e5ca Update memory peak stats while loading RDB / AOF. 2012-10-30 19:11:23 +01:00
antirez 5bf0997ff8 No longer used macro rdbIsOpcode() removed. 2012-10-30 19:10:56 +01:00
antirez ec0f483de2 help.h update (adds bitop, bitcount, evalsha...) 2012-10-30 18:57:28 +01:00
antirez 0dd6241314 Redis 2.6.2 2012-10-26 16:20:24 +02:00
antirez 3b1a41b172 Ctrl+w support in linenoise. 2012-10-26 16:06:36 +02:00
antirez 9b3b110004 Marginally more robust glibc version test for sync_file_range detection. 2012-10-26 15:55:10 +02:00
charsyam b25b0dc5f4 patch config.h for sync_file_range 2012-10-26 09:30:35 +02:00
antirez 2b5fc529a4 Redis 2.6.1 2012-10-25 22:07:27 +02:00
antirez be407c015c Fix compilation on Linux kernels or glibc versions lacking sync_file_range(). 
This fixes issue #667.

Many thanks to Didier Spezia for the fix.
 2012-10-25 22:01:07 +02:00
antirez 1eb9145089 Update memory peak stats while loading RDB / AOF. 2012-10-24 12:21:34 +02:00
antirez f7b42b1749 2.7.0 branch created as a fork of 2.6.0. 2012-10-22 23:32:26 +02:00
antirez 5eec376c2f Redis 2.6.0 2012-10-22 23:27:18 +02:00
antirez 99d7dbe669 A filed called slave_read_only added in INFO output. 
This was an important information missing from the INFO output in the
replication section.

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

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

The code is still compatible with Redis 2.4 generated dumps.

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

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

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

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

This commit introduces two different prefixes.

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

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

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

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

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

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

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

MULTI
EXEC

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

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

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

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

    1.6 seconds with djbhash
    2.0 seconds with murmurhash2

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

This commit introduces two things:

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

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

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

Conflicts:

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

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

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

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

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

The result in redis-cli will be:

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

After many requests for this feature it was finally implemented.

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

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

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

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

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

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

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

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

However later we implemented two things:

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

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

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

Scripting + synchronous blocking operations = Issue #614.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

    EVAL "redis.call()" 0

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

    REPLCONF listening-prot 6380
    SYNC

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

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

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

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

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

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

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

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

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

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

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

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

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

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

This commit fixes issue #547.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

The algorithm used is described here:

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

    GETBIT
    SETBIT
    BITOP
    BITCOUNT

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

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

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

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

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

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

BITCOUNT key [start end]

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

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

For this reason the BITOP command was introduced:

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

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

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

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

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

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

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

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

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

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

    addReplySds(slave,sdsempty());

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

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

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

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

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

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

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

Notable improvements from the point of view of Redis:

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

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

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

    (1000/REDIS_HZ)*2 milliseconds

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

2) Scripts just calling read-only commands will work against read only
slaves, when the server is out of memory, or when persistence is into an
error condition. Before the patch EVAL always failed in this condition.
 2012-03-22 10:09:43 +01:00
antirez 7c1cec2525 Redis cluster stuff removed from 2.6 redis.conf file. 2012-03-20 13:07:26 +01:00
antirez 0380c13bbe redis_init_script template updated. 2012-03-20 13:06:50 +01:00
antirez 518e7202b2 Fixed typo in 2.6 release notes. 2012-03-19 22:10:18 +01:00
antirez f597910121 Merge remote-tracking branch 'origin/2.6' into 2.6 2012-03-19 19:30:30 +01:00
antirez 24b0942275 Suppress warnings compiling redis-cli with certain gcc versions. 2012-03-19 19:29:06 +01:00
antirez 6e6bbac7a5 Read-only flag removed from PUBLISH command. 2012-03-19 19:18:51 +01:00
antirez a7ef5ce1b0 Memory addressing test implemented. 2012-03-19 14:02:57 +01:00
antirez ea693f0282 More memory tests implemented. Default number of iterations lowered to a more acceptable value of 50. 2012-03-18 18:03:49 +01:00
antirez 74760d3ccd Fixed typo. 2012-03-18 17:28:03 +01:00
antirez 32f62ed6d0 Number of iteration of --test-memory is now 300 (several minutes per gigabyte). Memtest86 and Memtester links are also displayed while running the test. 2012-03-18 17:24:48 +01:00
antirez d4e6ce3e97 On crash suggest to give --test-memory a try. 2012-03-18 11:43:09 +01:00
antirez d605fdabfa Memory test function now less boring thanks to screen-wide progress bar. 2012-03-18 11:43:02 +01:00
antirez f4df22d1c5 Hem... actual memtest.c file added. 2012-03-18 11:42:56 +01:00
antirez 78d6a02b0c First implementation of --test-memory. Still a work in progress. 2012-03-18 11:42:47 +01:00
antirez f1eaf57200 Fix for issue #391. 
Use a simple protocol between clientsCron() and helper functions to
understand if the client is still valind and clientsCron() should
continue processing or if the client was freed and we should continue
with the next one.
 2012-03-15 20:51:39 +01:00
antirez 749817b7c3 Version bumped to 2.5.2 2012-03-14 15:37:47 +01:00
antirez 9fa9ccb04e Reclaim space from the client querybuf if needed. 2012-03-14 15:36:27 +01:00
antirez 6934832e53 sds.c: sdsAllocSize() function added. 2012-03-14 15:36:24 +01:00
antirez f9322fb8ed sds.c new function sdsRemoveFreeSpace(). 
The new function is used in order to resize the string allocation so
that only the minimal allocation possible is used, removing all the free
space at the end of the string normally used to improve efficiency of
concatenation operations.
 2012-03-14 15:36:20 +01:00
antirez 5e473cd8fd Call all the helper functions needed by clientsCron() as clientsCronSomething() for clarity. 2012-03-14 15:36:16 +01:00
antirez 3e8fcb6d03 CLIENT LIST test modified to reflect the new output. 2012-03-13 18:08:59 +01:00
antirez cfa4b57cb0 Process async client checks like client timeouts and BLPOP timeouts incrementally using a circular list. 2012-03-13 18:08:53 +01:00
antirez 57a5e54ddc Added a qbuf-free field to CLIENT LIST output. 2012-03-13 18:08:50 +01:00
antirez 41e8e5cb8e Client creation time in redisClient structure. New age field in CLIENT LIST output. 2012-03-13 18:08:47 +01:00
antirez 56de4964ee c->bufpos initialization moved for aesthetics. 2012-03-13 18:08:43 +01:00
antirez a74ab6478c RDB hashes loading, fixed another bug in the loading of HT-encoded hashes: when the hash entry is too big for ziplist, add the field, then convert. The code used to break before the new entry was inserted, resulting into missing fields in the loaded Hash object. 2012-03-13 11:01:08 +01:00
antirez ee61a4b99e RDB hashes loading fixed removing the assertion that failed every time an HT-encoded hash was loaded. 2012-03-13 09:49:21 +01:00
quiver dfc2545470 fix typo of redis.conf 2012-03-10 13:23:31 +01:00
antirez b014c1f211 RDB4 support in redis-check-dump. 2012-03-10 12:38:59 +01:00
antirez 37180ed9cd RDB version is no 4, because small hashes are now encoded as ziplists, so older versions of Redis will not understand this format. 2012-03-10 12:35:15 +01:00
antirez 7551f2a0b1 Version is now 2.5.1, first unstable release of Redis 2.6 2012-03-10 12:29:47 +01:00
antirez 00b3336334 Build dependencies updated. 2012-03-10 12:28:14 +01:00
antirez 571e257db1 Redis 2.6 branch obtained from unstable removing all the cluster related code. 2012-03-10 12:26:37 +01:00
824 changed files with 74984 additions and 161175 deletions
Expand all files Collapse all files
.github/FUNDING.yml
-3
View File
@@ -1,3 +0,0 @@
# These are supported funding model platforms
github: [tporadowski]
.github/ISSUE_TEMPLATE/bug_report.md
-27
View File
@@ -1,27 +0,0 @@
---
name: Bug report
about: Create a report about a problem with Redis for Windows
title: ''
labels: ''
assignees: ''
---
**Describe the bug/issue**
A clear and concise description of what the bug/issue is.
**To Reproduce**
Please provide more information about:
- version of Redis for Windows you are using
- configuration of your instance if it's not out-of-the-box one (i.e. persistence type)
- typical workload handled by your instance
- any other details that might be helpful in reproducing the issue.
**Crash report**
If Redis generated a crash report - please paste it here.
**Event Log**
Please check your system's Event Log to see if there are additional details logged related to this issue.
**Additional context**
Please provide Windows version you are using, mode in which you are starting Redis (standalone, cluster, Sentinel), etc.
.gitignore
+27 -19
View File
@@ -1,22 +1,15 @@
.*.swp
*.o
*.log
*.bak
dump.rdb
redis-benchmark
redis-check-aof
redis-check-rdb
redis-check-dump
redis-cli
redis-sentinel
redis-server
doc-tools
release
bin/
obj/
Debug/
Release/
x64/
misc/*
src/release.h
appendonly.aof
@@ -26,7 +19,6 @@ src/transfer.sh
src/configs
redis.ds
src/redis.conf
src/nodes.conf
deps/lua/src/lua
deps/lua/src/luac
deps/lua/src/liblua.a
@@ -37,22 +29,38 @@ deps/lua/src/liblua.a
*.exe
*.sdf
*.suo
msvs/ReleasePackagingTool/bin/
msvs/ReleasePackagingTool/obj/
msvs/ReleasePackagingTool/packages/
msvs/Debug/
msvs/Release/
msvs/RedisBenchmark/Debug/
msvs/RedisBenchmark/Release/
msvs/RedisBenchmark/x64/
msvs/RedisCheckAof/Debug/
msvs/RedisCheckAof/Release/
msvs/RedisCheckAof/x64/
msvs/RedisCheckDump/Debug/
msvs/RedisCheckDump/Release/
msvs/RedisCheckDump/x64/
msvs/RedisCli/Debug/
msvs/RedisCli/Release/
msvs/RedisCli/x64/
msvs/hiredis/Debug/
msvs/hiredis/Release/
msvs/hiredis/x64/
msvs/lua/lua/Debug/
msvs/lua/lua/Release/
msvs/lua/lua/x64/
msvs/x64/
msvs/ipch
msvs/RedisServer.opensdf
!msvs/RedisWaInst/bin/*
!msvs/RedisWaInst/bin/Inst4WA/*
!msvs/RedisWaInst/bin/RedisPkgBin/*
msvs/RedisWAInst/src/RedisDeployCmdlets/bin/
msvs/RedisWAInst/src/RedisDeployCmdlets/obj/
msvs/RedisWAInst/src/RedisInstBin/
msvs/RedisWAInst/src/RedisInstWA/app.config
msvs/jemalloc
msvs/BuildRelease
msvs/Documentation
tests/tmp/
msvs/RedisWAInst/src/RedisInstWA/bin/
msvs/RedisWAInst/src/RedisInstWA/obj/
msvs/RedisWAInst/src/RedisServer/bin/
msvs/RedisWAInst/src/RedisServer/obj/
/msvs/.vs/RedisServer/v15
/msvs/msi/.vs/RedisMsi/v15/Server/sqlite3
.vs/
*.rdb
00-RELEASENOTES
+571 -3503
View File
File diff suppressed because it is too large Load Diff
CONTRIBUTING
+8 -24
View File
@@ -8,43 +8,27 @@ each source file that you contribute.
# IMPORTANT: HOW TO USE REDIS GITHUB ISSUES
* Github issues SHOULD ONLY BE USED to report bugs, and for DETAILED feature
requests. Everything else belongs to the Redis Google Group:
https://groups.google.com/forum/m/#!forum/Redis-db
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 mailing list.
all the support in the Redis Google Group.
There is also an active community of Redis users at Stack Overflow:
http://stackoverflow.com/questions/tagged/redis
Redis Google Group address:
https://groups.google.com/forum/?fromgroups#!forum/redis-db
# How to provide a patch for a new feature
1. If it is a major feature or a semantical change, please don't start coding
straight away: if your feature is not a conceptual fit you'll lose a lot of
time writing the code without any reason. Start by posting in the mailing list
and creating an issue at Github with the description of, exactly, what you want
to accomplish and why. Use cases are important for features to be accepted.
Here you'll see if there is consensus about your idea.
1. Drop a message to the Redis Google Group with a proposal of semantics/API.
2. If in step 1 you get an acknowledgment from the project leaders, use the
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)
c. Push to your branch (git push origin my_branch)
d. Initiate a pull request on github ( https://help.github.com/articles/creating-a-pull-request/ )
d. Initiate a pull request on github ( http://help.github.com/send-pull-requests/ )
e. Done :)
3. Keep in mind that we are very overloaded, so issues and PRs sometimes wait
for a *very* long time. However this is not lack of interest, as the project
gets more and more users, we find ourselves in a constant need to prioritize
certain issues/PRs over others. If you think your issue/PR is very important
try to popularize it, have other users commenting and sharing their point of
view and so forth. This helps.
4. For minor fixes just open a pull request on Github.
Thanks!
COPYING
+1 -1
View File
@@ -1,4 +1,4 @@
Copyright (c) 2006-2015, 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:
MANIFESTO
+7 -46
View File
@@ -34,21 +34,7 @@ Redis Manifesto
so that the complexity is obvious and more complex operations can be
performed as the sum of the basic operations.
4 - We believe in code efficiency. Computers get faster and faster, yet we
believe that abusing computing capabilities is not wise: the amount of
operations you can do for a given amount of energy remains anyway a
significant parameter: it allows to do more with less computers and, at
the same time, having a smaller environmental impact. Similarly Redis is
able to "scale down" to smaller devices. It is perfectly usable in a
Raspberry Pi and other small ARM based computers. Faster code having
just the layers of abstractions that are really needed will also result,
often, in more predictable performances. We think likewise about memory
usage, one of the fundamental goals of the Redis project is to
incrementally build more and more memory efficient data structures, so that
problems that were not approachable in RAM in the past will be perfectly
fine to handle in the future.
5 - Code is like a poem; it's not just something we write to reach some
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
@@ -59,48 +45,23 @@ Redis Manifesto
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.
6 - We're against complexity. We believe designing systems is a fight against
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. Complexity is also a form of lock-in: code that is
very hard to understand cannot be modified by users in an independent way
regardless of the license. One of the main Redis goals is to remain
understandable, enough for a single programmer to have a clear idea of how
it works in detail just reading the source code for a couple of weeks.
creating it at all.
7 - Threading is not a silver bullet. Instead of making Redis threaded we
believe on the idea of an efficient (mostly) single threaded Redis core.
Multiple of such cores, that may run in the same computer or may run
in multiple computers, are abstracted away as a single big system by
higher order protocols and features: Redis Cluster and the upcoming
Redis Proxy are our main goals. A shared nothing approach is not just
much simpler (see the previous point in this document), is also optimal
in NUMA systems. In the specific case of Redis it allows for each instance
to have a more limited amount of data, making the Redis persist-by-fork
approach more sounding. In the future we may explore parallelism only for
I/O, which is the low hanging fruit: minimal complexity could provide an
improved single process experience.
8 - Two levels of API. The Redis API has two levels: 1) a subset of the API fits
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
trade-offs to the user.
instance to another to perform multi-key operations and expose the tradeoffs
to the user.
9 - We optimize for joy. We believe writing code is a lot of hard work, and the
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.
10 - All the above points are put together in what we call opportunistic
programming: trying to get the most for the user with minimal increases
in complexity (hanging fruits). Solve 95% of the problem with 5% of the
code when it is acceptable. Avoid a fixed schedule but follow the flow of
user requests, inspiration, Redis internal readiness for certain features
(sometimes many past changes reach a critical point making a previously
complex feature very easy to obtain).
README
+154
View File
@@ -0,0 +1,154 @@
Where to find complete Redis documentation?
-------------------------------------------
This README is just a fast "quick start" document. You can find more detailed
documentation at http://redis.io
Building Redis
--------------
Redis can be compiled and used on Linux, OSX, OpenBSD, NetBSD, FreeBSD.
We support big endian and little endian architectures.
It may compile on Solaris derived systems (for instance SmartOS) but our
support for this platform is "best effort" and Redis is not guaranteed to
work as well as in Linux, OSX, and *BSD there.
It is as simple as:
% make
You can run a 32 bit Redis binary using:
% make 32bit
After building Redis is a good idea to test it, using:
% make test
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
---------
Selecting a non-default memory allocator when building Redis is done by setting
the `MALLOC` environment variable. Redis is compiled and linked against libc
malloc by default, with the exception of jemalloc being the default on Linux
systems. This default was picked because jemalloc has proven to have fewer
fragmentation problems than libc malloc.
To force compiling against libc malloc, use:
% make MALLOC=libc
To compile against jemalloc on Mac OS X systems, use:
% make MALLOC=jemalloc
Verbose build
-------------
Redis will build with a user friendly colorized output by default.
If you want to see a more verbose output use the following:
% make V=1
Running Redis
-------------
To run Redis with the default configuration just type:
% cd src
% ./redis-server
If you want to provide your redis.conf, you have to run it using an additional
parameter (the path of the configuration file):
% cd src
% ./redis-server /path/to/redis.conf
It is possible to alter the Redis configuration passing parameters directly
as options using the command line. Examples:
% ./redis-server --port 9999 --slaveof 127.0.0.1 6379
% ./redis-server /etc/redis/6379.conf --loglevel debug
All the options in redis.conf are also supported as options using the command
line, with exactly the same name.
Playing with Redis
------------------
You can use redis-cli to play with Redis. Start a redis-server instance,
then in another terminal try the following:
% cd src
% ./redis-cli
redis> ping
PONG
redis> set foo bar
OK
redis> get foo
"bar"
redis> incr mycounter
(integer) 1
redis> incr mycounter
(integer) 2
redis>
You can find the list of all the available commands here:
http://redis.io/commands
Installing Redis
-----------------
In order to install Redis binaries into /usr/local/bin just use:
% make install
You can use "make PREFIX=/some/other/directory install" if you wish to use a
different destination.
Make install will just install binaries in your system, but will not configure
init scripts and configuration files in the appropriate place. This is not
needed if you want just to play a bit with Redis, but if you are installing
it the proper way for a production system, we have a script doing this
for Ubuntu and Debian systems:
% cd utils
% ./install_server.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
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!
README.md
+61 -19
View File
@@ -1,30 +1,72 @@
# Redis 5.0.9 for Windows
[![Windows Status](http://img.shields.io/appveyor/ci/MSOpenTech-lab/redis.svg?style=flat-square)](https://ci.appveyor.com/project/MSOpenTech-lab/redis)
You can find the release of **Redis 5.0.9 for Windows** on [releases page](https://github.com/tporadowski/redis/releases). Please test it and [report any issues](https://github.com/tporadowski/redis/wiki/Submitting-an-Issue), thanks in advance!
[![NuGet version](http://img.shields.io/nuget/v/redis-64.svg?style=flat-square)](http://www.nuget.org/packages/redis-64/)
[![NuGet downloads](http://img.shields.io/nuget/dt/redis-64.svg?style=flat-square)](http://www.nuget.org/packages/redis-64/)
## Redis 4.0.14 for Windows
[![Chocolatey version](http://img.shields.io/chocolatey/v/redis-64.svg?style=flat-square)](http://www.chocolatey.org/packages/redis-64/)
[![Chocolatey downloads](http://img.shields.io/chocolatey/dt/redis-64.svg?style=flat-square)](http://www.chocolatey.org/packages/redis-64/)
You can find the release of **Redis 4.0.14 for Windows** on [releases page](https://github.com/tporadowski/redis/releases). Please test it and [report any issues](https://github.com/tporadowski/redis/wiki/Submitting-an-Issue), thanks in advance!
Redis on Windows
===
**DISCLAIMER**
## Redis 2.8 Branch
At the moment [win-4.0.14 branch](https://github.com/tporadowski/redis/tree/win-4.0.14) provides a **stable port of [Redis 4.0.14](https://github.com/antirez/redis/releases/tag/4.0.14) for Windows x64** and [win-5.0 branch](https://github.com/tporadowski/redis/tree/win-5.0) provides a **fairly stable port of [Redis 5.0.9](https://github.com/antirez/redis/releases/tag/5.0.9) for Windows x64**, both merged with archived port of [win-3.2.100 version](https://github.com/MicrosoftArchive/redis/releases/tag/win-3.2.100) from MS Open Tech team. Since the latter is no longer maintained - the sources were merged by hand, projects updated to Visual Studio 2019 (v16.2.5) and any findings (mostly via unit tests) were fixed.
- This is a port for Windows based on Redis 2.8.
- There is support for the 64-bit version. We have dropped official support for the 32-bit version, though you can build it from source if desired.
- You can download the latest binaries (unsigned) from [http://github.com/MSOpenTech/redis/releases](http://github.com/MSOpenTech/redis/releases "Release page"). For releases prior to 2.8.17.1, the binaries can found in a zip file inside the source archive, under the bin/release folder.
- Signed binaries can be downloaded using Nuget and Chocolatey.
- There is a replacement for the UNIX fork() API that simulates the copy-on-write behavior using a memory mapped file.
- Because Redis makes some assumptions about the values of File Descriptors, we have built a virtual file descriptor mapping layer.
- We are moving towards moving all Windows-specific changes into the Win32_Interop library.
- Redis can be installed as a Windows Service.
You can find the original description of what this fork provides, how it evolved, what are its requirements, etc. on Wiki: https://github.com/tporadowski/redis/wiki/Old-MSOpenTech-redis-README.md
## What's new since 2.8.12
**Building from source code**
- See the Redis release notes: http://download.redis.io/redis-stable/00-RELEASENOTES
In order to build this project from source code you need to have:
1. Visual Studio 2019 (i.e. Community Edition, version 16.2.5 or newer) with "C/C++ features" enabled,
1. Windows SDK 10,
1. [Git Bash](https://gitforwindows.org/) for Windows or [Cygwin](http://cygwin.com/) with Git - after cloning this repository you need to run `src/mkreleasehdr.sh` script that creates `src/release.h` with some information taken from Git; optionally you can create that file by hand.
## Important: More documentation is available
**Dependencies**
Please read the documentation in msvs\setups\documentation. This is the documentation that is bundled with the binaries, and contains vital information about configuring and deploying Redis on Windows.
This project depends on [`jemalloc`](https://github.com/jemalloc/jemalloc) memory allocator, which is slightly customized
in regard to calls to `VirtualAlloc` and `VirtualFree` WinAPI functions. They are being replaced with calls to `AllocHeapBlock/PurgePages`
and `FreeHeapBlock` from `src/Win32_Interop/Win32_QFork.cpp` in order to keep track which memory regions are to be made
available to child processes (for saving RDB/AOF).
## How to build Redis using Visual Studio
You can use the free Visual Studio Community edition available at http://www.visualstudio.com/products/visual-studio-community-vs.
- Open the solution file msvs\redisserver.sln in Visual Studio, select a build configuration (Debug or Release) and target (Win32 or x64) then build.
This should create the following executables in the msvs\$(Target)\$(Configuration) folder:
- redis-server.exe
- redis-benchmark.exe
- redis-cli.exe
- redis-check-dump.exe
- redis-check-aof.exe
## Unit Testing
To run the Redis test suite requires some manual work:
- The tests assume that the binaries are in the src folder. Use mklink to create a symbolic link to the files in the msvs\x64\Debug|Release folders. You will
need symbolic links for src\redis-server, src\redis-benchmark, src\redis-check-aof, src\redis-check-dump, src\redis-cli, and src\redis-sentinel.
- The tests make use of TCL. This must be installed separately.
- To run the tests you need to have a Unix shell on your machine, or MinGW tools in your path. To execute the tests, run the following command:
"tclsh8.5.exe tests/test_helper.tcl --clients N", where N is the number of parallel clients . If a Unix shell is not installed you may see the
following error message: "couldn't execute "cat": no such file or directory".
- By default the test suite launches 16 parallel tests. I will get time out errors on an iCore 7-2620m@2.7Ghz with some of the tests when the number of clients
is greater than 6.
## Known issues
Problem:
On versions of Windows prior to Windows 8/Server 2012, when an AOF or RDB operation is complete and the child process is being rejoined with the parent, it is
possible for Redis to unexpectedly terminate.
Cause:
The PAGE_REVERT_TO_FILE_MAP flag is not usable in VirtualProtect() in earlier versions of the OS. This flag allows for removing copy on write(COW) pages from
a memory mapped view without unmapping the view. In prior versions of the OS, the only way to purge COW pages is to unmap and then remap the memory mapped view.
Between the unmapping and remapping operations a third party process(e.g., an anti-virus program) could allocate virtual memory occupied by the memory mapped view.
As this view is used for the Redis heap, failure to remap is a fatal error. (See RejoinCOWPages() in src\Win32_Interop\Win32_QFork.cpp)
Solution:
If you encounter this problem, host Redis on Windows 8/Server 2012 or newer.
Changes to `jemalloc` that provide those customizations are being maintained in [tporadowski/jemalloc repository](https://github.com/tporadowski/jemalloc)
and are copied over to `deps/jemalloc`.
RELEASENOTES.txt
-90
View File
@@ -1,90 +0,0 @@
Redis for Windows - https://github.com/tporadowski/redis
========================================================
This file provides information about Windows-specific changes to Redis.
For release notes related to original Redis project - please see 00-RELEASENOTES.
--------------------------------------------------------
2020-11-08: Redis 5.0.10 for Windows
https://github.com/tporadowski/redis/releases/tag/v5.0.10
Bugfix/maintenance release of Redis for Windows, updated to be in sync with
redis/5.0.10 (https://github.com/redis/redis/releases/tag/5.0.10)
NOTE: active memory defragmentation feature ("activedefrag" option) is turned OFF.
--------------------------------------------------------
2020-05-02: Redis 5.0.9 for Windows
https://github.com/tporadowski/redis/releases/tag/v5.0.9
First release of Redis 5.x for Windows, updated to be in sync with antirez/5.0.9.
--------------------------------------------------------
2020-01-26: Redis 4.0.14.2 for Windows
https://github.com/tporadowski/redis/releases/tag/v4.0.14.2
This is a hotfix release of 4.0.14 branch that fixes #50 related to running in
Sentinel mode.
--------------------------------------------------------
2020-01-15: Redis 4.0.14.1 for Windows
https://github.com/tporadowski/redis/releases/tag/v4.0.14.1
This is a hotfix release of 4.0.14 branch that fixes 2 Windows-specific issues:
* #46 - added support for older Windows versions (prior Windows 8 and Windows Server 2012)
* #47 - fixed problem with parsing command-line arguments.
--------------------------------------------------------
2019-08-29: Redis 4.0.14 for Windows
https://github.com/tporadowski/redis/releases/tag/v4.0.14
Redis 4.0.14 for Windows is a merge of Windows-specific changes from latest (unsupported) 3.2.100 release from MSOpenTech and original Redis 4.0.14.
--------------------------------------------------------
2018-10-01: Redis for Windows 4.0.2.3 (alpha)
This 4.0.2.3 release is still an alpha version, but contains enhancements and fixes for:
* #14: decrease logging verbosity of some cluster-related messages
* #23: ZRANK/ZREVRANK bugfix (win-port only)
* failed unit tests (bdcf80e).
--------------------------------------------------------
2018-03-26: Redis for Windows 4.0.2.2 (alpha)
https://github.com/tporadowski/redis/releases/tag/v4.0.2.2-alpha
This 4.0.2.2 release is still an alpha version, but contains a fix to issue #12
(crash when rewriting AOF file - this issue was specific to Windows port only).
--------------------------------------------------------
2018-03-17: Redis for Windows 4.0.2.1 (alpha)
https://github.com/tporadowski/redis/releases/tag/v4.0.2.1-alpha
This 4.0.2.1 release is still an alpha version, but contains a fix to issue #11,
which was related to sending back larger amounts of data to Redis clients
(this issue was specific to Windows port only).
--------------------------------------------------------
2017-11-22: Redis 4.0.2 for Windows (alpha)
https://github.com/tporadowski/redis/releases/tag/v4.0.2-alpha
Alpha version of Redis 4.0.2 for Windows.
Redis 4.0.2 for Windows is a merge of Windows-specific changes from latest (unsupported) 3.2.100 release
from MSOpenTech and Redis 4.0.2 and this alpha release consists of:
* all Redis 4.0.2 features except modules,
* all executables of Redis (redis-server, redis-cli, redis-benchmark, redis-check-aof, redis-check-rdb).
Main difference to official Redis 4.0.2 (except no support for modules at the moment) is old version
of jemalloc-win dependency, which is planned to be updated to the same version in beta release.
Redis on Windows Release Notes.md
+30 -27
View File
@@ -1,32 +1,35 @@
MSOpenTech Redis on Windows 3.2 Release Notes
=============================================
--[ Redis on Windows 3.2.100 ] Release date: Jul 01 2016
MSOpenTech Redis 2.8.17 Release Notes
=====================================
This is the first stable release of Redis on Windows 3.2.
This release is based on antirez/redis/3.2.1 plus some Windows specific fixes.
It has passed all the standard tests but it hasn't been tested in a production
environment. Therefore before considering using this release in production
make sure to test it thoroughly in your own test environment.
Welcome to the binary release of Redis from Microsoft Open Technologies, Inc.
Changelog:
- [Portability] strtol and strtoul fixes.
- [Fix] Possible AV during background save.
- [Fix] Use overlapped sockets for cluster failover communication.
- [Cleanup cleanup] Minor changes.
- [Portability] Windows portability fixes.
- Merged tag 3.2.1 from antirez/3.2.
- [Setup] Removed subdir for log, the log is now saved in the main redis dir.
- [Cleanup] Removed unused project.
--[ Redis on Windows 3.2.000-preview ] Release date: Jun 14 2016
What is Redis?
--------------
This is a technical preview of Redis on Windows 3.2.
There are still known issues/bugs, in particular there is a bug that prevents
the cluster fail-over functionality to work properly in certain scenarios.
This release SHOULD NOT be used in production.
Redis is an open source, high performance, key-value store. Values may contain strings, hashes, lists, sets and sorted sets. Redis has been developed primarily for UNIX-like operating systems.
Changelog:
Porting Goals
-------------
Our goal is to provide a version of Redis that runs on Windows with a performance essentially equal to the performance of Redis on an equivalent UNIX machine.
What is new with the 2.8.17 release
-----------------------------------
Our last official release was 2.8.12. We have merged in the changes up to 2.8.17. Please see the [release notes for the UNIX 2.8 branch](http://download.redis.io/redis-stable/00-RELEASENOTES) to understand how this impacts Redis functionality.
### Network layer changes
There have been significant changes to the networking layer for this version. Likely there will be a few weeks before there is another official (Chocolatey and Nuget) release. Most of these changes target IPv6.
### persistence-available flag
If Redis is to be used as an in-memory-only cache without any kind of persistence, then the fork() mechanism used by the background AOF/RDB persistence is unnecessary. As an optimization, all persistence can be turned off in the Windows version of Redis in this scenario. This will disable the creation of the memory mapped heap file, redirect heap allocations to the system heap allocator, and disable commands that would otherwise cause fork() operations: BGSAVE and BGREWRITEAOF. This flag may not be combined with any of the other flags that configure AOF and RDB operations.
persistence-available [(yes)|no]
How to develop for Redis
------------------------
You will need a client library for accessing Redis. There are a wide variety of client libraries available as listed at <http://redis.io/clients>.
- Windows port of antirez/3.2.
Redis on Windows.md
+15 -15
View File
@@ -1,8 +1,6 @@
MSOpenTech’s Redis on Windows
=============================
Redis is an open source, high performance, key-value store. Values may contain strings, hashes, lists, sets and sorted sets. Redis has been developed primarily for UNIX-like operating systems.
Our goal is to provide a version of Redis that runs on Windows with a performance essentially equal to the performance of Redis on an equivalent UNIX machine.
We strive to have a stable, functionally equivalent and comparably performing version of Redis on Windows. We have achieved performance nearly identical to the POSIX version running head-to-head on identical hardware across the network. Aside from feature differences that help Redis take advantage of the Windows infrastructure, our version of Redis should work in most situations with the identical setup and configuration that one would use on a POSIX operating system.
How is Redis on Windows implemented?
@@ -23,7 +21,7 @@ Redis is a C code base that compiles under Visual Studio. Most of the code compi
Networking Differences
----------------------
The Windows networking stack is split between user mode code and kernel mode code. Transitions between user and kernel mode are expensive operations. The POSIX networking APIs on Windows utilize a programming model that incurs significant performance loss due to the kernel/user mode transitions. Efficient Windows networking code instead uses the IO Completion Port model to reduce the impact of this behavior. The APIs used and the programming model for IO Completion is different enough that we were forced to implement a new networking layer in Redis.
The Windows networking stack is split between user mode code and kernel mode code. Transitions between user and kernel mode are expensive operations. The POSIX networking APIs on Windows utilize a programming model that incurs significant performance loss due to the kernel/user mode transitions. Efficient Windows networking code instead uses the IO Completion Port model to reduce the impact of this behavior. The APIs used and the programming model for IO Completion is different enough that we were forced to implement a new networking layer in in Redis.
File Descriptors
----------------
@@ -35,25 +33,27 @@ fork()
The POSIX version of Redis uses the fork() API. There is no equivalent in Windows, and it is an exceedingly difficult API to completely simulate. For most of the uses of fork() we have used Windows specific programming idioms to bypass the need to use a fork()-like API. The one case where we could not do so was with the point-in-time heap snapshot behavior that the Redis persistence model is based on. We tried several different approaches to work around the need for a fork()-like API, but always ran into significant performance penalties and stability issues.
Our current approach is to simulate the point-in-time snapshot behavior aspect of fork() without doing a complete simulation of fork(). We do this with the system-paging file that contains the Redis heap. When a fork() operation is required we do the following:
Our current approach is to simulate the point-in-time snapshot behavior aspect of fork() without doing a complete simulation of fork(). We do this with a memory mapped file that contains the Redis heap. When a fork() operation is required we do the following:
- Mark every page in the system-paging file with the Copy on Write page protection
- Mark every page in the memory mapped file with the Copy on Write page protection
- Start a child process and pass it the handle to the system-paging file
- Start a child process and pass it the handle to the memory mapped file
- Signal the child to start the AOF or RDB persistence process on the memory shared via the system-paging file
- Signal the child to start the AOF or RDB persistence process on the memory shared via the memory mapped file
- Wait (asynchronously) for the child process to finish
- Map the changes in the Redis heap that occurred during the fork() operation back into the system-paging file.
- Map the changes in the Redis heap that occurred during the fork() operation back into the memory mapped file.
The upside with this implementation is that our performance and stability is now on par with the POSIX version of Redis. The down side is that we have a runtime disk space requirement for Redis equal to the size of the Redis memory heap. The disk space requirement defaults to:
The upside with this implementation is that our performance and stability is now on par with the POSIX version of Redis. The down side is that we have a runtime disk space requirement for Redis equal to the size of the Redis memory mapped heap. The disk space requirement defaults to:
- The size specified by the –maxheap flag if present, otherwise
- 50% more than the --maxmemory setting if present, otherwise
- The size of physical RAM
We also have a runtime page file commit requirement that varies depending on the amount data in the Redis heap during the quasi-fork operation. The maximum for this is about 3 times the size of the max-memory This is usually not a problem because the default configuration of Windows allows the page file to grow to 3.5 times the size of physical memory. There are scenarios where 3<sup>rd</sup> party programs also compete for system swap space at runtime.
We also have a runtime page file commit requirement that varies depending on the amount data in the Redis heap during the quasi-fork operation. The maximum for this is about 3 times the size of the memory mapped file. This is usually not a problem because the default configuration of Windows allows the page file to grow to 3.5 times the size of physical memory. There are scenarios where 3<sup>rd</sup> party programs also compete for system swap space at runtime.
Logging
-------
@@ -75,6 +75,10 @@ The GitHub repository should be considered a work in progress until we release t
For instance, the Windows Service feature has taken many iterations with community input to get right. The initial Windows service code was checked in on April 3. Since that time we have added the following to the service based on community input:
- Preshutdown notification in order to clean up the memory mapped file consistently.
- Code to identify and clean up orphaned memory mapped files left behind when a machine running Redis as a service crashes or loses power.
- Self elevation of the Redis executable so that service commands would work from a non-elevated command prompt.
- Service naming so that multiple instances of the Redis service could be installed on one machine.
@@ -88,7 +92,7 @@ The final service code has been released as of June 25 in the 2.8.9 Redis-64 pac
Heap Sizing
-----------
Native heaps are prone to fragmentation. If we are not able to allocate more heap space due to fragmentation Redis will flag the problem and exit. Unlike the POSIX version of Redis, our heap size is constrained by both by disk space and by swap file space. It is important to consider how much data you are expecting to put into Redis, and how much fragmentation you are likely to see in the Redis heap. With very high levels of fragmentation the 50% overhead that –maxmemory imposes may not be enough to prevent running out of heap space. In this case, the use of –maxheap will supersede the –maxmemory default heap setting.
Native heaps are prone to fragmentation. If we are not able to allocate more heap space due to fragmentation Redis will flag the problem and exit. Unlike the POSIX version of Redis, our heap size is constrained by both by disk space and by swap file space. It is important to consider the how much data you are expecting to put into Redis, and how much fragmentation you are likely to see in the Redis heap. With very high levels of fragmentation the 50% overhead that –maxmemory imposes may not be enough to prevent running out of heap space. In this case, the use of –maxheap will supersede the –maxmemory default heap setting.
Installation and Maintenance
----------------------------
@@ -102,7 +106,3 @@ Service Account
When using Redis as a Windows service, the default installation configures Redis to run under the system’s NETWORK SERVICE account. There are some environments where another account must be used (perhaps a domain service account). Configuration of this account needs to be done manually at this point with the service control manager. If this is done, it is also important to give read/write/create permission to the folder that the Redis executable is in to this user identity.
How to develop for Redis
========================
You will need a client library for accessing Redis. There are a wide variety of client libraries available as listed at <http://redis.io/clients>.
Windows Service Documentation.md
+8 -29
View File
@@ -1,28 +1,6 @@
Running Redis as a Service
Running Redis as a Service
==========================
If you installed Redis using the MSI package, then Redis was already installed as a Windows service. Nothing further to do. If you would like to change its settings, you can update the *redis.windows-service.conf* file and then restart the Redis service (Run -\> services.msc -\> Redis -\> Restart).
During installation of the MSI you can either use the installer’s user interface to update the port that Redis listens at and the firewall exception or run it silently without a UI. The following examples show how to install from the command line:
**default install (port 6379 and firewall exception ON):**
*msiexec /i Redis-x64.msi *
**set port and turn OFF firewall exception:**
*msiexec /i Redis-x64.msi PORT=1234 FIREWALL\_ON=""*
**set port and turn ON firewall exception:**
*msiexec /i Redis-x64.msi PORT=1234 FIREWALL\_ON=1*
**install with no user interface:**
*msiexec /quiet /i Redis-x64.msi*
If you did *not* install Redis using the MSI package, then you still run Redis as a Windows service by following these instructions:
In order to better integrate with the Windows Services model, new command line arguments have been introduced to Redis. These service arguments require an elevated user context in order to connect to the service control manager. If these commands are invoked from a non-elevated context, Redis will attempt to create an elevated context in which to execute these commands. This will cause a User Account Control dialog to be displayed by Windows and may require Administrative user credentials in order to proceed.
Installing the Service
@@ -82,14 +60,15 @@ This optional argument may be used with any of the preceding commands to set the
The following would install and start three separate instances of Redis as a service:
redis-server --service-install --service-name redisService1 --port 10001
redis-server --service-install –service-name redisService1 –port 10001
redis-server --service-start --service-name redisService1
redis-server --service-start –service-name redisService1
redis-server --service-install --service-name redisService2 --port 10002
redis-server --service-install –service-name redisService2 –port 10002
redis-server --service-start --service-name redisService2
redis-server --service-start –service-name redisService2
redis-server --service-install --service-name redisService3 --port 10003
redis-server --service-install –service-name redisService3 –port 10003
redis-server --service-start –service-name redisService3
redis-server --service-start --service-name redisService3
appveyor.yml
+2 -2
View File
@@ -1,9 +1,9 @@
version: 3.2.100.{build}
version: 2.8.13.{build}
branches:
# whitelist
only:
- '3.2'
- '2.8'
- appveyor
install:
bin/.gitignore
+2
View File
@@ -0,0 +1,2 @@
release/
debug/
deps/Makefile
Vendored
+2 -2
View File
@@ -58,7 +58,7 @@ ifeq ($(uname_S),SunOS)
LUA_CFLAGS= -D__C99FEATURES__=1
endif
LUA_CFLAGS+= -O2 -Wall -DLUA_ANSI -DENABLE_CJSON_GLOBAL -DREDIS_STATIC='' $(CFLAGS)
LUA_CFLAGS+= -O2 -Wall -DLUA_ANSI -DENABLE_CJSON_GLOBAL $(CFLAGS)
LUA_LDFLAGS+= $(LDFLAGS)
# lua's Makefile defines AR="ar rcu", which is unusual, and makes it more
# challenging to cross-compile lua (and redis). These defines make it easier
@@ -77,7 +77,7 @@ JEMALLOC_LDFLAGS= $(LDFLAGS)
jemalloc: .make-prerequisites
@printf '%b %b\n' $(MAKECOLOR)MAKE$(ENDCOLOR) $(BINCOLOR)$@$(ENDCOLOR)
cd jemalloc && ./configure --with-version=5.1.0-0-g0 --with-lg-quantum=3 --with-jemalloc-prefix=je_ --enable-cc-silence CFLAGS="$(JEMALLOC_CFLAGS)" LDFLAGS="$(JEMALLOC_LDFLAGS)"
cd jemalloc && ./configure --with-jemalloc-prefix=je_ --enable-cc-silence CFLAGS="$(JEMALLOC_CFLAGS)" LDFLAGS="$(JEMALLOC_LDFLAGS)"
cd jemalloc && $(MAKE) CFLAGS="$(JEMALLOC_CFLAGS)" LDFLAGS="$(JEMALLOC_LDFLAGS)" lib/libjemalloc.a
.PHONY: jemalloc
deps/README.md
Vendored
-101
View File
@@ -1,101 +0,0 @@
This directory contains all Redis dependencies, except for the libc that
should be provided by the operating system.
* **Jemalloc** is our memory allocator, used as replacement for libc malloc on Linux by default. It has good performances and excellent fragmentation behavior. This component is upgraded from time to time.
* **geohash-int** is inside the dependencies directory but is actually part of the Redis project, since it is our private fork (heavily modified) of a library initially developed for Ardb, which is in turn a fork of Redis.
* **hiredis** is the official C client library for Redis. It is used by redis-cli, redis-benchmark and Redis Sentinel. It is part of the Redis official ecosystem but is developed externally from the Redis repository, so we just upgrade it as needed.
* **linenoise** is a readline replacement. It is developed by the same authors of Redis but is managed as a separated project and updated as needed.
* **lua** is Lua 5.1 with minor changes for security and additional libraries.
How to upgrade the above dependencies
===
Jemalloc
---
Jemalloc is modified with changes that allow us to implement the Redis
active defragmentation logic. However this feature of Redis is not mandatory
and Redis is able to understand if the Jemalloc version it is compiled
against supports such Redis-specific modifications. So in theory, if you
are not interested in the active defragmentation, you can replace Jemalloc
just following tose steps:
1. Remove the jemalloc directory.
2. Substitute it with the new jemalloc source tree.
3. Edit the Makefile localted in the same directory as the README you are
reading, and change the --with-version in the Jemalloc configure script
options with the version you are using. This is required because otherwise
Jemalloc configuration script is broken and will not work nested in another
git repository.
However note that we change Jemalloc settings via the `configure` script of Jemalloc using the `--with-lg-quantum` option, setting it to the value of 3 instead of 4. This provides us with more size classes that better suit the Redis data structures, in order to gain memory efficiency.
If you want to upgrade Jemalloc while also providing support for
active defragmentation, in addition to the above steps you need to perform
the following additional steps:
5. In Jemalloc three, file `include/jemalloc/jemalloc_macros.h.in`, make sure
to add `#define JEMALLOC_FRAG_HINT`.
6. Implement the function `je_get_defrag_hint()` inside `src/jemalloc.c`. You
can see how it is implemented in the current Jemalloc source tree shipped
with Redis, and rewrite it according to the new Jemalloc internals, if they
changed, otherwise you could just copy the old implementation if you are
upgrading just to a similar version of Jemalloc.
Jemalloc for Redis for Windows
---
In order to reconfigure `jemalloc` for Redis for Windows - please follow the steps from `msvc/ReadMe.txt` file. Due to memory allocation tracking in Windows
port - options `--with-lg-page` and `--with-lg-hugepage` must be set to the same value (which is now `22` and it defines a `1 << 22` page size = 4MB).
`jemalloc` has been confiured with:
```
sh -c "CC=cl ./autogen.sh --with-lg-quantum=3 --with-lg-page=22 --with-lg-hugepage=22 --with-version=5.2.1-0-g6619f70ffc9b53ba5d3b38c8c3d5676905242c87"
```
**NOTE**: active defragmentation is **switched OFF** for now as the Redis-specific modification that supports that feature seems to not work properly (either in jemalooc 5.2.1 or together with memory allocation code specific to Windows port of Redis).
Geohash
---
This is never upgraded since it's part of the Redis project. If there are changes to merge from Ardb there is the need to manually check differences, but at this point the source code is pretty different.
Hiredis
---
Hiredis uses the SDS string library, that must be the same version used inside Redis itself. Hiredis is also very critical for Sentinel. Historically Redis often used forked versions of hiredis in a way or the other. In order to upgrade it is advised to take a lot of care:
1. Check with diff if hiredis API changed and what impact it could have in Redis.
2. Make sure thet the SDS library inside Hiredis and inside Redis are compatible.
3. After the upgrade, run the Redis Sentinel test.
4. Check manually that redis-cli and redis-benchmark behave as expecteed, since we have no tests for CLI utilities currently.
Linenoise
---
Linenoise is rarely upgraded as needed. The upgrade process is trivial since
Redis uses a non modified version of linenoise, so to upgrade just do the
following:
1. Remove the linenoise directory.
2. Substitute it with the new linenoise source tree.
Lua
---
We use Lua 5.1 and no upgrade is planned currently, since we don't want to break
Lua scripts for new Lua features: in the context of Redis Lua scripts the
capabilities of 5.1 are usually more than enough, the release is rock solid,
and we definitely don't want to break old scripts.
So upgrading of Lua is up to the Redis project maintainers and should be a
manual procedure performed by taking a diff between the different versions.
Currently we have at least the following differences between official Lua 5.1
and our version:
1. Makefile is modified to allow a different compiler than GCC.
2. We have the implementation source code, and directly link to the following external libraries: `lua_cjson.o`, `lua_struct.o`, `lua_cmsgpack.o` and `lua_bit.o`.
3. There is a security fix in `ldo.c`, line 498: The check for `LUA_SIGNATURE[0]` is removed in order toa void direct bytecode execution.
deps/geohash-int/Makefile
Vendored
-23
View File
@@ -1,23 +0,0 @@
STD=
WARN= -Wall
OPT= -O2
R_CFLAGS= $(STD) $(WARN) $(OPT) $(DEBUG) $(CFLAGS)
R_LDFLAGS= $(LDFLAGS)
DEBUG= -g
R_CC=$(CC) $(R_CFLAGS)
R_LD=$(CC) $(R_LDFLAGS)
all: geohash.o geohash_helper.o
.PHONY: all
geohash.o: geohash.h geohash.c
geohash_helper.o: geohash.h geohash_helper.h geohash_helper.c
.c.o:
$(R_CC) -c $<
clean:
rm -f *.o
deps/geohash-int/geohash.c
Vendored
-299
View File
@@ -1,299 +0,0 @@
/*
* Copyright (c) 2013-2014, yinqiwen <yinqiwen@gmail.com>
* Copyright (c) 2014, Matt Stancliff <matt@genges.com>.
* Copyright (c) 2015, Salvatore Sanfilippo <antirez@gmail.com>.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF
* THE POSSIBILITY OF SUCH DAMAGE.
*/
#include "geohash.h"
#ifdef _WIN32
#define inline __inline
#endif
/**
* Hashing works like this:
* Divide the world into 4 buckets. Label each one as such:
* -----------------
* | | |
* | | |
* | 0,1 | 1,1 |
* -----------------
* | | |
* | | |
* | 0,0 | 1,0 |
* -----------------
*/
/* Interleave lower bits of x and y, so the bits of x
* are in the even positions and bits from y in the odd;
* x and y must initially be less than 2**32 (65536).
* From: https://graphics.stanford.edu/~seander/bithacks.html#InterleaveBMN
*/
static inline uint64_t interleave64(uint32_t xlo, uint32_t ylo) {
static const uint64_t B[] = {0x5555555555555555ULL, 0x3333333333333333ULL,
0x0F0F0F0F0F0F0F0FULL, 0x00FF00FF00FF00FFULL,
0x0000FFFF0000FFFFULL};
static const unsigned int S[] = {1, 2, 4, 8, 16};
uint64_t x = xlo;
uint64_t y = ylo;
x = (x | (x << S[4])) & B[4];
y = (y | (y << S[4])) & B[4];
x = (x | (x << S[3])) & B[3];
y = (y | (y << S[3])) & B[3];
x = (x | (x << S[2])) & B[2];
y = (y | (y << S[2])) & B[2];
x = (x | (x << S[1])) & B[1];
y = (y | (y << S[1])) & B[1];
x = (x | (x << S[0])) & B[0];
y = (y | (y << S[0])) & B[0];
return x | (y << 1);
}
/* reverse the interleave process
* derived from http://stackoverflow.com/questions/4909263
*/
static inline uint64_t deinterleave64(uint64_t interleaved) {
static const uint64_t B[] = {0x5555555555555555ULL, 0x3333333333333333ULL,
0x0F0F0F0F0F0F0F0FULL, 0x00FF00FF00FF00FFULL,
0x0000FFFF0000FFFFULL, 0x00000000FFFFFFFFULL};
static const unsigned int S[] = {0, 1, 2, 4, 8, 16};
uint64_t x = interleaved;
uint64_t y = interleaved >> 1;
x = (x | (x >> S[0])) & B[0];
y = (y | (y >> S[0])) & B[0];
x = (x | (x >> S[1])) & B[1];
y = (y | (y >> S[1])) & B[1];
x = (x | (x >> S[2])) & B[2];
y = (y | (y >> S[2])) & B[2];
x = (x | (x >> S[3])) & B[3];
y = (y | (y >> S[3])) & B[3];
x = (x | (x >> S[4])) & B[4];
y = (y | (y >> S[4])) & B[4];
x = (x | (x >> S[5])) & B[5];
y = (y | (y >> S[5])) & B[5];
return x | (y << 32);
}
void geohashGetCoordRange(GeoHashRange *long_range, GeoHashRange *lat_range) {
/* These are constraints from EPSG:900913 / EPSG:3785 / OSGEO:41001 */
/* We can't geocode at the north/south pole. */
long_range->max = GEO_LONG_MAX;
long_range->min = GEO_LONG_MIN;
lat_range->max = GEO_LAT_MAX;
lat_range->min = GEO_LAT_MIN;
}
int geohashEncode(GeoHashRange *long_range, GeoHashRange *lat_range,
double longitude, double latitude, uint8_t step,
GeoHashBits *hash) {
/* Check basic arguments sanity. */
if (hash == NULL || step > 32 || step == 0 ||
RANGEPISZERO(lat_range) || RANGEPISZERO(long_range)) return 0;
/* Return an error when trying to index outside the supported
* constraints. */
if (longitude > 180 || longitude < -180 ||
latitude > 85.05112878 || latitude < -85.05112878) return 0;
hash->bits = 0;
hash->step = step;
if (latitude < lat_range->min || latitude > lat_range->max ||
longitude < long_range->min || longitude > long_range->max) {
return 0;
}
double lat_offset =
(latitude - lat_range->min) / (lat_range->max - lat_range->min);
double long_offset =
(longitude - long_range->min) / (long_range->max - long_range->min);
/* convert to fixed point based on the step size */
lat_offset *= (1 << step);
long_offset *= (1 << step);
hash->bits = interleave64(lat_offset, long_offset);
return 1;
}
int geohashEncodeType(double longitude, double latitude, uint8_t step, GeoHashBits *hash) {
GeoHashRange r[2] = { { 0 } };
geohashGetCoordRange(&r[0], &r[1]);
return geohashEncode(&r[0], &r[1], longitude, latitude, step, hash);
}
int geohashEncodeWGS84(double longitude, double latitude, uint8_t step,
GeoHashBits *hash) {
return geohashEncodeType(longitude, latitude, step, hash);
}
int geohashDecode(const GeoHashRange long_range, const GeoHashRange lat_range,
const GeoHashBits hash, GeoHashArea *area) {
if (HASHISZERO(hash) || NULL == area || RANGEISZERO(lat_range) ||
RANGEISZERO(long_range)) {
return 0;
}
area->hash = hash;
uint8_t step = hash.step;
uint64_t hash_sep = deinterleave64(hash.bits); /* hash = [LAT][LONG] */
double lat_scale = lat_range.max - lat_range.min;
double long_scale = long_range.max - long_range.min;
uint32_t ilato = hash_sep; /* get lat part of deinterleaved hash */
uint32_t ilono = hash_sep >> 32; /* shift over to get long part of hash */
/* divide by 2**step.
* Then, for 0-1 coordinate, multiply times scale and add
to the min to get the absolute coordinate. */
area->latitude.min =
lat_range.min + (ilato * 1.0 / (1ull << step)) * lat_scale;
area->latitude.max =
lat_range.min + ((ilato + 1) * 1.0 / (1ull << step)) * lat_scale;
area->longitude.min =
long_range.min + (ilono * 1.0 / (1ull << step)) * long_scale;
area->longitude.max =
long_range.min + ((ilono + 1) * 1.0 / (1ull << step)) * long_scale;
return 1;
}
int geohashDecodeType(const GeoHashBits hash, GeoHashArea *area) {
GeoHashRange r[2] = { { 0 } };
geohashGetCoordRange(&r[0], &r[1]);
return geohashDecode(r[0], r[1], hash, area);
}
int geohashDecodeWGS84(const GeoHashBits hash, GeoHashArea *area) {
return geohashDecodeType(hash, area);
}
int geohashDecodeAreaToLongLat(const GeoHashArea *area, double *xy) {
if (!xy) return 0;
xy[0] = (area->longitude.min + area->longitude.max) / 2;
xy[1] = (area->latitude.min + area->latitude.max) / 2;
return 1;
}
int geohashDecodeToLongLatType(const GeoHashBits hash, double *xy) {
GeoHashArea area = { { 0 } };
if (!xy || !geohashDecodeType(hash, &area))
return 0;
return geohashDecodeAreaToLongLat(&area, xy);
}
int geohashDecodeToLongLatWGS84(const GeoHashBits hash, double *xy) {
return geohashDecodeToLongLatType(hash, xy);
}
static void geohash_move_x(GeoHashBits *hash, int8_t d) {
if (d == 0)
return;
uint64_t x = hash->bits & 0xaaaaaaaaaaaaaaaaULL;
uint64_t y = hash->bits & 0x5555555555555555ULL;
uint64_t zz = 0x5555555555555555ULL >> (64 - hash->step * 2);
if (d > 0) {
x = x + (zz + 1);
} else {
x = x | zz;
x = x - (zz + 1);
}
x &= (0xaaaaaaaaaaaaaaaaULL >> (64 - hash->step * 2));
hash->bits = (x | y);
}
static void geohash_move_y(GeoHashBits *hash, int8_t d) {
if (d == 0)
return;
uint64_t x = hash->bits & 0xaaaaaaaaaaaaaaaaULL;
uint64_t y = hash->bits & 0x5555555555555555ULL;
uint64_t zz = 0xaaaaaaaaaaaaaaaaULL >> (64 - hash->step * 2);
if (d > 0) {
y = y + (zz + 1);
} else {
y = y | zz;
y = y - (zz + 1);
}
y &= (0x5555555555555555ULL >> (64 - hash->step * 2));
hash->bits = (x | y);
}
void geohashNeighbors(const GeoHashBits *hash, GeoHashNeighbors *neighbors) {
neighbors->east = *hash;
neighbors->west = *hash;
neighbors->north = *hash;
neighbors->south = *hash;
neighbors->south_east = *hash;
neighbors->south_west = *hash;
neighbors->north_east = *hash;
neighbors->north_west = *hash;
geohash_move_x(&neighbors->east, 1);
geohash_move_y(&neighbors->east, 0);
geohash_move_x(&neighbors->west, -1);
geohash_move_y(&neighbors->west, 0);
geohash_move_x(&neighbors->south, 0);
geohash_move_y(&neighbors->south, -1);
geohash_move_x(&neighbors->north, 0);
geohash_move_y(&neighbors->north, 1);
geohash_move_x(&neighbors->north_west, -1);
geohash_move_y(&neighbors->north_west, 1);
geohash_move_x(&neighbors->north_east, 1);
geohash_move_y(&neighbors->north_east, 1);
geohash_move_x(&neighbors->south_east, 1);
geohash_move_y(&neighbors->south_east, -1);
geohash_move_x(&neighbors->south_west, -1);
geohash_move_y(&neighbors->south_west, -1);
}
deps/geohash-int/geohash.h
Vendored
-118
View File
@@ -1,118 +0,0 @@
/*
* Copyright (c) 2013-2014, yinqiwen <yinqiwen@gmail.com>
* Copyright (c) 2014, Matt Stancliff <matt@genges.com>.
* Copyright (c) 2015, Salvatore Sanfilippo <antirez@gmail.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 GEOHASH_H_
#define GEOHASH_H_
#include <stddef.h>
#include <stdint.h>
#include <stdint.h>
#if defined(__cplusplus)
extern "C" {
#endif
#define HASHISZERO(r) (!(r).bits && !(r).step)
#define RANGEISZERO(r) (!(r).max && !(r).min)
#define RANGEPISZERO(r) (r == NULL || RANGEISZERO(*r))
#define GEO_STEP_MAX 26 /* 26*2 = 52 bits. */
/* Limits from EPSG:900913 / EPSG:3785 / OSGEO:41001 */
#define GEO_LAT_MIN -85.05112878
#define GEO_LAT_MAX 85.05112878
#define GEO_LONG_MIN -180
#define GEO_LONG_MAX 180
typedef enum {
GEOHASH_NORTH = 0,
GEOHASH_EAST,
GEOHASH_WEST,
GEOHASH_SOUTH,
GEOHASH_SOUTH_WEST,
GEOHASH_SOUTH_EAST,
GEOHASH_NORT_WEST,
GEOHASH_NORT_EAST
} GeoDirection;
typedef struct {
uint64_t bits;
uint8_t step;
} GeoHashBits;
typedef struct {
double min;
double max;
} GeoHashRange;
typedef struct {
GeoHashBits hash;
GeoHashRange longitude;
GeoHashRange latitude;
} GeoHashArea;
typedef struct {
GeoHashBits north;
GeoHashBits east;
GeoHashBits west;
GeoHashBits south;
GeoHashBits north_east;
GeoHashBits south_east;
GeoHashBits north_west;
GeoHashBits south_west;
} GeoHashNeighbors;
/*
* 0:success
* -1:failed
*/
void geohashGetCoordRange(GeoHashRange *long_range, GeoHashRange *lat_range);
int geohashEncode(GeoHashRange *long_range, GeoHashRange *lat_range,
double longitude, double latitude, uint8_t step,
GeoHashBits *hash);
int geohashEncodeType(double longitude, double latitude,
uint8_t step, GeoHashBits *hash);
int geohashEncodeWGS84(double longitude, double latitude, uint8_t step,
GeoHashBits *hash);
int geohashDecode(const GeoHashRange long_range, const GeoHashRange lat_range,
const GeoHashBits hash, GeoHashArea *area);
int geohashDecodeType(const GeoHashBits hash, GeoHashArea *area);
int geohashDecodeWGS84(const GeoHashBits hash, GeoHashArea *area);
int geohashDecodeAreaToLongLat(const GeoHashArea *area, double *xy);
int geohashDecodeToLongLatType(const GeoHashBits hash, double *xy);
int geohashDecodeToLongLatWGS84(const GeoHashBits hash, double *xy);
int geohashDecodeToLongLatMercator(const GeoHashBits hash, double *xy);
void geohashNeighbors(const GeoHashBits *hash, GeoHashNeighbors *neighbors);
#if defined(__cplusplus)
}
#endif
#endif /* GEOHASH_H_ */
deps/geohash-int/geohash_helper.c
Vendored
-203
View File
@@ -1,203 +0,0 @@
/*
* Copyright (c) 2013-2014, yinqiwen <yinqiwen@gmail.com>
* Copyright (c) 2014, Matt Stancliff <matt@genges.com>.
* Copyright (c) 2015, Salvatore Sanfilippo <antirez@gmail.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.
*/
/* This is a C++ to C conversion from the ardb project.
* This file started out as:
* https://github.com/yinqiwen/ardb/blob/d42503/src/geo/geohash_helper.cpp
*/
#include "geohash_helper.h"
#ifdef _WIN32
#define inline __inline
#define _USE_MATH_DEFINES // for C
#include <math.h>
#endif
#define D_R (M_PI / 180.0)
#define R_MAJOR 6378137.0
#define R_MINOR 6356752.3142
#define RATIO (R_MINOR / R_MAJOR)
#define ECCENT (sqrt(1.0 - (RATIO *RATIO)))
#define COM (0.5 * ECCENT)
/// @brief The usual PI/180 constant
const double DEG_TO_RAD = 0.017453292519943295769236907684886;
/// @brief Earth's quatratic mean radius for WGS-84
const double EARTH_RADIUS_IN_METERS = 6372797.560856;
const double MERCATOR_MAX = 20037726.37;
const double MERCATOR_MIN = -20037726.37;
static inline double deg_rad(double ang) { return ang * D_R; }
static inline double rad_deg(double ang) { return ang / D_R; }
/* You must *ONLY* estimate steps when you are encoding.
* If you are decoding, always decode to GEO_STEP_MAX (26). */
uint8_t geohashEstimateStepsByRadius(double range_meters, double lat) {
if (range_meters == 0) return 26;
int step = 1;
while (range_meters < MERCATOR_MAX) {
range_meters *= 2;
step++;
}
step -= 2; /* Make sure range is included in the worst case. */
/* Wider range torwards the poles... Note: it is possible to do better
* than this approximation by computing the distance between meridians
* at this latitude, but this does the trick for now. */
if (lat > 67 || lat < -67) step--;
if (lat > 80 || lat < -80) step--;
/* Frame to valid range. */
if (step < 1) step = 1;
if (step > 26) step = 26;
return step;
}
int geohashBitsComparator(const GeoHashBits *a, const GeoHashBits *b) {
/* If step not equal, compare on step. Else, compare on bits. */
return a->step != b->step ? a->step - b->step : a->bits - b->bits;
}
int geohashBoundingBox(double longitude, double latitude, double radius_meters,
double *bounds) {
if (!bounds) return 0;
double lonr, latr;
lonr = deg_rad(longitude);
latr = deg_rad(latitude);
if (radius_meters > EARTH_RADIUS_IN_METERS)
radius_meters = EARTH_RADIUS_IN_METERS;
double distance = radius_meters / EARTH_RADIUS_IN_METERS;
double min_latitude = latr - distance;
double max_latitude = latr + distance;
/* Note: we're being lazy and not accounting for coordinates near poles */
double min_longitude, max_longitude;
double difference_longitude = asin(sin(distance) / cos(latr));
min_longitude = lonr - difference_longitude;
max_longitude = lonr + difference_longitude;
bounds[0] = rad_deg(min_longitude);
bounds[1] = rad_deg(min_latitude);
bounds[2] = rad_deg(max_longitude);
bounds[3] = rad_deg(max_latitude);
return 1;
}
GeoHashRadius geohashGetAreasByRadius(double longitude, double latitude, double radius_meters) {
GeoHashRange long_range, lat_range;
GeoHashRadius radius = { { 0 } };
GeoHashBits hash = { 0 };
GeoHashNeighbors neighbors = { { 0 } };
GeoHashArea area = { { 0 } };
double min_lon, max_lon, min_lat, max_lat;
double bounds[4];
int steps;
geohashBoundingBox(longitude, latitude, radius_meters, bounds);
min_lon = bounds[0];
min_lat = bounds[1];
max_lon = bounds[2];
max_lat = bounds[3];
steps = geohashEstimateStepsByRadius(radius_meters,latitude);
geohashGetCoordRange(&long_range, &lat_range);
geohashEncode(&long_range, &lat_range, longitude, latitude, steps, &hash);
geohashNeighbors(&hash, &neighbors);
geohashGetCoordRange(&long_range, &lat_range);
geohashDecode(long_range, lat_range, hash, &area);
if (area.latitude.min < min_lat) {
GZERO(neighbors.south);
GZERO(neighbors.south_west);
GZERO(neighbors.south_east);
}
if (area.latitude.max > max_lat) {
GZERO(neighbors.north);
GZERO(neighbors.north_east);
GZERO(neighbors.north_west);
}
if (area.longitude.min < min_lon) {
GZERO(neighbors.west);
GZERO(neighbors.south_west);
GZERO(neighbors.north_west);
}
if (area.longitude.max > max_lon) {
GZERO(neighbors.east);
GZERO(neighbors.south_east);
GZERO(neighbors.north_east);
}
radius.hash = hash;
radius.neighbors = neighbors;
radius.area = area;
return radius;
}
GeoHashRadius geohashGetAreasByRadiusWGS84(double longitude, double latitude,
double radius_meters) {
return geohashGetAreasByRadius(longitude, latitude, radius_meters);
}
GeoHashFix52Bits geohashAlign52Bits(const GeoHashBits hash) {
uint64_t bits = hash.bits;
bits <<= (52 - hash.step * 2);
return bits;
}
/* Calculate distance using haversin great circle distance formula. */
double geohashGetDistance(double lon1d, double lat1d, double lon2d, double lat2d) {
double lat1r, lon1r, lat2r, lon2r, u, v;
lat1r = deg_rad(lat1d);
lon1r = deg_rad(lon1d);
lat2r = deg_rad(lat2d);
lon2r = deg_rad(lon2d);
u = sin((lat2r - lat1r) / 2);
v = sin((lon2r - lon1r) / 2);
return 2.0 * EARTH_RADIUS_IN_METERS *
asin(sqrt(u * u + cos(lat1r) * cos(lat2r) * v * v));
}
int geohashGetDistanceIfInRadius(double x1, double y1,
double x2, double y2, double radius,
double *distance) {
*distance = geohashGetDistance(x1, y1, x2, y2);
if (*distance > radius) return 0;
return 1;
}
int geohashGetDistanceIfInRadiusWGS84(double x1, double y1, double x2,
double y2, double radius,
double *distance) {
return geohashGetDistanceIfInRadius(x1, y1, x2, y2, radius, distance);
}
deps/geohash-int/geohash_helper.h
Vendored
-71
View File
@@ -1,71 +0,0 @@
/*
* Copyright (c) 2013-2014, yinqiwen <yinqiwen@gmail.com>
* Copyright (c) 2014, Matt Stancliff <matt@genges.com>.
* Copyright (c) 2015, Salvatore Sanfilippo <antirez@gmail.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 GEOHASH_HELPER_HPP_
#define GEOHASH_HELPER_HPP_
#include <math.h>
#include "geohash.h"
#define GZERO(s) s.bits = s.step = 0;
#define GISZERO(s) (!s.bits && !s.step)
#define GISNOTZERO(s) (s.bits || s.step)
typedef uint64_t GeoHashFix52Bits;
typedef uint64_t GeoHashVarBits;
typedef struct {
GeoHashBits hash;
GeoHashArea area;
GeoHashNeighbors neighbors;
} GeoHashRadius;
int GeoHashBitsComparator(const GeoHashBits *a, const GeoHashBits *b);
uint8_t geohashEstimateStepsByRadius(double range_meters, double lat);
int geohashBoundingBox(double longitude, double latitude, double radius_meters,
double *bounds);
GeoHashRadius geohashGetAreasByRadius(double longitude,
double latitude, double radius_meters);
GeoHashRadius geohashGetAreasByRadiusWGS84(double longitude, double latitude,
double radius_meters);
GeoHashRadius geohashGetAreasByRadiusMercator(double longitude, double latitude,
double radius_meters);
GeoHashFix52Bits geohashAlign52Bits(const GeoHashBits hash);
double geohashGetDistance(double lon1d, double lat1d,
double lon2d, double lat2d);
int geohashGetDistanceIfInRadius(double x1, double y1,
double x2, double y2, double radius,
double *distance);
int geohashGetDistanceIfInRadiusWGS84(double x1, double y1, double x2,
double y2, double radius,
double *distance);
#endif /* GEOHASH_HELPER_HPP_ */
deps/hiredis/README.md
Vendored
+2 -2
View File
@@ -98,7 +98,7 @@ was received:
* **`REDIS_REPLY_INTEGER`**:
* The command replied with an integer. The integer value can be accessed using the
`reply->integer` field of type `PORT_LONGLONG`.
`reply->integer` field of type `long long`.
* **`REDIS_REPLY_NIL`**:
* The command replied with a **nil** object. There is no data to access.
@@ -300,7 +300,7 @@ An asynchronous connection can be terminated using:
void redisAsyncDisconnect(redisAsyncContext *ac);
When this function is called, the connection is **not** immediately terminated. Instead, new
commands are no PORT_LONGer accepted and the connection is only terminated when all pending commands
commands are no longer accepted and the connection is only terminated when all pending commands
have been written to the socket, their respective replies have been read and their respective
callbacks have been executed. After this, the disconnection callback is executed with the
`REDIS_OK` status and the context object is free'd.
deps/hiredis/adapters/libevent.h
Vendored
+12 -12
View File
@@ -30,13 +30,13 @@
#ifndef __HIREDIS_LIBEVENT_H__
#define __HIREDIS_LIBEVENT_H__
#include <event2/event.h>
#include <event.h>
#include "../hiredis.h"
#include "../async.h"
typedef struct redisLibeventEvents {
redisAsyncContext *context;
struct event *rev, *wev;
struct event rev, wev;
} redisLibeventEvents;
static void redisLibeventReadEvent(int fd, short event, void *arg) {
@@ -53,28 +53,28 @@ static void redisLibeventWriteEvent(int fd, short event, void *arg) {
static void redisLibeventAddRead(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_add(e->rev,NULL);
event_add(&e->rev,NULL);
}
static void redisLibeventDelRead(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_del(e->rev);
event_del(&e->rev);
}
static void redisLibeventAddWrite(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_add(e->wev,NULL);
event_add(&e->wev,NULL);
}
static void redisLibeventDelWrite(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_del(e->wev);
event_del(&e->wev);
}
static void redisLibeventCleanup(void *privdata) {
redisLibeventEvents *e = (redisLibeventEvents*)privdata;
event_del(e->rev);
event_del(e->wev);
event_del(&e->rev);
event_del(&e->wev);
free(e);
}
@@ -99,10 +99,10 @@ static int redisLibeventAttach(redisAsyncContext *ac, struct event_base *base) {
ac->ev.data = e;
/* Initialize and install read/write events */
e->rev = event_new(base, c->fd, EV_READ, redisLibeventReadEvent, e);
e->wev = event_new(base, c->fd, EV_WRITE, redisLibeventWriteEvent, e);
event_add(e->rev, NULL);
event_add(e->wev, NULL);
event_set(&e->rev,c->fd,EV_READ,redisLibeventReadEvent,e);
event_set(&e->wev,c->fd,EV_WRITE,redisLibeventWriteEvent,e);
event_base_set(base,&e->rev);
event_base_set(base,&e->wev);
return REDIS_OK;
}
#endif
deps/hiredis/adapters/libuv.h
Vendored
+3 -4
View File
@@ -1,6 +1,5 @@
#ifndef __HIREDIS_LIBUV_H__
#define __HIREDIS_LIBUV_H__
#include <stdlib.h>
#include <uv.h>
#include "../hiredis.h"
#include "../async.h"
@@ -12,6 +11,7 @@ typedef struct redisLibuvEvents {
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;
@@ -20,10 +20,10 @@ static void redisLibuvPoll(uv_poll_t* handle, int status, int events) {
return;
}
if (p->context != NULL && (events & UV_READABLE)) {
if (events & UV_READABLE) {
redisAsyncHandleRead(p->context);
}
if (p->context != NULL && (events & UV_WRITABLE)) {
if (events & UV_WRITABLE) {
redisAsyncHandleWrite(p->context);
}
}
@@ -83,7 +83,6 @@ static void on_close(uv_handle_t* handle) {
static void redisLibuvCleanup(void *privdata) {
redisLibuvEvents* p = (redisLibuvEvents*)privdata;
p->context = NULL; // indicate that context might no PORT_LONGer exist
uv_close((uv_handle_t*)&p->handle, on_close);
}
deps/hiredis/async.c
Vendored
+486 -520
View File
File diff suppressed because it is too large Load Diff
deps/hiredis/async.h
Vendored
-3
View File
@@ -106,8 +106,6 @@ 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 *redisAsyncConnectBindWithReuse(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);
@@ -127,7 +125,6 @@ int redisAsyncHandleWriteComplete(redisAsyncContext *ac, int written);
int redisvAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *format, va_list ap);
int redisAsyncCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *format, ...);
int redisAsyncCommandArgv(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, int argc, const char **argv, const size_t *argvlen);
int redisAsyncFormattedCommand(redisAsyncContext *ac, redisCallbackFn *fn, void *privdata, const char *cmd, size_t len);
#ifdef __cplusplus
}
deps/hiredis/dict.c
Vendored
+74 -12
View File
@@ -32,9 +32,6 @@
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifdef _WIN32
#include "../../src/win32_Interop/win32_types.h"
#endif
#include "fmacros.h"
#include <stdlib.h>
@@ -45,7 +42,11 @@
/* -------------------------- private prototypes ---------------------------- */
static int _dictExpandIfNeeded(dict *ht);
static PORT_ULONG _dictNextPower(PORT_ULONG size);
#ifdef _WIN32
static size_t _dictNextPower(size_t size);
#else
static unsigned long _dictNextPower(unsigned long size);
#endif
static int _dictKeyIndex(dict *ht, const void *key);
static int _dictInit(dict *ht, dictType *type, void *privDataPtr);
@@ -88,9 +89,10 @@ static int _dictInit(dict *ht, dictType *type, void *privDataPtr) {
}
/* Expand or create the hashtable */
static int dictExpand(dict *ht, PORT_ULONG size) {
#ifdef _WIN32
static int dictExpand(dict *ht, size_t size) {
dict n; /* the new hashtable */
PORT_ULONG realsize = _dictNextPower(size), i;
size_t realsize = _dictNextPower(size), i;
/* the size is invalid if it is smaller than the number of
* elements already inside the hashtable */
@@ -133,6 +135,53 @@ static int dictExpand(dict *ht, PORT_ULONG size) {
*ht = n;
return DICT_OK;
}
#else
static int dictExpand(dict *ht, unsigned long size) {
dict n; /* the new hashtable */
unsigned long realsize = _dictNextPower(size), i;
/* the size is invalid if it is smaller than the number of
* elements already inside the hashtable */
if (ht->used > size)
return DICT_ERR;
_dictInit(&n, ht->type, ht->privdata);
n.size = realsize;
n.sizemask = realsize-1;
n.table = calloc(realsize,sizeof(dictEntry*));
/* Copy all the elements from the old to the new table:
* note that if the old hash table is empty ht->size is zero,
* so dictExpand just creates an hash table. */
n.used = ht->used;
for (i = 0; i < ht->size && ht->used > 0; i++) {
dictEntry *he, *nextHe;
if (ht->table[i] == NULL) continue;
/* For each hash entry on this slot... */
he = ht->table[i];
while(he) {
unsigned int h;
nextHe = he->next;
/* Get the new element index */
h = dictHashKey(ht, he->key) & n.sizemask;
he->next = n.table[h];
n.table[h] = he;
ht->used--;
/* Pass to the next element */
he = nextHe;
}
}
assert(ht->used == 0);
free(ht->table);
/* Remap the new hashtable in the old */
*ht = n;
return DICT_OK;
}
#endif
/* Add an element to the target hash table */
static int dictAdd(dict *ht, void *key, void *val) {
@@ -164,7 +213,7 @@ static int dictReplace(dict *ht, void *key, void *val) {
dictEntry *entry, auxentry;
/* Try to add the element. If the key
* does not exists dictAdd will succeed. */
* does not exists dictAdd will suceed. */
if (dictAdd(ht, key, val) == DICT_OK)
return 1;
/* It already exists, get the entry */
@@ -214,7 +263,7 @@ static int dictDelete(dict *ht, const void *key) {
/* Destroy an entire hash table */
static int _dictClear(dict *ht) {
PORT_ULONG i;
unsigned long i;
/* Free all the elements */
for (i = 0; i < ht->size && ht->used > 0; i++) {
@@ -296,7 +345,7 @@ static void dictReleaseIterator(dictIterator *iter) {
/* Expand the hash table if needed */
static int _dictExpandIfNeeded(dict *ht) {
/* If the hash table is empty expand it to the initial size,
/* If the hash table is empty expand it to the intial size,
* if the table is "full" dobule its size. */
if (ht->size == 0)
return dictExpand(ht, DICT_HT_INITIAL_SIZE);
@@ -306,16 +355,29 @@ static int _dictExpandIfNeeded(dict *ht) {
}
/* Our hash table capability is a power of two */
static PORT_ULONG _dictNextPower(PORT_ULONG size) {
PORT_ULONG i = DICT_HT_INITIAL_SIZE;
#ifdef _WIN32
static size_t _dictNextPower(size_t size) {
size_t i = DICT_HT_INITIAL_SIZE;
if (size >= PORT_LONG_MAX) return PORT_LONG_MAX;
if (size >= LONG_MAX) return LONG_MAX;
while(1) {
if (i >= size)
return i;
i *= 2;
}
}
#else
static unsigned long _dictNextPower(unsigned long size) {
unsigned long i = DICT_HT_INITIAL_SIZE;
if (size >= LONG_MAX) return LONG_MAX;
while(1) {
if (i >= size)
return i;
i *= 2;
}
}
#endif
/* Returns the index of a free slot that can be populated with
* an hash entry for the given 'key'.
deps/hiredis/dict.h
Vendored
+14 -4
View File
@@ -60,9 +60,15 @@ typedef struct dictType {
typedef struct dict {
dictEntry **table;
dictType *type;
PORT_ULONG size;
PORT_ULONG sizemask;
PORT_ULONG used;
#ifdef _WIN32
size_t size;
size_t sizemask;
size_t used;
#else
unsigned long size;
unsigned long sizemask;
unsigned long used;
#endif
void *privdata;
} dict;
@@ -113,7 +119,11 @@ typedef struct dictIterator {
/* API */
static unsigned int dictGenHashFunction(const unsigned char *buf, int len);
static dict *dictCreate(dictType *type, void *privDataPtr);
static int dictExpand(dict *ht, PORT_ULONG size);
#ifdef _WIN32
static int dictExpand(dict *ht, size_t size);
#else
static int dictExpand(dict *ht, unsigned long size);
#endif
static int dictAdd(dict *ht, void *key, void *val);
static int dictReplace(dict *ht, void *key, void *val);
static int dictDelete(dict *ht, const void *key);
deps/hiredis/fmacros.h
Vendored
+7 -8
View File
@@ -1,24 +1,23 @@
#ifndef __HIREDIS_FMACRO_H
#define __HIREDIS_FMACRO_H
#if defined(__linux__)
#if !defined(_BSD_SOURCE)
#define _BSD_SOURCE
#define _DEFAULT_SOURCE
#endif
#if defined(__CYGWIN__)
#include <sys/cdefs.h>
#if defined(_AIX)
#define _ALL_SOURCE
#endif
#if defined(__sun__)
#define _POSIX_C_SOURCE 200112L
#else
#if !(defined(__APPLE__) && defined(__MACH__))
#elif defined(__linux__) || defined(__OpenBSD__) || defined(__NetBSD__)
#define _XOPEN_SOURCE 600
#endif
#else
#define _XOPEN_SOURCE
#endif
#if defined(__APPLE__) && defined(__MACH__)
#if __APPLE__ && __MACH__
#define _OSX
#endif
deps/hiredis/hiredis.c
Vendored
+1099 -757
View File
File diff suppressed because it is too large Load Diff
deps/hiredis/hiredis.h
Vendored
+87 -80
View File
@@ -1,8 +1,6 @@
/*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2014, Pieter Noordhuis <pcnoordhuis at gmail dot com>
* Copyright (c) 2015, Matt Stancliff <matt at genges dot com>,
* Jan-Erik Rediger <janerik at fnordig dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
@@ -33,21 +31,36 @@
#ifndef __HIREDIS_H
#define __HIREDIS_H
#include "read.h"
#include <stdio.h> /* for size_t */
#include <stdarg.h> /* for va_list */
#ifndef _WIN32
#include <sys/time.h> /* for struct timeval */
#else
#include "../../src/Win32_Interop/win32_types_hiredis.h"
#endif
#include <stdint.h> /* uintXX_t, etc */
#include "sds.h" /* for sds */
#ifdef _WIN32
#include "..\..\src\win32_Interop\Win32_FDAPI.h"
#include <windows.h>
#ifndef va_copy
#define va_copy(d,s) d = (s)
#endif
#endif
#define HIREDIS_MAJOR 0
#define HIREDIS_MINOR 13
#define HIREDIS_PATCH 3
#define HIREDIS_SONAME 0.13
#define HIREDIS_MINOR 11
#define HIREDIS_PATCH 0
#define REDIS_ERR -1
#define REDIS_OK 0
/* When an error occurs, the err flag in a context is set to hold the type of
* error that occured. REDIS_ERR_IO means there was an I/O error and you
* should use the "errno" variable to find out what is wrong.
* For other values, the "errstr" field will hold a description. */
#define REDIS_ERR_IO 1 /* Error in read or write */
#define REDIS_ERR_EOF 3 /* End of file */
#define REDIS_ERR_PROTOCOL 4 /* Protocol error */
#define REDIS_ERR_OOM 5 /* Out of memory */
#define REDIS_ERR_OTHER 2 /* Everything else... */
/* Connection type can be blocking or non-blocking and is set in the
* least significant bit of the flags field in redisContext. */
@@ -76,39 +89,17 @@
/* Flag that is set when monitor mode is active */
#define REDIS_MONITORING 0x40
/* Flag that is set when we should set SO_REUSEADDR before calling bind() */
#define REDIS_REUSEADDR 0x80
#define REDIS_REPLY_STRING 1
#define REDIS_REPLY_ARRAY 2
#define REDIS_REPLY_INTEGER 3
#define REDIS_REPLY_NIL 4
#define REDIS_REPLY_STATUS 5
#define REDIS_REPLY_ERROR 6
#define REDIS_READER_MAX_BUF (1024*16) /* Default max unused reader buffer. */
#define REDIS_KEEPALIVE_INTERVAL 15 /* seconds */
/* number of times we retry to connect in the case of EADDRNOTAVAIL and
* SO_REUSEADDR is being used. */
#define REDIS_CONNECT_RETRIES 10
/* strerror_r has two completely different prototypes and behaviors
* depending on system issues, so we need to operate on the error buffer
* differently depending on which strerror_r we're using. */
#ifndef _GNU_SOURCE
/* "regular" POSIX strerror_r that does the right thing. */
#define __redis_strerror_r(errno, buf, len) \
do { \
strerror_r((errno), (buf), (len)); \
} while (0)
#else
/* "bad" GNU strerror_r we need to clean up after. */
#define __redis_strerror_r(errno, buf, len) \
do { \
char *err_str = strerror_r((errno), (buf), (len)); \
/* If return value _isn't_ the start of the buffer we passed in, \
* then GNU strerror_r returned an internal static buffer and we \
* need to copy the result into our private buffer. */ \
if (err_str != (buf)) { \
strncpy((buf), err_str, ((len) - 1)); \
buf[(len)-1] = '\0'; \
} \
} while (0)
#endif
#ifdef __cplusplus
extern "C" {
#endif
@@ -116,14 +107,62 @@ extern "C" {
/* This is the reply object returned by redisCommand() */
typedef struct redisReply {
int type; /* REDIS_REPLY_* */
PORT_LONGLONG integer; /* The integer when type is REDIS_REPLY_INTEGER */
long long integer; /* The integer when type is REDIS_REPLY_INTEGER */
int len; /* Length of string */
char *str; /* Used for both REDIS_REPLY_ERROR and REDIS_REPLY_STRING */
size_t elements; /* number of elements, for REDIS_REPLY_ARRAY */
struct redisReply **element; /* elements vector for REDIS_REPLY_ARRAY */
} redisReply;
typedef struct redisReadTask {
int type;
int elements; /* number of elements in multibulk container */
int idx; /* index in parent (array) object */
void *obj; /* holds user-generated value for a read task */
struct redisReadTask *parent; /* parent task */
void *privdata; /* user-settable arbitrary field */
} redisReadTask;
typedef struct redisReplyObjectFunctions {
void *(*createString)(const redisReadTask*, char*, size_t);
void *(*createArray)(const redisReadTask*, int);
void *(*createInteger)(const redisReadTask*, long long);
void *(*createNil)(const redisReadTask*);
void (*freeObject)(void*);
} redisReplyObjectFunctions;
/* State for the protocol parser */
typedef struct redisReader {
int err; /* Error flags, 0 when there is no error */
char errstr[128]; /* String representation of error when applicable */
char *buf; /* Read buffer */
size_t pos; /* Buffer cursor */
size_t len; /* Buffer length */
size_t maxbuf; /* Max length of unused buffer */
redisReadTask rstack[9];
int ridx; /* Index of current read task */
void *reply; /* Temporary reply pointer */
redisReplyObjectFunctions *fn;
void *privdata;
} redisReader;
/* Public API for the protocol parser. */
redisReader *redisReaderCreate(void);
void redisReaderFree(redisReader *r);
int redisReaderFeed(redisReader *r, const char *buf, size_t len);
int redisReaderGetReply(redisReader *r, void **reply);
/* Backwards compatibility, can be removed on big version bump. */
#define redisReplyReaderCreate redisReaderCreate
#define redisReplyReaderFree redisReaderFree
#define redisReplyReaderFeed redisReaderFeed
#define redisReplyReaderGetReply redisReaderGetReply
#define redisReplyReaderSetPrivdata(_r, _p) (int)(((redisReader*)(_r))->privdata = (_p))
#define redisReplyReaderGetObject(_r) (((redisReader*)(_r))->reply)
#define redisReplyReaderGetError(_r) (((redisReader*)(_r))->errstr)
/* Function to free the reply objects hiredis returns by default. */
void freeReplyObject(void *reply);
@@ -132,14 +171,6 @@ void freeReplyObject(void *reply);
int redisvFormatCommand(char **target, const char *format, va_list ap);
int redisFormatCommand(char **target, const char *format, ...);
int redisFormatCommandArgv(char **target, int argc, const char **argv, const size_t *argvlen);
int redisFormatSdsCommandArgv(sds *target, int argc, const char ** argv, const size_t *argvlen);
void redisFreeCommand(char *cmd);
void redisFreeSdsCommand(sds cmd);
enum redisConnectionType {
REDIS_CONN_TCP,
REDIS_CONN_UNIX
};
/* Context for a connection to Redis */
typedef struct redisContext {
@@ -149,51 +180,27 @@ typedef struct redisContext {
int flags;
char *obuf; /* Write buffer */
redisReader *reader; /* Protocol reader */
enum redisConnectionType connection_type;
struct timeval *timeout;
struct {
char *host;
char *source_addr;
int port;
} tcp;
struct {
char *path;
} unix_sock;
} redisContext;
redisContext *redisConnect(const char *ip, int port);
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 *redisConnectBindNonBlockWithReuse(const char *ip, int port,
const char *source_addr);
redisContext *redisConnectBindNonBlock(const char *ip, int port, const char *source_addr);
redisContext *redisConnectUnix(const char *path);
redisContext *redisConnectUnixWithTimeout(const char *path, const struct timeval tv);
redisContext *redisConnectUnixNonBlock(const char *path);
redisContext *redisConnectFd(int fd);
/**
* Reconnect the given context using the saved information.
*
* This re-uses the exact same connect options as in the initial connection.
* host, ip (or path), timeout and bind address are reused,
* flags are used unmodified from the existing context.
*
* Returns REDIS_OK on successful connect or REDIS_ERR otherwise.
*/
int redisReconnect(redisContext *c);
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);
#ifdef _WIN32
redisContext *redisPreConnectNonBlock(const char *ip, int port, SOCKADDR_STORAGE *sa);
int redisBufferReadDone(redisContext *c, char *buf, int nread);
int redisBufferWriteDone(redisContext *c, int nwritten, int *done);
#endif
/* In a blocking context, this function first checks if there are unconsumed
* replies to return and returns one if so. Otherwise, it flushes the output
deps/hiredis/net.c
Vendored
+293 -395
View File
@@ -1,9 +1,7 @@
/* Extracted from anet.c to work properly with Hiredis error reporting.
*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2014, Pieter Noordhuis <pcnoordhuis at gmail dot com>
* Copyright (c) 2015, Matt Stancliff <matt at genges dot com>,
* Jan-Erik Rediger <janerik at fnordig dot com>
* Copyright (c) 2006-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
@@ -60,498 +58,398 @@
#include "net.h"
#include "sds.h"
#ifdef _WIN32
#include "win32_hiredis.h"
#include "../../src/win32_Interop/win32fixes.h"
#include "mstcpip.h"
#endif
/* Defined in hiredis.c */
/* 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;
}
if (c && c->fd >= 0) {
close(c->fd);
c->fd = -1;
}
}
static void __redisSetErrorFromErrno(redisContext *c, int type, const char *prefix) {
char buf[128] = { 0 };
size_t len = 0;
char buf[128] = { 0 };
size_t len = 0;
if (prefix != NULL)
len = snprintf(buf, sizeof(buf), "%s: ", prefix);
__redis_strerror_r(errno, (char *)(buf + len), sizeof(buf) - len);
__redisSetError(c, type, buf);
if (prefix != NULL)
len = snprintf(buf,sizeof(buf),"%s: ",prefix);
strerror_r(errno,buf+len,sizeof(buf)-len);
__redisSetError(c,type,buf);
}
static int redisSetReuseAddr(redisContext *c) {
int on = 1;
if (setsockopt(c->fd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(on)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
int on = 1;
if (setsockopt(c->fd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(on)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
}
static int redisCreateSocket(redisContext *c, int type) {
int s;
if ((s = socket(type, SOCK_STREAM, 0)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
return REDIS_ERR;
}
c->fd = s;
if (type == AF_INET) {
if (redisSetReuseAddr(c) == REDIS_ERR) {
return REDIS_ERR;
}
}
return REDIS_OK;
int s;
if ((s = socket(type, SOCK_STREAM, 0)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
return REDIS_ERR;
}
c->fd = s;
if (type == AF_INET) {
if (redisSetReuseAddr(c) == REDIS_ERR) {
return REDIS_ERR;
}
}
return REDIS_OK;
}
static int redisSetBlocking(redisContext *c, int blocking) {
int flags;
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(c->fd, F_GETFL, 0)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "fcntl(F_GETFL)");
redisContextCloseFd(c);
return REDIS_ERR;
}
/* Set the socket nonblocking.
* Note that fcntl(2) for F_GETFL and F_SETFL can't be
* interrupted by a signal. */
if ((flags = fcntl(c->fd, F_GETFL,0)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"fcntl(F_GETFL)");
redisContextCloseFd(c);
return REDIS_ERR;
}
if (blocking)
flags &= ~O_NONBLOCK;
else
flags |= O_NONBLOCK;
if (blocking)
flags &= ~O_NONBLOCK;
else
flags |= O_NONBLOCK;
if (fcntl(c->fd, F_SETFL, flags) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "fcntl(F_SETFL)");
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
if (fcntl(c->fd, F_SETFL, flags) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"fcntl(F_SETFL)");
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
}
int redisKeepAlive(redisContext *c, int interval) {
int val = 1;
int fd = c->fd;
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;
}
if (setsockopt(fd, SOL_SOCKET, SO_KEEPALIVE, &val, sizeof(val)) == -1){
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
val = interval;
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;
}
if (setsockopt(fd, IPPROTO_TCP, TCP_KEEPALIVE, &val, sizeof(val)) < 0) {
__redisSetError(c,REDIS_ERR_OTHER,strerror(errno));
return REDIS_ERR;
}
#else
#ifndef __sun
#ifdef _WIN32
{
struct tcp_keepalive settings;
DWORD bytesReturned;
WSAOVERLAPPED overlapped;
settings.onoff = 1;
settings.keepalivetime = interval * 1000;
settings.keepaliveinterval = interval * 1000 / 3;
overlapped.hEvent = NULL;
FDAPI_WSAIoctl(fd,
SIO_KEEPALIVE_VALS,
&settings,
sizeof(struct tcp_keepalive),
NULL,
0,
&bytesReturned,
&overlapped,
NULL);
}
{
struct tcp_keepalive settings;
DWORD bytesReturned;
WSAOVERLAPPED overlapped;
settings.onoff = 1;
settings.keepalivetime = interval*1000;
settings.keepaliveinterval = interval*1000/3;
overlapped.hEvent = NULL;
WSAIoctl(
fd,
SIO_KEEPALIVE_VALS,
&settings,
sizeof(struct tcp_keepalive),
NULL,
0,
&bytesReturned,
&overlapped,
NULL);
}
#else
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;
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 = 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;
}
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
#endif
return REDIS_OK;
return REDIS_OK;
}
static int redisSetTcpNoDelay(redisContext *c) {
int yes = 1;
if (setsockopt(c->fd, IPPROTO_TCP, TCP_NODELAY, &yes, sizeof(yes)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "setsockopt(TCP_NODELAY)");
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
int yes = 1;
if (setsockopt(c->fd, IPPROTO_TCP, TCP_NODELAY, &yes, sizeof(yes)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(TCP_NODELAY)");
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
}
#define __MAX_MSEC (((LONG_MAX) - 999) / 1000)
static int redisContextTimeoutMsec(redisContext *c, PORT_LONG *result)
{
const struct timeval *timeout = c->timeout;
PORT_LONG msec = -1;
static int redisContextWaitReady(redisContext *c, const struct timeval *timeout) {
struct pollfd wfd[1];
long msec;
/* Only use timeout when not NULL. */
if (timeout != NULL) {
if (timeout->tv_usec > 1000000 || timeout->tv_sec > __MAX_MSEC) {
*result = msec;
return REDIS_ERR;
}
msec = -1;
wfd[0].fd = c->fd;
wfd[0].events = POLLOUT;
msec = (timeout->tv_sec * 1000) + ((timeout->tv_usec + 999) / 1000);
/* Only use timeout when not NULL. */
if (timeout != NULL) {
if (timeout->tv_usec > 1000000 || timeout->tv_sec > __MAX_MSEC) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
redisContextCloseFd(c);
return REDIS_ERR;
}
if (msec < 0 || msec > INT_MAX) {
msec = INT_MAX;
}
}
msec = (timeout->tv_sec * 1000) + ((timeout->tv_usec + 999) / 1000);
*result = msec;
return REDIS_OK;
}
if (msec < 0 || msec > INT_MAX) {
msec = INT_MAX;
}
}
static int redisContextWaitReady(redisContext *c, PORT_LONG msec) {
struct pollfd wfd[1];
if (errno == EINPROGRESS) {
int res;
wfd[0].fd = c->fd;
wfd[0].events = POLLOUT;
if ((res = poll(wfd, 1, msec)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "poll(2)");
redisContextCloseFd(c);
return REDIS_ERR;
} else if (res == 0) {
errno = ETIMEDOUT;
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
redisContextCloseFd(c);
return REDIS_ERR;
}
if (errno == EINPROGRESS) {
int res;
if (redisCheckSocketError(c) != REDIS_OK)
return REDIS_ERR;
if ((res = poll(wfd, 1, (int)msec)) == -1) {
WIN_PORT_FIX /* cast (int) */
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "poll(2)");
redisContextCloseFd(c);
return REDIS_ERR;
}
else if (res == 0) {
errno = ETIMEDOUT;
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
redisContextCloseFd(c);
return REDIS_ERR;
}
return REDIS_OK;
}
if (redisCheckSocketError(c) != REDIS_OK)
return REDIS_ERR;
return REDIS_OK;
}
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
redisContextCloseFd(c);
return REDIS_ERR;
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
redisContextCloseFd(c);
return REDIS_ERR;
}
int redisCheckSocketError(redisContext *c) {
int err = 0;
socklen_t errlen = sizeof(err);
int err = 0;
socklen_t errlen = sizeof(err);
if (getsockopt(c->fd, SOL_SOCKET, SO_ERROR, &err, &errlen) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "getsockopt(SO_ERROR)");
return REDIS_ERR;
}
if (getsockopt(c->fd, SOL_SOCKET, SO_ERROR, &err, &errlen) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"getsockopt(SO_ERROR)");
return REDIS_ERR;
}
if (err) {
errno = err;
__redisSetErrorFromErrno(c, REDIS_ERR_IO, NULL);
return REDIS_ERR;
}
if (err) {
errno = err;
__redisSetErrorFromErrno(c,REDIS_ERR_IO,NULL);
return REDIS_ERR;
}
return REDIS_OK;
return REDIS_OK;
}
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;
}
if (setsockopt(c->fd, SOL_SOCKET, SO_SNDTIMEO, &tv, sizeof(tv)) == -1) {
__redisSetErrorFromErrno(c, REDIS_ERR_IO, "setsockopt(SO_SNDTIMEO)");
return REDIS_ERR;
}
return REDIS_OK;
if (setsockopt(c->fd,SOL_SOCKET,SO_RCVTIMEO,&tv,sizeof(tv)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(SO_RCVTIMEO)");
return REDIS_ERR;
}
if (setsockopt(c->fd,SOL_SOCKET,SO_SNDTIMEO,&tv,sizeof(tv)) == -1) {
__redisSetErrorFromErrno(c,REDIS_ERR_IO,"setsockopt(SO_SNDTIMEO)");
return REDIS_ERR;
}
return REDIS_OK;
}
#ifdef _WIN32
int redisContextPreConnectTcp(
redisContext *c,
const char *addr,
int port,
struct timeval *timeout,
SOCKADDR_STORAGE* ss) {
int blocking = (c->flags & REDIS_BLOCK);
redisContext *c,
const char *addr,
int port,
struct timeval *timeout,
SOCKADDR_STORAGE* ss) {
int blocking = (c->flags & REDIS_BLOCK);
if (ParseStorageAddress(addr, port, ss) == FALSE) {
return REDIS_ERR;
}
if (ParseStorageAddress(addr, port, ss) == FALSE) {
DebugBreak();
}
if (REDIS_OK != redisCreateSocket(c, ss->ss_family)) {
return REDIS_ERR;
}
if (REDIS_OK != redisCreateSocket(c, ss->ss_family)) {
return REDIS_ERR;
}
if (redisSetTcpNoDelay(c) != REDIS_OK)
return REDIS_ERR;
if (redisSetTcpNoDelay(c) != REDIS_OK)
return REDIS_ERR;
if (blocking == 0) {
if (redisSetBlocking(c, 0) != REDIS_OK)
return REDIS_ERR;
}
if (blocking == 0) {
if (redisSetBlocking(c, 0) != REDIS_OK)
return REDIS_ERR;
}
return REDIS_OK;
return REDIS_OK;
}
#endif
static int _redisContextConnectTcp(redisContext *c, const char *addr, int port,
const struct timeval *timeout,
const char *source_addr) {
int s, rv, n;
char _port[6]; /* strlen("65535"); */
struct addrinfo hints, *servinfo, *bservinfo, *p, *b;
int blocking = (c->flags & REDIS_BLOCK);
int reuseaddr = (c->flags & REDIS_REUSEADDR);
int reuses = 0;
PORT_LONG timeout_msec = -1;
const struct timeval *timeout,
const char *source_addr) {
int s, rv;
char _port[6]; /* strlen("65535"); */
struct addrinfo hints, *servinfo, *bservinfo, *p, *b;
int blocking = (c->flags & REDIS_BLOCK);
servinfo = NULL;
c->connection_type = REDIS_CONN_TCP;
c->tcp.port = port;
snprintf(_port, 6, "%d", port);
memset(&hints,0,sizeof(hints));
hints.ai_family = AF_INET;
hints.ai_socktype = SOCK_STREAM;
/* We need to take possession of the passed parameters
* to make them reusable for a reconnect.
* We also carefully check we don't free data we already own,
* as in the case of the reconnect method.
*
* This is a bit ugly, but atleast it works and doesn't leak memory.
**/
if (c->tcp.host != addr) {
if (c->tcp.host)
free(c->tcp.host);
/* 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) {
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;
c->tcp.host = strdup(addr);
}
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,(socklen_t)b->ai_addrlen) != -1) {
bound = 1;
break;
}
}
freeaddrinfo(bservinfo);
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) {
redisContextCloseFd(c);
continue;
} else if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
} else {
if (redisContextWaitReady(c,timeout) != REDIS_OK)
goto error;
}
}
if (blocking && redisSetBlocking(c,1) != REDIS_OK)
goto error;
if (redisSetTcpNoDelay(c) != REDIS_OK)
goto error;
if (timeout) {
if (c->timeout != timeout) {
if (c->timeout == NULL)
c->timeout = malloc(sizeof(struct timeval));
memcpy(c->timeout, timeout, sizeof(struct timeval));
}
}
else {
if (c->timeout)
free(c->timeout);
c->timeout = NULL;
}
if (redisContextTimeoutMsec(c, &timeout_msec) != REDIS_OK) {
__redisSetError(c, REDIS_ERR_IO, "Invalid timeout specified");
goto error;
}
if (source_addr == NULL) {
free(c->tcp.source_addr);
c->tcp.source_addr = NULL;
}
else if (c->tcp.source_addr != source_addr) {
free(c->tcp.source_addr);
c->tcp.source_addr = strdup(source_addr);
}
snprintf(_port, 6, "%d", port);
memset(&hints, 0, sizeof(hints));
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(c->tcp.host, _port, &hints, &servinfo)) != 0) {
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) {
addrretry:
if ((s = socket(p->ai_family, p->ai_socktype, p->ai_protocol)) == -1)
continue;
c->fd = s;
if (redisSetBlocking(c, 0) != REDIS_OK)
goto error;
if (c->tcp.source_addr) {
int bound = 0;
/* Using getaddrinfo saves us from self-determining IPv4 vs IPv6 */
if ((rv = getaddrinfo(c->tcp.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;
}
if (reuseaddr) {
n = 1;
if (setsockopt(s, SOL_SOCKET, SO_REUSEADDR, (char*)&n,
sizeof(n)) < 0) {
goto error;
}
}
for (b = bservinfo; b != NULL; b = b->ai_next) {
if (bind(s, b->ai_addr, b->ai_addrlen) != -1) {
bound = 1;
break;
}
}
freeaddrinfo(bservinfo);
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) {
redisContextCloseFd(c);
continue;
}
else if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
}
else if (errno == EADDRNOTAVAIL && reuseaddr) {
if (++reuses >= REDIS_CONNECT_RETRIES) {
goto error;
}
else {
redisContextCloseFd(c);
goto addrretry;
}
}
else {
if (redisContextWaitReady(c, timeout_msec) != REDIS_OK)
goto error;
}
}
if (blocking && redisSetBlocking(c, 1) != REDIS_OK)
goto error;
if (redisSetTcpNoDelay(c) != REDIS_OK)
goto error;
c->flags |= REDIS_CONNECTED;
rv = REDIS_OK;
goto end;
}
if (p == NULL) {
char buf[128];
snprintf(buf, sizeof(buf), "Can't create socket: %s", strerror(errno));
__redisSetError(c, REDIS_ERR_OTHER, buf);
goto error;
}
c->flags |= REDIS_CONNECTED;
rv = REDIS_OK;
goto end;
}
if (p == NULL) {
char buf[128];
snprintf(buf,sizeof(buf),"Can't create socket: %s",strerror(errno));
__redisSetError(c,REDIS_ERR_OTHER,buf);
goto error;
}
error:
rv = REDIS_ERR;
rv = REDIS_ERR;
end:
freeaddrinfo(servinfo);
return rv; // Need to return REDIS_OK if alright
freeaddrinfo(servinfo);
return rv; // Need to return REDIS_OK if alright
}
int redisContextConnectTcp(redisContext *c, const char *addr, int port,
const struct timeval *timeout) {
return _redisContextConnectTcp(c, addr, port, timeout, NULL);
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);
const struct timeval *timeout,
const char *source_addr) {
return _redisContextConnectTcp(c, addr, port, timeout, source_addr);
}
#ifdef _WIN32
int redisContextConnectUnix(redisContext *c, const char *path, const struct timeval *timeout) {
(void)timeout;
__redisSetError(c, REDIS_ERR_IO,
sdscatprintf(sdsempty(), "Unix sockets are not suported on Windows platform. (%s)\n", path));
(void) timeout;
__redisSetError(c,REDIS_ERR_IO,
sdscatprintf(sdsempty(),"Unix sockets are not suported on Windows platform. (%s)\n", path));
return REDIS_ERR;
return REDIS_ERR;
}
#else
int redisContextConnectUnix(redisContext *c, const char *path, const struct timeval *timeout) {
int blocking = (c->flags & REDIS_BLOCK);
struct sockaddr_un sa;
PORT_LONG timeout_msec = -1;
int blocking = (c->flags & REDIS_BLOCK);
struct sockaddr_un sa;
if (redisCreateSocket(c, AF_LOCAL) < 0)
return REDIS_ERR;
if (redisSetBlocking(c, 0) != REDIS_OK)
return REDIS_ERR;
if (redisCreateSocket(c,AF_LOCAL) < 0)
return REDIS_ERR;
if (redisSetBlocking(c,0) != REDIS_OK)
return REDIS_ERR;
c->connection_type = REDIS_CONN_UNIX;
if (c->unix_sock.path != path)
c->unix_sock.path = strdup(path);
sa.sun_family = AF_LOCAL;
strncpy(sa.sun_path,path,sizeof(sa.sun_path)-1);
if (connect(c->fd, (struct sockaddr*)&sa, sizeof(sa)) == -1) {
if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
} else {
if (redisContextWaitReady(c,timeout) != REDIS_OK)
return REDIS_ERR;
}
}
if (timeout) {
if (c->timeout != timeout) {
if (c->timeout == NULL)
c->timeout = malloc(sizeof(struct timeval));
/* Reset socket to be blocking after connect(2). */
if (blocking && redisSetBlocking(c,1) != REDIS_OK)
return REDIS_ERR;
memcpy(c->timeout, timeout, sizeof(struct timeval));
}
}
else {
if (c->timeout)
free(c->timeout);
c->timeout = NULL;
}
if (redisContextTimeoutMsec(c, &timeout_msec) != REDIS_OK)
return REDIS_ERR;
sa.sun_family = AF_LOCAL;
strncpy(sa.sun_path, path, sizeof(sa.sun_path) - 1);
if (connect(c->fd, (struct sockaddr*)&sa, sizeof(sa)) == -1) {
if (errno == EINPROGRESS && !blocking) {
/* This is ok. */
}
else {
if (redisContextWaitReady(c, timeout_msec) != REDIS_OK)
return REDIS_ERR;
}
}
/* Reset socket to be blocking after connect(2). */
if (blocking && redisSetBlocking(c, 1) != REDIS_OK)
return REDIS_ERR;
c->flags |= REDIS_CONNECTED;
return REDIS_OK;
c->flags |= REDIS_CONNECTED;
return REDIS_OK;
}
#endif
#endif
deps/hiredis/net.h
Vendored
+7 -5
View File
@@ -1,9 +1,7 @@
/* Extracted from anet.c to work properly with Hiredis error reporting.
*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2014, Pieter Noordhuis <pcnoordhuis at gmail dot com>
* Copyright (c) 2015, Matt Stancliff <matt at genges dot com>,
* Jan-Erik Rediger <janerik at fnordig dot com>
* Copyright (c) 2006-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
@@ -37,7 +35,7 @@
#include "hiredis.h"
#if defined(__sun)
#if defined(__sun) || defined(_AIX)
#define AF_LOCAL AF_UNIX
#endif
@@ -50,4 +48,8 @@ int redisContextConnectBindTcp(redisContext *c, const char *addr, int port,
int redisContextConnectUnix(redisContext *c, const char *path, const struct timeval *timeout);
int redisKeepAlive(redisContext *c, int interval);
#ifdef _WIN32
int redisContextPreConnectTcp(redisContext *c, const char *addr, int port, struct timeval *timeout, SOCKADDR_STORAGE *ss);
#endif
#endif
deps/hiredis/read.c
Vendored
-525
View File
@@ -1,525 +0,0 @@
/*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#include "fmacros.h"
#include <string.h>
#include <stdlib.h>
#ifndef _MSC_VER
#include <unistd.h>
#endif
#include <assert.h>
#include <errno.h>
#include <ctype.h>
#include "read.h"
#include "sds.h"
static void __redisReaderSetError(redisReader *r, int type, const char *str) {
size_t len;
if (r->reply != NULL && r->fn && r->fn->freeObject) {
r->fn->freeObject(r->reply);
r->reply = NULL;
}
/* Clear input buffer on errors. */
if (r->buf != NULL) {
sdsfree(r->buf);
r->buf = NULL;
r->pos = r->len = 0;
}
/* Reset task stack. */
r->ridx = -1;
/* Set error. */
r->err = type;
len = strlen(str);
len = len < (sizeof(r->errstr)-1) ? len : (sizeof(r->errstr)-1);
memcpy(r->errstr,str,len);
r->errstr[len] = '\0';
}
static size_t chrtos(char *buf, size_t size, char byte) {
size_t len = 0;
switch(byte) {
case '\\':
case '"':
len = snprintf(buf,size,"\"\\%c\"",byte);
break;
case '\n': len = snprintf(buf,size,"\"\\n\""); break;
case '\r': len = snprintf(buf,size,"\"\\r\""); break;
case '\t': len = snprintf(buf,size,"\"\\t\""); break;
case '\a': len = snprintf(buf,size,"\"\\a\""); break;
case '\b': len = snprintf(buf,size,"\"\\b\""); break;
default:
if (isprint(byte))
len = snprintf(buf,size,"\"%c\"",byte);
else
len = snprintf(buf,size,"\"\\x%02x\"",(unsigned char)byte);
break;
}
return len;
}
static void __redisReaderSetErrorProtocolByte(redisReader *r, char byte) {
char cbuf[8], sbuf[128];
chrtos(cbuf,sizeof(cbuf),byte);
snprintf(sbuf,sizeof(sbuf),
"Protocol error, got %s as reply type byte", cbuf);
__redisReaderSetError(r,REDIS_ERR_PROTOCOL,sbuf);
}
static void __redisReaderSetErrorOOM(redisReader *r) {
__redisReaderSetError(r,REDIS_ERR_OOM,"Out of memory");
}
static char *readBytes(redisReader *r, unsigned int bytes) {
char *p;
if (r->len-r->pos >= bytes) {
p = r->buf+r->pos;
r->pos += bytes;
return p;
}
return NULL;
}
/* Find pointer to \r\n. */
static char *seekNewline(char *s, size_t len) {
int pos = 0;
int _len = len-1;
/* Position should be < len-1 because the character at "pos" should be
* followed by a \n. Note that strchr cannot be used because it doesn't
* allow to search a limited length and the buffer that is being searched
* might not have a trailing NULL character. */
while (pos < _len) {
while(pos < _len && s[pos] != '\r') pos++;
if (pos==_len) {
/* Not found. */
return NULL;
} else {
if (s[pos+1] == '\n') {
/* Found. */
return s+pos;
} else {
/* Continue searching. */
pos++;
}
}
}
return NULL;
}
/* Read a PORT_LONGLONG value starting at *s, under the assumption that it will be
* terminated by \r\n. Ambiguously returns -1 for unexpected input. */
static PORT_LONGLONG readLongLong(char *s) {
PORT_LONGLONG v = 0;
int dec, mult = 1;
char c;
if (*s == '-') {
mult = -1;
s++;
} else if (*s == '+') {
mult = 1;
s++;
}
while ((c = *(s++)) != '\r') {
dec = c - '0';
if (dec >= 0 && dec < 10) {
v *= 10;
v += dec;
} else {
/* Should not happen... */
return -1;
}
}
return mult*v;
}
static char *readLine(redisReader *r, int *_len) {
char *p, *s;
int len;
p = r->buf+r->pos;
s = seekNewline(p,(r->len-r->pos));
if (s != NULL) {
len = s-(r->buf+r->pos);
r->pos += len+2; /* skip \r\n */
if (_len) *_len = len;
return p;
}
return NULL;
}
static void moveToNextTask(redisReader *r) {
redisReadTask *cur, *prv;
while (r->ridx >= 0) {
/* Return a.s.a.p. when the stack is now empty. */
if (r->ridx == 0) {
r->ridx--;
return;
}
cur = &(r->rstack[r->ridx]);
prv = &(r->rstack[r->ridx-1]);
assert(prv->type == REDIS_REPLY_ARRAY);
if (cur->idx == prv->elements-1) {
r->ridx--;
} else {
/* Reset the type because the next item can be anything */
assert(cur->idx < prv->elements);
cur->type = -1;
cur->elements = -1;
cur->idx++;
return;
}
}
}
static int processLineItem(redisReader *r) {
redisReadTask *cur = &(r->rstack[r->ridx]);
void *obj;
char *p;
int len;
if ((p = readLine(r,&len)) != NULL) {
if (cur->type == REDIS_REPLY_INTEGER) {
if (r->fn && r->fn->createInteger)
obj = r->fn->createInteger(cur,readLongLong(p));
else
obj = (void*)REDIS_REPLY_INTEGER;
} else {
/* Type will be error or status. */
if (r->fn && r->fn->createString)
obj = r->fn->createString(cur,p,len);
else
obj = (void*)(size_t)(cur->type);
}
if (obj == NULL) {
__redisReaderSetErrorOOM(r);
return REDIS_ERR;
}
/* Set reply if this is the root object. */
if (r->ridx == 0) r->reply = obj;
moveToNextTask(r);
return REDIS_OK;
}
return REDIS_ERR;
}
static int processBulkItem(redisReader *r) {
redisReadTask *cur = &(r->rstack[r->ridx]);
void *obj = NULL;
char *p, *s;
PORT_LONG len;
PORT_ULONG bytelen;
int success = 0;
p = r->buf+r->pos;
s = seekNewline(p,r->len-r->pos);
if (s != NULL) {
p = r->buf+r->pos;
bytelen = s-(r->buf+r->pos)+2; /* include \r\n */
len = readLongLong(p);
if (len < 0) {
/* The nil object can always be created. */
if (r->fn && r->fn->createNil)
obj = r->fn->createNil(cur);
else
obj = (void*)REDIS_REPLY_NIL;
success = 1;
} else {
/* Only continue when the buffer contains the entire bulk item. */
bytelen += len+2; /* include \r\n */
if (r->pos+bytelen <= r->len) {
if (r->fn && r->fn->createString)
obj = r->fn->createString(cur,s+2,len);
else
obj = (void*)REDIS_REPLY_STRING;
success = 1;
}
}
/* Proceed when obj was created. */
if (success) {
if (obj == NULL) {
__redisReaderSetErrorOOM(r);
return REDIS_ERR;
}
r->pos += bytelen;
/* Set reply if this is the root object. */
if (r->ridx == 0) r->reply = obj;
moveToNextTask(r);
return REDIS_OK;
}
}
return REDIS_ERR;
}
static int processMultiBulkItem(redisReader *r) {
redisReadTask *cur = &(r->rstack[r->ridx]);
void *obj;
char *p;
PORT_LONG elements;
int root = 0;
/* 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");
return REDIS_ERR;
}
if ((p = readLine(r,NULL)) != NULL) {
elements = readLongLong(p);
root = (r->ridx == 0);
if (elements == -1) {
if (r->fn && r->fn->createNil)
obj = r->fn->createNil(cur);
else
obj = (void*)REDIS_REPLY_NIL;
if (obj == NULL) {
__redisReaderSetErrorOOM(r);
return REDIS_ERR;
}
moveToNextTask(r);
} else {
if (r->fn && r->fn->createArray)
obj = r->fn->createArray(cur,elements);
else
obj = (void*)REDIS_REPLY_ARRAY;
if (obj == NULL) {
__redisReaderSetErrorOOM(r);
return REDIS_ERR;
}
/* Modify task stack when there are more than 0 elements. */
if (elements > 0) {
cur->elements = elements;
cur->obj = obj;
r->ridx++;
r->rstack[r->ridx].type = -1;
r->rstack[r->ridx].elements = -1;
r->rstack[r->ridx].idx = 0;
r->rstack[r->ridx].obj = NULL;
r->rstack[r->ridx].parent = cur;
r->rstack[r->ridx].privdata = r->privdata;
} else {
moveToNextTask(r);
}
}
/* Set reply if this is the root object. */
if (root) r->reply = obj;
return REDIS_OK;
}
return REDIS_ERR;
}
static int processItem(redisReader *r) {
redisReadTask *cur = &(r->rstack[r->ridx]);
char *p;
/* check if we need to read type */
if (cur->type < 0) {
if ((p = readBytes(r,1)) != NULL) {
switch (p[0]) {
case '-':
cur->type = REDIS_REPLY_ERROR;
break;
case '+':
cur->type = REDIS_REPLY_STATUS;
break;
case ':':
cur->type = REDIS_REPLY_INTEGER;
break;
case '$':
cur->type = REDIS_REPLY_STRING;
break;
case '*':
cur->type = REDIS_REPLY_ARRAY;
break;
default:
__redisReaderSetErrorProtocolByte(r,*p);
return REDIS_ERR;
}
} else {
/* could not consume 1 byte */
return REDIS_ERR;
}
}
/* process typed item */
switch(cur->type) {
case REDIS_REPLY_ERROR:
case REDIS_REPLY_STATUS:
case REDIS_REPLY_INTEGER:
return processLineItem(r);
case REDIS_REPLY_STRING:
return processBulkItem(r);
case REDIS_REPLY_ARRAY:
return processMultiBulkItem(r);
default:
assert(NULL);
return REDIS_ERR; /* Avoid warning. */
}
}
redisReader *redisReaderCreateWithFunctions(redisReplyObjectFunctions *fn) {
redisReader *r;
r = calloc(sizeof(redisReader),1);
if (r == NULL)
return NULL;
r->err = 0;
r->errstr[0] = '\0';
r->fn = fn;
r->buf = sdsempty();
r->maxbuf = REDIS_READER_MAX_BUF;
if (r->buf == NULL) {
free(r);
return NULL;
}
r->ridx = -1;
return r;
}
void redisReaderFree(redisReader *r) {
if (r->reply != NULL && r->fn && r->fn->freeObject)
r->fn->freeObject(r->reply);
if (r->buf != NULL)
sdsfree(r->buf);
free(r);
}
int redisReaderFeed(redisReader *r, const char *buf, size_t len) {
sds newbuf;
/* Return early when this reader is in an erroneous state. */
if (r->err)
return REDIS_ERR;
/* Copy the provided buffer. */
if (buf != NULL && len >= 1) {
/* Destroy internal buffer when it is empty and is quite large. */
if (r->len == 0 && r->maxbuf != 0 && sdsavail(r->buf) > r->maxbuf) {
sdsfree(r->buf);
r->buf = sdsempty();
r->pos = 0;
/* r->buf should not be NULL since we just free'd a larger one. */
assert(r->buf != NULL);
}
newbuf = sdscatlen(r->buf,buf,len);
if (newbuf == NULL) {
__redisReaderSetErrorOOM(r);
return REDIS_ERR;
}
r->buf = newbuf;
r->len = sdslen(r->buf);
}
return REDIS_OK;
}
int redisReaderGetReply(redisReader *r, void **reply) {
/* Default target pointer to NULL. */
if (reply != NULL)
*reply = NULL;
/* Return early when this reader is in an erroneous state. */
if (r->err)
return REDIS_ERR;
/* When the buffer is empty, there will never be a reply. */
if (r->len == 0)
return REDIS_OK;
/* Set first item to process when the stack is empty. */
if (r->ridx == -1) {
r->rstack[0].type = -1;
r->rstack[0].elements = -1;
r->rstack[0].idx = -1;
r->rstack[0].obj = NULL;
r->rstack[0].parent = NULL;
r->rstack[0].privdata = r->privdata;
r->ridx = 0;
}
/* Process items in reply. */
while (r->ridx >= 0)
if (processItem(r) != REDIS_OK)
break;
/* Return ASAP when an error occurred. */
if (r->err)
return REDIS_ERR;
/* 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) {
sdsrange(r->buf,r->pos,-1);
r->pos = 0;
r->len = sdslen(r->buf);
}
/* Emit a reply when there is one. */
if (r->ridx == -1) {
if (reply != NULL)
*reply = r->reply;
r->reply = NULL;
}
return REDIS_OK;
}
deps/hiredis/read.h
Vendored
-114
View File
@@ -1,114 +0,0 @@
/*
* Copyright (c) 2009-2011, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2011, Pieter Noordhuis <pcnoordhuis at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Redis nor the names of its contributors may be used
* to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
* LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef __HIREDIS_READ_H
#define __HIREDIS_READ_H
#include <stdio.h> /* for size_t */
#ifdef _WIN32
#include "../../src/Win32_Interop/win32_types_hiredis.h"
#endif
#define REDIS_ERR -1
#define REDIS_OK 0
/* When an error occurs, the err flag in a context is set to hold the type of
* error that occurred. REDIS_ERR_IO means there was an I/O error and you
* should use the "errno" variable to find out what is wrong.
* For other values, the "errstr" field will hold a description. */
#define REDIS_ERR_IO 1 /* Error in read or write */
#define REDIS_ERR_EOF 3 /* End of file */
#define REDIS_ERR_PROTOCOL 4 /* Protocol error */
#define REDIS_ERR_OOM 5 /* Out of memory */
#define REDIS_ERR_OTHER 2 /* Everything else... */
#define REDIS_REPLY_STRING 1
#define REDIS_REPLY_ARRAY 2
#define REDIS_REPLY_INTEGER 3
#define REDIS_REPLY_NIL 4
#define REDIS_REPLY_STATUS 5
#define REDIS_REPLY_ERROR 6
#define REDIS_READER_MAX_BUF (1024*16) /* Default max unused reader buffer. */
#ifdef __cplusplus
extern "C" {
#endif
typedef struct redisReadTask {
int type;
int elements; /* number of elements in multibulk container */
int idx; /* index in parent (array) object */
void *obj; /* holds user-generated value for a read task */
struct redisReadTask *parent; /* parent task */
void *privdata; /* user-settable arbitrary field */
} redisReadTask;
typedef struct redisReplyObjectFunctions {
void *(*createString)(const redisReadTask*, char*, size_t);
void *(*createArray)(const redisReadTask*, int);
void *(*createInteger)(const redisReadTask*, PORT_LONGLONG);
void *(*createNil)(const redisReadTask*);
void (*freeObject)(void*);
} redisReplyObjectFunctions;
typedef struct redisReader {
int err; /* Error flags, 0 when there is no error */
char errstr[128]; /* String representation of error when applicable */
char *buf; /* Read buffer */
size_t pos; /* Buffer cursor */
size_t len; /* Buffer length */
size_t maxbuf; /* Max length of unused buffer */
redisReadTask rstack[9];
int ridx; /* Index of current read task */
void *reply; /* Temporary reply pointer */
redisReplyObjectFunctions *fn;
void *privdata;
} redisReader;
/* Public API for the protocol parser. */
redisReader *redisReaderCreateWithFunctions(redisReplyObjectFunctions *fn);
void redisReaderFree(redisReader *r);
int redisReaderFeed(redisReader *r, const char *buf, size_t len);
int redisReaderGetReply(redisReader *r, void **reply);
#define redisReaderSetPrivdata(_r, _p) (int)(((redisReader*)(_r))->privdata = (_p))
#define redisReaderGetObject(_r) (((redisReader*)(_r))->reply)
#define redisReaderGetError(_r) (((redisReader*)(_r))->errstr)
#ifdef __cplusplus
}
#endif
#endif
deps/hiredis/sds.c
Vendored
+663 -852
View File
File diff suppressed because it is too large Load Diff
deps/hiredis/sds.h
Vendored
+18 -234
View File
@@ -1,8 +1,6 @@
/* SDSLib 2.0 -- A C dynamic strings library
/* SDSLib, A C dynamic strings library
*
* Copyright (c) 2006-2015, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2015, Oran Agra
* Copyright (c) 2015, Redis Labs, Inc
* Copyright (c) 2006-2010, Salvatore Sanfilippo <antirez at gmail dot com>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
@@ -35,238 +33,38 @@
#define SDS_MAX_PREALLOC (1024*1024)
#ifdef _WIN32
#include "../../src/Win32_Interop/Win32_Portability.h"
#include "../../src/Win32_Interop/win32_types_hiredis.h"
#endif
#include <sys/types.h>
#include <stdarg.h>
#include <stdint.h>
#ifdef _WIN32
#define inline __inline
#endif
typedef char *sds;
#ifdef _WIN32
#define PACK( __Declaration__ ) __pragma( pack(push, 1) ) __Declaration__ __pragma( pack(pop) )
/* Note: sdshdr5 is never used, we just access the flags byte directly.
* However is here to document the layout of type 5 SDS strings. */
PACK(
struct sdshdr5 {
unsigned char flags; /* 3 lsb of type, and 5 msb of string length */
char buf[];
};)
PACK(
struct sdshdr8 {
uint8_t len; /* used */
uint8_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};)
PACK(
struct sdshdr16 {
uint16_t len; /* used */
uint16_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};)
PACK(
struct sdshdr32 {
uint32_t len; /* used */
uint32_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};)
PACK(
struct sdshdr64 {
uint64_t len; /* used */
uint64_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};)
#else
/* Note: sdshdr5 is never used, we just access the flags byte directly.
* However is here to document the layout of type 5 SDS strings. */
struct __attribute__((__packed__)) sdshdr5 {
unsigned char flags; /* 3 lsb of type, and 5 msb of string length */
char buf[];
struct sdshdr {
unsigned int len;
unsigned int free;
char buf[];
};
struct __attribute__((__packed__)) sdshdr8 {
uint8_t len; /* used */
uint8_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
struct __attribute__((__packed__)) sdshdr16 {
uint16_t len; /* used */
uint16_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
struct __attribute__((__packed__)) sdshdr32 {
uint32_t len; /* used */
uint32_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
struct __attribute__((__packed__)) sdshdr64 {
uint64_t len; /* used */
uint64_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
#endif
#define SDS_TYPE_5 0
#define SDS_TYPE_8 1
#define SDS_TYPE_16 2
#define SDS_TYPE_32 3
#define SDS_TYPE_64 4
#define SDS_TYPE_MASK 7
#define SDS_TYPE_BITS 3
#define SDS_HDR_VAR(T,s) struct sdshdr##T *sh = (struct sdshdr##T *)((s)-(sizeof(struct sdshdr##T)));
#define SDS_HDR(T,s) ((struct sdshdr##T *)((s)-(sizeof(struct sdshdr##T))))
#define SDS_TYPE_5_LEN(f) ((f)>>SDS_TYPE_BITS)
#ifdef _WIN32
#define inline __inline
#endif
static inline size_t sdslen(const sds s) {
unsigned char flags = s[-1];
switch (flags&SDS_TYPE_MASK) {
case SDS_TYPE_5:
return SDS_TYPE_5_LEN(flags);
case SDS_TYPE_8:
return SDS_HDR(8, s)->len;
case SDS_TYPE_16:
return SDS_HDR(16, s)->len;
case SDS_TYPE_32:
return SDS_HDR(32, s)->len;
case SDS_TYPE_64:
return SDS_HDR(64, s)->len;
}
return 0;
struct sdshdr *sh = (struct sdshdr *)(s-(sizeof(struct sdshdr)));
return sh->len;
}
static inline size_t sdsavail(const sds s) {
unsigned char flags = s[-1];
switch (flags&SDS_TYPE_MASK) {
case SDS_TYPE_5: {
return 0;
}
case SDS_TYPE_8: {
SDS_HDR_VAR(8, s);
return sh->alloc - sh->len;
}
case SDS_TYPE_16: {
SDS_HDR_VAR(16, s);
return sh->alloc - sh->len;
}
case SDS_TYPE_32: {
SDS_HDR_VAR(32, s);
return sh->alloc - sh->len;
}
case SDS_TYPE_64: {
SDS_HDR_VAR(64, s);
return sh->alloc - sh->len;
}
}
return 0;
}
static inline void sdssetlen(sds s, size_t newlen) {
unsigned char flags = s[-1];
switch (flags&SDS_TYPE_MASK) {
case SDS_TYPE_5:
{
unsigned char *fp = ((unsigned char*)s) - 1;
*fp = SDS_TYPE_5 | (newlen << SDS_TYPE_BITS);
}
break;
case SDS_TYPE_8:
SDS_HDR(8, s)->len = newlen;
break;
case SDS_TYPE_16:
SDS_HDR(16, s)->len = newlen;
break;
case SDS_TYPE_32:
SDS_HDR(32, s)->len = newlen;
break;
case SDS_TYPE_64:
SDS_HDR(64, s)->len = newlen;
break;
}
}
static inline void sdsinclen(sds s, size_t inc) {
unsigned char flags = s[-1];
switch (flags&SDS_TYPE_MASK) {
case SDS_TYPE_5:
{
unsigned char *fp = ((unsigned char*)s) - 1;
unsigned char newlen = SDS_TYPE_5_LEN(flags) + inc;
*fp = SDS_TYPE_5 | (newlen << SDS_TYPE_BITS);
}
break;
case SDS_TYPE_8:
SDS_HDR(8, s)->len += inc;
break;
case SDS_TYPE_16:
SDS_HDR(16, s)->len += inc;
break;
case SDS_TYPE_32:
SDS_HDR(32, s)->len += inc;
break;
case SDS_TYPE_64:
SDS_HDR(64, s)->len += inc;
break;
}
}
/* sdsalloc() = sdsavail() + sdslen() */
static inline size_t sdsalloc(const sds s) {
unsigned char flags = s[-1];
switch (flags&SDS_TYPE_MASK) {
case SDS_TYPE_5:
return SDS_TYPE_5_LEN(flags);
case SDS_TYPE_8:
return SDS_HDR(8, s)->alloc;
case SDS_TYPE_16:
return SDS_HDR(16, s)->alloc;
case SDS_TYPE_32:
return SDS_HDR(32, s)->alloc;
case SDS_TYPE_64:
return SDS_HDR(64, s)->alloc;
}
return 0;
}
static inline void sdssetalloc(sds s, size_t newlen) {
unsigned char flags = s[-1];
switch (flags&SDS_TYPE_MASK) {
case SDS_TYPE_5:
/* Nothing to do, this type has no total allocation info. */
break;
case SDS_TYPE_8:
SDS_HDR(8, s)->alloc = newlen;
break;
case SDS_TYPE_16:
SDS_HDR(16, s)->alloc = newlen;
break;
case SDS_TYPE_32:
SDS_HDR(32, s)->alloc = newlen;
break;
case SDS_TYPE_64:
SDS_HDR(64, s)->alloc = newlen;
break;
}
struct sdshdr *sh = (struct sdshdr *)(s-(sizeof(struct sdshdr)));
return sh->free;
}
sds sdsnewlen(const void *init, size_t initlen);
sds sdsnew(const char *init);
sds sdsempty(void);
size_t sdslen(const sds s);
sds sdsdup(const sds s);
void sdsfree(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);
@@ -277,7 +75,7 @@ sds sdscpy(sds s, const char *t);
sds sdscatvprintf(sds s, const char *fmt, va_list ap);
#ifdef __GNUC__
sds sdscatprintf(sds s, const char *fmt, ...)
__attribute__((format(printf, 2, 3)));
__attribute__((format(printf, 2, 3)));
#else
sds sdscatprintf(sds s, const char *fmt, ...);
#endif
@@ -292,30 +90,16 @@ 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(PORT_LONGLONG value);
sds sdsfromlonglong(long long value);
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);
sds sdsjoinsds(sds *argv, int argc, const char *sep, size_t seplen);
/* 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);
void *sdsAllocPtr(sds s);
/* Export the allocator used by SDS to the program using SDS.
* Sometimes the program SDS is linked to, may use a different set of
* allocators, but may want to allocate or free things that SDS will
* respectively free or allocate. */
void *sds_malloc(size_t size);
void *sds_realloc(void *ptr, size_t size);
void sds_free(void *ptr);
#ifdef REDIS_TEST
int sdsTest(int argc, char *argv[]);
#endif
#endif
deps/hiredis/sdsalloc.h
Vendored
-42
View File
@@ -1,42 +0,0 @@
/* SDSLib 2.0 -- A C dynamic strings library
*
* Copyright (c) 2006-2015, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2015, Oran Agra
* Copyright (c) 2015, Redis Labs, 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 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.
*/
/* SDS allocator selection.
*
* This file is used in order to change the SDS allocator at compile time.
* Just define the following defines to what you want to use. Also add
* the include of your alternate allocator if needed (not needed in order
* to use the default libc allocator). */
#define s_malloc malloc
#define s_realloc realloc
#define s_free free
deps/hiredis/test.c
Vendored
+29 -136
View File
@@ -2,30 +2,16 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#ifndef _WIN32
#include <strings.h>
#include <sys/time.h>
#else
#include <WinSock2.h>
#include "../../src/Win32_Interop/Win32_Time.h"
#endif
#include <assert.h>
#ifndef _WIN32
#include <unistd.h>
#endif
#include <signal.h>
#include <errno.h>
#include <limits.h>
#include "hiredis.h"
#ifdef _WIN32
#define strcasecmp _stricmp
#define strncasecmp _strnicmp
#define SIGPIPE 13
#endif
enum connection_type {
CONN_TCP,
CONN_UNIX,
@@ -43,7 +29,7 @@ struct config {
struct {
const char *path;
} unix_sock;
} unix;
};
/* The following lines make up our testing "framework" :) */
@@ -51,19 +37,12 @@ static int tests = 0, fails = 0;
#define test(_s) { printf("#%02d ", ++tests); printf(_s); }
#define test_cond(_c) if(_c) printf("\033[0;32mPASSED\033[0;0m\n"); else {printf("\033[0;31mFAILED\033[0;0m\n"); fails++;}
static PORT_LONGLONG usec(void) {
static long long usec(void) {
struct timeval tv;
gettimeofday(&tv,NULL);
return (((PORT_LONGLONG)tv.tv_sec)*1000000)+tv.tv_usec;
return (((long long)tv.tv_sec)*1000000)+tv.tv_usec;
}
/* The assert() calls below have side effects, so we need assert()
* even if we are compiling without asserts (-DNDEBUG). */
#ifdef NDEBUG
#undef assert
#define assert(e) (void)(e)
#endif
static redisContext *select_database(redisContext *c) {
redisReply *reply;
@@ -104,20 +83,16 @@ static int disconnect(redisContext *c, int keep_fd) {
return -1;
}
#ifdef _WIN32
static redisContext *_connect(struct config config) {
#else
static redisContext *connect(struct config config) {
#endif
redisContext *c = NULL;
if (config.type == CONN_TCP) {
c = redisConnect(config.tcp.host, config.tcp.port);
} else if (config.type == CONN_UNIX) {
c = redisConnectUnix(config.unix_sock.path);
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_sock.path);
redisContext *dummy_ctx = redisConnectUnix(config.unix.path);
if (dummy_ctx) {
int fd = disconnect(dummy_ctx, 1);
printf("Connecting to inherited fd %d\n", fd);
@@ -132,7 +107,6 @@ static redisContext *connect(struct config config) {
exit(1);
} else if (c->err) {
printf("Connection error: %s\n", c->errstr);
redisFree(c);
exit(1);
}
@@ -209,13 +183,13 @@ static void test_format_commands(void) {
INTEGER_WIDTH_TEST("d", int);
INTEGER_WIDTH_TEST("hhd", char);
INTEGER_WIDTH_TEST("hd", short);
INTEGER_WIDTH_TEST("ld", PORT_LONG);
INTEGER_WIDTH_TEST("lld", PORT_LONGLONG);
INTEGER_WIDTH_TEST("ld", long);
INTEGER_WIDTH_TEST("lld", long long);
INTEGER_WIDTH_TEST("u", unsigned int);
INTEGER_WIDTH_TEST("hhu", unsigned char);
INTEGER_WIDTH_TEST("hu", unsigned short);
INTEGER_WIDTH_TEST("lu", PORT_ULONG);
INTEGER_WIDTH_TEST("llu", PORT_ULONGLONG);
INTEGER_WIDTH_TEST("lu", unsigned long);
INTEGER_WIDTH_TEST("llu", unsigned long long);
FLOAT_WIDTH_TEST(float);
FLOAT_WIDTH_TEST(double);
@@ -241,22 +215,6 @@ static void test_format_commands(void) {
test_cond(strncmp(cmd,"*3\r\n$3\r\nSET\r\n$7\r\nfoo\0xxx\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(7+2)+4+(3+2));
free(cmd);
sds sds_cmd;
sds_cmd = sdsempty();
test("Format command into sds by passing argc/argv without lengths: ");
len = redisFormatSdsCommandArgv(&sds_cmd,argc,argv,NULL);
test_cond(strncmp(sds_cmd,"*3\r\n$3\r\nSET\r\n$3\r\nfoo\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(3+2)+4+(3+2));
sdsfree(sds_cmd);
sds_cmd = sdsempty();
test("Format command into sds by passing argc/argv with lengths: ");
len = redisFormatSdsCommandArgv(&sds_cmd,argc,argv,lens);
test_cond(strncmp(sds_cmd,"*3\r\n$3\r\nSET\r\n$7\r\nfoo\0xxx\r\n$3\r\nbar\r\n",len) == 0 &&
len == 4+4+(3+2)+4+(7+2)+4+(3+2));
sdsfree(sds_cmd);
}
static void test_append_formatted_commands(struct config config) {
@@ -265,7 +223,7 @@ static void test_append_formatted_commands(struct config config) {
char *cmd;
int len;
c = IF_WIN32(_connect,connect)(config);
c = connect(config);
test("Append format command: ");
@@ -360,31 +318,16 @@ static void test_reply_reader(void) {
redisReaderFree(reader);
}
static void test_free_null(void) {
void *redisCtx = NULL;
void *reply = NULL;
test("Don't fail when redisFree is passed a NULL value: ");
redisFree(redisCtx);
test_cond(redisCtx == NULL);
test("Don't fail when freeReplyObject is passed a NULL value: ");
freeReplyObject(reply);
test_cond(reply == NULL);
}
static void test_blocking_connection_errors(void) {
redisContext *c;
test("Returns error when host cannot be resolved: ");
c = redisConnect((char*)"idontexist.test", 6379);
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.test") == 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,"Temporary failure in name resolution") == 0 ||
strcmp(c->errstr,"hostname nor servname provided, or not known") == 0 ||
strcmp(c->errstr,"no address associated with name") == 0));
redisFree(c);
@@ -394,7 +337,7 @@ static void test_blocking_connection_errors(void) {
strcmp(c->errstr,"Connection refused") == 0);
redisFree(c);
test("Returns error when the unix_sock socket path doesn't accept connections: ");
test("Returns error when the unix socket path doesn't accept connections: ");
c = redisConnectUnix((char*)"/tmp/idontexist.sock");
test_cond(c->err == REDIS_ERR_IO); /* Don't care about the message... */
redisFree(c);
@@ -404,7 +347,7 @@ static void test_blocking_connection(struct config config) {
redisContext *c;
redisReply *reply;
c = IF_WIN32(_connect,connect)(config);
c = connect(config);
test("Is able to deliver commands: ");
reply = redisCommand(c,"PING");
@@ -478,52 +421,6 @@ static void test_blocking_connection(struct config config) {
disconnect(c, 0);
}
static void test_blocking_connection_timeouts(struct config config) {
redisContext *c;
redisReply *reply;
ssize_t s;
const char *cmd = "DEBUG SLEEP 3\r\n";
struct timeval tv;
c = IF_WIN32(_connect,connect)(config);
test("Successfully completes a command when the timeout is not exceeded: ");
reply = redisCommand(c,"SET foo fast");
freeReplyObject(reply);
tv.tv_sec = 0;
tv.tv_usec = 10000;
redisSetTimeout(c, tv);
reply = redisCommand(c, "GET foo");
test_cond(reply != NULL && reply->type == REDIS_REPLY_STRING && memcmp(reply->str, "fast", 4) == 0);
freeReplyObject(reply);
disconnect(c, 0);
c = IF_WIN32(_connect,connect)(config);
test("Does not return a reply when the command times out: ");
s = write(c->fd, cmd, strlen(cmd));
tv.tv_sec = 0;
tv.tv_usec = 10000;
redisSetTimeout(c, tv);
reply = redisCommand(c, "GET foo");
test_cond(s > 0 && reply == NULL && c->err == REDIS_ERR_IO && strcmp(c->errstr, "Resource temporarily unavailable") == 0);
freeReplyObject(reply);
test("Reconnect properly reconnects after a timeout: ");
redisReconnect(c);
reply = redisCommand(c, "PING");
test_cond(reply != NULL && reply->type == REDIS_REPLY_STATUS && strcmp(reply->str, "PONG") == 0);
freeReplyObject(reply);
test("Reconnect properly uses owned parameters: ");
config.tcp.host = "foo";
config.unix_sock.path = "foo";
redisReconnect(c);
reply = redisCommand(c, "PING");
test_cond(reply != NULL && reply->type == REDIS_REPLY_STATUS && strcmp(reply->str, "PONG") == 0);
freeReplyObject(reply);
disconnect(c, 0);
}
static void test_blocking_io_errors(struct config config) {
redisContext *c;
redisReply *reply;
@@ -531,7 +428,7 @@ static void test_blocking_io_errors(struct config config) {
int major, minor;
/* Connect to target given by config. */
c = IF_WIN32(_connect,connect)(config);
c = connect(config);
{
/* Find out Redis version to determine the path for the next test */
const char *field = "redis_version:";
@@ -547,7 +444,7 @@ static void test_blocking_io_errors(struct config config) {
test("Returns I/O error when the connection is lost: ");
reply = redisCommand(c,"QUIT");
if (major > 2 || (major == 2 && minor > 0)) {
if (major >= 2 && minor > 0) {
/* > 2.0 returns OK on QUIT and read() should be issued once more
* to know the descriptor is at EOF. */
test_cond(strcasecmp(reply->str,"OK") == 0 &&
@@ -566,7 +463,7 @@ static void test_blocking_io_errors(struct config config) {
strcmp(c->errstr,"Server closed the connection") == 0);
redisFree(c);
c = IF_WIN32(_connect,connect)(config);
c = connect(config);
test("Returns I/O error on socket timeout: ");
struct timeval tv = { 0, 1000 };
assert(redisSetTimeout(c,tv) == REDIS_OK);
@@ -585,8 +482,7 @@ static void test_invalid_timeout_errors(struct config config) {
c = redisConnectWithTimeout(config.tcp.host, config.tcp.port, config.tcp.timeout);
test_cond(c->err == REDIS_ERR_IO && strcmp(c->errstr, "Invalid timeout specified") == 0);
redisFree(c);
test_cond(c->err == REDIS_ERR_IO);
test("Set error when an invalid timeout sec value is given to redisConnectWithTimeout: ");
@@ -595,15 +491,16 @@ static void test_invalid_timeout_errors(struct config config) {
c = redisConnectWithTimeout(config.tcp.host, config.tcp.port, config.tcp.timeout);
test_cond(c->err == REDIS_ERR_IO && strcmp(c->errstr, "Invalid timeout specified") == 0);
test_cond(c->err == REDIS_ERR_IO);
redisFree(c);
}
static void test_throughput(struct config config) {
redisContext *c = IF_WIN32(_connect,connect)(config);
redisContext *c = connect(config);
redisReply **replies;
int i, num;
PORT_LONGLONG t1, t2;
long long t1, t2;
test("Throughput:\n");
for (i = 0; i < 500; i++)
@@ -664,19 +561,19 @@ static void test_throughput(struct config config) {
disconnect(c, 0);
}
// static PORT_LONG __test_callback_flags = 0;
// static long __test_callback_flags = 0;
// static void __test_callback(redisContext *c, void *privdata) {
// ((void)c);
// /* Shift to detect execution order */
// __test_callback_flags <<= 8;
// __test_callback_flags |= (PORT_LONG)privdata;
// __test_callback_flags |= (long)privdata;
// }
//
// static void __test_reply_callback(redisContext *c, redisReply *reply, void *privdata) {
// ((void)c);
// /* Shift to detect execution order */
// __test_callback_flags <<= 8;
// __test_callback_flags |= (PORT_LONG)privdata;
// __test_callback_flags |= (long)privdata;
// if (reply) freeReplyObject(reply);
// }
//
@@ -769,7 +666,7 @@ int main(int argc, char **argv) {
.host = "127.0.0.1",
.port = 6379
},
.unix_sock = {
.unix = {
.path = "/tmp/redis.sock"
}
};
@@ -790,7 +687,7 @@ int main(int argc, char **argv) {
cfg.tcp.port = atoi(argv[0]);
} else if (argc >= 2 && !strcmp(argv[0],"-s")) {
argv++; argc--;
cfg.unix_sock.path = argv[0];
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")) {
@@ -805,31 +702,27 @@ int main(int argc, char **argv) {
test_format_commands();
test_reply_reader();
test_blocking_connection_errors();
test_free_null();
printf("\nTesting against TCP connection (%s:%d):\n", cfg.tcp.host, cfg.tcp.port);
cfg.type = CONN_TCP;
test_blocking_connection(cfg);
test_blocking_connection_timeouts(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_sock.path);
printf("\nTesting against Unix socket connection (%s):\n", cfg.unix.path);
cfg.type = CONN_UNIX;
test_blocking_connection(cfg);
test_blocking_connection_timeouts(cfg);
test_blocking_io_errors(cfg);
if (throughput) test_throughput(cfg);
if (test_inherit_fd) {
printf("\nTesting against inherited fd (%s):\n", cfg.unix_sock.path);
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/win32_hiredis.h
Vendored
-73
View File
@@ -1,73 +0,0 @@
/*
* Copyright (c), Microsoft Open Technologies, 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.
* 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 HOLDER 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 WIN32_HIREDIS_H
#define WIN32_HIREDIS_H
#include "../../src/Win32_Interop/Win32_Portability.h"
#include "../../src/Win32_Interop/Win32_types_hiredis.h"
#include "../../src/Win32_Interop/Win32_Error.h"
#include "../../src/Win32_Interop/Win32_FDAPI.h"
#define INCL_WINSOCK_API_PROTOTYPES 0 // Important! Do not include Winsock API definitions to avoid conflicts with API entry points defined below.
#include <WinSock2.h> // For SOCKADDR_STORAGE
#include "hiredis.h"
#ifndef snprintf
#define snprintf c99_snprintf
__inline int c99_vsnprintf(char* str, size_t size, const char* format, va_list ap)
{
int count = -1;
if (size != 0)
count = _vsnprintf_s(str, size, _TRUNCATE, format, ap);
if (count == -1)
count = _vscprintf(format, ap);
return count;
}
__inline int c99_snprintf(char* str, size_t size, const char* format, ...)
{
int count;
va_list ap;
va_start(ap, format);
count = c99_vsnprintf(str, size, format, ap);
va_end(ap);
return count;
}
#endif
#ifndef va_copy
#define va_copy(d,s) ((d) = (s))
#endif
redisContext *redisPreConnectNonBlock(const char *ip, int port, SOCKADDR_STORAGE *sa);
int redisBufferReadDone(redisContext *c, char *buf, ssize_t nread);
int redisBufferWriteDone(redisContext *c, int nwritten, int *done);
int redisContextPreConnectTcp(redisContext *c, const char *addr, int port, struct timeval *timeout, SOCKADDR_STORAGE *ss);
#endif
deps/jemalloc-5.2.1/.appveyor.yml
Vendored
-42
View File
@@ -1,42 +0,0 @@
version: '{build}'
environment:
matrix:
- MSYSTEM: MINGW64
CPU: x86_64
MSVC: amd64
CONFIG_FLAGS: --enable-debug
- MSYSTEM: MINGW64
CPU: x86_64
CONFIG_FLAGS: --enable-debug
- MSYSTEM: MINGW32
CPU: i686
MSVC: x86
CONFIG_FLAGS: --enable-debug
- MSYSTEM: MINGW32
CPU: i686
CONFIG_FLAGS: --enable-debug
- MSYSTEM: MINGW64
CPU: x86_64
MSVC: amd64
- MSYSTEM: MINGW64
CPU: x86_64
- MSYSTEM: MINGW32
CPU: i686
MSVC: x86
- MSYSTEM: MINGW32
CPU: i686
install:
- set PATH=c:\msys64\%MSYSTEM%\bin;c:\msys64\usr\bin;%PATH%
- if defined MSVC call "c:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat" %MSVC%
- if defined MSVC pacman --noconfirm -Rsc mingw-w64-%CPU%-gcc gcc
- pacman --noconfirm -Suy mingw-w64-%CPU%-make
build_script:
- bash -c "autoconf"
- bash -c "./configure $CONFIG_FLAGS"
- mingw32-make
- file lib/jemalloc.dll
- mingw32-make tests
- mingw32-make -k check
deps/jemalloc-5.2.1/.autom4te.cfg
Vendored
-3
View File
@@ -1,3 +0,0 @@
begin-language: "Autoconf-without-aclocal-m4"
args: --no-cache
end-language: "Autoconf-without-aclocal-m4"
deps/jemalloc-5.2.1/.cirrus.yml
Vendored
-21
View File
@@ -1,21 +0,0 @@
env:
CIRRUS_CLONE_DEPTH: 1
ARCH: amd64
task:
freebsd_instance:
matrix:
image: freebsd-12-0-release-amd64
image: freebsd-11-2-release-amd64
install_script:
- sed -i.bak -e 's,pkg+http://pkg.FreeBSD.org/\${ABI}/quarterly,pkg+http://pkg.FreeBSD.org/\${ABI}/latest,' /etc/pkg/FreeBSD.conf
- pkg upgrade -y
- pkg install -y autoconf gmake
script:
- autoconf
#- ./configure ${COMPILER_FLAGS:+ CC="$CC $COMPILER_FLAGS" CXX="$CXX $COMPILER_FLAGS" } $CONFIGURE_FLAGS
- ./configure
- export JFLAG=`sysctl -n kern.smp.cpus`
- gmake -j${JFLAG}
- gmake -j${JFLAG} tests
- gmake check
deps/jemalloc-5.2.1/.gitattributes
-1
View File
@@ -1 +0,0 @@
* text=auto eol=lf
deps/jemalloc-5.2.1/.gitignore
-71
View File
@@ -1,71 +0,0 @@
/bin/jemalloc-config
/bin/jemalloc.sh
/bin/jeprof
/config.stamp
/config.log
/config.status
/configure
/doc/html.xsl
/doc/manpages.xsl
/doc/jemalloc.xml
/doc/jemalloc.html
/doc/jemalloc.3
/jemalloc.pc
/lib/
/Makefile
/src/*.[od]
/src/*.sym
/run_tests.out/
/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/*.out
/test/integration/cpp/[A-Za-z]*
!/test/integration/cpp/[A-Za-z]*.*
/test/integration/cpp/*.[od]
/test/integration/cpp/*.out
/test/src/*.[od]
/test/stress/[A-Za-z]*
!/test/stress/[A-Za-z]*.*
/test/stress/*.[od]
/test/stress/*.out
/test/unit/[A-Za-z]*
!/test/unit/[A-Za-z]*.*
/test/unit/*.[od]
/test/unit/*.out
/VERSION
*.pdb
*.sdf
*.opendb
*.VC.db
*.opensdf
*.cachefile
*.suo
*.user
*.sln.docstates
*.tmp
.vs/
/msvc/Win32/
/msvc/x64/
/msvc/projects/*/*/Debug*/
/msvc/projects/*/*/Release*/
/msvc/projects/*/*/Win32/
/msvc/projects/*/*/x64/
deps/jemalloc-5.2.1/.travis.yml
Vendored
-195
View File
@@ -1,195 +0,0 @@
language: generic
dist: precise
matrix:
include:
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: &gcc_multilib
apt:
packages:
- gcc-multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: osx
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=clang CXX=clang++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--enable-debug" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="-m32" CONFIGURE_FLAGS="--with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
addons: *gcc_multilib
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --disable-stats" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-prof --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --disable-libdl" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-stats --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl --enable-opt-safety-checks" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--disable-libdl --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks --with-malloc-conf=tcache:false" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks --with-malloc-conf=dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks --with-malloc-conf=percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-opt-safety-checks --with-malloc-conf=background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false,dss:primary" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false,percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=tcache:false,background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary,percpu_arena:percpu" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=dss:primary,background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--with-malloc-conf=percpu_arena:percpu,background_thread:true" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
# Development build
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --disable-cache-oblivious --enable-stats --enable-log --enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
# --enable-expermental-smallocx:
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="--enable-debug --enable-experimental-smallocx --enable-stats --enable-prof" EXTRA_CFLAGS="-Werror -Wno-array-bounds"
# Valgrind
- os: linux
env: CC=gcc CXX=g++ COMPILER_FLAGS="" CONFIGURE_FLAGS="" EXTRA_CFLAGS="-Werror -Wno-array-bounds" JEMALLOC_TEST_PREFIX="valgrind"
addons:
apt:
packages:
- valgrind
before_script:
- autoconf
- scripts/gen_travis.py > travis_script && diff .travis.yml travis_script
- ./configure ${COMPILER_FLAGS:+ CC="$CC $COMPILER_FLAGS" CXX="$CXX $COMPILER_FLAGS" } $CONFIGURE_FLAGS
- make -j3
- make -j3 tests
script:
- make check
deps/jemalloc-5.2.1/ChangeLog
Vendored
-1518
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc-5.2.1/INSTALL.md
Vendored
-421
View File
@@ -1,421 +0,0 @@
Building and installing a packaged release of jemalloc can be as simple as
typing the following while in the root directory of the source tree:
./configure
make
make install
If building from unpackaged developer sources, the simplest command sequence
that might work is:
./autogen.sh
make dist
make
make install
Note that documentation is not built by the default target because doing so
would create a dependency on xsltproc in packaged releases, hence the
requirement to either run 'make dist' or avoid installing docs via the various
install_* targets documented below.
## Advanced configuration
The 'configure' script supports numerous options that allow control of which
functionality is enabled, where jemalloc is installed, etc. Optionally, pass
any of the following arguments (not a definitive list) to 'configure':
* `--help`
Print a definitive list of options.
* `--prefix=<install-root-dir>`
Set the base directory in which to install. For example:
./configure --prefix=/usr/local
will cause files to be installed into /usr/local/include, /usr/local/lib,
and /usr/local/man.
* `--with-version=(<major>.<minor>.<bugfix>-<nrev>-g<gid>|VERSION)`
The VERSION file is mandatory for successful configuration, and the
following steps are taken to assure its presence:
1) If --with-version=<major>.<minor>.<bugfix>-<nrev>-g<gid> is specified,
generate VERSION using the specified value.
2) If --with-version is not specified in either form and the source
directory is inside a git repository, try to generate VERSION via 'git
describe' invocations that pattern-match release tags.
3) If VERSION is missing, generate it with a bogus version:
0.0.0-0-g0000000000000000000000000000000000000000
Note that --with-version=VERSION bypasses (1) and (2), which simplifies
VERSION configuration when embedding a jemalloc release into another
project's git repository.
* `--with-rpath=<colon-separated-rpath>`
Embed one or more library paths, so that libjemalloc can find the libraries
it is linked to. This works only on ELF-based systems.
* `--with-mangling=<map>`
Mangle public symbols specified in <map> which is a comma-separated list of
name:mangled pairs.
For example, to use ld's --wrap option as an alternative method for
overriding libc's malloc implementation, specify something like:
--with-mangling=malloc:__wrap_malloc,free:__wrap_free[...]
Note that mangling happens prior to application of the prefix specified by
--with-jemalloc-prefix, and mangled symbols are then ignored when applying
the prefix.
* `--with-jemalloc-prefix=<prefix>`
Prefix all public APIs with <prefix>. For example, if <prefix> is
"prefix_", API changes like the following occur:
malloc() --> prefix_malloc()
malloc_conf --> prefix_malloc_conf
/etc/malloc.conf --> /etc/prefix_malloc.conf
MALLOC_CONF --> PREFIX_MALLOC_CONF
This makes it possible to use jemalloc at the same time as the system
allocator, or even to use multiple copies of jemalloc simultaneously.
By default, the prefix is "", except on OS X, where it is "je_". On OS X,
jemalloc overlays the default malloc zone, but makes no attempt to actually
replace the "malloc", "calloc", etc. symbols.
* `--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>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, <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
versions of jemalloc can coexist in the same installation directory. For
example, libjemalloc.so.0 becomes libjemalloc<suffix>.so.0.
* `--with-malloc-conf=<malloc_conf>`
Embed `<malloc_conf>` as a run-time options string that is processed prior to
the malloc_conf global variable, the /etc/malloc.conf symlink, and the
MALLOC_CONF environment variable. For example, to change the default decay
time to 30 seconds:
--with-malloc-conf=decay_ms:30000
* `--enable-debug`
Enable assertions and validation code. This incurs a substantial
performance hit, but is very useful during application development.
* `--disable-stats`
Disable statistics gathering functionality. See the "opt.stats_print"
option documentation for usage details.
* `--enable-prof`
Enable heap profiling and leak detection functionality. See the "opt.prof"
option documentation for usage details. When enabled, there are several
approaches to backtracing, and the configure script chooses the first one
in the following list that appears to function correctly:
+ libunwind (requires --enable-prof-libunwind)
+ libgcc (unless --disable-prof-libgcc)
+ gcc intrinsics (unless --disable-prof-gcc)
* `--enable-prof-libunwind`
Use the libunwind library (http://www.nongnu.org/libunwind/) for stack
backtracing.
* `--disable-prof-libgcc`
Disable the use of libgcc's backtracing functionality.
* `--disable-prof-gcc`
Disable the use of gcc intrinsics for backtracing.
* `--with-static-libunwind=<libunwind.a>`
Statically link against the specified libunwind.a rather than dynamically
linking with -lunwind.
* `--disable-fill`
Disable support for junk/zero filling of memory. See the "opt.junk" and
"opt.zero" option documentation for usage details.
* `--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-lazy-lock`
Enable code that wraps pthread_create() to detect when an application
switches from single-threaded to multi-threaded mode, so that it can avoid
mutex locking/unlocking operations while in single-threaded mode. In
practice, this feature usually has little impact on performance unless
thread-specific caching is disabled.
* `--disable-cache-oblivious`
Disable cache-oblivious large allocation alignment for large allocation
requests with no alignment constraints. If this feature is disabled, all
large allocations are page-aligned as an implementation artifact, which can
severely harm CPU cache utilization. However, the cache-oblivious layout
comes at the cost of one extra page per large allocation, which in the
most extreme case increases physical memory usage for the 16 KiB size class
to 20 KiB.
* `--disable-syscall`
Disable use of syscall(2) rather than {open,read,write,close}(2). This is
intended as a workaround for systems that place security limitations on
syscall(2).
* `--disable-cxx`
Disable C++ integration. This will cause new and delete operator
implementations to be omitted.
* `--with-xslroot=<path>`
Specify where to find DocBook XSL stylesheets when building the
documentation.
* `--with-lg-page=<lg-page>`
Specify the base 2 log of the allocator page size, which must in turn be at
least as large as the system page size. By default the configure script
determines the host's page size and sets the allocator page size equal to
the system page size, so this option need not be specified unless the
system page size may change between configuration and execution, e.g. when
cross compiling.
* `--with-lg-hugepage=<lg-hugepage>`
Specify the base 2 log of the system huge page size. This option is useful
when cross compiling, or when overriding the default for systems that do
not explicitly support huge pages.
* `--with-lg-quantum=<lg-quantum>`
Specify the base 2 log of the minimum allocation alignment. jemalloc needs
to know the minimum alignment that meets the following C standard
requirement (quoted from the April 12, 2011 draft of the C11 standard):
> The pointer returned if the allocation succeeds is suitably aligned so
that it may be assigned to a pointer to any type of object with a
fundamental alignment requirement and then used to access such an object
or an array of such objects in the space allocated [...]
This setting is architecture-specific, and although jemalloc includes known
safe values for the most commonly used modern architectures, there is a
wrinkle related to GNU libc (glibc) that may impact your choice of
<lg-quantum>. On most modern architectures, this mandates 16-byte
alignment (<lg-quantum>=4), but the glibc developers chose not to meet this
requirement for performance reasons. An old discussion can be found at
<https://sourceware.org/bugzilla/show_bug.cgi?id=206> . Unlike glibc,
jemalloc does follow the C standard by default (caveat: jemalloc
technically cheats for size classes smaller than the quantum), but the fact
that Linux systems already work around this allocator noncompliance means
that it is generally safe in practice to let jemalloc's minimum alignment
follow glibc's lead. If you specify `--with-lg-quantum=3` during
configuration, jemalloc will provide additional size classes that are not
16-byte-aligned (24, 40, and 56).
* `--with-lg-vaddr=<lg-vaddr>`
Specify the number of significant virtual address bits. By default, the
configure script attempts to detect virtual address size on those platforms
where it knows how, and picks a default otherwise. This option may be
useful when cross-compiling.
* `--disable-initial-exec-tls`
Disable the initial-exec TLS model for jemalloc's internal thread-local
storage (on those platforms that support explicit settings). This can allow
jemalloc to be dynamically loaded after program startup (e.g. using dlopen).
Note that in this case, there will be two malloc implementations operating
in the same process, which will almost certainly result in confusing runtime
crashes if pointers leak from one implementation to the other.
* `--disable-libdl`
Disable the usage of libdl, namely dlsym(3) which is required by the lazy
lock option. This can allow building static binaries.
The following environment variables (not a definitive list) impact configure's
behavior:
* `CFLAGS="?"`
* `CXXFLAGS="?"`
Pass these flags to the C/C++ compiler. Any flags set by the configure
script are prepended, which means explicitly set flags generally take
precedence. Take care when specifying flags such as -Werror, because
configure tests may be affected in undesirable ways.
* `EXTRA_CFLAGS="?"`
* `EXTRA_CXXFLAGS="?"`
Append these flags to CFLAGS/CXXFLAGS, without passing them to the
compiler(s) during configuration. This makes it possible to add flags such
as -Werror, while allowing the configure script to determine what other
flags are appropriate for the specified configuration.
* `CPPFLAGS="?"`
Pass these flags to the C preprocessor. Note that CFLAGS is not passed to
'cpp' when 'configure' is looking for include files, so you must use
CPPFLAGS instead if you need to help 'configure' find header files.
* `LD_LIBRARY_PATH="?"`
'ld' uses this colon-separated list to find libraries.
* `LDFLAGS="?"`
Pass these flags when linking.
* `PATH="?"`
'configure' uses this to find programs.
In some cases it may be necessary to work around configuration results that do
not match reality. For example, Linux 4.5 added support for the MADV_FREE flag
to madvise(2), which can cause problems if building on a host with MADV_FREE
support and deploying to a target without. To work around this, use a cache
file to override the relevant configuration variable defined in configure.ac,
e.g.:
echo "je_cv_madv_free=no" > config.cache && ./configure -C
## 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_pc
install_lib
install_doc_html
install_doc_man
install_doc
To clean up build results to varying degrees, use the following make targets:
clean
distclean
relclean
## Advanced installation
Optionally, define make variables when invoking make, including (not
exclusively):
* `INCLUDEDIR="?"`
Use this as the installation prefix for header files.
* `LIBDIR="?"`
Use this as the installation prefix for libraries.
* `MANDIR="?"`
Use this as the installation prefix for man pages.
* `DESTDIR="?"`
Prepend DESTDIR to INCLUDEDIR, LIBDIR, DATADIR, and MANDIR. This is useful
when installing to a different path than was specified via --prefix.
* `CC="?"`
Use this to invoke the C compiler.
* `CFLAGS="?"`
Pass these flags to the compiler.
* `CPPFLAGS="?"`
Pass these flags to the C preprocessor.
* `LDFLAGS="?"`
Pass these flags when linking.
* `PATH="?"`
Use this to search for programs used during configuration and building.
## Development
If you intend to make non-trivial changes to jemalloc, use the 'autogen.sh'
script rather than 'configure'. This re-generates 'configure', enables
configuration dependency rules, and enables re-generation of automatically
generated source files.
The build system supports using an object directory separate from the source
tree. For example, you can create an 'obj' directory, and from within that
directory, issue configuration and build commands:
autoconf
mkdir obj
cd obj
../configure --enable-autogen
make
## Documentation
The manual page is generated in both html and roff formats. Any web browser
can be used to view the html manual. The roff manual page can be formatted
prior to installation via the following command:
nroff -man -t doc/jemalloc.3
deps/jemalloc-5.2.1/Makefile.in
Vendored
-623
View File
@@ -1,623 +0,0 @@
# Clear out all vpaths, then set just one (default vpath) for the main build
# directory.
vpath
vpath % .
# Clear the default suffixes, so that built-in rules are not used.
.SUFFIXES :
SHELL := /bin/sh
CC := @CC@
CXX := @CXX@
# Configuration parameters.
DESTDIR =
BINDIR := $(DESTDIR)@BINDIR@
INCLUDEDIR := $(DESTDIR)@INCLUDEDIR@
LIBDIR := $(DESTDIR)@LIBDIR@
DATADIR := $(DESTDIR)@DATADIR@
MANDIR := $(DESTDIR)@MANDIR@
srcroot := @srcroot@
objroot := @objroot@
abs_srcroot := @abs_srcroot@
abs_objroot := @abs_objroot@
# Build parameters.
CPPFLAGS := @CPPFLAGS@ -I$(objroot)include -I$(srcroot)include
CONFIGURE_CFLAGS := @CONFIGURE_CFLAGS@
SPECIFIED_CFLAGS := @SPECIFIED_CFLAGS@
EXTRA_CFLAGS := @EXTRA_CFLAGS@
CFLAGS := $(strip $(CONFIGURE_CFLAGS) $(SPECIFIED_CFLAGS) $(EXTRA_CFLAGS))
CONFIGURE_CXXFLAGS := @CONFIGURE_CXXFLAGS@
SPECIFIED_CXXFLAGS := @SPECIFIED_CXXFLAGS@
EXTRA_CXXFLAGS := @EXTRA_CXXFLAGS@
CXXFLAGS := $(strip $(CONFIGURE_CXXFLAGS) $(SPECIFIED_CXXFLAGS) $(EXTRA_CXXFLAGS))
LDFLAGS := @LDFLAGS@
EXTRA_LDFLAGS := @EXTRA_LDFLAGS@
LIBS := @LIBS@
RPATH_EXTRA := @RPATH_EXTRA@
SO := @so@
IMPORTLIB := @importlib@
O := @o@
A := @a@
EXE := @exe@
LIBPREFIX := @libprefix@
REV := @rev@
install_suffix := @install_suffix@
ABI := @abi@
XSLTPROC := @XSLTPROC@
XSLROOT := @XSLROOT@
AUTOCONF := @AUTOCONF@
_RPATH = @RPATH@
RPATH = $(if $(1),$(call _RPATH,$(1)))
cfghdrs_in := $(addprefix $(srcroot),@cfghdrs_in@)
cfghdrs_out := @cfghdrs_out@
cfgoutputs_in := $(addprefix $(srcroot),@cfgoutputs_in@)
cfgoutputs_out := @cfgoutputs_out@
enable_autogen := @enable_autogen@
enable_doc := @enable_doc@
enable_shared := @enable_shared@
enable_static := @enable_static@
enable_prof := @enable_prof@
enable_zone_allocator := @enable_zone_allocator@
enable_experimental_smallocx := @enable_experimental_smallocx@
MALLOC_CONF := @JEMALLOC_CPREFIX@MALLOC_CONF
link_whole_archive := @link_whole_archive@
DSO_LDFLAGS = @DSO_LDFLAGS@
SOREV = @SOREV@
PIC_CFLAGS = @PIC_CFLAGS@
CTARGET = @CTARGET@
LDTARGET = @LDTARGET@
TEST_LD_MODE = @TEST_LD_MODE@
MKLIB = @MKLIB@
AR = @AR@
ARFLAGS = @ARFLAGS@
DUMP_SYMS = @DUMP_SYMS@
AWK := @AWK@
CC_MM = @CC_MM@
LM := @LM@
INSTALL = @INSTALL@
ifeq (macho, $(ABI))
TEST_LIBRARY_PATH := DYLD_FALLBACK_LIBRARY_PATH="$(objroot)lib"
else
ifeq (pecoff, $(ABI))
TEST_LIBRARY_PATH := PATH="$(PATH):$(objroot)lib"
else
TEST_LIBRARY_PATH :=
endif
endif
LIBJEMALLOC := $(LIBPREFIX)jemalloc$(install_suffix)
# Lists of files.
BINS := $(objroot)bin/jemalloc-config $(objroot)bin/jemalloc.sh $(objroot)bin/jeprof
C_HDRS := $(objroot)include/jemalloc/jemalloc$(install_suffix).h
C_SRCS := $(srcroot)src/jemalloc.c \
$(srcroot)src/arena.c \
$(srcroot)src/background_thread.c \
$(srcroot)src/base.c \
$(srcroot)src/bin.c \
$(srcroot)src/bitmap.c \
$(srcroot)src/ckh.c \
$(srcroot)src/ctl.c \
$(srcroot)src/div.c \
$(srcroot)src/extent.c \
$(srcroot)src/extent_dss.c \
$(srcroot)src/extent_mmap.c \
$(srcroot)src/hash.c \
$(srcroot)src/hook.c \
$(srcroot)src/large.c \
$(srcroot)src/log.c \
$(srcroot)src/malloc_io.c \
$(srcroot)src/mutex.c \
$(srcroot)src/mutex_pool.c \
$(srcroot)src/nstime.c \
$(srcroot)src/pages.c \
$(srcroot)src/prng.c \
$(srcroot)src/prof.c \
$(srcroot)src/rtree.c \
$(srcroot)src/safety_check.c \
$(srcroot)src/stats.c \
$(srcroot)src/sc.c \
$(srcroot)src/sz.c \
$(srcroot)src/tcache.c \
$(srcroot)src/test_hooks.c \
$(srcroot)src/ticker.c \
$(srcroot)src/tsd.c \
$(srcroot)src/witness.c
ifeq ($(enable_zone_allocator), 1)
C_SRCS += $(srcroot)src/zone.c
endif
ifeq ($(IMPORTLIB),$(SO))
STATIC_LIBS := $(objroot)lib/$(LIBJEMALLOC).$(A)
endif
ifdef PIC_CFLAGS
STATIC_LIBS += $(objroot)lib/$(LIBJEMALLOC)_pic.$(A)
else
STATIC_LIBS += $(objroot)lib/$(LIBJEMALLOC)_s.$(A)
endif
DSOS := $(objroot)lib/$(LIBJEMALLOC).$(SOREV)
ifneq ($(SOREV),$(SO))
DSOS += $(objroot)lib/$(LIBJEMALLOC).$(SO)
endif
ifeq (1, $(link_whole_archive))
LJEMALLOC := -Wl,--whole-archive -L$(objroot)lib -l$(LIBJEMALLOC) -Wl,--no-whole-archive
else
LJEMALLOC := $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB)
endif
PC := $(objroot)jemalloc.pc
MAN3 := $(objroot)doc/jemalloc$(install_suffix).3
DOCS_XML := $(objroot)doc/jemalloc$(install_suffix).xml
DOCS_HTML := $(DOCS_XML:$(objroot)%.xml=$(objroot)%.html)
DOCS_MAN3 := $(DOCS_XML:$(objroot)%.xml=$(objroot)%.3)
DOCS := $(DOCS_HTML) $(DOCS_MAN3)
C_TESTLIB_SRCS := $(srcroot)test/src/btalloc.c $(srcroot)test/src/btalloc_0.c \
$(srcroot)test/src/btalloc_1.c $(srcroot)test/src/math.c \
$(srcroot)test/src/mtx.c $(srcroot)test/src/mq.c \
$(srcroot)test/src/SFMT.c $(srcroot)test/src/test.c \
$(srcroot)test/src/thd.c $(srcroot)test/src/timer.c
ifeq (1, $(link_whole_archive))
C_UTIL_INTEGRATION_SRCS :=
C_UTIL_CPP_SRCS :=
else
C_UTIL_INTEGRATION_SRCS := $(srcroot)src/nstime.c $(srcroot)src/malloc_io.c
C_UTIL_CPP_SRCS := $(srcroot)src/nstime.c $(srcroot)src/malloc_io.c
endif
TESTS_UNIT := \
$(srcroot)test/unit/a0.c \
$(srcroot)test/unit/arena_reset.c \
$(srcroot)test/unit/atomic.c \
$(srcroot)test/unit/background_thread.c \
$(srcroot)test/unit/background_thread_enable.c \
$(srcroot)test/unit/base.c \
$(srcroot)test/unit/bitmap.c \
$(srcroot)test/unit/bit_util.c \
$(srcroot)test/unit/binshard.c \
$(srcroot)test/unit/ckh.c \
$(srcroot)test/unit/decay.c \
$(srcroot)test/unit/div.c \
$(srcroot)test/unit/emitter.c \
$(srcroot)test/unit/extent_quantize.c \
$(srcroot)test/unit/extent_util.c \
$(srcroot)test/unit/fork.c \
$(srcroot)test/unit/hash.c \
$(srcroot)test/unit/hook.c \
$(srcroot)test/unit/huge.c \
$(srcroot)test/unit/junk.c \
$(srcroot)test/unit/junk_alloc.c \
$(srcroot)test/unit/junk_free.c \
$(srcroot)test/unit/log.c \
$(srcroot)test/unit/mallctl.c \
$(srcroot)test/unit/malloc_io.c \
$(srcroot)test/unit/math.c \
$(srcroot)test/unit/mq.c \
$(srcroot)test/unit/mtx.c \
$(srcroot)test/unit/pack.c \
$(srcroot)test/unit/pages.c \
$(srcroot)test/unit/ph.c \
$(srcroot)test/unit/prng.c \
$(srcroot)test/unit/prof_accum.c \
$(srcroot)test/unit/prof_active.c \
$(srcroot)test/unit/prof_gdump.c \
$(srcroot)test/unit/prof_idump.c \
$(srcroot)test/unit/prof_log.c \
$(srcroot)test/unit/prof_reset.c \
$(srcroot)test/unit/prof_tctx.c \
$(srcroot)test/unit/prof_thread_name.c \
$(srcroot)test/unit/ql.c \
$(srcroot)test/unit/qr.c \
$(srcroot)test/unit/rb.c \
$(srcroot)test/unit/retained.c \
$(srcroot)test/unit/rtree.c \
$(srcroot)test/unit/safety_check.c \
$(srcroot)test/unit/seq.c \
$(srcroot)test/unit/SFMT.c \
$(srcroot)test/unit/sc.c \
$(srcroot)test/unit/size_classes.c \
$(srcroot)test/unit/slab.c \
$(srcroot)test/unit/smoothstep.c \
$(srcroot)test/unit/spin.c \
$(srcroot)test/unit/stats.c \
$(srcroot)test/unit/stats_print.c \
$(srcroot)test/unit/test_hooks.c \
$(srcroot)test/unit/ticker.c \
$(srcroot)test/unit/nstime.c \
$(srcroot)test/unit/tsd.c \
$(srcroot)test/unit/witness.c \
$(srcroot)test/unit/zero.c
ifeq (@enable_prof@, 1)
TESTS_UNIT += \
$(srcroot)test/unit/arena_reset_prof.c
endif
TESTS_INTEGRATION := $(srcroot)test/integration/aligned_alloc.c \
$(srcroot)test/integration/allocated.c \
$(srcroot)test/integration/extent.c \
$(srcroot)test/integration/malloc.c \
$(srcroot)test/integration/mallocx.c \
$(srcroot)test/integration/MALLOCX_ARENA.c \
$(srcroot)test/integration/overflow.c \
$(srcroot)test/integration/posix_memalign.c \
$(srcroot)test/integration/rallocx.c \
$(srcroot)test/integration/sdallocx.c \
$(srcroot)test/integration/slab_sizes.c \
$(srcroot)test/integration/thread_arena.c \
$(srcroot)test/integration/thread_tcache_enabled.c \
$(srcroot)test/integration/xallocx.c
ifeq (@enable_experimental_smallocx@, 1)
TESTS_INTEGRATION += \
$(srcroot)test/integration/smallocx.c
endif
ifeq (@enable_cxx@, 1)
CPP_SRCS := $(srcroot)src/jemalloc_cpp.cpp
TESTS_INTEGRATION_CPP := $(srcroot)test/integration/cpp/basic.cpp
else
CPP_SRCS :=
TESTS_INTEGRATION_CPP :=
endif
TESTS_STRESS := $(srcroot)test/stress/microbench.c \
$(srcroot)test/stress/hookbench.c
TESTS := $(TESTS_UNIT) $(TESTS_INTEGRATION) $(TESTS_INTEGRATION_CPP) $(TESTS_STRESS)
PRIVATE_NAMESPACE_HDRS := $(objroot)include/jemalloc/internal/private_namespace.h $(objroot)include/jemalloc/internal/private_namespace_jet.h
PRIVATE_NAMESPACE_GEN_HDRS := $(PRIVATE_NAMESPACE_HDRS:%.h=%.gen.h)
C_SYM_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.sym.$(O))
C_SYMS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.sym)
C_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.$(O))
CPP_OBJS := $(CPP_SRCS:$(srcroot)%.cpp=$(objroot)%.$(O))
C_PIC_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.pic.$(O))
CPP_PIC_OBJS := $(CPP_SRCS:$(srcroot)%.cpp=$(objroot)%.pic.$(O))
C_JET_SYM_OBJS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.jet.sym.$(O))
C_JET_SYMS := $(C_SRCS:$(srcroot)%.c=$(objroot)%.jet.sym)
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_INTEGRATION_OBJS := $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%.$(O))
TESTS_INTEGRATION_CPP_OBJS := $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%.$(O))
TESTS_STRESS_OBJS := $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%.$(O))
TESTS_OBJS := $(TESTS_UNIT_OBJS) $(TESTS_INTEGRATION_OBJS) $(TESTS_STRESS_OBJS)
TESTS_CPP_OBJS := $(TESTS_INTEGRATION_CPP_OBJS)
.PHONY: all dist build_doc_html build_doc_man build_doc
.PHONY: install_bin install_include install_lib
.PHONY: install_doc_html install_doc_man install_doc install
.PHONY: tests check clean distclean relclean
.SECONDARY : $(PRIVATE_NAMESPACE_GEN_HDRS) $(TESTS_OBJS) $(TESTS_CPP_OBJS)
# Default target.
all: build_lib
dist: build_doc
$(objroot)doc/%.html : $(objroot)doc/%.xml $(srcroot)doc/stylesheet.xsl $(objroot)doc/html.xsl
ifneq ($(XSLROOT),)
$(XSLTPROC) -o $@ $(objroot)doc/html.xsl $<
else
ifeq ($(wildcard $(DOCS_HTML)),)
@echo "<p>Missing xsltproc. Doc not built.</p>" > $@
endif
@echo "Missing xsltproc. "$@" not (re)built."
endif
$(objroot)doc/%.3 : $(objroot)doc/%.xml $(srcroot)doc/stylesheet.xsl $(objroot)doc/manpages.xsl
ifneq ($(XSLROOT),)
$(XSLTPROC) -o $@ $(objroot)doc/manpages.xsl $<
else
ifeq ($(wildcard $(DOCS_MAN3)),)
@echo "Missing xsltproc. Doc not built." > $@
endif
@echo "Missing xsltproc. "$@" not (re)built."
endif
build_doc_html: $(DOCS_HTML)
build_doc_man: $(DOCS_MAN3)
build_doc: $(DOCS)
#
# Include generated dependency files.
#
ifdef CC_MM
-include $(C_SYM_OBJS:%.$(O)=%.d)
-include $(C_OBJS:%.$(O)=%.d)
-include $(CPP_OBJS:%.$(O)=%.d)
-include $(C_PIC_OBJS:%.$(O)=%.d)
-include $(CPP_PIC_OBJS:%.$(O)=%.d)
-include $(C_JET_SYM_OBJS:%.$(O)=%.d)
-include $(C_JET_OBJS:%.$(O)=%.d)
-include $(C_TESTLIB_OBJS:%.$(O)=%.d)
-include $(TESTS_OBJS:%.$(O)=%.d)
-include $(TESTS_CPP_OBJS:%.$(O)=%.d)
endif
$(C_SYM_OBJS): $(objroot)src/%.sym.$(O): $(srcroot)src/%.c
$(C_SYM_OBJS): CPPFLAGS += -DJEMALLOC_NO_PRIVATE_NAMESPACE
$(C_SYMS): $(objroot)src/%.sym: $(objroot)src/%.sym.$(O)
$(C_OBJS): $(objroot)src/%.$(O): $(srcroot)src/%.c
$(CPP_OBJS): $(objroot)src/%.$(O): $(srcroot)src/%.cpp
$(C_PIC_OBJS): $(objroot)src/%.pic.$(O): $(srcroot)src/%.c
$(C_PIC_OBJS): CFLAGS += $(PIC_CFLAGS)
$(CPP_PIC_OBJS): $(objroot)src/%.pic.$(O): $(srcroot)src/%.cpp
$(CPP_PIC_OBJS): CXXFLAGS += $(PIC_CFLAGS)
$(C_JET_SYM_OBJS): $(objroot)src/%.jet.sym.$(O): $(srcroot)src/%.c
$(C_JET_SYM_OBJS): CPPFLAGS += -DJEMALLOC_JET -DJEMALLOC_NO_PRIVATE_NAMESPACE
$(C_JET_SYMS): $(objroot)src/%.jet.sym: $(objroot)src/%.jet.sym.$(O)
$(C_JET_OBJS): $(objroot)src/%.jet.$(O): $(srcroot)src/%.c
$(C_JET_OBJS): CPPFLAGS += -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_INTEGRATION_OBJS): CPPFLAGS += -DJEMALLOC_INTEGRATION_TEST
$(TESTS_INTEGRATION_CPP_OBJS): CPPFLAGS += -DJEMALLOC_INTEGRATION_CPP_TEST
$(TESTS_STRESS_OBJS): CPPFLAGS += -DJEMALLOC_STRESS_TEST
$(TESTS_OBJS): $(objroot)test/%.$(O): $(srcroot)test/%.c
$(TESTS_CPP_OBJS): $(objroot)test/%.$(O): $(srcroot)test/%.cpp
$(TESTS_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include
$(TESTS_CPP_OBJS): CPPFLAGS += -I$(srcroot)test/include -I$(objroot)test/include
ifneq ($(IMPORTLIB),$(SO))
$(CPP_OBJS) $(C_SYM_OBJS) $(C_OBJS) $(C_JET_SYM_OBJS) $(C_JET_OBJS): CPPFLAGS += -DDLLEXPORT
endif
# Dependencies.
ifndef CC_MM
HEADER_DIRS = $(srcroot)include/jemalloc/internal \
$(objroot)include/jemalloc $(objroot)include/jemalloc/internal
HEADERS = $(filter-out $(PRIVATE_NAMESPACE_HDRS),$(wildcard $(foreach dir,$(HEADER_DIRS),$(dir)/*.h)))
$(C_SYM_OBJS) $(C_OBJS) $(CPP_OBJS) $(C_PIC_OBJS) $(CPP_PIC_OBJS) $(C_JET_SYM_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS) $(TESTS_CPP_OBJS): $(HEADERS)
$(TESTS_OBJS) $(TESTS_CPP_OBJS): $(objroot)test/include/test/jemalloc_test.h
endif
$(C_OBJS) $(CPP_OBJS) $(C_PIC_OBJS) $(CPP_PIC_OBJS) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(TESTS_INTEGRATION_OBJS) $(TESTS_INTEGRATION_CPP_OBJS): $(objroot)include/jemalloc/internal/private_namespace.h
$(C_JET_OBJS) $(C_TESTLIB_UNIT_OBJS) $(C_TESTLIB_STRESS_OBJS) $(TESTS_UNIT_OBJS) $(TESTS_STRESS_OBJS): $(objroot)include/jemalloc/internal/private_namespace_jet.h
$(C_SYM_OBJS) $(C_OBJS) $(C_PIC_OBJS) $(C_JET_SYM_OBJS) $(C_JET_OBJS) $(C_TESTLIB_OBJS) $(TESTS_OBJS): %.$(O):
@mkdir -p $(@D)
$(CC) $(CFLAGS) -c $(CPPFLAGS) $(CTARGET) $<
ifdef CC_MM
@$(CC) -MM $(CPPFLAGS) -MT $@ -o $(@:%.$(O)=%.d) $<
endif
$(C_SYMS): %.sym:
@mkdir -p $(@D)
$(DUMP_SYMS) $< | $(AWK) -f $(objroot)include/jemalloc/internal/private_symbols.awk > $@
$(C_JET_SYMS): %.sym:
@mkdir -p $(@D)
$(DUMP_SYMS) $< | $(AWK) -f $(objroot)include/jemalloc/internal/private_symbols_jet.awk > $@
$(objroot)include/jemalloc/internal/private_namespace.gen.h: $(C_SYMS)
$(SHELL) $(srcroot)include/jemalloc/internal/private_namespace.sh $^ > $@
$(objroot)include/jemalloc/internal/private_namespace_jet.gen.h: $(C_JET_SYMS)
$(SHELL) $(srcroot)include/jemalloc/internal/private_namespace.sh $^ > $@
%.h: %.gen.h
@if ! `cmp -s $< $@` ; then echo "cp $< $<"; cp $< $@ ; fi
$(CPP_OBJS) $(CPP_PIC_OBJS) $(TESTS_CPP_OBJS): %.$(O):
@mkdir -p $(@D)
$(CXX) $(CXXFLAGS) -c $(CPPFLAGS) $(CTARGET) $<
ifdef CC_MM
@$(CXX) -MM $(CPPFLAGS) -MT $@ -o $(@:%.$(O)=%.d) $<
endif
ifneq ($(SOREV),$(SO))
%.$(SO) : %.$(SOREV)
@mkdir -p $(@D)
ln -sf $(<F) $@
endif
$(objroot)lib/$(LIBJEMALLOC).$(SOREV) : $(if $(PIC_CFLAGS),$(C_PIC_OBJS),$(C_OBJS)) $(if $(PIC_CFLAGS),$(CPP_PIC_OBJS),$(CPP_OBJS))
@mkdir -p $(@D)
$(CC) $(DSO_LDFLAGS) $(call RPATH,$(RPATH_EXTRA)) $(LDTARGET) $+ $(LDFLAGS) $(LIBS) $(EXTRA_LDFLAGS)
$(objroot)lib/$(LIBJEMALLOC)_pic.$(A) : $(C_PIC_OBJS) $(CPP_PIC_OBJS)
$(objroot)lib/$(LIBJEMALLOC).$(A) : $(C_OBJS) $(CPP_OBJS)
$(objroot)lib/$(LIBJEMALLOC)_s.$(A) : $(C_OBJS) $(CPP_OBJS)
$(STATIC_LIBS):
@mkdir -p $(@D)
$(AR) $(ARFLAGS)@AROUT@ $+
$(objroot)test/unit/%$(EXE): $(objroot)test/unit/%.$(O) $(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) $(TEST_LD_MODE) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(LJEMALLOC) $(LDFLAGS) $(filter-out -lm,$(filter -lrt -pthread -lstdc++,$(LIBS))) $(LM) $(EXTRA_LDFLAGS)
$(objroot)test/integration/cpp/%$(EXE): $(objroot)test/integration/cpp/%.$(O) $(C_TESTLIB_INTEGRATION_OBJS) $(C_UTIL_INTEGRATION_OBJS) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB)
@mkdir -p $(@D)
$(CXX) $(LDTARGET) $(filter %.$(O),$^) $(call RPATH,$(objroot)lib) $(objroot)lib/$(LIBJEMALLOC).$(IMPORTLIB) $(LDFLAGS) $(filter-out -lm,$(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) $(TEST_LD_MODE) $(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: $(STATIC_LIBS)
ifeq ($(enable_shared), 1)
build_lib: build_lib_shared
endif
ifeq ($(enable_static), 1)
build_lib: build_lib_static
endif
install_bin:
$(INSTALL) -d $(BINDIR)
@for b in $(BINS); do \
echo "$(INSTALL) -m 755 $$b $(BINDIR)"; \
$(INSTALL) -m 755 $$b $(BINDIR); \
done
install_include:
$(INSTALL) -d $(INCLUDEDIR)/jemalloc
@for h in $(C_HDRS); do \
echo "$(INSTALL) -m 644 $$h $(INCLUDEDIR)/jemalloc"; \
$(INSTALL) -m 644 $$h $(INCLUDEDIR)/jemalloc; \
done
install_lib_shared: $(DSOS)
$(INSTALL) -d $(LIBDIR)
$(INSTALL) -m 755 $(objroot)lib/$(LIBJEMALLOC).$(SOREV) $(LIBDIR)
ifneq ($(SOREV),$(SO))
ln -sf $(LIBJEMALLOC).$(SOREV) $(LIBDIR)/$(LIBJEMALLOC).$(SO)
endif
install_lib_static: $(STATIC_LIBS)
$(INSTALL) -d $(LIBDIR)
@for l in $(STATIC_LIBS); do \
echo "$(INSTALL) -m 755 $$l $(LIBDIR)"; \
$(INSTALL) -m 755 $$l $(LIBDIR); \
done
install_lib_pc: $(PC)
$(INSTALL) -d $(LIBDIR)/pkgconfig
@for l in $(PC); do \
echo "$(INSTALL) -m 644 $$l $(LIBDIR)/pkgconfig"; \
$(INSTALL) -m 644 $$l $(LIBDIR)/pkgconfig; \
done
ifeq ($(enable_shared), 1)
install_lib: install_lib_shared
endif
ifeq ($(enable_static), 1)
install_lib: install_lib_static
endif
install_lib: install_lib_pc
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 $(MANDIR)/man3; \
done
install_doc: build_doc install_doc_html install_doc_man
install: install_bin install_include install_lib
ifeq ($(enable_doc), 1)
install: install_doc
endif
tests_unit: $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%$(EXE))
tests_integration: $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%$(EXE)) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%$(EXE))
tests_stress: $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%$(EXE))
tests: tests_unit tests_integration tests_stress
check_unit_dir:
@mkdir -p $(objroot)test/unit
check_integration_dir:
@mkdir -p $(objroot)test/integration
stress_dir:
@mkdir -p $(objroot)test/stress
check_dir: check_unit_dir check_integration_dir
check_unit: tests_unit check_unit_dir
$(SHELL) $(objroot)test/test.sh $(TESTS_UNIT:$(srcroot)%.c=$(objroot)%)
check_integration_prof: tests_integration check_integration_dir
ifeq ($(enable_prof), 1)
$(MALLOC_CONF)="prof:true" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%)
$(MALLOC_CONF)="prof:true,prof_active:false" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%)
endif
check_integration_decay: tests_integration check_integration_dir
$(MALLOC_CONF)="dirty_decay_ms:-1,muzzy_decay_ms:-1" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%)
$(MALLOC_CONF)="dirty_decay_ms:0,muzzy_decay_ms:0" $(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%)
check_integration: tests_integration check_integration_dir
$(SHELL) $(objroot)test/test.sh $(TESTS_INTEGRATION:$(srcroot)%.c=$(objroot)%) $(TESTS_INTEGRATION_CPP:$(srcroot)%.cpp=$(objroot)%)
stress: tests_stress stress_dir
$(SHELL) $(objroot)test/test.sh $(TESTS_STRESS:$(srcroot)%.c=$(objroot)%)
check: check_unit check_integration check_integration_decay check_integration_prof
clean:
rm -f $(PRIVATE_NAMESPACE_HDRS)
rm -f $(PRIVATE_NAMESPACE_GEN_HDRS)
rm -f $(C_SYM_OBJS)
rm -f $(C_SYMS)
rm -f $(C_OBJS)
rm -f $(CPP_OBJS)
rm -f $(C_PIC_OBJS)
rm -f $(CPP_PIC_OBJS)
rm -f $(C_JET_SYM_OBJS)
rm -f $(C_JET_SYMS)
rm -f $(C_JET_OBJS)
rm -f $(C_TESTLIB_OBJS)
rm -f $(C_SYM_OBJS:%.$(O)=%.d)
rm -f $(C_OBJS:%.$(O)=%.d)
rm -f $(CPP_OBJS:%.$(O)=%.d)
rm -f $(C_PIC_OBJS:%.$(O)=%.d)
rm -f $(CPP_PIC_OBJS:%.$(O)=%.d)
rm -f $(C_JET_SYM_OBJS:%.$(O)=%.d)
rm -f $(C_JET_OBJS:%.$(O)=%.d)
rm -f $(C_TESTLIB_OBJS:%.$(O)=%.d)
rm -f $(TESTS_OBJS:%.$(O)=%$(EXE))
rm -f $(TESTS_OBJS)
rm -f $(TESTS_OBJS:%.$(O)=%.d)
rm -f $(TESTS_OBJS:%.$(O)=%.out)
rm -f $(TESTS_CPP_OBJS:%.$(O)=%$(EXE))
rm -f $(TESTS_CPP_OBJS)
rm -f $(TESTS_CPP_OBJS:%.$(O)=%.d)
rm -f $(TESTS_CPP_OBJS:%.$(O)=%.out)
rm -f $(DSOS) $(STATIC_LIBS)
distclean: clean
rm -f $(objroot)bin/jemalloc-config
rm -f $(objroot)bin/jemalloc.sh
rm -f $(objroot)bin/jeprof
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 $(objroot)VERSION
rm -f $(DOCS_HTML)
rm -f $(DOCS_MAN3)
#===============================================================================
# Re-configuration rules.
ifeq ($(enable_autogen), 1)
$(srcroot)configure : $(srcroot)configure.ac
cd ./$(srcroot) && $(AUTOCONF)
$(objroot)config.status : $(srcroot)configure
./$(objroot)config.status --recheck
$(srcroot)config.stamp.in : $(srcroot)configure.ac
echo stamp > $(srcroot)config.stamp.in
$(objroot)config.stamp : $(cfgoutputs_in) $(cfghdrs_in) $(srcroot)configure
./$(objroot)config.status
@touch $@
# There must be some action in order for make to re-read Makefile when it is
# out of date.
$(cfgoutputs_out) $(cfghdrs_out) : $(objroot)config.stamp
@true
endif
deps/jemalloc-5.2.1/TUNING.md
Vendored
-129
View File
@@ -1,129 +0,0 @@
This document summarizes the common approaches for performance fine tuning with
jemalloc (as of 5.1.0). The default configuration of jemalloc tends to work
reasonably well in practice, and most applications should not have to tune any
options. However, in order to cover a wide range of applications and avoid
pathological cases, the default setting is sometimes kept conservative and
suboptimal, even for many common workloads. When jemalloc is properly tuned for
a specific application / workload, it is common to improve system level metrics
by a few percent, or make favorable trade-offs.
## Notable runtime options for performance tuning
Runtime options can be set via
[malloc_conf](http://jemalloc.net/jemalloc.3.html#tuning).
* [background_thread](http://jemalloc.net/jemalloc.3.html#background_thread)
Enabling jemalloc background threads generally improves the tail latency for
application threads, since unused memory purging is shifted to the dedicated
background threads. In addition, unintended purging delay caused by
application inactivity is avoided with background threads.
Suggested: `background_thread:true` when jemalloc managed threads can be
allowed.
* [metadata_thp](http://jemalloc.net/jemalloc.3.html#opt.metadata_thp)
Allowing jemalloc to utilize transparent huge pages for its internal
metadata usually reduces TLB misses significantly, especially for programs
with large memory footprint and frequent allocation / deallocation
activities. Metadata memory usage may increase due to the use of huge
pages.
Suggested for allocation intensive programs: `metadata_thp:auto` or
`metadata_thp:always`, which is expected to improve CPU utilization at a
small memory cost.
* [dirty_decay_ms](http://jemalloc.net/jemalloc.3.html#opt.dirty_decay_ms) and
[muzzy_decay_ms](http://jemalloc.net/jemalloc.3.html#opt.muzzy_decay_ms)
Decay time determines how fast jemalloc returns unused pages back to the
operating system, and therefore provides a fairly straightforward trade-off
between CPU and memory usage. Shorter decay time purges unused pages faster
to reduces memory usage (usually at the cost of more CPU cycles spent on
purging), and vice versa.
Suggested: tune the values based on the desired trade-offs.
* [narenas](http://jemalloc.net/jemalloc.3.html#opt.narenas)
By default jemalloc uses multiple arenas to reduce internal lock contention.
However high arena count may also increase overall memory fragmentation,
since arenas manage memory independently. When high degree of parallelism
is not expected at the allocator level, lower number of arenas often
improves memory usage.
Suggested: if low parallelism is expected, try lower arena count while
monitoring CPU and memory usage.
* [percpu_arena](http://jemalloc.net/jemalloc.3.html#opt.percpu_arena)
Enable dynamic thread to arena association based on running CPU. This has
the potential to improve locality, e.g. when thread to CPU affinity is
present.
Suggested: try `percpu_arena:percpu` or `percpu_arena:phycpu` if
thread migration between processors is expected to be infrequent.
Examples:
* High resource consumption application, prioritizing CPU utilization:
`background_thread:true,metadata_thp:auto` combined with relaxed decay time
(increased `dirty_decay_ms` and / or `muzzy_decay_ms`,
e.g. `dirty_decay_ms:30000,muzzy_decay_ms:30000`).
* High resource consumption application, prioritizing memory usage:
`background_thread:true` combined with shorter decay time (decreased
`dirty_decay_ms` and / or `muzzy_decay_ms`,
e.g. `dirty_decay_ms:5000,muzzy_decay_ms:5000`), and lower arena count
(e.g. number of CPUs).
* Low resource consumption application:
`narenas:1,lg_tcache_max:13` combined with shorter decay time (decreased
`dirty_decay_ms` and / or `muzzy_decay_ms`,e.g.
`dirty_decay_ms:1000,muzzy_decay_ms:0`).
* Extremely conservative -- minimize memory usage at all costs, only suitable when
allocation activity is very rare:
`narenas:1,tcache:false,dirty_decay_ms:0,muzzy_decay_ms:0`
Note that it is recommended to combine the options with `abort_conf:true` which
aborts immediately on illegal options.
## Beyond runtime options
In addition to the runtime options, there are a number of programmatic ways to
improve application performance with jemalloc.
* [Explicit arenas](http://jemalloc.net/jemalloc.3.html#arenas.create)
Manually created arenas can help performance in various ways, e.g. by
managing locality and contention for specific usages. For example,
applications can explicitly allocate frequently accessed objects from a
dedicated arena with
[mallocx()](http://jemalloc.net/jemalloc.3.html#MALLOCX_ARENA) to improve
locality. In addition, explicit arenas often benefit from individually
tuned options, e.g. relaxed [decay
time](http://jemalloc.net/jemalloc.3.html#arena.i.dirty_decay_ms) if
frequent reuse is expected.
* [Extent hooks](http://jemalloc.net/jemalloc.3.html#arena.i.extent_hooks)
Extent hooks allow customization for managing underlying memory. One use
case for performance purpose is to utilize huge pages -- for example,
[HHVM](https://github.com/facebook/hhvm/blob/master/hphp/util/alloc.cpp)
uses explicit arenas with customized extent hooks to manage 1GB huge pages
for frequently accessed data, which reduces TLB misses significantly.
* [Explicit thread-to-arena
binding](http://jemalloc.net/jemalloc.3.html#thread.arena)
It is common for some threads in an application to have different memory
access / allocation patterns. Threads with heavy workloads often benefit
from explicit binding, e.g. binding very active threads to dedicated arenas
may reduce contention at the allocator level.
deps/jemalloc-5.2.1/autogen.sh
Vendored
-17
View File
@@ -1,17 +0,0 @@
#!/bin/sh
for i in autoconf; do
echo "$i"
$i
if [ $? -ne 0 ]; then
echo "Error $? in $i"
exit 1
fi
done
echo "./configure --enable-autogen $@"
./configure --enable-autogen $@
if [ $? -ne 0 ]; then
echo "Error $? in ./configure"
exit 1
fi
deps/jemalloc-5.2.1/bin/jemalloc-config.in
Vendored
-83
View File
@@ -1,83 +0,0 @@
#!/bin/sh
usage() {
cat <<EOF
Usage:
@BINDIR@/jemalloc-config <option>
Options:
--help | -h : Print usage.
--version : Print jemalloc version.
--revision : Print shared library revision number.
--config : Print configure options used to build jemalloc.
--prefix : Print installation directory prefix.
--bindir : Print binary installation directory.
--datadir : Print data installation directory.
--includedir : Print include installation directory.
--libdir : Print library installation directory.
--mandir : Print manual page installation directory.
--cc : Print compiler used to build jemalloc.
--cflags : Print compiler flags used to build jemalloc.
--cppflags : Print preprocessor flags used to build jemalloc.
--cxxflags : Print C++ compiler flags used to build jemalloc.
--ldflags : Print library flags used to build jemalloc.
--libs : Print libraries jemalloc was linked against.
EOF
}
prefix="@prefix@"
exec_prefix="@exec_prefix@"
case "$1" in
--help | -h)
usage
exit 0
;;
--version)
echo "@jemalloc_version@"
;;
--revision)
echo "@rev@"
;;
--config)
echo "@CONFIG@"
;;
--prefix)
echo "@PREFIX@"
;;
--bindir)
echo "@BINDIR@"
;;
--datadir)
echo "@DATADIR@"
;;
--includedir)
echo "@INCLUDEDIR@"
;;
--libdir)
echo "@LIBDIR@"
;;
--mandir)
echo "@MANDIR@"
;;
--cc)
echo "@CC@"
;;
--cflags)
echo "@CFLAGS@"
;;
--cppflags)
echo "@CPPFLAGS@"
;;
--cxxflags)
echo "@CXXFLAGS@"
;;
--ldflags)
echo "@LDFLAGS@ @EXTRA_LDFLAGS@"
;;
--libs)
echo "@LIBS@"
;;
*)
usage
exit 1
esac
deps/jemalloc-5.2.1/bin/jemalloc.sh.in
Vendored
-9
View File
@@ -1,9 +0,0 @@
#!/bin/sh
prefix=@prefix@
exec_prefix=@exec_prefix@
libdir=@libdir@
@LD_PRELOAD_VAR@=${libdir}/libjemalloc.@SOREV@
export @LD_PRELOAD_VAR@
exec "$@"
deps/jemalloc-5.2.1/build-aux/install-sh
Vendored
-250
View File
@@ -1,250 +0,0 @@
#! /bin/sh
#
# install - install a program, script, or datafile
# This comes from X11R5 (mit/util/scripts/install.sh).
#
# Copyright 1991 by the Massachusetts Institute of Technology
#
# Permission to use, copy, modify, distribute, and sell this software and its
# documentation for any purpose is hereby granted without fee, provided that
# the above copyright notice appear in all copies and that both that
# copyright notice and this permission notice appear in supporting
# documentation, and that the name of M.I.T. not be used in advertising or
# publicity pertaining to distribution of the software without specific,
# written prior permission. M.I.T. makes no representations about the
# suitability of this software for any purpose. It is provided "as is"
# without express or implied warranty.
#
# Calling this script install-sh is preferred over install.sh, to prevent
# `make' implicit rules from creating a file called install from it
# when there is no Makefile.
#
# This script is compatible with the BSD install script, but was written
# from scratch. It can only install one file at a time, a restriction
# shared with many OS's install programs.
# set DOITPROG to echo to test this script
# Don't use :- since 4.3BSD and earlier shells don't like it.
doit="${DOITPROG-}"
# put in absolute paths if you don't have them in your path; or use env. vars.
mvprog="${MVPROG-mv}"
cpprog="${CPPROG-cp}"
chmodprog="${CHMODPROG-chmod}"
chownprog="${CHOWNPROG-chown}"
chgrpprog="${CHGRPPROG-chgrp}"
stripprog="${STRIPPROG-strip}"
rmprog="${RMPROG-rm}"
mkdirprog="${MKDIRPROG-mkdir}"
transformbasename=""
transform_arg=""
instcmd="$mvprog"
chmodcmd="$chmodprog 0755"
chowncmd=""
chgrpcmd=""
stripcmd=""
rmcmd="$rmprog -f"
mvcmd="$mvprog"
src=""
dst=""
dir_arg=""
while [ x"$1" != x ]; do
case $1 in
-c) instcmd="$cpprog"
shift
continue;;
-d) dir_arg=true
shift
continue;;
-m) chmodcmd="$chmodprog $2"
shift
shift
continue;;
-o) chowncmd="$chownprog $2"
shift
shift
continue;;
-g) chgrpcmd="$chgrpprog $2"
shift
shift
continue;;
-s) stripcmd="$stripprog"
shift
continue;;
-t=*) transformarg=`echo $1 | sed 's/-t=//'`
shift
continue;;
-b=*) transformbasename=`echo $1 | sed 's/-b=//'`
shift
continue;;
*) if [ x"$src" = x ]
then
src=$1
else
# this colon is to work around a 386BSD /bin/sh bug
:
dst=$1
fi
shift
continue;;
esac
done
if [ x"$src" = x ]
then
echo "install: no input file specified"
exit 1
else
true
fi
if [ x"$dir_arg" != x ]; then
dst=$src
src=""
if [ -d $dst ]; then
instcmd=:
else
instcmd=mkdir
fi
else
# Waiting for this to be detected by the "$instcmd $src $dsttmp" command
# might cause directories to be created, which would be especially bad
# if $src (and thus $dsttmp) contains '*'.
if [ -f $src -o -d $src ]
then
true
else
echo "install: $src does not exist"
exit 1
fi
if [ x"$dst" = x ]
then
echo "install: no destination specified"
exit 1
else
true
fi
# If destination is a directory, append the input filename; if your system
# does not like double slashes in filenames, you may need to add some logic
if [ -d $dst ]
then
dst="$dst"/`basename $src`
else
true
fi
fi
## this sed command emulates the dirname command
dstdir=`echo $dst | sed -e 's,[^/]*$,,;s,/$,,;s,^$,.,'`
# Make sure that the destination directory exists.
# this part is taken from Noah Friedman's mkinstalldirs script
# Skip lots of stat calls in the usual case.
if [ ! -d "$dstdir" ]; then
defaultIFS='
'
IFS="${IFS-${defaultIFS}}"
oIFS="${IFS}"
# Some sh's can't handle IFS=/ for some reason.
IFS='%'
set - `echo ${dstdir} | sed -e 's@/@%@g' -e 's@^%@/@'`
IFS="${oIFS}"
pathcomp=''
while [ $# -ne 0 ] ; do
pathcomp="${pathcomp}${1}"
shift
if [ ! -d "${pathcomp}" ] ;
then
$mkdirprog "${pathcomp}"
else
true
fi
pathcomp="${pathcomp}/"
done
fi
if [ x"$dir_arg" != x ]
then
$doit $instcmd $dst &&
if [ x"$chowncmd" != x ]; then $doit $chowncmd $dst; else true ; fi &&
if [ x"$chgrpcmd" != x ]; then $doit $chgrpcmd $dst; else true ; fi &&
if [ x"$stripcmd" != x ]; then $doit $stripcmd $dst; else true ; fi &&
if [ x"$chmodcmd" != x ]; then $doit $chmodcmd $dst; else true ; fi
else
# If we're going to rename the final executable, determine the name now.
if [ x"$transformarg" = x ]
then
dstfile=`basename $dst`
else
dstfile=`basename $dst $transformbasename |
sed $transformarg`$transformbasename
fi
# don't allow the sed command to completely eliminate the filename
if [ x"$dstfile" = x ]
then
dstfile=`basename $dst`
else
true
fi
# Make a temp file name in the proper directory.
dsttmp=$dstdir/#inst.$$#
# Move or copy the file name to the temp name
$doit $instcmd $src $dsttmp &&
trap "rm -f ${dsttmp}" 0 &&
# and set any options; do chmod last to preserve setuid bits
# If any of these fail, we abort the whole thing. If we want to
# ignore errors from any of these, just make sure not to ignore
# errors from the above "$doit $instcmd $src $dsttmp" command.
if [ x"$chowncmd" != x ]; then $doit $chowncmd $dsttmp; else true;fi &&
if [ x"$chgrpcmd" != x ]; then $doit $chgrpcmd $dsttmp; else true;fi &&
if [ x"$stripcmd" != x ]; then $doit $stripcmd $dsttmp; else true;fi &&
if [ x"$chmodcmd" != x ]; then $doit $chmodcmd $dsttmp; else true;fi &&
# Now rename the file to the real destination.
$doit $rmcmd -f $dstdir/$dstfile &&
$doit $mvcmd $dsttmp $dstdir/$dstfile
fi &&
exit 0
deps/jemalloc-5.2.1/config.stamp.in
Vendored
View File
deps/jemalloc-5.2.1/configure.ac
Vendored
-2406
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc-5.2.1/doc/html.xsl.in
Vendored
-5
View File
@@ -1,5 +0,0 @@
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:import href="@XSLROOT@/html/docbook.xsl"/>
<xsl:import href="@abs_srcroot@doc/stylesheet.xsl"/>
<xsl:output method="xml" encoding="utf-8"/>
</xsl:stylesheet>
deps/jemalloc-5.2.1/doc/jemalloc.xml.in
Vendored
-3513
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc-5.2.1/doc/manpages.xsl.in
Vendored
-4
View File
@@ -1,4 +0,0 @@
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:import href="@XSLROOT@/manpages/docbook.xsl"/>
<xsl:import href="@abs_srcroot@doc/stylesheet.xsl"/>
</xsl:stylesheet>
deps/jemalloc-5.2.1/doc/stylesheet.xsl
Vendored
-10
View File
@@ -1,10 +0,0 @@
<xsl:stylesheet xmlns:xsl="http://www.w3.org/1999/XSL/Transform" version="1.0">
<xsl:param name="funcsynopsis.style">ansi</xsl:param>
<xsl:param name="function.parens" select="0"/>
<xsl:template match="function">
<xsl:call-template name="inline.monoseq"/>
</xsl:template>
<xsl:template match="mallctl">
<quote><xsl:call-template name="inline.monoseq"/></quote>
</xsl:template>
</xsl:stylesheet>
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_externs.h
Vendored
-104
View File
@@ -1,104 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_EXTERNS_H
#define JEMALLOC_INTERNAL_ARENA_EXTERNS_H
#include "jemalloc/internal/bin.h"
#include "jemalloc/internal/extent_dss.h"
#include "jemalloc/internal/hook.h"
#include "jemalloc/internal/pages.h"
#include "jemalloc/internal/stats.h"
extern ssize_t opt_dirty_decay_ms;
extern ssize_t opt_muzzy_decay_ms;
extern percpu_arena_mode_t opt_percpu_arena;
extern const char *percpu_arena_mode_names[];
extern const uint64_t h_steps[SMOOTHSTEP_NSTEPS];
extern malloc_mutex_t arenas_lock;
extern size_t opt_oversize_threshold;
extern size_t oversize_threshold;
void arena_basic_stats_merge(tsdn_t *tsdn, arena_t *arena,
unsigned *nthreads, const char **dss, ssize_t *dirty_decay_ms,
ssize_t *muzzy_decay_ms, size_t *nactive, size_t *ndirty, size_t *nmuzzy);
void arena_stats_merge(tsdn_t *tsdn, arena_t *arena, unsigned *nthreads,
const char **dss, ssize_t *dirty_decay_ms, ssize_t *muzzy_decay_ms,
size_t *nactive, size_t *ndirty, size_t *nmuzzy, arena_stats_t *astats,
bin_stats_t *bstats, arena_stats_large_t *lstats,
arena_stats_extents_t *estats);
void arena_extents_dirty_dalloc(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent);
#ifdef JEMALLOC_JET
size_t arena_slab_regind(extent_t *slab, szind_t binind, const void *ptr);
#endif
extent_t *arena_extent_alloc_large(tsdn_t *tsdn, arena_t *arena,
size_t usize, size_t alignment, bool *zero);
void arena_extent_dalloc_large_prep(tsdn_t *tsdn, arena_t *arena,
extent_t *extent);
void arena_extent_ralloc_large_shrink(tsdn_t *tsdn, arena_t *arena,
extent_t *extent, size_t oldsize);
void arena_extent_ralloc_large_expand(tsdn_t *tsdn, arena_t *arena,
extent_t *extent, size_t oldsize);
ssize_t arena_dirty_decay_ms_get(arena_t *arena);
bool arena_dirty_decay_ms_set(tsdn_t *tsdn, arena_t *arena, ssize_t decay_ms);
ssize_t arena_muzzy_decay_ms_get(arena_t *arena);
bool arena_muzzy_decay_ms_set(tsdn_t *tsdn, arena_t *arena, ssize_t decay_ms);
void arena_decay(tsdn_t *tsdn, arena_t *arena, bool is_background_thread,
bool all);
void arena_reset(tsd_t *tsd, arena_t *arena);
void arena_destroy(tsd_t *tsd, arena_t *arena);
void arena_tcache_fill_small(tsdn_t *tsdn, arena_t *arena, tcache_t *tcache,
cache_bin_t *tbin, szind_t binind, uint64_t prof_accumbytes);
void arena_alloc_junk_small(void *ptr, const bin_info_t *bin_info,
bool zero);
typedef void (arena_dalloc_junk_small_t)(void *, const bin_info_t *);
extern arena_dalloc_junk_small_t *JET_MUTABLE arena_dalloc_junk_small;
void *arena_malloc_hard(tsdn_t *tsdn, arena_t *arena, size_t size,
szind_t ind, bool zero);
void *arena_palloc(tsdn_t *tsdn, arena_t *arena, size_t usize,
size_t alignment, bool zero, tcache_t *tcache);
void arena_prof_promote(tsdn_t *tsdn, void *ptr, size_t usize);
void arena_dalloc_promoted(tsdn_t *tsdn, void *ptr, tcache_t *tcache,
bool slow_path);
void arena_dalloc_bin_junked_locked(tsdn_t *tsdn, arena_t *arena, bin_t *bin,
szind_t binind, extent_t *extent, void *ptr);
void arena_dalloc_small(tsdn_t *tsdn, void *ptr);
bool arena_ralloc_no_move(tsdn_t *tsdn, void *ptr, size_t oldsize, size_t size,
size_t extra, bool zero, size_t *newsize);
void *arena_ralloc(tsdn_t *tsdn, arena_t *arena, void *ptr, size_t oldsize,
size_t size, size_t alignment, bool zero, tcache_t *tcache,
hook_ralloc_args_t *hook_args);
dss_prec_t arena_dss_prec_get(arena_t *arena);
bool arena_dss_prec_set(arena_t *arena, dss_prec_t dss_prec);
ssize_t arena_dirty_decay_ms_default_get(void);
bool arena_dirty_decay_ms_default_set(ssize_t decay_ms);
ssize_t arena_muzzy_decay_ms_default_get(void);
bool arena_muzzy_decay_ms_default_set(ssize_t decay_ms);
bool arena_retain_grow_limit_get_set(tsd_t *tsd, arena_t *arena,
size_t *old_limit, size_t *new_limit);
unsigned arena_nthreads_get(arena_t *arena, bool internal);
void arena_nthreads_inc(arena_t *arena, bool internal);
void arena_nthreads_dec(arena_t *arena, bool internal);
size_t arena_extent_sn_next(arena_t *arena);
arena_t *arena_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks);
bool arena_init_huge(void);
bool arena_is_huge(unsigned arena_ind);
arena_t *arena_choose_huge(tsd_t *tsd);
bin_t *arena_bin_choose_lock(tsdn_t *tsdn, arena_t *arena, szind_t binind,
unsigned *binshard);
void arena_boot(sc_data_t *sc_data);
void arena_prefork0(tsdn_t *tsdn, arena_t *arena);
void arena_prefork1(tsdn_t *tsdn, arena_t *arena);
void arena_prefork2(tsdn_t *tsdn, arena_t *arena);
void arena_prefork3(tsdn_t *tsdn, arena_t *arena);
void arena_prefork4(tsdn_t *tsdn, arena_t *arena);
void arena_prefork5(tsdn_t *tsdn, arena_t *arena);
void arena_prefork6(tsdn_t *tsdn, arena_t *arena);
void arena_prefork7(tsdn_t *tsdn, arena_t *arena);
void arena_postfork_parent(tsdn_t *tsdn, arena_t *arena);
void arena_postfork_child(tsdn_t *tsdn, arena_t *arena);
#endif /* JEMALLOC_INTERNAL_ARENA_EXTERNS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_inlines_a.h
Vendored
-57
View File
@@ -1,57 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_INLINES_A_H
#define JEMALLOC_INTERNAL_ARENA_INLINES_A_H
static inline unsigned
arena_ind_get(const arena_t *arena) {
return base_ind_get(arena->base);
}
static inline void
arena_internal_add(arena_t *arena, size_t size) {
atomic_fetch_add_zu(&arena->stats.internal, size, ATOMIC_RELAXED);
}
static inline void
arena_internal_sub(arena_t *arena, size_t size) {
atomic_fetch_sub_zu(&arena->stats.internal, size, ATOMIC_RELAXED);
}
static inline size_t
arena_internal_get(arena_t *arena) {
return atomic_load_zu(&arena->stats.internal, ATOMIC_RELAXED);
}
static inline bool
arena_prof_accum(tsdn_t *tsdn, arena_t *arena, uint64_t accumbytes) {
cassert(config_prof);
if (likely(prof_interval == 0 || !prof_active_get_unlocked())) {
return false;
}
return prof_accum_add(tsdn, &arena->prof_accum, accumbytes);
}
static inline void
percpu_arena_update(tsd_t *tsd, unsigned cpu) {
assert(have_percpu_arena);
arena_t *oldarena = tsd_arena_get(tsd);
assert(oldarena != NULL);
unsigned oldind = arena_ind_get(oldarena);
if (oldind != cpu) {
unsigned newind = cpu;
arena_t *newarena = arena_get(tsd_tsdn(tsd), newind, true);
assert(newarena != NULL);
/* Set new arena/tcache associations. */
arena_migrate(tsd, oldind, newind);
tcache_t *tcache = tcache_get(tsd);
if (tcache != NULL) {
tcache_arena_reassociate(tsd_tsdn(tsd), tcache,
newarena);
}
}
}
#endif /* JEMALLOC_INTERNAL_ARENA_INLINES_A_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_inlines_b.h
Vendored
-427
View File
@@ -1,427 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_INLINES_B_H
#define JEMALLOC_INTERNAL_ARENA_INLINES_B_H
#include "jemalloc/internal/jemalloc_internal_types.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/rtree.h"
#include "jemalloc/internal/sc.h"
#include "jemalloc/internal/sz.h"
#include "jemalloc/internal/ticker.h"
JEMALLOC_ALWAYS_INLINE bool
arena_has_default_hooks(arena_t *arena) {
return (extent_hooks_get(arena) == &extent_hooks_default);
}
JEMALLOC_ALWAYS_INLINE arena_t *
arena_choose_maybe_huge(tsd_t *tsd, arena_t *arena, size_t size) {
if (arena != NULL) {
return arena;
}
/*
* For huge allocations, use the dedicated huge arena if both are true:
* 1) is using auto arena selection (i.e. arena == NULL), and 2) the
* thread is not assigned to a manual arena.
*/
if (unlikely(size >= oversize_threshold)) {
arena_t *tsd_arena = tsd_arena_get(tsd);
if (tsd_arena == NULL || arena_is_auto(tsd_arena)) {
return arena_choose_huge(tsd);
}
}
return arena_choose(tsd, NULL);
}
JEMALLOC_ALWAYS_INLINE prof_tctx_t *
arena_prof_tctx_get(tsdn_t *tsdn, const void *ptr, alloc_ctx_t *alloc_ctx) {
cassert(config_prof);
assert(ptr != NULL);
/* Static check. */
if (alloc_ctx == NULL) {
const extent_t *extent = iealloc(tsdn, ptr);
if (unlikely(!extent_slab_get(extent))) {
return large_prof_tctx_get(tsdn, extent);
}
} else {
if (unlikely(!alloc_ctx->slab)) {
return large_prof_tctx_get(tsdn, iealloc(tsdn, ptr));
}
}
return (prof_tctx_t *)(uintptr_t)1U;
}
JEMALLOC_ALWAYS_INLINE void
arena_prof_tctx_set(tsdn_t *tsdn, const void *ptr, size_t usize,
alloc_ctx_t *alloc_ctx, prof_tctx_t *tctx) {
cassert(config_prof);
assert(ptr != NULL);
/* Static check. */
if (alloc_ctx == NULL) {
extent_t *extent = iealloc(tsdn, ptr);
if (unlikely(!extent_slab_get(extent))) {
large_prof_tctx_set(tsdn, extent, tctx);
}
} else {
if (unlikely(!alloc_ctx->slab)) {
large_prof_tctx_set(tsdn, iealloc(tsdn, ptr), tctx);
}
}
}
static inline void
arena_prof_tctx_reset(tsdn_t *tsdn, const void *ptr, prof_tctx_t *tctx) {
cassert(config_prof);
assert(ptr != NULL);
extent_t *extent = iealloc(tsdn, ptr);
assert(!extent_slab_get(extent));
large_prof_tctx_reset(tsdn, extent);
}
JEMALLOC_ALWAYS_INLINE nstime_t
arena_prof_alloc_time_get(tsdn_t *tsdn, const void *ptr,
alloc_ctx_t *alloc_ctx) {
cassert(config_prof);
assert(ptr != NULL);
extent_t *extent = iealloc(tsdn, ptr);
/*
* Unlike arena_prof_prof_tctx_{get, set}, we only call this once we're
* sure we have a sampled allocation.
*/
assert(!extent_slab_get(extent));
return large_prof_alloc_time_get(extent);
}
JEMALLOC_ALWAYS_INLINE void
arena_prof_alloc_time_set(tsdn_t *tsdn, const void *ptr, alloc_ctx_t *alloc_ctx,
nstime_t t) {
cassert(config_prof);
assert(ptr != NULL);
extent_t *extent = iealloc(tsdn, ptr);
assert(!extent_slab_get(extent));
large_prof_alloc_time_set(extent, t);
}
JEMALLOC_ALWAYS_INLINE void
arena_decay_ticks(tsdn_t *tsdn, arena_t *arena, unsigned nticks) {
tsd_t *tsd;
ticker_t *decay_ticker;
if (unlikely(tsdn_null(tsdn))) {
return;
}
tsd = tsdn_tsd(tsdn);
decay_ticker = decay_ticker_get(tsd, arena_ind_get(arena));
if (unlikely(decay_ticker == NULL)) {
return;
}
if (unlikely(ticker_ticks(decay_ticker, nticks))) {
arena_decay(tsdn, arena, false, false);
}
}
JEMALLOC_ALWAYS_INLINE void
arena_decay_tick(tsdn_t *tsdn, arena_t *arena) {
malloc_mutex_assert_not_owner(tsdn, &arena->decay_dirty.mtx);
malloc_mutex_assert_not_owner(tsdn, &arena->decay_muzzy.mtx);
arena_decay_ticks(tsdn, arena, 1);
}
/* Purge a single extent to retained / unmapped directly. */
JEMALLOC_ALWAYS_INLINE void
arena_decay_extent(tsdn_t *tsdn,arena_t *arena, extent_hooks_t **r_extent_hooks,
extent_t *extent) {
size_t extent_size = extent_size_get(extent);
extent_dalloc_wrapper(tsdn, arena,
r_extent_hooks, extent);
if (config_stats) {
/* Update stats accordingly. */
arena_stats_lock(tsdn, &arena->stats);
arena_stats_add_u64(tsdn, &arena->stats,
&arena->decay_dirty.stats->nmadvise, 1);
arena_stats_add_u64(tsdn, &arena->stats,
&arena->decay_dirty.stats->purged, extent_size >> LG_PAGE);
arena_stats_sub_zu(tsdn, &arena->stats, &arena->stats.mapped,
extent_size);
arena_stats_unlock(tsdn, &arena->stats);
}
}
JEMALLOC_ALWAYS_INLINE void *
arena_malloc(tsdn_t *tsdn, arena_t *arena, size_t size, szind_t ind, bool zero,
tcache_t *tcache, bool slow_path) {
assert(!tsdn_null(tsdn) || tcache == NULL);
if (likely(tcache != NULL)) {
if (likely(size <= SC_SMALL_MAXCLASS)) {
return tcache_alloc_small(tsdn_tsd(tsdn), arena,
tcache, size, ind, zero, slow_path);
}
if (likely(size <= tcache_maxclass)) {
return tcache_alloc_large(tsdn_tsd(tsdn), arena,
tcache, size, ind, zero, slow_path);
}
/* (size > tcache_maxclass) case falls through. */
assert(size > tcache_maxclass);
}
return arena_malloc_hard(tsdn, arena, size, ind, zero);
}
JEMALLOC_ALWAYS_INLINE arena_t *
arena_aalloc(tsdn_t *tsdn, const void *ptr) {
return extent_arena_get(iealloc(tsdn, ptr));
}
JEMALLOC_ALWAYS_INLINE size_t
arena_salloc(tsdn_t *tsdn, const void *ptr) {
assert(ptr != NULL);
rtree_ctx_t rtree_ctx_fallback;
rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback);
szind_t szind = rtree_szind_read(tsdn, &extents_rtree, rtree_ctx,
(uintptr_t)ptr, true);
assert(szind != SC_NSIZES);
return sz_index2size(szind);
}
JEMALLOC_ALWAYS_INLINE size_t
arena_vsalloc(tsdn_t *tsdn, const void *ptr) {
/*
* Return 0 if ptr is not within an extent managed by jemalloc. This
* function has two extra costs relative to isalloc():
* - The rtree calls cannot claim to be dependent lookups, which induces
* rtree lookup load dependencies.
* - The lookup may fail, so there is an extra branch to check for
* failure.
*/
rtree_ctx_t rtree_ctx_fallback;
rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback);
extent_t *extent;
szind_t szind;
if (rtree_extent_szind_read(tsdn, &extents_rtree, rtree_ctx,
(uintptr_t)ptr, false, &extent, &szind)) {
return 0;
}
if (extent == NULL) {
return 0;
}
assert(extent_state_get(extent) == extent_state_active);
/* Only slab members should be looked up via interior pointers. */
assert(extent_addr_get(extent) == ptr || extent_slab_get(extent));
assert(szind != SC_NSIZES);
return sz_index2size(szind);
}
static inline void
arena_dalloc_large_no_tcache(tsdn_t *tsdn, void *ptr, szind_t szind) {
if (config_prof && unlikely(szind < SC_NBINS)) {
arena_dalloc_promoted(tsdn, ptr, NULL, true);
} else {
extent_t *extent = iealloc(tsdn, ptr);
large_dalloc(tsdn, extent);
}
}
static inline void
arena_dalloc_no_tcache(tsdn_t *tsdn, void *ptr) {
assert(ptr != NULL);
rtree_ctx_t rtree_ctx_fallback;
rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn, &rtree_ctx_fallback);
szind_t szind;
bool slab;
rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx, (uintptr_t)ptr,
true, &szind, &slab);
if (config_debug) {
extent_t *extent = rtree_extent_read(tsdn, &extents_rtree,
rtree_ctx, (uintptr_t)ptr, true);
assert(szind == extent_szind_get(extent));
assert(szind < SC_NSIZES);
assert(slab == extent_slab_get(extent));
}
if (likely(slab)) {
/* Small allocation. */
arena_dalloc_small(tsdn, ptr);
} else {
arena_dalloc_large_no_tcache(tsdn, ptr, szind);
}
}
JEMALLOC_ALWAYS_INLINE void
arena_dalloc_large(tsdn_t *tsdn, void *ptr, tcache_t *tcache, szind_t szind,
bool slow_path) {
if (szind < nhbins) {
if (config_prof && unlikely(szind < SC_NBINS)) {
arena_dalloc_promoted(tsdn, ptr, tcache, slow_path);
} else {
tcache_dalloc_large(tsdn_tsd(tsdn), tcache, ptr, szind,
slow_path);
}
} else {
extent_t *extent = iealloc(tsdn, ptr);
large_dalloc(tsdn, extent);
}
}
JEMALLOC_ALWAYS_INLINE void
arena_dalloc(tsdn_t *tsdn, void *ptr, tcache_t *tcache,
alloc_ctx_t *alloc_ctx, bool slow_path) {
assert(!tsdn_null(tsdn) || tcache == NULL);
assert(ptr != NULL);
if (unlikely(tcache == NULL)) {
arena_dalloc_no_tcache(tsdn, ptr);
return;
}
szind_t szind;
bool slab;
rtree_ctx_t *rtree_ctx;
if (alloc_ctx != NULL) {
szind = alloc_ctx->szind;
slab = alloc_ctx->slab;
assert(szind != SC_NSIZES);
} else {
rtree_ctx = tsd_rtree_ctx(tsdn_tsd(tsdn));
rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx,
(uintptr_t)ptr, true, &szind, &slab);
}
if (config_debug) {
rtree_ctx = tsd_rtree_ctx(tsdn_tsd(tsdn));
extent_t *extent = rtree_extent_read(tsdn, &extents_rtree,
rtree_ctx, (uintptr_t)ptr, true);
assert(szind == extent_szind_get(extent));
assert(szind < SC_NSIZES);
assert(slab == extent_slab_get(extent));
}
if (likely(slab)) {
/* Small allocation. */
tcache_dalloc_small(tsdn_tsd(tsdn), tcache, ptr, szind,
slow_path);
} else {
arena_dalloc_large(tsdn, ptr, tcache, szind, slow_path);
}
}
static inline void
arena_sdalloc_no_tcache(tsdn_t *tsdn, void *ptr, size_t size) {
assert(ptr != NULL);
assert(size <= SC_LARGE_MAXCLASS);
szind_t szind;
bool slab;
if (!config_prof || !opt_prof) {
/*
* There is no risk of being confused by a promoted sampled
* object, so base szind and slab on the given size.
*/
szind = sz_size2index(size);
slab = (szind < SC_NBINS);
}
if ((config_prof && opt_prof) || config_debug) {
rtree_ctx_t rtree_ctx_fallback;
rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn,
&rtree_ctx_fallback);
rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx,
(uintptr_t)ptr, true, &szind, &slab);
assert(szind == sz_size2index(size));
assert((config_prof && opt_prof) || slab == (szind < SC_NBINS));
if (config_debug) {
extent_t *extent = rtree_extent_read(tsdn,
&extents_rtree, rtree_ctx, (uintptr_t)ptr, true);
assert(szind == extent_szind_get(extent));
assert(slab == extent_slab_get(extent));
}
}
if (likely(slab)) {
/* Small allocation. */
arena_dalloc_small(tsdn, ptr);
} else {
arena_dalloc_large_no_tcache(tsdn, ptr, szind);
}
}
JEMALLOC_ALWAYS_INLINE void
arena_sdalloc(tsdn_t *tsdn, void *ptr, size_t size, tcache_t *tcache,
alloc_ctx_t *alloc_ctx, bool slow_path) {
assert(!tsdn_null(tsdn) || tcache == NULL);
assert(ptr != NULL);
assert(size <= SC_LARGE_MAXCLASS);
if (unlikely(tcache == NULL)) {
arena_sdalloc_no_tcache(tsdn, ptr, size);
return;
}
szind_t szind;
bool slab;
alloc_ctx_t local_ctx;
if (config_prof && opt_prof) {
if (alloc_ctx == NULL) {
/* Uncommon case and should be a static check. */
rtree_ctx_t rtree_ctx_fallback;
rtree_ctx_t *rtree_ctx = tsdn_rtree_ctx(tsdn,
&rtree_ctx_fallback);
rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx,
(uintptr_t)ptr, true, &local_ctx.szind,
&local_ctx.slab);
assert(local_ctx.szind == sz_size2index(size));
alloc_ctx = &local_ctx;
}
slab = alloc_ctx->slab;
szind = alloc_ctx->szind;
} else {
/*
* There is no risk of being confused by a promoted sampled
* object, so base szind and slab on the given size.
*/
szind = sz_size2index(size);
slab = (szind < SC_NBINS);
}
if (config_debug) {
rtree_ctx_t *rtree_ctx = tsd_rtree_ctx(tsdn_tsd(tsdn));
rtree_szind_slab_read(tsdn, &extents_rtree, rtree_ctx,
(uintptr_t)ptr, true, &szind, &slab);
extent_t *extent = rtree_extent_read(tsdn,
&extents_rtree, rtree_ctx, (uintptr_t)ptr, true);
assert(szind == extent_szind_get(extent));
assert(slab == extent_slab_get(extent));
}
if (likely(slab)) {
/* Small allocation. */
tcache_dalloc_small(tsdn_tsd(tsdn), tcache, ptr, szind,
slow_path);
} else {
arena_dalloc_large(tsdn, ptr, tcache, szind, slow_path);
}
}
#endif /* JEMALLOC_INTERNAL_ARENA_INLINES_B_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_stats.h
Vendored
-271
View File
@@ -1,271 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_STATS_H
#define JEMALLOC_INTERNAL_ARENA_STATS_H
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/mutex_prof.h"
#include "jemalloc/internal/sc.h"
JEMALLOC_DIAGNOSTIC_DISABLE_SPURIOUS
/*
* In those architectures that support 64-bit atomics, we use atomic updates for
* our 64-bit values. Otherwise, we use a plain uint64_t and synchronize
* externally.
*/
#ifdef JEMALLOC_ATOMIC_U64
typedef atomic_u64_t arena_stats_u64_t;
#else
/* Must hold the arena stats mutex while reading atomically. */
typedef uint64_t arena_stats_u64_t;
#endif
typedef struct arena_stats_large_s arena_stats_large_t;
struct arena_stats_large_s {
/*
* Total number of allocation/deallocation requests served directly by
* the arena.
*/
arena_stats_u64_t nmalloc;
arena_stats_u64_t ndalloc;
/*
* Number of allocation requests that correspond to this size class.
* This includes requests served by tcache, though tcache only
* periodically merges into this counter.
*/
arena_stats_u64_t nrequests; /* Partially derived. */
/*
* Number of tcache fills / flushes for large (similarly, periodically
* merged). Note that there is no large tcache batch-fill currently
* (i.e. only fill 1 at a time); however flush may be batched.
*/
arena_stats_u64_t nfills; /* Partially derived. */
arena_stats_u64_t nflushes; /* Partially derived. */
/* Current number of allocations of this size class. */
size_t curlextents; /* Derived. */
};
typedef struct arena_stats_decay_s arena_stats_decay_t;
struct arena_stats_decay_s {
/* Total number of purge sweeps. */
arena_stats_u64_t npurge;
/* Total number of madvise calls made. */
arena_stats_u64_t nmadvise;
/* Total number of pages purged. */
arena_stats_u64_t purged;
};
typedef struct arena_stats_extents_s arena_stats_extents_t;
struct arena_stats_extents_s {
/*
* Stats for a given index in the range [0, SC_NPSIZES] in an extents_t.
* We track both bytes and # of extents: two extents in the same bucket
* may have different sizes if adjacent size classes differ by more than
* a page, so bytes cannot always be derived from # of extents.
*/
atomic_zu_t ndirty;
atomic_zu_t dirty_bytes;
atomic_zu_t nmuzzy;
atomic_zu_t muzzy_bytes;
atomic_zu_t nretained;
atomic_zu_t retained_bytes;
};
/*
* Arena stats. Note that fields marked "derived" are not directly maintained
* within the arena code; rather their values are derived during stats merge
* requests.
*/
typedef struct arena_stats_s arena_stats_t;
struct arena_stats_s {
#ifndef JEMALLOC_ATOMIC_U64
malloc_mutex_t mtx;
#endif
/* Number of bytes currently mapped, excluding retained memory. */
atomic_zu_t mapped; /* Partially derived. */
/*
* Number of unused virtual memory bytes currently retained. Retained
* bytes are technically mapped (though always decommitted or purged),
* but they are excluded from the mapped statistic (above).
*/
atomic_zu_t retained; /* Derived. */
/* Number of extent_t structs allocated by base, but not being used. */
atomic_zu_t extent_avail;
arena_stats_decay_t decay_dirty;
arena_stats_decay_t decay_muzzy;
atomic_zu_t base; /* Derived. */
atomic_zu_t internal;
atomic_zu_t resident; /* Derived. */
atomic_zu_t metadata_thp;
atomic_zu_t allocated_large; /* Derived. */
arena_stats_u64_t nmalloc_large; /* Derived. */
arena_stats_u64_t ndalloc_large; /* Derived. */
arena_stats_u64_t nfills_large; /* Derived. */
arena_stats_u64_t nflushes_large; /* Derived. */
arena_stats_u64_t nrequests_large; /* Derived. */
/* VM space had to be leaked (undocumented). Normally 0. */
atomic_zu_t abandoned_vm;
/* Number of bytes cached in tcache associated with this arena. */
atomic_zu_t tcache_bytes; /* Derived. */
mutex_prof_data_t mutex_prof_data[mutex_prof_num_arena_mutexes];
/* One element for each large size class. */
arena_stats_large_t lstats[SC_NSIZES - SC_NBINS];
/* Arena uptime. */
nstime_t uptime;
};
static inline bool
arena_stats_init(tsdn_t *tsdn, arena_stats_t *arena_stats) {
if (config_debug) {
for (size_t i = 0; i < sizeof(arena_stats_t); i++) {
assert(((char *)arena_stats)[i] == 0);
}
}
#ifndef JEMALLOC_ATOMIC_U64
if (malloc_mutex_init(&arena_stats->mtx, "arena_stats",
WITNESS_RANK_ARENA_STATS, malloc_mutex_rank_exclusive)) {
return true;
}
#endif
/* Memory is zeroed, so there is no need to clear stats. */
return false;
}
static inline void
arena_stats_lock(tsdn_t *tsdn, arena_stats_t *arena_stats) {
#ifndef JEMALLOC_ATOMIC_U64
malloc_mutex_lock(tsdn, &arena_stats->mtx);
#endif
}
static inline void
arena_stats_unlock(tsdn_t *tsdn, arena_stats_t *arena_stats) {
#ifndef JEMALLOC_ATOMIC_U64
malloc_mutex_unlock(tsdn, &arena_stats->mtx);
#endif
}
static inline uint64_t
arena_stats_read_u64(tsdn_t *tsdn, arena_stats_t *arena_stats,
arena_stats_u64_t *p) {
#ifdef JEMALLOC_ATOMIC_U64
return atomic_load_u64(p, ATOMIC_RELAXED);
#else
malloc_mutex_assert_owner(tsdn, &arena_stats->mtx);
return *p;
#endif
}
static inline void
arena_stats_add_u64(tsdn_t *tsdn, arena_stats_t *arena_stats,
arena_stats_u64_t *p, uint64_t x) {
#ifdef JEMALLOC_ATOMIC_U64
atomic_fetch_add_u64(p, x, ATOMIC_RELAXED);
#else
malloc_mutex_assert_owner(tsdn, &arena_stats->mtx);
*p += x;
#endif
}
static inline void
arena_stats_sub_u64(tsdn_t *tsdn, arena_stats_t *arena_stats,
arena_stats_u64_t *p, uint64_t x) {
#ifdef JEMALLOC_ATOMIC_U64
uint64_t r = atomic_fetch_sub_u64(p, x, ATOMIC_RELAXED);
assert(r - x <= r);
#else
malloc_mutex_assert_owner(tsdn, &arena_stats->mtx);
*p -= x;
assert(*p + x >= *p);
#endif
}
/*
* Non-atomically sets *dst += src. *dst needs external synchronization.
* This lets us avoid the cost of a fetch_add when its unnecessary (note that
* the types here are atomic).
*/
static inline void
arena_stats_accum_u64(arena_stats_u64_t *dst, uint64_t src) {
#ifdef JEMALLOC_ATOMIC_U64
uint64_t cur_dst = atomic_load_u64(dst, ATOMIC_RELAXED);
atomic_store_u64(dst, src + cur_dst, ATOMIC_RELAXED);
#else
*dst += src;
#endif
}
static inline size_t
arena_stats_read_zu(tsdn_t *tsdn, arena_stats_t *arena_stats,
atomic_zu_t *p) {
#ifdef JEMALLOC_ATOMIC_U64
return atomic_load_zu(p, ATOMIC_RELAXED);
#else
malloc_mutex_assert_owner(tsdn, &arena_stats->mtx);
return atomic_load_zu(p, ATOMIC_RELAXED);
#endif
}
static inline void
arena_stats_add_zu(tsdn_t *tsdn, arena_stats_t *arena_stats,
atomic_zu_t *p, size_t x) {
#ifdef JEMALLOC_ATOMIC_U64
atomic_fetch_add_zu(p, x, ATOMIC_RELAXED);
#else
malloc_mutex_assert_owner(tsdn, &arena_stats->mtx);
size_t cur = atomic_load_zu(p, ATOMIC_RELAXED);
atomic_store_zu(p, cur + x, ATOMIC_RELAXED);
#endif
}
static inline void
arena_stats_sub_zu(tsdn_t *tsdn, arena_stats_t *arena_stats,
atomic_zu_t *p, size_t x) {
#ifdef JEMALLOC_ATOMIC_U64
size_t r = atomic_fetch_sub_zu(p, x, ATOMIC_RELAXED);
assert(r - x <= r);
#else
malloc_mutex_assert_owner(tsdn, &arena_stats->mtx);
size_t cur = atomic_load_zu(p, ATOMIC_RELAXED);
atomic_store_zu(p, cur - x, ATOMIC_RELAXED);
#endif
}
/* Like the _u64 variant, needs an externally synchronized *dst. */
static inline void
arena_stats_accum_zu(atomic_zu_t *dst, size_t src) {
size_t cur_dst = atomic_load_zu(dst, ATOMIC_RELAXED);
atomic_store_zu(dst, src + cur_dst, ATOMIC_RELAXED);
}
static inline void
arena_stats_large_flush_nrequests_add(tsdn_t *tsdn, arena_stats_t *arena_stats,
szind_t szind, uint64_t nrequests) {
arena_stats_lock(tsdn, arena_stats);
arena_stats_large_t *lstats = &arena_stats->lstats[szind - SC_NBINS];
arena_stats_add_u64(tsdn, arena_stats, &lstats->nrequests, nrequests);
arena_stats_add_u64(tsdn, arena_stats, &lstats->nflushes, 1);
arena_stats_unlock(tsdn, arena_stats);
}
static inline void
arena_stats_mapped_add(tsdn_t *tsdn, arena_stats_t *arena_stats, size_t size) {
arena_stats_lock(tsdn, arena_stats);
arena_stats_add_zu(tsdn, arena_stats, &arena_stats->mapped, size);
arena_stats_unlock(tsdn, arena_stats);
}
#endif /* JEMALLOC_INTERNAL_ARENA_STATS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_structs_a.h
Vendored
-11
View File
@@ -1,11 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_STRUCTS_A_H
#define JEMALLOC_INTERNAL_ARENA_STRUCTS_A_H
#include "jemalloc/internal/bitmap.h"
struct arena_slab_data_s {
/* Per region allocated/deallocated bitmap. */
bitmap_t bitmap[BITMAP_GROUPS_MAX];
};
#endif /* JEMALLOC_INTERNAL_ARENA_STRUCTS_A_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_structs_b.h
Vendored
-232
View File
@@ -1,232 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_STRUCTS_B_H
#define JEMALLOC_INTERNAL_ARENA_STRUCTS_B_H
#include "jemalloc/internal/arena_stats.h"
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/bin.h"
#include "jemalloc/internal/bitmap.h"
#include "jemalloc/internal/extent_dss.h"
#include "jemalloc/internal/jemalloc_internal_types.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/nstime.h"
#include "jemalloc/internal/ql.h"
#include "jemalloc/internal/sc.h"
#include "jemalloc/internal/smoothstep.h"
#include "jemalloc/internal/ticker.h"
struct arena_decay_s {
/* Synchronizes all non-atomic fields. */
malloc_mutex_t mtx;
/*
* True if a thread is currently purging the extents associated with
* this decay structure.
*/
bool purging;
/*
* Approximate time in milliseconds from the creation of a set of unused
* dirty pages until an equivalent set of unused dirty pages is purged
* and/or reused.
*/
atomic_zd_t time_ms;
/* time / SMOOTHSTEP_NSTEPS. */
nstime_t interval;
/*
* Time at which the current decay interval logically started. We do
* not actually advance to a new epoch until sometime after it starts
* because of scheduling and computation delays, and it is even possible
* to completely skip epochs. In all cases, during epoch advancement we
* merge all relevant activity into the most recently recorded epoch.
*/
nstime_t epoch;
/* Deadline randomness generator. */
uint64_t jitter_state;
/*
* Deadline for current epoch. This is the sum of interval and per
* epoch jitter which is a uniform random variable in [0..interval).
* Epochs always advance by precise multiples of interval, but we
* randomize the deadline to reduce the likelihood of arenas purging in
* lockstep.
*/
nstime_t deadline;
/*
* Number of unpurged pages at beginning of current epoch. During epoch
* advancement we use the delta between arena->decay_*.nunpurged and
* extents_npages_get(&arena->extents_*) to determine how many dirty
* pages, if any, were generated.
*/
size_t nunpurged;
/*
* Trailing log of how many unused dirty pages were generated during
* each of the past SMOOTHSTEP_NSTEPS decay epochs, where the last
* element is the most recent epoch. Corresponding epoch times are
* relative to epoch.
*/
size_t backlog[SMOOTHSTEP_NSTEPS];
/*
* Pointer to associated stats. These stats are embedded directly in
* the arena's stats due to how stats structures are shared between the
* arena and ctl code.
*
* Synchronization: Same as associated arena's stats field. */
arena_stats_decay_t *stats;
/* Peak number of pages in associated extents. Used for debug only. */
uint64_t ceil_npages;
};
struct arena_s {
/*
* Number of threads currently assigned to this arena. Each thread has
* two distinct assignments, one for application-serving allocation, and
* the other for internal metadata allocation. Internal metadata must
* not be allocated from arenas explicitly created via the arenas.create
* mallctl, because the arena.<i>.reset mallctl indiscriminately
* discards all allocations for the affected arena.
*
* 0: Application allocation.
* 1: Internal metadata allocation.
*
* Synchronization: atomic.
*/
atomic_u_t nthreads[2];
/* Next bin shard for binding new threads. Synchronization: atomic. */
atomic_u_t binshard_next;
/*
* When percpu_arena is enabled, to amortize the cost of reading /
* updating the current CPU id, track the most recent thread accessing
* this arena, and only read CPU if there is a mismatch.
*/
tsdn_t *last_thd;
/* Synchronization: internal. */
arena_stats_t stats;
/*
* Lists of tcaches and cache_bin_array_descriptors for extant threads
* associated with this arena. Stats from these are merged
* incrementally, and at exit if opt_stats_print is enabled.
*
* Synchronization: tcache_ql_mtx.
*/
ql_head(tcache_t) tcache_ql;
ql_head(cache_bin_array_descriptor_t) cache_bin_array_descriptor_ql;
malloc_mutex_t tcache_ql_mtx;
/* Synchronization: internal. */
prof_accum_t prof_accum;
/*
* PRNG state for cache index randomization of large allocation base
* pointers.
*
* Synchronization: atomic.
*/
atomic_zu_t offset_state;
/*
* Extent serial number generator state.
*
* Synchronization: atomic.
*/
atomic_zu_t extent_sn_next;
/*
* Represents a dss_prec_t, but atomically.
*
* Synchronization: atomic.
*/
atomic_u_t dss_prec;
/*
* Number of pages in active extents.
*
* Synchronization: atomic.
*/
atomic_zu_t nactive;
/*
* Extant large allocations.
*
* Synchronization: large_mtx.
*/
extent_list_t large;
/* Synchronizes all large allocation/update/deallocation. */
malloc_mutex_t large_mtx;
/*
* Collections of extents that were previously allocated. These are
* used when allocating extents, in an attempt to re-use address space.
*
* Synchronization: internal.
*/
extents_t extents_dirty;
extents_t extents_muzzy;
extents_t extents_retained;
/*
* Decay-based purging state, responsible for scheduling extent state
* transitions.
*
* Synchronization: internal.
*/
arena_decay_t decay_dirty; /* dirty --> muzzy */
arena_decay_t decay_muzzy; /* muzzy --> retained */
/*
* Next extent size class in a growing series to use when satisfying a
* request via the extent hooks (only if opt_retain). This limits the
* number of disjoint virtual memory ranges so that extent merging can
* be effective even if multiple arenas' extent allocation requests are
* highly interleaved.
*
* retain_grow_limit is the max allowed size ind to expand (unless the
* required size is greater). Default is no limit, and controlled
* through mallctl only.
*
* Synchronization: extent_grow_mtx
*/
pszind_t extent_grow_next;
pszind_t retain_grow_limit;
malloc_mutex_t extent_grow_mtx;
/*
* Available extent structures that were allocated via
* base_alloc_extent().
*
* Synchronization: extent_avail_mtx.
*/
extent_tree_t extent_avail;
atomic_zu_t extent_avail_cnt;
malloc_mutex_t extent_avail_mtx;
/*
* bins is used to store heaps of free regions.
*
* Synchronization: internal.
*/
bins_t bins[SC_NBINS];
/*
* Base allocator, from which arena metadata are allocated.
*
* Synchronization: internal.
*/
base_t *base;
/* Used to determine uptime. Read-only after initialization. */
nstime_t create_time;
};
/* Used in conjunction with tsd for fast arena-related context lookup. */
struct arena_tdata_s {
ticker_t decay_ticker;
};
/* Used to pass rtree lookup context down the path. */
struct alloc_ctx_s {
szind_t szind;
bool slab;
};
#endif /* JEMALLOC_INTERNAL_ARENA_STRUCTS_B_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/arena_types.h
Vendored
-51
View File
@@ -1,51 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ARENA_TYPES_H
#define JEMALLOC_INTERNAL_ARENA_TYPES_H
#include "jemalloc/internal/sc.h"
/* Maximum number of regions in one slab. */
#define LG_SLAB_MAXREGS (LG_PAGE - SC_LG_TINY_MIN)
#define SLAB_MAXREGS (1U << LG_SLAB_MAXREGS)
/* Default decay times in milliseconds. */
#define DIRTY_DECAY_MS_DEFAULT ZD(10 * 1000)
#define MUZZY_DECAY_MS_DEFAULT (0)
/* Number of event ticks between time checks. */
#define DECAY_NTICKS_PER_UPDATE 1000
typedef struct arena_slab_data_s arena_slab_data_t;
typedef struct arena_decay_s arena_decay_t;
typedef struct arena_s arena_t;
typedef struct arena_tdata_s arena_tdata_t;
typedef struct alloc_ctx_s alloc_ctx_t;
typedef enum {
percpu_arena_mode_names_base = 0, /* Used for options processing. */
/*
* *_uninit are used only during bootstrapping, and must correspond
* to initialized variant plus percpu_arena_mode_enabled_base.
*/
percpu_arena_uninit = 0,
per_phycpu_arena_uninit = 1,
/* All non-disabled modes must come after percpu_arena_disabled. */
percpu_arena_disabled = 2,
percpu_arena_mode_names_limit = 3, /* Used for options processing. */
percpu_arena_mode_enabled_base = 3,
percpu_arena = 3,
per_phycpu_arena = 4 /* Hyper threads share arena. */
} percpu_arena_mode_t;
#define PERCPU_ARENA_ENABLED(m) ((m) >= percpu_arena_mode_enabled_base)
#define PERCPU_ARENA_DEFAULT percpu_arena_disabled
/*
* When allocation_size >= oversize_threshold, use the dedicated huge arena
* (unless have explicitly spicified arena index). 0 disables the feature.
*/
#define OVERSIZE_THRESHOLD_DEFAULT (8 << 20)
#endif /* JEMALLOC_INTERNAL_ARENA_TYPES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/assert.h
Vendored
-56
View File
@@ -1,56 +0,0 @@
#include "jemalloc/internal/malloc_io.h"
#include "jemalloc/internal/util.h"
/*
* Define a custom assert() in order to reduce the chances of deadlock during
* assertion failure.
*/
#ifndef assert
#define assert(e) do { \
if (unlikely(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(); \
} \
unreachable(); \
} 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 (unlikely(config_debug && !(e))) { \
not_implemented(); \
} \
} while (0)
#endif
/* Use to assert a particular configuration, e.g., cassert(config_debug). */
#ifndef cassert
#define cassert(c) do { \
if (unlikely(!(c))) { \
not_reached(); \
} \
} while (0)
#endif
deps/jemalloc-5.2.1/include/jemalloc/internal/atomic.h
Vendored
-86
View File
@@ -1,86 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ATOMIC_H
#define JEMALLOC_INTERNAL_ATOMIC_H
#define ATOMIC_INLINE JEMALLOC_ALWAYS_INLINE
#define JEMALLOC_U8_ATOMICS
#if defined(JEMALLOC_GCC_ATOMIC_ATOMICS)
# include "jemalloc/internal/atomic_gcc_atomic.h"
# if !defined(JEMALLOC_GCC_U8_ATOMIC_ATOMICS)
# undef JEMALLOC_U8_ATOMICS
# endif
#elif defined(JEMALLOC_GCC_SYNC_ATOMICS)
# include "jemalloc/internal/atomic_gcc_sync.h"
# if !defined(JEMALLOC_GCC_U8_SYNC_ATOMICS)
# undef JEMALLOC_U8_ATOMICS
# endif
#elif defined(_MSC_VER)
# include "jemalloc/internal/atomic_msvc.h"
#elif defined(JEMALLOC_C11_ATOMICS)
# include "jemalloc/internal/atomic_c11.h"
#else
# error "Don't have atomics implemented on this platform."
#endif
/*
* This header gives more or less a backport of C11 atomics. The user can write
* JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_sizeof_type); to generate
* counterparts of the C11 atomic functions for type, as so:
* JEMALLOC_GENERATE_ATOMICS(int *, pi, 3);
* and then write things like:
* int *some_ptr;
* atomic_pi_t atomic_ptr_to_int;
* atomic_store_pi(&atomic_ptr_to_int, some_ptr, ATOMIC_RELAXED);
* int *prev_value = atomic_exchange_pi(&ptr_to_int, NULL, ATOMIC_ACQ_REL);
* assert(some_ptr == prev_value);
* and expect things to work in the obvious way.
*
* Also included (with naming differences to avoid conflicts with the standard
* library):
* atomic_fence(atomic_memory_order_t) (mimics C11's atomic_thread_fence).
* ATOMIC_INIT (mimics C11's ATOMIC_VAR_INIT).
*/
/*
* Pure convenience, so that we don't have to type "atomic_memory_order_"
* quite so often.
*/
#define ATOMIC_RELAXED atomic_memory_order_relaxed
#define ATOMIC_ACQUIRE atomic_memory_order_acquire
#define ATOMIC_RELEASE atomic_memory_order_release
#define ATOMIC_ACQ_REL atomic_memory_order_acq_rel
#define ATOMIC_SEQ_CST atomic_memory_order_seq_cst
/*
* Not all platforms have 64-bit atomics. If we do, this #define exposes that
* fact.
*/
#if (LG_SIZEOF_PTR == 3 || LG_SIZEOF_INT == 3)
# define JEMALLOC_ATOMIC_U64
#endif
JEMALLOC_GENERATE_ATOMICS(void *, p, LG_SIZEOF_PTR)
/*
* There's no actual guarantee that sizeof(bool) == 1, but it's true on the only
* platform that actually needs to know the size, MSVC.
*/
JEMALLOC_GENERATE_ATOMICS(bool, b, 0)
JEMALLOC_GENERATE_INT_ATOMICS(unsigned, u, LG_SIZEOF_INT)
JEMALLOC_GENERATE_INT_ATOMICS(size_t, zu, LG_SIZEOF_PTR)
JEMALLOC_GENERATE_INT_ATOMICS(ssize_t, zd, LG_SIZEOF_PTR)
JEMALLOC_GENERATE_INT_ATOMICS(uint8_t, u8, 0)
JEMALLOC_GENERATE_INT_ATOMICS(uint32_t, u32, 2)
#ifdef JEMALLOC_ATOMIC_U64
JEMALLOC_GENERATE_INT_ATOMICS(uint64_t, u64, 3)
#endif
#undef ATOMIC_INLINE
#endif /* JEMALLOC_INTERNAL_ATOMIC_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/atomic_c11.h
Vendored
-97
View File
@@ -1,97 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ATOMIC_C11_H
#define JEMALLOC_INTERNAL_ATOMIC_C11_H
#include <stdatomic.h>
#define ATOMIC_INIT(...) ATOMIC_VAR_INIT(__VA_ARGS__)
#define atomic_memory_order_t memory_order
#define atomic_memory_order_relaxed memory_order_relaxed
#define atomic_memory_order_acquire memory_order_acquire
#define atomic_memory_order_release memory_order_release
#define atomic_memory_order_acq_rel memory_order_acq_rel
#define atomic_memory_order_seq_cst memory_order_seq_cst
#define atomic_fence atomic_thread_fence
#define JEMALLOC_GENERATE_ATOMICS(type, short_type, \
/* unused */ lg_size) \
typedef _Atomic(type) atomic_##short_type##_t; \
\
ATOMIC_INLINE type \
atomic_load_##short_type(const atomic_##short_type##_t *a, \
atomic_memory_order_t mo) { \
/* \
* A strict interpretation of the C standard prevents \
* atomic_load from taking a const argument, but it's \
* convenient for our purposes. This cast is a workaround. \
*/ \
atomic_##short_type##_t* a_nonconst = \
(atomic_##short_type##_t*)a; \
return atomic_load_explicit(a_nonconst, mo); \
} \
\
ATOMIC_INLINE void \
atomic_store_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
atomic_store_explicit(a, val, mo); \
} \
\
ATOMIC_INLINE type \
atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return atomic_exchange_explicit(a, val, mo); \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \
type *expected, type desired, atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
return atomic_compare_exchange_weak_explicit(a, expected, \
desired, success_mo, failure_mo); \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \
type *expected, type desired, atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
return atomic_compare_exchange_strong_explicit(a, expected, \
desired, success_mo, failure_mo); \
}
/*
* Integral types have some special operations available that non-integral ones
* lack.
*/
#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, \
/* unused */ lg_size) \
JEMALLOC_GENERATE_ATOMICS(type, short_type, /* unused */ lg_size) \
\
ATOMIC_INLINE type \
atomic_fetch_add_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return atomic_fetch_add_explicit(a, val, mo); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return atomic_fetch_sub_explicit(a, val, mo); \
} \
ATOMIC_INLINE type \
atomic_fetch_and_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return atomic_fetch_and_explicit(a, val, mo); \
} \
ATOMIC_INLINE type \
atomic_fetch_or_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return atomic_fetch_or_explicit(a, val, mo); \
} \
ATOMIC_INLINE type \
atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return atomic_fetch_xor_explicit(a, val, mo); \
}
#endif /* JEMALLOC_INTERNAL_ATOMIC_C11_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/atomic_gcc_atomic.h
Vendored
-129
View File
@@ -1,129 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ATOMIC_GCC_ATOMIC_H
#define JEMALLOC_INTERNAL_ATOMIC_GCC_ATOMIC_H
#include "jemalloc/internal/assert.h"
#define ATOMIC_INIT(...) {__VA_ARGS__}
typedef enum {
atomic_memory_order_relaxed,
atomic_memory_order_acquire,
atomic_memory_order_release,
atomic_memory_order_acq_rel,
atomic_memory_order_seq_cst
} atomic_memory_order_t;
ATOMIC_INLINE int
atomic_enum_to_builtin(atomic_memory_order_t mo) {
switch (mo) {
case atomic_memory_order_relaxed:
return __ATOMIC_RELAXED;
case atomic_memory_order_acquire:
return __ATOMIC_ACQUIRE;
case atomic_memory_order_release:
return __ATOMIC_RELEASE;
case atomic_memory_order_acq_rel:
return __ATOMIC_ACQ_REL;
case atomic_memory_order_seq_cst:
return __ATOMIC_SEQ_CST;
}
/* Can't happen; the switch is exhaustive. */
not_reached();
}
ATOMIC_INLINE void
atomic_fence(atomic_memory_order_t mo) {
__atomic_thread_fence(atomic_enum_to_builtin(mo));
}
#define JEMALLOC_GENERATE_ATOMICS(type, short_type, \
/* unused */ lg_size) \
typedef struct { \
type repr; \
} atomic_##short_type##_t; \
\
ATOMIC_INLINE type \
atomic_load_##short_type(const atomic_##short_type##_t *a, \
atomic_memory_order_t mo) { \
type result; \
__atomic_load(&a->repr, &result, atomic_enum_to_builtin(mo)); \
return result; \
} \
\
ATOMIC_INLINE void \
atomic_store_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
__atomic_store(&a->repr, &val, atomic_enum_to_builtin(mo)); \
} \
\
ATOMIC_INLINE type \
atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
type result; \
__atomic_exchange(&a->repr, &val, &result, \
atomic_enum_to_builtin(mo)); \
return result; \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \
UNUSED type *expected, type desired, \
atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
return __atomic_compare_exchange(&a->repr, expected, &desired, \
true, atomic_enum_to_builtin(success_mo), \
atomic_enum_to_builtin(failure_mo)); \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \
UNUSED type *expected, type desired, \
atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
return __atomic_compare_exchange(&a->repr, expected, &desired, \
false, \
atomic_enum_to_builtin(success_mo), \
atomic_enum_to_builtin(failure_mo)); \
}
#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, \
/* unused */ lg_size) \
JEMALLOC_GENERATE_ATOMICS(type, short_type, /* unused */ lg_size) \
\
ATOMIC_INLINE type \
atomic_fetch_add_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __atomic_fetch_add(&a->repr, val, \
atomic_enum_to_builtin(mo)); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __atomic_fetch_sub(&a->repr, val, \
atomic_enum_to_builtin(mo)); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_and_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __atomic_fetch_and(&a->repr, val, \
atomic_enum_to_builtin(mo)); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_or_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __atomic_fetch_or(&a->repr, val, \
atomic_enum_to_builtin(mo)); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __atomic_fetch_xor(&a->repr, val, \
atomic_enum_to_builtin(mo)); \
}
#endif /* JEMALLOC_INTERNAL_ATOMIC_GCC_ATOMIC_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/atomic_gcc_sync.h
Vendored
-195
View File
@@ -1,195 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ATOMIC_GCC_SYNC_H
#define JEMALLOC_INTERNAL_ATOMIC_GCC_SYNC_H
#define ATOMIC_INIT(...) {__VA_ARGS__}
typedef enum {
atomic_memory_order_relaxed,
atomic_memory_order_acquire,
atomic_memory_order_release,
atomic_memory_order_acq_rel,
atomic_memory_order_seq_cst
} atomic_memory_order_t;
ATOMIC_INLINE void
atomic_fence(atomic_memory_order_t mo) {
/* Easy cases first: no barrier, and full barrier. */
if (mo == atomic_memory_order_relaxed) {
asm volatile("" ::: "memory");
return;
}
if (mo == atomic_memory_order_seq_cst) {
asm volatile("" ::: "memory");
__sync_synchronize();
asm volatile("" ::: "memory");
return;
}
asm volatile("" ::: "memory");
# if defined(__i386__) || defined(__x86_64__)
/* This is implicit on x86. */
# elif defined(__ppc64__)
asm volatile("lwsync");
# elif defined(__ppc__)
asm volatile("sync");
# elif defined(__sparc__) && defined(__arch64__)
if (mo == atomic_memory_order_acquire) {
asm volatile("membar #LoadLoad | #LoadStore");
} else if (mo == atomic_memory_order_release) {
asm volatile("membar #LoadStore | #StoreStore");
} else {
asm volatile("membar #LoadLoad | #LoadStore | #StoreStore");
}
# else
__sync_synchronize();
# endif
asm volatile("" ::: "memory");
}
/*
* A correct implementation of seq_cst loads and stores on weakly ordered
* architectures could do either of the following:
* 1. store() is weak-fence -> store -> strong fence, load() is load ->
* strong-fence.
* 2. store() is strong-fence -> store, load() is strong-fence -> load ->
* weak-fence.
* The tricky thing is, load() and store() above can be the load or store
* portions of a gcc __sync builtin, so we have to follow GCC's lead, which
* means going with strategy 2.
* On strongly ordered architectures, the natural strategy is to stick a strong
* fence after seq_cst stores, and have naked loads. So we want the strong
* fences in different places on different architectures.
* atomic_pre_sc_load_fence and atomic_post_sc_store_fence allow us to
* accomplish this.
*/
ATOMIC_INLINE void
atomic_pre_sc_load_fence() {
# if defined(__i386__) || defined(__x86_64__) || \
(defined(__sparc__) && defined(__arch64__))
atomic_fence(atomic_memory_order_relaxed);
# else
atomic_fence(atomic_memory_order_seq_cst);
# endif
}
ATOMIC_INLINE void
atomic_post_sc_store_fence() {
# if defined(__i386__) || defined(__x86_64__) || \
(defined(__sparc__) && defined(__arch64__))
atomic_fence(atomic_memory_order_seq_cst);
# else
atomic_fence(atomic_memory_order_relaxed);
# endif
}
#define JEMALLOC_GENERATE_ATOMICS(type, short_type, \
/* unused */ lg_size) \
typedef struct { \
type volatile repr; \
} atomic_##short_type##_t; \
\
ATOMIC_INLINE type \
atomic_load_##short_type(const atomic_##short_type##_t *a, \
atomic_memory_order_t mo) { \
if (mo == atomic_memory_order_seq_cst) { \
atomic_pre_sc_load_fence(); \
} \
type result = a->repr; \
if (mo != atomic_memory_order_relaxed) { \
atomic_fence(atomic_memory_order_acquire); \
} \
return result; \
} \
\
ATOMIC_INLINE void \
atomic_store_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
if (mo != atomic_memory_order_relaxed) { \
atomic_fence(atomic_memory_order_release); \
} \
a->repr = val; \
if (mo == atomic_memory_order_seq_cst) { \
atomic_post_sc_store_fence(); \
} \
} \
\
ATOMIC_INLINE type \
atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
/* \
* Because of FreeBSD, we care about gcc 4.2, which doesn't have\
* an atomic exchange builtin. We fake it with a CAS loop. \
*/ \
while (true) { \
type old = a->repr; \
if (__sync_bool_compare_and_swap(&a->repr, old, val)) { \
return old; \
} \
} \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \
type *expected, type desired, \
atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
type prev = __sync_val_compare_and_swap(&a->repr, *expected, \
desired); \
if (prev == *expected) { \
return true; \
} else { \
*expected = prev; \
return false; \
} \
} \
ATOMIC_INLINE bool \
atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \
type *expected, type desired, \
atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
type prev = __sync_val_compare_and_swap(&a->repr, *expected, \
desired); \
if (prev == *expected) { \
return true; \
} else { \
*expected = prev; \
return false; \
} \
}
#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, \
/* unused */ lg_size) \
JEMALLOC_GENERATE_ATOMICS(type, short_type, /* unused */ lg_size) \
\
ATOMIC_INLINE type \
atomic_fetch_add_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __sync_fetch_and_add(&a->repr, val); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __sync_fetch_and_sub(&a->repr, val); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_and_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __sync_fetch_and_and(&a->repr, val); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_or_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __sync_fetch_and_or(&a->repr, val); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return __sync_fetch_and_xor(&a->repr, val); \
}
#endif /* JEMALLOC_INTERNAL_ATOMIC_GCC_SYNC_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/atomic_msvc.h
Vendored
-158
View File
@@ -1,158 +0,0 @@
#ifndef JEMALLOC_INTERNAL_ATOMIC_MSVC_H
#define JEMALLOC_INTERNAL_ATOMIC_MSVC_H
#define ATOMIC_INIT(...) {__VA_ARGS__}
typedef enum {
atomic_memory_order_relaxed,
atomic_memory_order_acquire,
atomic_memory_order_release,
atomic_memory_order_acq_rel,
atomic_memory_order_seq_cst
} atomic_memory_order_t;
typedef char atomic_repr_0_t;
typedef short atomic_repr_1_t;
typedef long atomic_repr_2_t;
typedef __int64 atomic_repr_3_t;
ATOMIC_INLINE void
atomic_fence(atomic_memory_order_t mo) {
_ReadWriteBarrier();
# if defined(_M_ARM) || defined(_M_ARM64)
/* ARM needs a barrier for everything but relaxed. */
if (mo != atomic_memory_order_relaxed) {
MemoryBarrier();
}
# elif defined(_M_IX86) || defined (_M_X64)
/* x86 needs a barrier only for seq_cst. */
if (mo == atomic_memory_order_seq_cst) {
MemoryBarrier();
}
# else
# error "Don't know how to create atomics for this platform for MSVC."
# endif
_ReadWriteBarrier();
}
#define ATOMIC_INTERLOCKED_REPR(lg_size) atomic_repr_ ## lg_size ## _t
#define ATOMIC_CONCAT(a, b) ATOMIC_RAW_CONCAT(a, b)
#define ATOMIC_RAW_CONCAT(a, b) a ## b
#define ATOMIC_INTERLOCKED_NAME(base_name, lg_size) ATOMIC_CONCAT( \
base_name, ATOMIC_INTERLOCKED_SUFFIX(lg_size))
#define ATOMIC_INTERLOCKED_SUFFIX(lg_size) \
ATOMIC_CONCAT(ATOMIC_INTERLOCKED_SUFFIX_, lg_size)
#define ATOMIC_INTERLOCKED_SUFFIX_0 8
#define ATOMIC_INTERLOCKED_SUFFIX_1 16
#define ATOMIC_INTERLOCKED_SUFFIX_2
#define ATOMIC_INTERLOCKED_SUFFIX_3 64
#define JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_size) \
typedef struct { \
ATOMIC_INTERLOCKED_REPR(lg_size) repr; \
} atomic_##short_type##_t; \
\
ATOMIC_INLINE type \
atomic_load_##short_type(const atomic_##short_type##_t *a, \
atomic_memory_order_t mo) { \
ATOMIC_INTERLOCKED_REPR(lg_size) ret = a->repr; \
if (mo != atomic_memory_order_relaxed) { \
atomic_fence(atomic_memory_order_acquire); \
} \
return (type) ret; \
} \
\
ATOMIC_INLINE void \
atomic_store_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
if (mo != atomic_memory_order_relaxed) { \
atomic_fence(atomic_memory_order_release); \
} \
a->repr = (ATOMIC_INTERLOCKED_REPR(lg_size)) val; \
if (mo == atomic_memory_order_seq_cst) { \
atomic_fence(atomic_memory_order_seq_cst); \
} \
} \
\
ATOMIC_INLINE type \
atomic_exchange_##short_type(atomic_##short_type##_t *a, type val, \
atomic_memory_order_t mo) { \
return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedExchange, \
lg_size)(&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_weak_##short_type(atomic_##short_type##_t *a, \
type *expected, type desired, atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
ATOMIC_INTERLOCKED_REPR(lg_size) e = \
(ATOMIC_INTERLOCKED_REPR(lg_size))*expected; \
ATOMIC_INTERLOCKED_REPR(lg_size) d = \
(ATOMIC_INTERLOCKED_REPR(lg_size))desired; \
ATOMIC_INTERLOCKED_REPR(lg_size) old = \
ATOMIC_INTERLOCKED_NAME(_InterlockedCompareExchange, \
lg_size)(&a->repr, d, e); \
if (old == e) { \
return true; \
} else { \
*expected = (type)old; \
return false; \
} \
} \
\
ATOMIC_INLINE bool \
atomic_compare_exchange_strong_##short_type(atomic_##short_type##_t *a, \
type *expected, type desired, atomic_memory_order_t success_mo, \
atomic_memory_order_t failure_mo) { \
/* We implement the weak version with strong semantics. */ \
return atomic_compare_exchange_weak_##short_type(a, expected, \
desired, success_mo, failure_mo); \
}
#define JEMALLOC_GENERATE_INT_ATOMICS(type, short_type, lg_size) \
JEMALLOC_GENERATE_ATOMICS(type, short_type, lg_size) \
\
ATOMIC_INLINE type \
atomic_fetch_add_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedExchangeAdd, \
lg_size)(&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \
} \
\
ATOMIC_INLINE type \
atomic_fetch_sub_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
/* \
* MSVC warns on negation of unsigned operands, but for us it \
* gives exactly the right semantics (MAX_TYPE + 1 - operand). \
*/ \
__pragma(warning(push)) \
__pragma(warning(disable: 4146)) \
return atomic_fetch_add_##short_type(a, -val, mo); \
__pragma(warning(pop)) \
} \
ATOMIC_INLINE type \
atomic_fetch_and_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedAnd, lg_size)( \
&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \
} \
ATOMIC_INLINE type \
atomic_fetch_or_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedOr, lg_size)( \
&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \
} \
ATOMIC_INLINE type \
atomic_fetch_xor_##short_type(atomic_##short_type##_t *a, \
type val, atomic_memory_order_t mo) { \
return (type)ATOMIC_INTERLOCKED_NAME(_InterlockedXor, lg_size)( \
&a->repr, (ATOMIC_INTERLOCKED_REPR(lg_size))val); \
}
#endif /* JEMALLOC_INTERNAL_ATOMIC_MSVC_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/background_thread_externs.h
Vendored
-32
View File
@@ -1,32 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BACKGROUND_THREAD_EXTERNS_H
#define JEMALLOC_INTERNAL_BACKGROUND_THREAD_EXTERNS_H
extern bool opt_background_thread;
extern size_t opt_max_background_threads;
extern malloc_mutex_t background_thread_lock;
extern atomic_b_t background_thread_enabled_state;
extern size_t n_background_threads;
extern size_t max_background_threads;
extern background_thread_info_t *background_thread_info;
bool background_thread_create(tsd_t *tsd, unsigned arena_ind);
bool background_threads_enable(tsd_t *tsd);
bool background_threads_disable(tsd_t *tsd);
void background_thread_interval_check(tsdn_t *tsdn, arena_t *arena,
arena_decay_t *decay, size_t npages_new);
void background_thread_prefork0(tsdn_t *tsdn);
void background_thread_prefork1(tsdn_t *tsdn);
void background_thread_postfork_parent(tsdn_t *tsdn);
void background_thread_postfork_child(tsdn_t *tsdn);
bool background_thread_stats_read(tsdn_t *tsdn,
background_thread_stats_t *stats);
void background_thread_ctl_init(tsdn_t *tsdn);
#ifdef JEMALLOC_PTHREAD_CREATE_WRAPPER
extern int pthread_create_wrapper(pthread_t *__restrict, const pthread_attr_t *,
void *(*)(void *), void *__restrict);
#endif
bool background_thread_boot0(void);
bool background_thread_boot1(tsdn_t *tsdn);
#endif /* JEMALLOC_INTERNAL_BACKGROUND_THREAD_EXTERNS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/background_thread_inlines.h
Vendored
-62
View File
@@ -1,62 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BACKGROUND_THREAD_INLINES_H
#define JEMALLOC_INTERNAL_BACKGROUND_THREAD_INLINES_H
JEMALLOC_ALWAYS_INLINE bool
background_thread_enabled(void) {
return atomic_load_b(&background_thread_enabled_state, ATOMIC_RELAXED);
}
JEMALLOC_ALWAYS_INLINE void
background_thread_enabled_set(tsdn_t *tsdn, bool state) {
malloc_mutex_assert_owner(tsdn, &background_thread_lock);
atomic_store_b(&background_thread_enabled_state, state, ATOMIC_RELAXED);
}
JEMALLOC_ALWAYS_INLINE background_thread_info_t *
arena_background_thread_info_get(arena_t *arena) {
unsigned arena_ind = arena_ind_get(arena);
return &background_thread_info[arena_ind % max_background_threads];
}
JEMALLOC_ALWAYS_INLINE background_thread_info_t *
background_thread_info_get(size_t ind) {
return &background_thread_info[ind % max_background_threads];
}
JEMALLOC_ALWAYS_INLINE uint64_t
background_thread_wakeup_time_get(background_thread_info_t *info) {
uint64_t next_wakeup = nstime_ns(&info->next_wakeup);
assert(atomic_load_b(&info->indefinite_sleep, ATOMIC_ACQUIRE) ==
(next_wakeup == BACKGROUND_THREAD_INDEFINITE_SLEEP));
return next_wakeup;
}
JEMALLOC_ALWAYS_INLINE void
background_thread_wakeup_time_set(tsdn_t *tsdn, background_thread_info_t *info,
uint64_t wakeup_time) {
malloc_mutex_assert_owner(tsdn, &info->mtx);
atomic_store_b(&info->indefinite_sleep,
wakeup_time == BACKGROUND_THREAD_INDEFINITE_SLEEP, ATOMIC_RELEASE);
nstime_init(&info->next_wakeup, wakeup_time);
}
JEMALLOC_ALWAYS_INLINE bool
background_thread_indefinite_sleep(background_thread_info_t *info) {
return atomic_load_b(&info->indefinite_sleep, ATOMIC_ACQUIRE);
}
JEMALLOC_ALWAYS_INLINE void
arena_background_thread_inactivity_check(tsdn_t *tsdn, arena_t *arena,
bool is_background_thread) {
if (!background_thread_enabled() || is_background_thread) {
return;
}
background_thread_info_t *info =
arena_background_thread_info_get(arena);
if (background_thread_indefinite_sleep(info)) {
background_thread_interval_check(tsdn, arena,
&arena->decay_dirty, 0);
}
}
#endif /* JEMALLOC_INTERNAL_BACKGROUND_THREAD_INLINES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/background_thread_structs.h
Vendored
-54
View File
@@ -1,54 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BACKGROUND_THREAD_STRUCTS_H
#define JEMALLOC_INTERNAL_BACKGROUND_THREAD_STRUCTS_H
/* This file really combines "structs" and "types", but only transitionally. */
#if defined(JEMALLOC_BACKGROUND_THREAD) || defined(JEMALLOC_LAZY_LOCK)
# define JEMALLOC_PTHREAD_CREATE_WRAPPER
#endif
#define BACKGROUND_THREAD_INDEFINITE_SLEEP UINT64_MAX
#define MAX_BACKGROUND_THREAD_LIMIT MALLOCX_ARENA_LIMIT
#define DEFAULT_NUM_BACKGROUND_THREAD 4
typedef enum {
background_thread_stopped,
background_thread_started,
/* Thread waits on the global lock when paused (for arena_reset). */
background_thread_paused,
} background_thread_state_t;
struct background_thread_info_s {
#ifdef JEMALLOC_BACKGROUND_THREAD
/* Background thread is pthread specific. */
pthread_t thread;
pthread_cond_t cond;
#endif
malloc_mutex_t mtx;
background_thread_state_t state;
/* When true, it means no wakeup scheduled. */
atomic_b_t indefinite_sleep;
/* Next scheduled wakeup time (absolute time in ns). */
nstime_t next_wakeup;
/*
* Since the last background thread run, newly added number of pages
* that need to be purged by the next wakeup. This is adjusted on
* epoch advance, and is used to determine whether we should signal the
* background thread to wake up earlier.
*/
size_t npages_to_purge_new;
/* Stats: total number of runs since started. */
uint64_t tot_n_runs;
/* Stats: total sleep time since started. */
nstime_t tot_sleep_time;
};
typedef struct background_thread_info_s background_thread_info_t;
struct background_thread_stats_s {
size_t num_threads;
uint64_t num_runs;
nstime_t run_interval;
};
typedef struct background_thread_stats_s background_thread_stats_t;
#endif /* JEMALLOC_INTERNAL_BACKGROUND_THREAD_STRUCTS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/base_externs.h
Vendored
-22
View File
@@ -1,22 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BASE_EXTERNS_H
#define JEMALLOC_INTERNAL_BASE_EXTERNS_H
extern metadata_thp_mode_t opt_metadata_thp;
extern const char *metadata_thp_mode_names[];
base_t *b0get(void);
base_t *base_new(tsdn_t *tsdn, unsigned ind, extent_hooks_t *extent_hooks);
void base_delete(tsdn_t *tsdn, base_t *base);
extent_hooks_t *base_extent_hooks_get(base_t *base);
extent_hooks_t *base_extent_hooks_set(base_t *base,
extent_hooks_t *extent_hooks);
void *base_alloc(tsdn_t *tsdn, base_t *base, size_t size, size_t alignment);
extent_t *base_alloc_extent(tsdn_t *tsdn, base_t *base);
void base_stats_get(tsdn_t *tsdn, base_t *base, size_t *allocated,
size_t *resident, size_t *mapped, size_t *n_thp);
void base_prefork(tsdn_t *tsdn, base_t *base);
void base_postfork_parent(tsdn_t *tsdn, base_t *base);
void base_postfork_child(tsdn_t *tsdn, base_t *base);
bool base_boot(tsdn_t *tsdn);
#endif /* JEMALLOC_INTERNAL_BASE_EXTERNS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/base_inlines.h
Vendored
-13
View File
@@ -1,13 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BASE_INLINES_H
#define JEMALLOC_INTERNAL_BASE_INLINES_H
static inline unsigned
base_ind_get(const base_t *base) {
return base->ind;
}
static inline bool
metadata_thp_enabled(void) {
return (opt_metadata_thp != metadata_thp_disabled);
}
#endif /* JEMALLOC_INTERNAL_BASE_INLINES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/base_structs.h
Vendored
-59
View File
@@ -1,59 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BASE_STRUCTS_H
#define JEMALLOC_INTERNAL_BASE_STRUCTS_H
#include "jemalloc/internal/jemalloc_internal_types.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/sc.h"
/* Embedded at the beginning of every block of base-managed virtual memory. */
struct base_block_s {
/* Total size of block's virtual memory mapping. */
size_t size;
/* Next block in list of base's blocks. */
base_block_t *next;
/* Tracks unused trailing space. */
extent_t extent;
};
struct base_s {
/* Associated arena's index within the arenas array. */
unsigned ind;
/*
* User-configurable extent hook functions. Points to an
* extent_hooks_t.
*/
atomic_p_t extent_hooks;
/* Protects base_alloc() and base_stats_get() operations. */
malloc_mutex_t mtx;
/* Using THP when true (metadata_thp auto mode). */
bool auto_thp_switched;
/*
* Most recent size class in the series of increasingly large base
* extents. Logarithmic spacing between subsequent allocations ensures
* that the total number of distinct mappings remains small.
*/
pszind_t pind_last;
/* Serial number generation state. */
size_t extent_sn_next;
/* Chain of all blocks associated with base. */
base_block_t *blocks;
/* Heap of extents that track unused trailing space within blocks. */
extent_heap_t avail[SC_NSIZES];
/* Stats, only maintained if config_stats. */
size_t allocated;
size_t resident;
size_t mapped;
/* Number of THP regions touched. */
size_t n_thp;
};
#endif /* JEMALLOC_INTERNAL_BASE_STRUCTS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/base_types.h
Vendored
-33
View File
@@ -1,33 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BASE_TYPES_H
#define JEMALLOC_INTERNAL_BASE_TYPES_H
typedef struct base_block_s base_block_t;
typedef struct base_s base_t;
#define METADATA_THP_DEFAULT metadata_thp_disabled
/*
* In auto mode, arenas switch to huge pages for the base allocator on the
* second base block. a0 switches to thp on the 5th block (after 20 megabytes
* of metadata), since more metadata (e.g. rtree nodes) come from a0's base.
*/
#define BASE_AUTO_THP_THRESHOLD 2
#define BASE_AUTO_THP_THRESHOLD_A0 5
typedef enum {
metadata_thp_disabled = 0,
/*
* Lazily enable hugepage for metadata. To avoid high RSS caused by THP
* + low usage arena (i.e. THP becomes a significant percentage), the
* "auto" option only starts using THP after a base allocator used up
* the first THP region. Starting from the second hugepage (in a single
* arena), "auto" behaves the same as "always", i.e. madvise hugepage
* right away.
*/
metadata_thp_auto = 1,
metadata_thp_always = 2,
metadata_thp_mode_limit = 3
} metadata_thp_mode_t;
#endif /* JEMALLOC_INTERNAL_BASE_TYPES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/bin.h
Vendored
-123
View File
@@ -1,123 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BIN_H
#define JEMALLOC_INTERNAL_BIN_H
#include "jemalloc/internal/bin_stats.h"
#include "jemalloc/internal/bin_types.h"
#include "jemalloc/internal/extent_types.h"
#include "jemalloc/internal/extent_structs.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/sc.h"
/*
* A bin contains a set of extents that are currently being used for slab
* allocations.
*/
/*
* Read-only information associated with each element of arena_t's bins array
* is stored separately, partly to reduce memory usage (only one copy, rather
* than one per arena), but mainly to avoid false cacheline sharing.
*
* Each slab has the following layout:
*
* /--------------------\
* | region 0 |
* |--------------------|
* | region 1 |
* |--------------------|
* | ... |
* | ... |
* | ... |
* |--------------------|
* | region nregs-1 |
* \--------------------/
*/
typedef struct bin_info_s bin_info_t;
struct bin_info_s {
/* Size of regions in a slab for this bin's size class. */
size_t reg_size;
/* Total size of a slab for this bin's size class. */
size_t slab_size;
/* Total number of regions in a slab for this bin's size class. */
uint32_t nregs;
/* Number of sharded bins in each arena for this size class. */
uint32_t n_shards;
/*
* Metadata used to manipulate bitmaps for slabs associated with this
* bin.
*/
bitmap_info_t bitmap_info;
};
extern bin_info_t bin_infos[SC_NBINS];
typedef struct bin_s bin_t;
struct bin_s {
/* All operations on bin_t fields require lock ownership. */
malloc_mutex_t lock;
/*
* Current slab being used to service allocations of this bin's size
* class. slabcur is independent of slabs_{nonfull,full}; whenever
* slabcur is reassigned, the previous slab must be deallocated or
* inserted into slabs_{nonfull,full}.
*/
extent_t *slabcur;
/*
* Heap of non-full slabs. This heap is used to assure that new
* allocations come from the non-full slab that is oldest/lowest in
* memory.
*/
extent_heap_t slabs_nonfull;
/* List used to track full slabs. */
extent_list_t slabs_full;
/* Bin statistics. */
bin_stats_t stats;
};
/* A set of sharded bins of the same size class. */
typedef struct bins_s bins_t;
struct bins_s {
/* Sharded bins. Dynamically sized. */
bin_t *bin_shards;
};
void bin_shard_sizes_boot(unsigned bin_shards[SC_NBINS]);
bool bin_update_shard_size(unsigned bin_shards[SC_NBINS], size_t start_size,
size_t end_size, size_t nshards);
void bin_boot(sc_data_t *sc_data, unsigned bin_shard_sizes[SC_NBINS]);
/* Initializes a bin to empty. Returns true on error. */
bool bin_init(bin_t *bin);
/* Forking. */
void bin_prefork(tsdn_t *tsdn, bin_t *bin);
void bin_postfork_parent(tsdn_t *tsdn, bin_t *bin);
void bin_postfork_child(tsdn_t *tsdn, bin_t *bin);
/* Stats. */
static inline void
bin_stats_merge(tsdn_t *tsdn, bin_stats_t *dst_bin_stats, bin_t *bin) {
malloc_mutex_lock(tsdn, &bin->lock);
malloc_mutex_prof_accum(tsdn, &dst_bin_stats->mutex_data, &bin->lock);
dst_bin_stats->nmalloc += bin->stats.nmalloc;
dst_bin_stats->ndalloc += bin->stats.ndalloc;
dst_bin_stats->nrequests += bin->stats.nrequests;
dst_bin_stats->curregs += bin->stats.curregs;
dst_bin_stats->nfills += bin->stats.nfills;
dst_bin_stats->nflushes += bin->stats.nflushes;
dst_bin_stats->nslabs += bin->stats.nslabs;
dst_bin_stats->reslabs += bin->stats.reslabs;
dst_bin_stats->curslabs += bin->stats.curslabs;
dst_bin_stats->nonfull_slabs += bin->stats.nonfull_slabs;
malloc_mutex_unlock(tsdn, &bin->lock);
}
#endif /* JEMALLOC_INTERNAL_BIN_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/bin_stats.h
Vendored
-54
View File
@@ -1,54 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BIN_STATS_H
#define JEMALLOC_INTERNAL_BIN_STATS_H
#include "jemalloc/internal/mutex_prof.h"
typedef struct bin_stats_s bin_stats_t;
struct bin_stats_s {
/*
* Total number of allocation/deallocation requests served directly by
* the bin. Note that tcache may allocate an object, then recycle it
* many times, resulting many increments to nrequests, but only one
* each to nmalloc and ndalloc.
*/
uint64_t nmalloc;
uint64_t ndalloc;
/*
* Number of allocation requests that correspond to the size of this
* bin. This includes requests served by tcache, though tcache only
* periodically merges into this counter.
*/
uint64_t nrequests;
/*
* Current number of regions of this size class, including regions
* currently cached by tcache.
*/
size_t curregs;
/* Number of tcache fills from this bin. */
uint64_t nfills;
/* Number of tcache flushes to this bin. */
uint64_t nflushes;
/* Total number of slabs created for this bin's size class. */
uint64_t nslabs;
/*
* Total number of slabs reused by extracting them from the slabs heap
* for this bin's size class.
*/
uint64_t reslabs;
/* Current number of slabs in this bin. */
size_t curslabs;
/* Current size of nonfull slabs heap in this bin. */
size_t nonfull_slabs;
mutex_prof_data_t mutex_data;
};
#endif /* JEMALLOC_INTERNAL_BIN_STATS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/bin_types.h
Vendored
-17
View File
@@ -1,17 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BIN_TYPES_H
#define JEMALLOC_INTERNAL_BIN_TYPES_H
#include "jemalloc/internal/sc.h"
#define BIN_SHARDS_MAX (1 << EXTENT_BITS_BINSHARD_WIDTH)
#define N_BIN_SHARDS_DEFAULT 1
/* Used in TSD static initializer only. Real init in arena_bind(). */
#define TSD_BINSHARDS_ZERO_INITIALIZER {{UINT8_MAX}}
typedef struct tsd_binshards_s tsd_binshards_t;
struct tsd_binshards_s {
uint8_t binshard[SC_NBINS];
};
#endif /* JEMALLOC_INTERNAL_BIN_TYPES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/bit_util.h
Vendored
-239
View File
@@ -1,239 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BIT_UTIL_H
#define JEMALLOC_INTERNAL_BIT_UTIL_H
#include "jemalloc/internal/assert.h"
#define BIT_UTIL_INLINE static inline
/* Sanity check. */
#if !defined(JEMALLOC_INTERNAL_FFSLL) || !defined(JEMALLOC_INTERNAL_FFSL) \
|| !defined(JEMALLOC_INTERNAL_FFS)
# error JEMALLOC_INTERNAL_FFS{,L,LL} should have been defined by configure
#endif
BIT_UTIL_INLINE unsigned
ffs_llu(unsigned long long bitmap) {
return JEMALLOC_INTERNAL_FFSLL(bitmap);
}
BIT_UTIL_INLINE unsigned
ffs_lu(unsigned long bitmap) {
return JEMALLOC_INTERNAL_FFSL(bitmap);
}
BIT_UTIL_INLINE unsigned
ffs_u(unsigned bitmap) {
return JEMALLOC_INTERNAL_FFS(bitmap);
}
#ifdef JEMALLOC_INTERNAL_POPCOUNTL
BIT_UTIL_INLINE unsigned
popcount_lu(unsigned long bitmap) {
return JEMALLOC_INTERNAL_POPCOUNTL(bitmap);
}
#endif
/*
* Clears first unset bit in bitmap, and returns
* place of bit. bitmap *must not* be 0.
*/
BIT_UTIL_INLINE size_t
cfs_lu(unsigned long* bitmap) {
size_t bit = ffs_lu(*bitmap) - 1;
*bitmap ^= ZU(1) << bit;
return bit;
}
BIT_UTIL_INLINE unsigned
ffs_zu(size_t bitmap) {
#if LG_SIZEOF_PTR == LG_SIZEOF_INT
return ffs_u(bitmap);
#elif LG_SIZEOF_PTR == LG_SIZEOF_LONG
return ffs_lu(bitmap);
#elif LG_SIZEOF_PTR == LG_SIZEOF_LONG_LONG
return ffs_llu(bitmap);
#else
#error No implementation for size_t ffs()
#endif
}
BIT_UTIL_INLINE unsigned
ffs_u64(uint64_t bitmap) {
#if LG_SIZEOF_LONG == 3
return ffs_lu(bitmap);
#elif LG_SIZEOF_LONG_LONG == 3
return ffs_llu(bitmap);
#else
#error No implementation for 64-bit ffs()
#endif
}
BIT_UTIL_INLINE unsigned
ffs_u32(uint32_t bitmap) {
#if LG_SIZEOF_INT == 2
return ffs_u(bitmap);
#else
#error No implementation for 32-bit ffs()
#endif
return ffs_u(bitmap);
}
BIT_UTIL_INLINE uint64_t
pow2_ceil_u64(uint64_t x) {
#if (defined(__amd64__) || defined(__x86_64__) || defined(JEMALLOC_HAVE_BUILTIN_CLZ))
if(unlikely(x <= 1)) {
return x;
}
size_t msb_on_index;
#if (defined(__amd64__) || defined(__x86_64__))
asm ("bsrq %1, %0"
: "=r"(msb_on_index) // Outputs.
: "r"(x-1) // Inputs.
);
#elif (defined(JEMALLOC_HAVE_BUILTIN_CLZ))
msb_on_index = (63 ^ __builtin_clzll(x - 1));
#endif
assert(msb_on_index < 63);
return 1ULL << (msb_on_index + 1);
#else
x--;
x |= x >> 1;
x |= x >> 2;
x |= x >> 4;
x |= x >> 8;
x |= x >> 16;
x |= x >> 32;
x++;
return x;
#endif
}
BIT_UTIL_INLINE uint32_t
pow2_ceil_u32(uint32_t x) {
#if ((defined(__i386__) || defined(JEMALLOC_HAVE_BUILTIN_CLZ)) && (!defined(__s390__)))
if(unlikely(x <= 1)) {
return x;
}
size_t msb_on_index;
#if (defined(__i386__))
asm ("bsr %1, %0"
: "=r"(msb_on_index) // Outputs.
: "r"(x-1) // Inputs.
);
#elif (defined(JEMALLOC_HAVE_BUILTIN_CLZ))
msb_on_index = (31 ^ __builtin_clz(x - 1));
#endif
assert(msb_on_index < 31);
return 1U << (msb_on_index + 1);
#else
x--;
x |= x >> 1;
x |= x >> 2;
x |= x >> 4;
x |= x >> 8;
x |= x >> 16;
x++;
return x;
#endif
}
/* Compute the smallest power of 2 that is >= x. */
BIT_UTIL_INLINE size_t
pow2_ceil_zu(size_t x) {
#if (LG_SIZEOF_PTR == 3)
return pow2_ceil_u64(x);
#else
return pow2_ceil_u32(x);
#endif
}
#if (defined(__i386__) || defined(__amd64__) || defined(__x86_64__))
BIT_UTIL_INLINE unsigned
lg_floor(size_t x) {
size_t ret;
assert(x != 0);
asm ("bsr %1, %0"
: "=r"(ret) // Outputs.
: "r"(x) // Inputs.
);
assert(ret < UINT_MAX);
return (unsigned)ret;
}
#elif (defined(_MSC_VER))
BIT_UTIL_INLINE unsigned
lg_floor(size_t x) {
unsigned long ret;
assert(x != 0);
#if (LG_SIZEOF_PTR == 3)
_BitScanReverse64(&ret, x);
#elif (LG_SIZEOF_PTR == 2)
_BitScanReverse(&ret, x);
#else
# error "Unsupported type size for lg_floor()"
#endif
assert(ret < UINT_MAX);
return (unsigned)ret;
}
#elif (defined(JEMALLOC_HAVE_BUILTIN_CLZ))
BIT_UTIL_INLINE unsigned
lg_floor(size_t x) {
assert(x != 0);
#if (LG_SIZEOF_PTR == LG_SIZEOF_INT)
return ((8 << LG_SIZEOF_PTR) - 1) - __builtin_clz(x);
#elif (LG_SIZEOF_PTR == LG_SIZEOF_LONG)
return ((8 << LG_SIZEOF_PTR) - 1) - __builtin_clzl(x);
#else
# error "Unsupported type size for lg_floor()"
#endif
}
#else
BIT_UTIL_INLINE unsigned
lg_floor(size_t x) {
assert(x != 0);
x |= (x >> 1);
x |= (x >> 2);
x |= (x >> 4);
x |= (x >> 8);
x |= (x >> 16);
#if (LG_SIZEOF_PTR == 3)
x |= (x >> 32);
#endif
if (x == SIZE_T_MAX) {
return (8 << LG_SIZEOF_PTR) - 1;
}
x++;
return ffs_zu(x) - 2;
}
#endif
BIT_UTIL_INLINE unsigned
lg_ceil(size_t x) {
return lg_floor(x) + ((x & (x - 1)) == 0 ? 0 : 1);
}
#undef BIT_UTIL_INLINE
/* A compile-time version of lg_floor and lg_ceil. */
#define LG_FLOOR_1(x) 0
#define LG_FLOOR_2(x) (x < (1ULL << 1) ? LG_FLOOR_1(x) : 1 + LG_FLOOR_1(x >> 1))
#define LG_FLOOR_4(x) (x < (1ULL << 2) ? LG_FLOOR_2(x) : 2 + LG_FLOOR_2(x >> 2))
#define LG_FLOOR_8(x) (x < (1ULL << 4) ? LG_FLOOR_4(x) : 4 + LG_FLOOR_4(x >> 4))
#define LG_FLOOR_16(x) (x < (1ULL << 8) ? LG_FLOOR_8(x) : 8 + LG_FLOOR_8(x >> 8))
#define LG_FLOOR_32(x) (x < (1ULL << 16) ? LG_FLOOR_16(x) : 16 + LG_FLOOR_16(x >> 16))
#define LG_FLOOR_64(x) (x < (1ULL << 32) ? LG_FLOOR_32(x) : 32 + LG_FLOOR_32(x >> 32))
#if LG_SIZEOF_PTR == 2
# define LG_FLOOR(x) LG_FLOOR_32((x))
#else
# define LG_FLOOR(x) LG_FLOOR_64((x))
#endif
#define LG_CEIL(x) (LG_FLOOR(x) + (((x) & ((x) - 1)) == 0 ? 0 : 1))
#endif /* JEMALLOC_INTERNAL_BIT_UTIL_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/bitmap.h
Vendored
-369
View File
@@ -1,369 +0,0 @@
#ifndef JEMALLOC_INTERNAL_BITMAP_H
#define JEMALLOC_INTERNAL_BITMAP_H
#include "jemalloc/internal/arena_types.h"
#include "jemalloc/internal/bit_util.h"
#include "jemalloc/internal/sc.h"
typedef unsigned long bitmap_t;
#define LG_SIZEOF_BITMAP LG_SIZEOF_LONG
/* Maximum bitmap bit count is 2^LG_BITMAP_MAXBITS. */
#if LG_SLAB_MAXREGS > LG_CEIL(SC_NSIZES)
/* Maximum bitmap bit count is determined by maximum regions per slab. */
# define LG_BITMAP_MAXBITS LG_SLAB_MAXREGS
#else
/* Maximum bitmap bit count is determined by number of extent size classes. */
# define LG_BITMAP_MAXBITS LG_CEIL(SC_NSIZES)
#endif
#define BITMAP_MAXBITS (ZU(1) << LG_BITMAP_MAXBITS)
/* Number of bits per group. */
#define LG_BITMAP_GROUP_NBITS (LG_SIZEOF_BITMAP + 3)
#define BITMAP_GROUP_NBITS (1U << LG_BITMAP_GROUP_NBITS)
#define BITMAP_GROUP_NBITS_MASK (BITMAP_GROUP_NBITS-1)
/*
* Do some analysis on how big the bitmap is before we use a tree. For a brute
* force linear search, if we would have to call ffs_lu() more than 2^3 times,
* use a tree instead.
*/
#if LG_BITMAP_MAXBITS - LG_BITMAP_GROUP_NBITS > 3
# define BITMAP_USE_TREE
#endif
/* Number of groups required to store a given number of bits. */
#define BITMAP_BITS2GROUPS(nbits) \
(((nbits) + BITMAP_GROUP_NBITS_MASK) >> LG_BITMAP_GROUP_NBITS)
/*
* Number of groups required at a particular level for a given number of bits.
*/
#define BITMAP_GROUPS_L0(nbits) \
BITMAP_BITS2GROUPS(nbits)
#define BITMAP_GROUPS_L1(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(nbits))
#define BITMAP_GROUPS_L2(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS((nbits))))
#define BITMAP_GROUPS_L3(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS( \
BITMAP_BITS2GROUPS((nbits)))))
#define BITMAP_GROUPS_L4(nbits) \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS( \
BITMAP_BITS2GROUPS(BITMAP_BITS2GROUPS((nbits))))))
/*
* Assuming the number of levels, number of groups required for a given number
* of bits.
*/
#define BITMAP_GROUPS_1_LEVEL(nbits) \
BITMAP_GROUPS_L0(nbits)
#define BITMAP_GROUPS_2_LEVEL(nbits) \
(BITMAP_GROUPS_1_LEVEL(nbits) + BITMAP_GROUPS_L1(nbits))
#define BITMAP_GROUPS_3_LEVEL(nbits) \
(BITMAP_GROUPS_2_LEVEL(nbits) + BITMAP_GROUPS_L2(nbits))
#define BITMAP_GROUPS_4_LEVEL(nbits) \
(BITMAP_GROUPS_3_LEVEL(nbits) + BITMAP_GROUPS_L3(nbits))
#define BITMAP_GROUPS_5_LEVEL(nbits) \
(BITMAP_GROUPS_4_LEVEL(nbits) + BITMAP_GROUPS_L4(nbits))
/*
* Maximum number of groups required to support LG_BITMAP_MAXBITS.
*/
#ifdef BITMAP_USE_TREE
#if LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_1_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_1_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 2
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_2_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_2_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 3
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_3_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_3_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 4
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_4_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_4_LEVEL(BITMAP_MAXBITS)
#elif LG_BITMAP_MAXBITS <= LG_BITMAP_GROUP_NBITS * 5
# define BITMAP_GROUPS(nbits) BITMAP_GROUPS_5_LEVEL(nbits)
# define BITMAP_GROUPS_MAX BITMAP_GROUPS_5_LEVEL(BITMAP_MAXBITS)
#else
# error "Unsupported bitmap size"
#endif
/*
* Maximum number of levels possible. This could be statically computed based
* on LG_BITMAP_MAXBITS:
*
* #define BITMAP_MAX_LEVELS \
* (LG_BITMAP_MAXBITS / LG_SIZEOF_BITMAP) \
* + !!(LG_BITMAP_MAXBITS % LG_SIZEOF_BITMAP)
*
* However, that would not allow the generic BITMAP_INFO_INITIALIZER() macro, so
* instead hardcode BITMAP_MAX_LEVELS to the largest number supported by the
* various cascading macros. The only additional cost this incurs is some
* unused trailing entries in bitmap_info_t structures; the bitmaps themselves
* are not impacted.
*/
#define BITMAP_MAX_LEVELS 5
#define BITMAP_INFO_INITIALIZER(nbits) { \
/* nbits. */ \
nbits, \
/* nlevels. */ \
(BITMAP_GROUPS_L0(nbits) > BITMAP_GROUPS_L1(nbits)) + \
(BITMAP_GROUPS_L1(nbits) > BITMAP_GROUPS_L2(nbits)) + \
(BITMAP_GROUPS_L2(nbits) > BITMAP_GROUPS_L3(nbits)) + \
(BITMAP_GROUPS_L3(nbits) > BITMAP_GROUPS_L4(nbits)) + 1, \
/* levels. */ \
{ \
{0}, \
{BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L1(nbits) + BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L2(nbits) + BITMAP_GROUPS_L1(nbits) + \
BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L3(nbits) + BITMAP_GROUPS_L2(nbits) + \
BITMAP_GROUPS_L1(nbits) + BITMAP_GROUPS_L0(nbits)}, \
{BITMAP_GROUPS_L4(nbits) + BITMAP_GROUPS_L3(nbits) + \
BITMAP_GROUPS_L2(nbits) + BITMAP_GROUPS_L1(nbits) \
+ BITMAP_GROUPS_L0(nbits)} \
} \
}
#else /* BITMAP_USE_TREE */
#define BITMAP_GROUPS(nbits) BITMAP_BITS2GROUPS(nbits)
#define BITMAP_GROUPS_MAX BITMAP_BITS2GROUPS(BITMAP_MAXBITS)
#define BITMAP_INFO_INITIALIZER(nbits) { \
/* nbits. */ \
nbits, \
/* ngroups. */ \
BITMAP_BITS2GROUPS(nbits) \
}
#endif /* BITMAP_USE_TREE */
typedef struct bitmap_level_s {
/* Offset of this level's groups within the array of groups. */
size_t group_offset;
} bitmap_level_t;
typedef struct bitmap_info_s {
/* Logical number of bits in bitmap (stored at bottom level). */
size_t nbits;
#ifdef BITMAP_USE_TREE
/* Number of levels necessary for nbits. */
unsigned nlevels;
/*
* Only the first (nlevels+1) elements are used, and levels are ordered
* bottom to top (e.g. the bottom level is stored in levels[0]).
*/
bitmap_level_t levels[BITMAP_MAX_LEVELS+1];
#else /* BITMAP_USE_TREE */
/* Number of groups necessary for nbits. */
size_t ngroups;
#endif /* BITMAP_USE_TREE */
} bitmap_info_t;
void bitmap_info_init(bitmap_info_t *binfo, size_t nbits);
void bitmap_init(bitmap_t *bitmap, const bitmap_info_t *binfo, bool fill);
size_t bitmap_size(const bitmap_info_t *binfo);
static inline bool
bitmap_full(bitmap_t *bitmap, const bitmap_info_t *binfo) {
#ifdef BITMAP_USE_TREE
size_t rgoff = binfo->levels[binfo->nlevels].group_offset - 1;
bitmap_t rg = bitmap[rgoff];
/* The bitmap is full iff the root group is 0. */
return (rg == 0);
#else
size_t i;
for (i = 0; i < binfo->ngroups; i++) {
if (bitmap[i] != 0) {
return false;
}
}
return true;
#endif
}
static inline bool
bitmap_get(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) {
size_t goff;
bitmap_t g;
assert(bit < binfo->nbits);
goff = bit >> LG_BITMAP_GROUP_NBITS;
g = bitmap[goff];
return !(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)));
}
static inline void
bitmap_set(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) {
size_t goff;
bitmap_t *gp;
bitmap_t g;
assert(bit < binfo->nbits);
assert(!bitmap_get(bitmap, binfo, bit));
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[goff];
g = *gp;
assert(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
assert(bitmap_get(bitmap, binfo, bit));
#ifdef BITMAP_USE_TREE
/* Propagate group state transitions up the tree. */
if (g == 0) {
unsigned i;
for (i = 1; i < binfo->nlevels; i++) {
bit = goff;
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[binfo->levels[i].group_offset + goff];
g = *gp;
assert(g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)));
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
if (g != 0) {
break;
}
}
}
#endif
}
/* ffu: find first unset >= bit. */
static inline size_t
bitmap_ffu(const bitmap_t *bitmap, const bitmap_info_t *binfo, size_t min_bit) {
assert(min_bit < binfo->nbits);
#ifdef BITMAP_USE_TREE
size_t bit = 0;
for (unsigned level = binfo->nlevels; level--;) {
size_t lg_bits_per_group = (LG_BITMAP_GROUP_NBITS * (level +
1));
bitmap_t group = bitmap[binfo->levels[level].group_offset + (bit
>> lg_bits_per_group)];
unsigned group_nmask = (unsigned)(((min_bit > bit) ? (min_bit -
bit) : 0) >> (lg_bits_per_group - LG_BITMAP_GROUP_NBITS));
assert(group_nmask <= BITMAP_GROUP_NBITS);
bitmap_t group_mask = ~((1LU << group_nmask) - 1);
bitmap_t group_masked = group & group_mask;
if (group_masked == 0LU) {
if (group == 0LU) {
return binfo->nbits;
}
/*
* min_bit was preceded by one or more unset bits in
* this group, but there are no other unset bits in this
* group. Try again starting at the first bit of the
* next sibling. This will recurse at most once per
* non-root level.
*/
size_t sib_base = bit + (ZU(1) << lg_bits_per_group);
assert(sib_base > min_bit);
assert(sib_base > bit);
if (sib_base >= binfo->nbits) {
return binfo->nbits;
}
return bitmap_ffu(bitmap, binfo, sib_base);
}
bit += ((size_t)(ffs_lu(group_masked) - 1)) <<
(lg_bits_per_group - LG_BITMAP_GROUP_NBITS);
}
assert(bit >= min_bit);
assert(bit < binfo->nbits);
return bit;
#else
size_t i = min_bit >> LG_BITMAP_GROUP_NBITS;
bitmap_t g = bitmap[i] & ~((1LU << (min_bit & BITMAP_GROUP_NBITS_MASK))
- 1);
size_t bit;
do {
bit = ffs_lu(g);
if (bit != 0) {
return (i << LG_BITMAP_GROUP_NBITS) + (bit - 1);
}
i++;
g = bitmap[i];
} while (i < binfo->ngroups);
return binfo->nbits;
#endif
}
/* sfu: set first unset. */
static inline size_t
bitmap_sfu(bitmap_t *bitmap, const bitmap_info_t *binfo) {
size_t bit;
bitmap_t g;
unsigned i;
assert(!bitmap_full(bitmap, binfo));
#ifdef BITMAP_USE_TREE
i = binfo->nlevels - 1;
g = bitmap[binfo->levels[i].group_offset];
bit = ffs_lu(g) - 1;
while (i > 0) {
i--;
g = bitmap[binfo->levels[i].group_offset + bit];
bit = (bit << LG_BITMAP_GROUP_NBITS) + (ffs_lu(g) - 1);
}
#else
i = 0;
g = bitmap[0];
while ((bit = ffs_lu(g)) == 0) {
i++;
g = bitmap[i];
}
bit = (i << LG_BITMAP_GROUP_NBITS) + (bit - 1);
#endif
bitmap_set(bitmap, binfo, bit);
return bit;
}
static inline void
bitmap_unset(bitmap_t *bitmap, const bitmap_info_t *binfo, size_t bit) {
size_t goff;
bitmap_t *gp;
bitmap_t g;
UNUSED bool propagate;
assert(bit < binfo->nbits);
assert(bitmap_get(bitmap, binfo, bit));
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[goff];
g = *gp;
propagate = (g == 0);
assert((g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK))) == 0);
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
assert(!bitmap_get(bitmap, binfo, bit));
#ifdef BITMAP_USE_TREE
/* Propagate group state transitions up the tree. */
if (propagate) {
unsigned i;
for (i = 1; i < binfo->nlevels; i++) {
bit = goff;
goff = bit >> LG_BITMAP_GROUP_NBITS;
gp = &bitmap[binfo->levels[i].group_offset + goff];
g = *gp;
propagate = (g == 0);
assert((g & (ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK)))
== 0);
g ^= ZU(1) << (bit & BITMAP_GROUP_NBITS_MASK);
*gp = g;
if (!propagate) {
break;
}
}
}
#endif /* BITMAP_USE_TREE */
}
#endif /* JEMALLOC_INTERNAL_BITMAP_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/cache_bin.h
Vendored
-131
View File
@@ -1,131 +0,0 @@
#ifndef JEMALLOC_INTERNAL_CACHE_BIN_H
#define JEMALLOC_INTERNAL_CACHE_BIN_H
#include "jemalloc/internal/ql.h"
/*
* The cache_bins are the mechanism that the tcache and the arena use to
* communicate. The tcache fills from and flushes to the arena by passing a
* cache_bin_t to fill/flush. When the arena needs to pull stats from the
* tcaches associated with it, it does so by iterating over its
* cache_bin_array_descriptor_t objects and reading out per-bin stats it
* contains. This makes it so that the arena need not know about the existence
* of the tcache at all.
*/
/*
* The count of the number of cached allocations in a bin. We make this signed
* so that negative numbers can encode "invalid" states (e.g. a low water mark
* of -1 for a cache that has been depleted).
*/
typedef int32_t cache_bin_sz_t;
typedef struct cache_bin_stats_s cache_bin_stats_t;
struct cache_bin_stats_s {
/*
* Number of allocation requests that corresponded to the size of this
* bin.
*/
uint64_t nrequests;
};
/*
* Read-only information associated with each element of tcache_t's tbins array
* is stored separately, mainly to reduce memory usage.
*/
typedef struct cache_bin_info_s cache_bin_info_t;
struct cache_bin_info_s {
/* Upper limit on ncached. */
cache_bin_sz_t ncached_max;
};
typedef struct cache_bin_s cache_bin_t;
struct cache_bin_s {
/* Min # cached since last GC. */
cache_bin_sz_t low_water;
/* # of cached objects. */
cache_bin_sz_t ncached;
/*
* ncached and stats are both modified frequently. Let's keep them
* close so that they have a higher chance of being on the same
* cacheline, thus less write-backs.
*/
cache_bin_stats_t tstats;
/*
* Stack of available objects.
*
* To make use of adjacent cacheline prefetch, the items in the avail
* stack goes to higher address for newer allocations. avail points
* just above the available space, which means that
* avail[-ncached, ... -1] are available items and the lowest item will
* be allocated first.
*/
void **avail;
};
typedef struct cache_bin_array_descriptor_s cache_bin_array_descriptor_t;
struct cache_bin_array_descriptor_s {
/*
* The arena keeps a list of the cache bins associated with it, for
* stats collection.
*/
ql_elm(cache_bin_array_descriptor_t) link;
/* Pointers to the tcache bins. */
cache_bin_t *bins_small;
cache_bin_t *bins_large;
};
static inline void
cache_bin_array_descriptor_init(cache_bin_array_descriptor_t *descriptor,
cache_bin_t *bins_small, cache_bin_t *bins_large) {
ql_elm_new(descriptor, link);
descriptor->bins_small = bins_small;
descriptor->bins_large = bins_large;
}
JEMALLOC_ALWAYS_INLINE void *
cache_bin_alloc_easy(cache_bin_t *bin, bool *success) {
void *ret;
bin->ncached--;
/*
* Check for both bin->ncached == 0 and ncached < low_water
* in a single branch.
*/
if (unlikely(bin->ncached <= bin->low_water)) {
bin->low_water = bin->ncached;
if (bin->ncached == -1) {
bin->ncached = 0;
*success = false;
return NULL;
}
}
/*
* success (instead of ret) should be checked upon the return of this
* function. We avoid checking (ret == NULL) because there is never a
* null stored on the avail stack (which is unknown to the compiler),
* and eagerly checking ret would cause pipeline stall (waiting for the
* cacheline).
*/
*success = true;
ret = *(bin->avail - (bin->ncached + 1));
return ret;
}
JEMALLOC_ALWAYS_INLINE bool
cache_bin_dalloc_easy(cache_bin_t *bin, cache_bin_info_t *bin_info, void *ptr) {
if (unlikely(bin->ncached == bin_info->ncached_max)) {
return false;
}
assert(bin->ncached < bin_info->ncached_max);
bin->ncached++;
*(bin->avail - bin->ncached) = ptr;
return true;
}
#endif /* JEMALLOC_INTERNAL_CACHE_BIN_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/ckh.h
Vendored
-101
View File
@@ -1,101 +0,0 @@
#ifndef JEMALLOC_INTERNAL_CKH_H
#define JEMALLOC_INTERNAL_CKH_H
#include "jemalloc/internal/tsd.h"
/* Cuckoo hashing implementation. Skip to the end for the interface. */
/******************************************************************************/
/* INTERNAL DEFINITIONS -- IGNORE */
/******************************************************************************/
/* Maintain counters used to get an idea of performance. */
/* #define CKH_COUNT */
/* Print counter values in ckh_delete() (requires CKH_COUNT). */
/* #define CKH_VERBOSE */
/*
* There are 2^LG_CKH_BUCKET_CELLS cells in each hash table bucket. Try to fit
* one bucket per L1 cache line.
*/
#define LG_CKH_BUCKET_CELLS (LG_CACHELINE - LG_SIZEOF_PTR - 1)
/* Typedefs to allow easy function pointer passing. */
typedef void ckh_hash_t (const void *, size_t[2]);
typedef bool ckh_keycomp_t (const void *, const void *);
/* Hash table cell. */
typedef struct {
const void *key;
const void *data;
} ckhc_t;
/* The hash table itself. */
typedef struct {
#ifdef CKH_COUNT
/* Counters used to get an idea of performance. */
uint64_t ngrows;
uint64_t nshrinks;
uint64_t nshrinkfails;
uint64_t ninserts;
uint64_t nrelocs;
#endif
/* Used for pseudo-random number generation. */
uint64_t prng_state;
/* Total number of items. */
size_t count;
/*
* Minimum and current number of hash table buckets. There are
* 2^LG_CKH_BUCKET_CELLS cells per bucket.
*/
unsigned lg_minbuckets;
unsigned lg_curbuckets;
/* Hash and comparison functions. */
ckh_hash_t *hash;
ckh_keycomp_t *keycomp;
/* Hash table with 2^lg_curbuckets buckets. */
ckhc_t *tab;
} ckh_t;
/******************************************************************************/
/* BEGIN PUBLIC API */
/******************************************************************************/
/* Lifetime management. Minitems is the initial capacity. */
bool ckh_new(tsd_t *tsd, ckh_t *ckh, size_t minitems, ckh_hash_t *hash,
ckh_keycomp_t *keycomp);
void ckh_delete(tsd_t *tsd, ckh_t *ckh);
/* Get the number of elements in the set. */
size_t ckh_count(ckh_t *ckh);
/*
* To iterate over the elements in the table, initialize *tabind to 0 and call
* this function until it returns true. Each call that returns false will
* update *key and *data to the next element in the table, assuming the pointers
* are non-NULL.
*/
bool ckh_iter(ckh_t *ckh, size_t *tabind, void **key, void **data);
/*
* Basic hash table operations -- insert, removal, lookup. For ckh_remove and
* ckh_search, key or data can be NULL. The hash-table only stores pointers to
* the key and value, and doesn't do any lifetime management.
*/
bool ckh_insert(tsd_t *tsd, ckh_t *ckh, const void *key, const void *data);
bool ckh_remove(tsd_t *tsd, ckh_t *ckh, const void *searchkey, void **key,
void **data);
bool ckh_search(ckh_t *ckh, const void *searchkey, void **key, void **data);
/* Some useful hash and comparison functions for strings and pointers. */
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, size_t r_hash[2]);
bool ckh_pointer_keycomp(const void *k1, const void *k2);
#endif /* JEMALLOC_INTERNAL_CKH_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/ctl.h
Vendored
-134
View File
@@ -1,134 +0,0 @@
#ifndef JEMALLOC_INTERNAL_CTL_H
#define JEMALLOC_INTERNAL_CTL_H
#include "jemalloc/internal/jemalloc_internal_types.h"
#include "jemalloc/internal/malloc_io.h"
#include "jemalloc/internal/mutex_prof.h"
#include "jemalloc/internal/ql.h"
#include "jemalloc/internal/sc.h"
#include "jemalloc/internal/stats.h"
/* Maximum ctl tree depth. */
#define CTL_MAX_DEPTH 7
typedef struct ctl_node_s {
bool named;
} ctl_node_t;
typedef struct ctl_named_node_s {
ctl_node_t node;
const char *name;
/* If (nchildren == 0), this is a terminal node. */
size_t nchildren;
const ctl_node_t *children;
int (*ctl)(tsd_t *, const size_t *, size_t, void *, size_t *, void *,
size_t);
} ctl_named_node_t;
typedef struct ctl_indexed_node_s {
struct ctl_node_s node;
const ctl_named_node_t *(*index)(tsdn_t *, const size_t *, size_t,
size_t);
} ctl_indexed_node_t;
typedef struct ctl_arena_stats_s {
arena_stats_t astats;
/* Aggregate stats for small size classes, based on bin stats. */
size_t allocated_small;
uint64_t nmalloc_small;
uint64_t ndalloc_small;
uint64_t nrequests_small;
uint64_t nfills_small;
uint64_t nflushes_small;
bin_stats_t bstats[SC_NBINS];
arena_stats_large_t lstats[SC_NSIZES - SC_NBINS];
arena_stats_extents_t estats[SC_NPSIZES];
} ctl_arena_stats_t;
typedef struct ctl_stats_s {
size_t allocated;
size_t active;
size_t metadata;
size_t metadata_thp;
size_t resident;
size_t mapped;
size_t retained;
background_thread_stats_t background_thread;
mutex_prof_data_t mutex_prof_data[mutex_prof_num_global_mutexes];
} ctl_stats_t;
typedef struct ctl_arena_s ctl_arena_t;
struct ctl_arena_s {
unsigned arena_ind;
bool initialized;
ql_elm(ctl_arena_t) destroyed_link;
/* Basic stats, supported even if !config_stats. */
unsigned nthreads;
const char *dss;
ssize_t dirty_decay_ms;
ssize_t muzzy_decay_ms;
size_t pactive;
size_t pdirty;
size_t pmuzzy;
/* NULL if !config_stats. */
ctl_arena_stats_t *astats;
};
typedef struct ctl_arenas_s {
uint64_t epoch;
unsigned narenas;
ql_head(ctl_arena_t) destroyed;
/*
* Element 0 corresponds to merged stats for extant arenas (accessed via
* MALLCTL_ARENAS_ALL), element 1 corresponds to merged stats for
* destroyed arenas (accessed via MALLCTL_ARENAS_DESTROYED), and the
* remaining MALLOCX_ARENA_LIMIT elements correspond to arenas.
*/
ctl_arena_t *arenas[2 + MALLOCX_ARENA_LIMIT];
} ctl_arenas_t;
int ctl_byname(tsd_t *tsd, const char *name, void *oldp, size_t *oldlenp,
void *newp, size_t newlen);
int ctl_nametomib(tsd_t *tsd, const char *name, size_t *mibp, size_t *miblenp);
int ctl_bymib(tsd_t *tsd, const size_t *mib, size_t miblen, void *oldp,
size_t *oldlenp, void *newp, size_t newlen);
bool ctl_boot(void);
void ctl_prefork(tsdn_t *tsdn);
void ctl_postfork_parent(tsdn_t *tsdn);
void ctl_postfork_child(tsdn_t *tsdn);
#define xmallctl(name, oldp, oldlenp, newp, newlen) do { \
if (je_mallctl(name, oldp, oldlenp, newp, newlen) \
!= 0) { \
malloc_printf( \
"<jemalloc>: Failure in xmallctl(\"%s\", ...)\n", \
name); \
abort(); \
} \
} while (0)
#define xmallctlnametomib(name, mibp, miblenp) do { \
if (je_mallctlnametomib(name, mibp, miblenp) != 0) { \
malloc_printf("<jemalloc>: Failure in " \
"xmallctlnametomib(\"%s\", ...)\n", name); \
abort(); \
} \
} while (0)
#define xmallctlbymib(mib, miblen, oldp, oldlenp, newp, newlen) do { \
if (je_mallctlbymib(mib, miblen, oldp, oldlenp, newp, \
newlen) != 0) { \
malloc_write( \
"<jemalloc>: Failure in xmallctlbymib()\n"); \
abort(); \
} \
} while (0)
#endif /* JEMALLOC_INTERNAL_CTL_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/div.h
Vendored
-41
View File
@@ -1,41 +0,0 @@
#ifndef JEMALLOC_INTERNAL_DIV_H
#define JEMALLOC_INTERNAL_DIV_H
#include "jemalloc/internal/assert.h"
/*
* This module does the division that computes the index of a region in a slab,
* given its offset relative to the base.
* That is, given a divisor d, an n = i * d (all integers), we'll return i.
* We do some pre-computation to do this more quickly than a CPU division
* instruction.
* We bound n < 2^32, and don't support dividing by one.
*/
typedef struct div_info_s div_info_t;
struct div_info_s {
uint32_t magic;
#ifdef JEMALLOC_DEBUG
size_t d;
#endif
};
void div_init(div_info_t *div_info, size_t divisor);
static inline size_t
div_compute(div_info_t *div_info, size_t n) {
assert(n <= (uint32_t)-1);
/*
* This generates, e.g. mov; imul; shr on x86-64. On a 32-bit machine,
* the compilers I tried were all smart enough to turn this into the
* appropriate "get the high 32 bits of the result of a multiply" (e.g.
* mul; mov edx eax; on x86, umull on arm, etc.).
*/
size_t i = ((uint64_t)n * (uint64_t)div_info->magic) >> 32;
#ifdef JEMALLOC_DEBUG
assert(i * div_info->d == n);
#endif
return i;
}
#endif /* JEMALLOC_INTERNAL_DIV_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/emitter.h
Vendored
-486
View File
@@ -1,486 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EMITTER_H
#define JEMALLOC_INTERNAL_EMITTER_H
#include "jemalloc/internal/ql.h"
typedef enum emitter_output_e emitter_output_t;
enum emitter_output_e {
emitter_output_json,
emitter_output_table
};
typedef enum emitter_justify_e emitter_justify_t;
enum emitter_justify_e {
emitter_justify_left,
emitter_justify_right,
/* Not for users; just to pass to internal functions. */
emitter_justify_none
};
typedef enum emitter_type_e emitter_type_t;
enum emitter_type_e {
emitter_type_bool,
emitter_type_int,
emitter_type_unsigned,
emitter_type_uint32,
emitter_type_uint64,
emitter_type_size,
emitter_type_ssize,
emitter_type_string,
/*
* A title is a column title in a table; it's just a string, but it's
* not quoted.
*/
emitter_type_title,
};
typedef struct emitter_col_s emitter_col_t;
struct emitter_col_s {
/* Filled in by the user. */
emitter_justify_t justify;
int width;
emitter_type_t type;
union {
bool bool_val;
int int_val;
unsigned unsigned_val;
uint32_t uint32_val;
uint32_t uint32_t_val;
uint64_t uint64_val;
uint64_t uint64_t_val;
size_t size_val;
ssize_t ssize_val;
const char *str_val;
};
/* Filled in by initialization. */
ql_elm(emitter_col_t) link;
};
typedef struct emitter_row_s emitter_row_t;
struct emitter_row_s {
ql_head(emitter_col_t) cols;
};
typedef struct emitter_s emitter_t;
struct emitter_s {
emitter_output_t output;
/* The output information. */
void (*write_cb)(void *, const char *);
void *cbopaque;
int nesting_depth;
/* True if we've already emitted a value at the given depth. */
bool item_at_depth;
/* True if we emitted a key and will emit corresponding value next. */
bool emitted_key;
};
/* Internal convenience function. Write to the emitter the given string. */
JEMALLOC_FORMAT_PRINTF(2, 3)
static inline void
emitter_printf(emitter_t *emitter, const char *format, ...) {
va_list ap;
va_start(ap, format);
malloc_vcprintf(emitter->write_cb, emitter->cbopaque, format, ap);
va_end(ap);
}
static inline const char * JEMALLOC_FORMAT_ARG(3)
emitter_gen_fmt(char *out_fmt, size_t out_size, const char *fmt_specifier,
emitter_justify_t justify, int width) {
size_t written;
fmt_specifier++;
if (justify == emitter_justify_none) {
written = malloc_snprintf(out_fmt, out_size,
"%%%s", fmt_specifier);
} else if (justify == emitter_justify_left) {
written = malloc_snprintf(out_fmt, out_size,
"%%-%d%s", width, fmt_specifier);
} else {
written = malloc_snprintf(out_fmt, out_size,
"%%%d%s", width, fmt_specifier);
}
/* Only happens in case of bad format string, which *we* choose. */
assert(written < out_size);
return out_fmt;
}
/*
* Internal. Emit the given value type in the relevant encoding (so that the
* bool true gets mapped to json "true", but the string "true" gets mapped to
* json "\"true\"", for instance.
*
* Width is ignored if justify is emitter_justify_none.
*/
static inline void
emitter_print_value(emitter_t *emitter, emitter_justify_t justify, int width,
emitter_type_t value_type, const void *value) {
size_t str_written;
#define BUF_SIZE 256
#define FMT_SIZE 10
/*
* We dynamically generate a format string to emit, to let us use the
* snprintf machinery. This is kinda hacky, but gets the job done
* quickly without having to think about the various snprintf edge
* cases.
*/
char fmt[FMT_SIZE];
char buf[BUF_SIZE];
#define EMIT_SIMPLE(type, format) \
emitter_printf(emitter, \
emitter_gen_fmt(fmt, FMT_SIZE, format, justify, width), \
*(const type *)value);
switch (value_type) {
case emitter_type_bool:
emitter_printf(emitter,
emitter_gen_fmt(fmt, FMT_SIZE, "%s", justify, width),
*(const bool *)value ? "true" : "false");
break;
case emitter_type_int:
EMIT_SIMPLE(int, "%d")
break;
case emitter_type_unsigned:
EMIT_SIMPLE(unsigned, "%u")
break;
case emitter_type_ssize:
EMIT_SIMPLE(ssize_t, "%zd")
break;
case emitter_type_size:
EMIT_SIMPLE(size_t, "%zu")
break;
case emitter_type_string:
str_written = malloc_snprintf(buf, BUF_SIZE, "\"%s\"",
*(const char *const *)value);
/*
* We control the strings we output; we shouldn't get anything
* anywhere near the fmt size.
*/
assert(str_written < BUF_SIZE);
emitter_printf(emitter,
emitter_gen_fmt(fmt, FMT_SIZE, "%s", justify, width), buf);
break;
case emitter_type_uint32:
EMIT_SIMPLE(uint32_t, "%" FMTu32)
break;
case emitter_type_uint64:
EMIT_SIMPLE(uint64_t, "%" FMTu64)
break;
case emitter_type_title:
EMIT_SIMPLE(char *const, "%s");
break;
default:
unreachable();
}
#undef BUF_SIZE
#undef FMT_SIZE
}
/* Internal functions. In json mode, tracks nesting state. */
static inline void
emitter_nest_inc(emitter_t *emitter) {
emitter->nesting_depth++;
emitter->item_at_depth = false;
}
static inline void
emitter_nest_dec(emitter_t *emitter) {
emitter->nesting_depth--;
emitter->item_at_depth = true;
}
static inline void
emitter_indent(emitter_t *emitter) {
int amount = emitter->nesting_depth;
const char *indent_str;
if (emitter->output == emitter_output_json) {
indent_str = "\t";
} else {
amount *= 2;
indent_str = " ";
}
for (int i = 0; i < amount; i++) {
emitter_printf(emitter, "%s", indent_str);
}
}
static inline void
emitter_json_key_prefix(emitter_t *emitter) {
if (emitter->emitted_key) {
emitter->emitted_key = false;
return;
}
emitter_printf(emitter, "%s\n", emitter->item_at_depth ? "," : "");
emitter_indent(emitter);
}
/******************************************************************************/
/* Public functions for emitter_t. */
static inline void
emitter_init(emitter_t *emitter, emitter_output_t emitter_output,
void (*write_cb)(void *, const char *), void *cbopaque) {
emitter->output = emitter_output;
emitter->write_cb = write_cb;
emitter->cbopaque = cbopaque;
emitter->item_at_depth = false;
emitter->emitted_key = false;
emitter->nesting_depth = 0;
}
/******************************************************************************/
/* JSON public API. */
/*
* Emits a key (e.g. as appears in an object). The next json entity emitted will
* be the corresponding value.
*/
static inline void
emitter_json_key(emitter_t *emitter, const char *json_key) {
if (emitter->output == emitter_output_json) {
emitter_json_key_prefix(emitter);
emitter_printf(emitter, "\"%s\": ", json_key);
emitter->emitted_key = true;
}
}
static inline void
emitter_json_value(emitter_t *emitter, emitter_type_t value_type,
const void *value) {
if (emitter->output == emitter_output_json) {
emitter_json_key_prefix(emitter);
emitter_print_value(emitter, emitter_justify_none, -1,
value_type, value);
emitter->item_at_depth = true;
}
}
/* Shorthand for calling emitter_json_key and then emitter_json_value. */
static inline void
emitter_json_kv(emitter_t *emitter, const char *json_key,
emitter_type_t value_type, const void *value) {
emitter_json_key(emitter, json_key);
emitter_json_value(emitter, value_type, value);
}
static inline void
emitter_json_array_begin(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
emitter_json_key_prefix(emitter);
emitter_printf(emitter, "[");
emitter_nest_inc(emitter);
}
}
/* Shorthand for calling emitter_json_key and then emitter_json_array_begin. */
static inline void
emitter_json_array_kv_begin(emitter_t *emitter, const char *json_key) {
emitter_json_key(emitter, json_key);
emitter_json_array_begin(emitter);
}
static inline void
emitter_json_array_end(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
assert(emitter->nesting_depth > 0);
emitter_nest_dec(emitter);
emitter_printf(emitter, "\n");
emitter_indent(emitter);
emitter_printf(emitter, "]");
}
}
static inline void
emitter_json_object_begin(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
emitter_json_key_prefix(emitter);
emitter_printf(emitter, "{");
emitter_nest_inc(emitter);
}
}
/* Shorthand for calling emitter_json_key and then emitter_json_object_begin. */
static inline void
emitter_json_object_kv_begin(emitter_t *emitter, const char *json_key) {
emitter_json_key(emitter, json_key);
emitter_json_object_begin(emitter);
}
static inline void
emitter_json_object_end(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
assert(emitter->nesting_depth > 0);
emitter_nest_dec(emitter);
emitter_printf(emitter, "\n");
emitter_indent(emitter);
emitter_printf(emitter, "}");
}
}
/******************************************************************************/
/* Table public API. */
static inline void
emitter_table_dict_begin(emitter_t *emitter, const char *table_key) {
if (emitter->output == emitter_output_table) {
emitter_indent(emitter);
emitter_printf(emitter, "%s\n", table_key);
emitter_nest_inc(emitter);
}
}
static inline void
emitter_table_dict_end(emitter_t *emitter) {
if (emitter->output == emitter_output_table) {
emitter_nest_dec(emitter);
}
}
static inline void
emitter_table_kv_note(emitter_t *emitter, const char *table_key,
emitter_type_t value_type, const void *value,
const char *table_note_key, emitter_type_t table_note_value_type,
const void *table_note_value) {
if (emitter->output == emitter_output_table) {
emitter_indent(emitter);
emitter_printf(emitter, "%s: ", table_key);
emitter_print_value(emitter, emitter_justify_none, -1,
value_type, value);
if (table_note_key != NULL) {
emitter_printf(emitter, " (%s: ", table_note_key);
emitter_print_value(emitter, emitter_justify_none, -1,
table_note_value_type, table_note_value);
emitter_printf(emitter, ")");
}
emitter_printf(emitter, "\n");
}
emitter->item_at_depth = true;
}
static inline void
emitter_table_kv(emitter_t *emitter, const char *table_key,
emitter_type_t value_type, const void *value) {
emitter_table_kv_note(emitter, table_key, value_type, value, NULL,
emitter_type_bool, NULL);
}
/* Write to the emitter the given string, but only in table mode. */
JEMALLOC_FORMAT_PRINTF(2, 3)
static inline void
emitter_table_printf(emitter_t *emitter, const char *format, ...) {
if (emitter->output == emitter_output_table) {
va_list ap;
va_start(ap, format);
malloc_vcprintf(emitter->write_cb, emitter->cbopaque, format, ap);
va_end(ap);
}
}
static inline void
emitter_table_row(emitter_t *emitter, emitter_row_t *row) {
if (emitter->output != emitter_output_table) {
return;
}
emitter_col_t *col;
ql_foreach(col, &row->cols, link) {
emitter_print_value(emitter, col->justify, col->width,
col->type, (const void *)&col->bool_val);
}
emitter_table_printf(emitter, "\n");
}
static inline void
emitter_row_init(emitter_row_t *row) {
ql_new(&row->cols);
}
static inline void
emitter_col_init(emitter_col_t *col, emitter_row_t *row) {
ql_elm_new(col, link);
ql_tail_insert(&row->cols, col, link);
}
/******************************************************************************/
/*
* Generalized public API. Emits using either JSON or table, according to
* settings in the emitter_t. */
/*
* Note emits a different kv pair as well, but only in table mode. Omits the
* note if table_note_key is NULL.
*/
static inline void
emitter_kv_note(emitter_t *emitter, const char *json_key, const char *table_key,
emitter_type_t value_type, const void *value,
const char *table_note_key, emitter_type_t table_note_value_type,
const void *table_note_value) {
if (emitter->output == emitter_output_json) {
emitter_json_key(emitter, json_key);
emitter_json_value(emitter, value_type, value);
} else {
emitter_table_kv_note(emitter, table_key, value_type, value,
table_note_key, table_note_value_type, table_note_value);
}
emitter->item_at_depth = true;
}
static inline void
emitter_kv(emitter_t *emitter, const char *json_key, const char *table_key,
emitter_type_t value_type, const void *value) {
emitter_kv_note(emitter, json_key, table_key, value_type, value, NULL,
emitter_type_bool, NULL);
}
static inline void
emitter_dict_begin(emitter_t *emitter, const char *json_key,
const char *table_header) {
if (emitter->output == emitter_output_json) {
emitter_json_key(emitter, json_key);
emitter_json_object_begin(emitter);
} else {
emitter_table_dict_begin(emitter, table_header);
}
}
static inline void
emitter_dict_end(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
emitter_json_object_end(emitter);
} else {
emitter_table_dict_end(emitter);
}
}
static inline void
emitter_begin(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
assert(emitter->nesting_depth == 0);
emitter_printf(emitter, "{");
emitter_nest_inc(emitter);
} else {
/*
* This guarantees that we always call write_cb at least once.
* This is useful if some invariant is established by each call
* to write_cb, but doesn't hold initially: e.g., some buffer
* holds a null-terminated string.
*/
emitter_printf(emitter, "%s", "");
}
}
static inline void
emitter_end(emitter_t *emitter) {
if (emitter->output == emitter_output_json) {
assert(emitter->nesting_depth == 1);
emitter_nest_dec(emitter);
emitter_printf(emitter, "\n}\n");
}
}
#endif /* JEMALLOC_INTERNAL_EMITTER_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/extent_dss.h
Vendored
-26
View File
@@ -1,26 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EXTENT_DSS_H
#define JEMALLOC_INTERNAL_EXTENT_DSS_H
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"
extern const char *dss_prec_names[];
extern const char *opt_dss;
dss_prec_t extent_dss_prec_get(void);
bool extent_dss_prec_set(dss_prec_t dss_prec);
void *extent_alloc_dss(tsdn_t *tsdn, arena_t *arena, void *new_addr,
size_t size, size_t alignment, bool *zero, bool *commit);
bool extent_in_dss(void *addr);
bool extent_dss_mergeable(void *addr_a, void *addr_b);
void extent_dss_boot(void);
#endif /* JEMALLOC_INTERNAL_EXTENT_DSS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/extent_externs.h
Vendored
-83
View File
@@ -1,83 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EXTENT_EXTERNS_H
#define JEMALLOC_INTERNAL_EXTENT_EXTERNS_H
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/mutex_pool.h"
#include "jemalloc/internal/ph.h"
#include "jemalloc/internal/rtree.h"
extern size_t opt_lg_extent_max_active_fit;
extern rtree_t extents_rtree;
extern const extent_hooks_t extent_hooks_default;
extern mutex_pool_t extent_mutex_pool;
extent_t *extent_alloc(tsdn_t *tsdn, arena_t *arena);
void extent_dalloc(tsdn_t *tsdn, arena_t *arena, extent_t *extent);
extent_hooks_t *extent_hooks_get(arena_t *arena);
extent_hooks_t *extent_hooks_set(tsd_t *tsd, arena_t *arena,
extent_hooks_t *extent_hooks);
#ifdef JEMALLOC_JET
size_t extent_size_quantize_floor(size_t size);
size_t extent_size_quantize_ceil(size_t size);
#endif
ph_proto(, extent_avail_, extent_tree_t, extent_t)
ph_proto(, extent_heap_, extent_heap_t, extent_t)
bool extents_init(tsdn_t *tsdn, extents_t *extents, extent_state_t state,
bool delay_coalesce);
extent_state_t extents_state_get(const extents_t *extents);
size_t extents_npages_get(extents_t *extents);
/* Get the number of extents in the given page size index. */
size_t extents_nextents_get(extents_t *extents, pszind_t ind);
/* Get the sum total bytes of the extents in the given page size index. */
size_t extents_nbytes_get(extents_t *extents, pszind_t ind);
extent_t *extents_alloc(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extents_t *extents, void *new_addr,
size_t size, size_t pad, size_t alignment, bool slab, szind_t szind,
bool *zero, bool *commit);
void extents_dalloc(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extents_t *extents, extent_t *extent);
extent_t *extents_evict(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extents_t *extents, size_t npages_min);
void extents_prefork(tsdn_t *tsdn, extents_t *extents);
void extents_postfork_parent(tsdn_t *tsdn, extents_t *extents);
void extents_postfork_child(tsdn_t *tsdn, extents_t *extents);
extent_t *extent_alloc_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, void *new_addr, size_t size, size_t pad,
size_t alignment, bool slab, szind_t szind, bool *zero, bool *commit);
void extent_dalloc_gap(tsdn_t *tsdn, arena_t *arena, extent_t *extent);
void extent_dalloc_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent);
void extent_destroy_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent);
bool extent_commit_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset,
size_t length);
bool extent_decommit_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset,
size_t length);
bool extent_purge_lazy_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset,
size_t length);
bool extent_purge_forced_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent, size_t offset,
size_t length);
extent_t *extent_split_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *extent, size_t size_a,
szind_t szind_a, bool slab_a, size_t size_b, szind_t szind_b, bool slab_b);
bool extent_merge_wrapper(tsdn_t *tsdn, arena_t *arena,
extent_hooks_t **r_extent_hooks, extent_t *a, extent_t *b);
bool extent_boot(void);
void extent_util_stats_get(tsdn_t *tsdn, const void *ptr,
size_t *nfree, size_t *nregs, size_t *size);
void extent_util_stats_verbose_get(tsdn_t *tsdn, const void *ptr,
size_t *nfree, size_t *nregs, size_t *size,
size_t *bin_nfree, size_t *bin_nregs, void **slabcur_addr);
#endif /* JEMALLOC_INTERNAL_EXTENT_EXTERNS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/extent_inlines.h
Vendored
-501
View File
@@ -1,501 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EXTENT_INLINES_H
#define JEMALLOC_INTERNAL_EXTENT_INLINES_H
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/mutex_pool.h"
#include "jemalloc/internal/pages.h"
#include "jemalloc/internal/prng.h"
#include "jemalloc/internal/ql.h"
#include "jemalloc/internal/sc.h"
#include "jemalloc/internal/sz.h"
static inline void
extent_lock(tsdn_t *tsdn, extent_t *extent) {
assert(extent != NULL);
mutex_pool_lock(tsdn, &extent_mutex_pool, (uintptr_t)extent);
}
static inline void
extent_unlock(tsdn_t *tsdn, extent_t *extent) {
assert(extent != NULL);
mutex_pool_unlock(tsdn, &extent_mutex_pool, (uintptr_t)extent);
}
static inline void
extent_lock2(tsdn_t *tsdn, extent_t *extent1, extent_t *extent2) {
assert(extent1 != NULL && extent2 != NULL);
mutex_pool_lock2(tsdn, &extent_mutex_pool, (uintptr_t)extent1,
(uintptr_t)extent2);
}
static inline void
extent_unlock2(tsdn_t *tsdn, extent_t *extent1, extent_t *extent2) {
assert(extent1 != NULL && extent2 != NULL);
mutex_pool_unlock2(tsdn, &extent_mutex_pool, (uintptr_t)extent1,
(uintptr_t)extent2);
}
static inline unsigned
extent_arena_ind_get(const extent_t *extent) {
unsigned arena_ind = (unsigned)((extent->e_bits &
EXTENT_BITS_ARENA_MASK) >> EXTENT_BITS_ARENA_SHIFT);
assert(arena_ind < MALLOCX_ARENA_LIMIT);
return arena_ind;
}
static inline arena_t *
extent_arena_get(const extent_t *extent) {
unsigned arena_ind = extent_arena_ind_get(extent);
return (arena_t *)atomic_load_p(&arenas[arena_ind], ATOMIC_ACQUIRE);
}
static inline szind_t
extent_szind_get_maybe_invalid(const extent_t *extent) {
szind_t szind = (szind_t)((extent->e_bits & EXTENT_BITS_SZIND_MASK) >>
EXTENT_BITS_SZIND_SHIFT);
assert(szind <= SC_NSIZES);
return szind;
}
static inline szind_t
extent_szind_get(const extent_t *extent) {
szind_t szind = extent_szind_get_maybe_invalid(extent);
assert(szind < SC_NSIZES); /* Never call when "invalid". */
return szind;
}
static inline size_t
extent_usize_get(const extent_t *extent) {
return sz_index2size(extent_szind_get(extent));
}
static inline unsigned
extent_binshard_get(const extent_t *extent) {
unsigned binshard = (unsigned)((extent->e_bits &
EXTENT_BITS_BINSHARD_MASK) >> EXTENT_BITS_BINSHARD_SHIFT);
assert(binshard < bin_infos[extent_szind_get(extent)].n_shards);
return binshard;
}
static inline size_t
extent_sn_get(const extent_t *extent) {
return (size_t)((extent->e_bits & EXTENT_BITS_SN_MASK) >>
EXTENT_BITS_SN_SHIFT);
}
static inline extent_state_t
extent_state_get(const extent_t *extent) {
return (extent_state_t)((extent->e_bits & EXTENT_BITS_STATE_MASK) >>
EXTENT_BITS_STATE_SHIFT);
}
static inline bool
extent_zeroed_get(const extent_t *extent) {
return (bool)((extent->e_bits & EXTENT_BITS_ZEROED_MASK) >>
EXTENT_BITS_ZEROED_SHIFT);
}
static inline bool
extent_committed_get(const extent_t *extent) {
return (bool)((extent->e_bits & EXTENT_BITS_COMMITTED_MASK) >>
EXTENT_BITS_COMMITTED_SHIFT);
}
static inline bool
extent_dumpable_get(const extent_t *extent) {
return (bool)((extent->e_bits & EXTENT_BITS_DUMPABLE_MASK) >>
EXTENT_BITS_DUMPABLE_SHIFT);
}
static inline bool
extent_slab_get(const extent_t *extent) {
return (bool)((extent->e_bits & EXTENT_BITS_SLAB_MASK) >>
EXTENT_BITS_SLAB_SHIFT);
}
static inline unsigned
extent_nfree_get(const extent_t *extent) {
assert(extent_slab_get(extent));
return (unsigned)((extent->e_bits & EXTENT_BITS_NFREE_MASK) >>
EXTENT_BITS_NFREE_SHIFT);
}
static inline void *
extent_base_get(const extent_t *extent) {
assert(extent->e_addr == PAGE_ADDR2BASE(extent->e_addr) ||
!extent_slab_get(extent));
return PAGE_ADDR2BASE(extent->e_addr);
}
static inline void *
extent_addr_get(const extent_t *extent) {
assert(extent->e_addr == PAGE_ADDR2BASE(extent->e_addr) ||
!extent_slab_get(extent));
return extent->e_addr;
}
static inline size_t
extent_size_get(const extent_t *extent) {
return (extent->e_size_esn & EXTENT_SIZE_MASK);
}
static inline size_t
extent_esn_get(const extent_t *extent) {
return (extent->e_size_esn & EXTENT_ESN_MASK);
}
static inline size_t
extent_bsize_get(const extent_t *extent) {
return extent->e_bsize;
}
static inline void *
extent_before_get(const extent_t *extent) {
return (void *)((uintptr_t)extent_base_get(extent) - PAGE);
}
static inline void *
extent_last_get(const extent_t *extent) {
return (void *)((uintptr_t)extent_base_get(extent) +
extent_size_get(extent) - PAGE);
}
static inline void *
extent_past_get(const extent_t *extent) {
return (void *)((uintptr_t)extent_base_get(extent) +
extent_size_get(extent));
}
static inline arena_slab_data_t *
extent_slab_data_get(extent_t *extent) {
assert(extent_slab_get(extent));
return &extent->e_slab_data;
}
static inline const arena_slab_data_t *
extent_slab_data_get_const(const extent_t *extent) {
assert(extent_slab_get(extent));
return &extent->e_slab_data;
}
static inline prof_tctx_t *
extent_prof_tctx_get(const extent_t *extent) {
return (prof_tctx_t *)atomic_load_p(&extent->e_prof_tctx,
ATOMIC_ACQUIRE);
}
static inline nstime_t
extent_prof_alloc_time_get(const extent_t *extent) {
return extent->e_alloc_time;
}
static inline void
extent_arena_set(extent_t *extent, arena_t *arena) {
unsigned arena_ind = (arena != NULL) ? arena_ind_get(arena) : ((1U <<
MALLOCX_ARENA_BITS) - 1);
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_ARENA_MASK) |
((uint64_t)arena_ind << EXTENT_BITS_ARENA_SHIFT);
}
static inline void
extent_binshard_set(extent_t *extent, unsigned binshard) {
/* The assertion assumes szind is set already. */
assert(binshard < bin_infos[extent_szind_get(extent)].n_shards);
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_BINSHARD_MASK) |
((uint64_t)binshard << EXTENT_BITS_BINSHARD_SHIFT);
}
static inline void
extent_addr_set(extent_t *extent, void *addr) {
extent->e_addr = addr;
}
static inline void
extent_addr_randomize(tsdn_t *tsdn, extent_t *extent, size_t alignment) {
assert(extent_base_get(extent) == extent_addr_get(extent));
if (alignment < PAGE) {
unsigned lg_range = LG_PAGE -
lg_floor(CACHELINE_CEILING(alignment));
size_t r;
if (!tsdn_null(tsdn)) {
tsd_t *tsd = tsdn_tsd(tsdn);
r = (size_t)prng_lg_range_u64(
tsd_offset_statep_get(tsd), lg_range);
} else {
r = prng_lg_range_zu(
&extent_arena_get(extent)->offset_state,
lg_range, true);
}
uintptr_t random_offset = ((uintptr_t)r) << (LG_PAGE -
lg_range);
extent->e_addr = (void *)((uintptr_t)extent->e_addr +
random_offset);
assert(ALIGNMENT_ADDR2BASE(extent->e_addr, alignment) ==
extent->e_addr);
}
}
static inline void
extent_size_set(extent_t *extent, size_t size) {
assert((size & ~EXTENT_SIZE_MASK) == 0);
extent->e_size_esn = size | (extent->e_size_esn & ~EXTENT_SIZE_MASK);
}
static inline void
extent_esn_set(extent_t *extent, size_t esn) {
extent->e_size_esn = (extent->e_size_esn & ~EXTENT_ESN_MASK) | (esn &
EXTENT_ESN_MASK);
}
static inline void
extent_bsize_set(extent_t *extent, size_t bsize) {
extent->e_bsize = bsize;
}
static inline void
extent_szind_set(extent_t *extent, szind_t szind) {
assert(szind <= SC_NSIZES); /* SC_NSIZES means "invalid". */
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_SZIND_MASK) |
((uint64_t)szind << EXTENT_BITS_SZIND_SHIFT);
}
static inline void
extent_nfree_set(extent_t *extent, unsigned nfree) {
assert(extent_slab_get(extent));
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_NFREE_MASK) |
((uint64_t)nfree << EXTENT_BITS_NFREE_SHIFT);
}
static inline void
extent_nfree_binshard_set(extent_t *extent, unsigned nfree, unsigned binshard) {
/* The assertion assumes szind is set already. */
assert(binshard < bin_infos[extent_szind_get(extent)].n_shards);
extent->e_bits = (extent->e_bits &
(~EXTENT_BITS_NFREE_MASK & ~EXTENT_BITS_BINSHARD_MASK)) |
((uint64_t)binshard << EXTENT_BITS_BINSHARD_SHIFT) |
((uint64_t)nfree << EXTENT_BITS_NFREE_SHIFT);
}
static inline void
extent_nfree_inc(extent_t *extent) {
assert(extent_slab_get(extent));
extent->e_bits += ((uint64_t)1U << EXTENT_BITS_NFREE_SHIFT);
}
static inline void
extent_nfree_dec(extent_t *extent) {
assert(extent_slab_get(extent));
extent->e_bits -= ((uint64_t)1U << EXTENT_BITS_NFREE_SHIFT);
}
static inline void
extent_nfree_sub(extent_t *extent, uint64_t n) {
assert(extent_slab_get(extent));
extent->e_bits -= (n << EXTENT_BITS_NFREE_SHIFT);
}
static inline void
extent_sn_set(extent_t *extent, size_t sn) {
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_SN_MASK) |
((uint64_t)sn << EXTENT_BITS_SN_SHIFT);
}
static inline void
extent_state_set(extent_t *extent, extent_state_t state) {
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_STATE_MASK) |
((uint64_t)state << EXTENT_BITS_STATE_SHIFT);
}
static inline void
extent_zeroed_set(extent_t *extent, bool zeroed) {
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_ZEROED_MASK) |
((uint64_t)zeroed << EXTENT_BITS_ZEROED_SHIFT);
}
static inline void
extent_committed_set(extent_t *extent, bool committed) {
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_COMMITTED_MASK) |
((uint64_t)committed << EXTENT_BITS_COMMITTED_SHIFT);
}
static inline void
extent_dumpable_set(extent_t *extent, bool dumpable) {
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_DUMPABLE_MASK) |
((uint64_t)dumpable << EXTENT_BITS_DUMPABLE_SHIFT);
}
static inline void
extent_slab_set(extent_t *extent, bool slab) {
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_SLAB_MASK) |
((uint64_t)slab << EXTENT_BITS_SLAB_SHIFT);
}
static inline void
extent_prof_tctx_set(extent_t *extent, prof_tctx_t *tctx) {
atomic_store_p(&extent->e_prof_tctx, tctx, ATOMIC_RELEASE);
}
static inline void
extent_prof_alloc_time_set(extent_t *extent, nstime_t t) {
nstime_copy(&extent->e_alloc_time, &t);
}
static inline bool
extent_is_head_get(extent_t *extent) {
if (maps_coalesce) {
not_reached();
}
return (bool)((extent->e_bits & EXTENT_BITS_IS_HEAD_MASK) >>
EXTENT_BITS_IS_HEAD_SHIFT);
}
static inline void
extent_is_head_set(extent_t *extent, bool is_head) {
if (maps_coalesce) {
not_reached();
}
extent->e_bits = (extent->e_bits & ~EXTENT_BITS_IS_HEAD_MASK) |
((uint64_t)is_head << EXTENT_BITS_IS_HEAD_SHIFT);
}
static inline void
extent_init(extent_t *extent, arena_t *arena, void *addr, size_t size,
bool slab, szind_t szind, size_t sn, extent_state_t state, bool zeroed,
bool committed, bool dumpable, extent_head_state_t is_head) {
assert(addr == PAGE_ADDR2BASE(addr) || !slab);
extent_arena_set(extent, arena);
extent_addr_set(extent, addr);
extent_size_set(extent, size);
extent_slab_set(extent, slab);
extent_szind_set(extent, szind);
extent_sn_set(extent, sn);
extent_state_set(extent, state);
extent_zeroed_set(extent, zeroed);
extent_committed_set(extent, committed);
extent_dumpable_set(extent, dumpable);
ql_elm_new(extent, ql_link);
if (!maps_coalesce) {
extent_is_head_set(extent, (is_head == EXTENT_IS_HEAD) ? true :
false);
}
if (config_prof) {
extent_prof_tctx_set(extent, NULL);
}
}
static inline void
extent_binit(extent_t *extent, void *addr, size_t bsize, size_t sn) {
extent_arena_set(extent, NULL);
extent_addr_set(extent, addr);
extent_bsize_set(extent, bsize);
extent_slab_set(extent, false);
extent_szind_set(extent, SC_NSIZES);
extent_sn_set(extent, sn);
extent_state_set(extent, extent_state_active);
extent_zeroed_set(extent, true);
extent_committed_set(extent, true);
extent_dumpable_set(extent, true);
}
static inline void
extent_list_init(extent_list_t *list) {
ql_new(list);
}
static inline extent_t *
extent_list_first(const extent_list_t *list) {
return ql_first(list);
}
static inline extent_t *
extent_list_last(const extent_list_t *list) {
return ql_last(list, ql_link);
}
static inline void
extent_list_append(extent_list_t *list, extent_t *extent) {
ql_tail_insert(list, extent, ql_link);
}
static inline void
extent_list_prepend(extent_list_t *list, extent_t *extent) {
ql_head_insert(list, extent, ql_link);
}
static inline void
extent_list_replace(extent_list_t *list, extent_t *to_remove,
extent_t *to_insert) {
ql_after_insert(to_remove, to_insert, ql_link);
ql_remove(list, to_remove, ql_link);
}
static inline void
extent_list_remove(extent_list_t *list, extent_t *extent) {
ql_remove(list, extent, ql_link);
}
static inline int
extent_sn_comp(const extent_t *a, const extent_t *b) {
size_t a_sn = extent_sn_get(a);
size_t b_sn = extent_sn_get(b);
return (a_sn > b_sn) - (a_sn < b_sn);
}
static inline int
extent_esn_comp(const extent_t *a, const extent_t *b) {
size_t a_esn = extent_esn_get(a);
size_t b_esn = extent_esn_get(b);
return (a_esn > b_esn) - (a_esn < b_esn);
}
static inline int
extent_ad_comp(const extent_t *a, const extent_t *b) {
uintptr_t a_addr = (uintptr_t)extent_addr_get(a);
uintptr_t b_addr = (uintptr_t)extent_addr_get(b);
return (a_addr > b_addr) - (a_addr < b_addr);
}
static inline int
extent_ead_comp(const extent_t *a, const extent_t *b) {
uintptr_t a_eaddr = (uintptr_t)a;
uintptr_t b_eaddr = (uintptr_t)b;
return (a_eaddr > b_eaddr) - (a_eaddr < b_eaddr);
}
static inline int
extent_snad_comp(const extent_t *a, const extent_t *b) {
int ret;
ret = extent_sn_comp(a, b);
if (ret != 0) {
return ret;
}
ret = extent_ad_comp(a, b);
return ret;
}
static inline int
extent_esnead_comp(const extent_t *a, const extent_t *b) {
int ret;
ret = extent_esn_comp(a, b);
if (ret != 0) {
return ret;
}
ret = extent_ead_comp(a, b);
return ret;
}
#endif /* JEMALLOC_INTERNAL_EXTENT_INLINES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/extent_mmap.h
Vendored
-10
View File
@@ -1,10 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EXTENT_MMAP_EXTERNS_H
#define JEMALLOC_INTERNAL_EXTENT_MMAP_EXTERNS_H
extern bool opt_retain;
void *extent_alloc_mmap(void *new_addr, size_t size, size_t alignment,
bool *zero, bool *commit);
bool extent_dalloc_mmap(void *addr, size_t size);
#endif /* JEMALLOC_INTERNAL_EXTENT_MMAP_EXTERNS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/extent_structs.h
Vendored
-256
View File
@@ -1,256 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EXTENT_STRUCTS_H
#define JEMALLOC_INTERNAL_EXTENT_STRUCTS_H
#include "jemalloc/internal/atomic.h"
#include "jemalloc/internal/bit_util.h"
#include "jemalloc/internal/bitmap.h"
#include "jemalloc/internal/mutex.h"
#include "jemalloc/internal/ql.h"
#include "jemalloc/internal/ph.h"
#include "jemalloc/internal/sc.h"
typedef enum {
extent_state_active = 0,
extent_state_dirty = 1,
extent_state_muzzy = 2,
extent_state_retained = 3
} extent_state_t;
/* Extent (span of pages). Use accessor functions for e_* fields. */
struct extent_s {
/*
* Bitfield containing several fields:
*
* a: arena_ind
* b: slab
* c: committed
* d: dumpable
* z: zeroed
* t: state
* i: szind
* f: nfree
* s: bin_shard
* n: sn
*
* nnnnnnnn ... nnnnnnss ssssffff ffffffii iiiiiitt zdcbaaaa aaaaaaaa
*
* arena_ind: Arena from which this extent came, or all 1 bits if
* unassociated.
*
* slab: The slab flag indicates whether the extent is used for a slab
* of small regions. This helps differentiate small size classes,
* and it indicates whether interior pointers can be looked up via
* iealloc().
*
* committed: The committed flag indicates whether physical memory is
* committed to the extent, whether explicitly or implicitly
* as on a system that overcommits and satisfies physical
* memory needs on demand via soft page faults.
*
* dumpable: The dumpable flag indicates whether or not we've set the
* memory in question to be dumpable. Note that this
* interacts somewhat subtly with user-specified extent hooks,
* since we don't know if *they* are fiddling with
* dumpability (in which case, we don't want to undo whatever
* they're doing). To deal with this scenario, we:
* - Make dumpable false only for memory allocated with the
* default hooks.
* - Only allow memory to go from non-dumpable to dumpable,
* and only once.
* - Never make the OS call to allow dumping when the
* dumpable bit is already set.
* These three constraints mean that we will never
* accidentally dump user memory that the user meant to set
* nondumpable with their extent hooks.
*
*
* zeroed: The zeroed flag is used by extent recycling code to track
* whether memory is zero-filled.
*
* state: The state flag is an extent_state_t.
*
* szind: The szind flag indicates usable size class index for
* allocations residing in this extent, regardless of whether the
* extent is a slab. Extent size and usable size often differ
* even for non-slabs, either due to sz_large_pad or promotion of
* sampled small regions.
*
* nfree: Number of free regions in slab.
*
* bin_shard: the shard of the bin from which this extent came.
*
* sn: Serial number (potentially non-unique).
*
* Serial numbers may wrap around if !opt_retain, but as long as
* comparison functions fall back on address comparison for equal
* serial numbers, stable (if imperfect) ordering is maintained.
*
* Serial numbers may not be unique even in the absence of
* wrap-around, e.g. when splitting an extent and assigning the same
* serial number to both resulting adjacent extents.
*/
uint64_t e_bits;
#define MASK(CURRENT_FIELD_WIDTH, CURRENT_FIELD_SHIFT) ((((((uint64_t)0x1U) << (CURRENT_FIELD_WIDTH)) - 1)) << (CURRENT_FIELD_SHIFT))
#define EXTENT_BITS_ARENA_WIDTH MALLOCX_ARENA_BITS
#define EXTENT_BITS_ARENA_SHIFT 0
#define EXTENT_BITS_ARENA_MASK MASK(EXTENT_BITS_ARENA_WIDTH, EXTENT_BITS_ARENA_SHIFT)
#define EXTENT_BITS_SLAB_WIDTH 1
#define EXTENT_BITS_SLAB_SHIFT (EXTENT_BITS_ARENA_WIDTH + EXTENT_BITS_ARENA_SHIFT)
#define EXTENT_BITS_SLAB_MASK MASK(EXTENT_BITS_SLAB_WIDTH, EXTENT_BITS_SLAB_SHIFT)
#define EXTENT_BITS_COMMITTED_WIDTH 1
#define EXTENT_BITS_COMMITTED_SHIFT (EXTENT_BITS_SLAB_WIDTH + EXTENT_BITS_SLAB_SHIFT)
#define EXTENT_BITS_COMMITTED_MASK MASK(EXTENT_BITS_COMMITTED_WIDTH, EXTENT_BITS_COMMITTED_SHIFT)
#define EXTENT_BITS_DUMPABLE_WIDTH 1
#define EXTENT_BITS_DUMPABLE_SHIFT (EXTENT_BITS_COMMITTED_WIDTH + EXTENT_BITS_COMMITTED_SHIFT)
#define EXTENT_BITS_DUMPABLE_MASK MASK(EXTENT_BITS_DUMPABLE_WIDTH, EXTENT_BITS_DUMPABLE_SHIFT)
#define EXTENT_BITS_ZEROED_WIDTH 1
#define EXTENT_BITS_ZEROED_SHIFT (EXTENT_BITS_DUMPABLE_WIDTH + EXTENT_BITS_DUMPABLE_SHIFT)
#define EXTENT_BITS_ZEROED_MASK MASK(EXTENT_BITS_ZEROED_WIDTH, EXTENT_BITS_ZEROED_SHIFT)
#define EXTENT_BITS_STATE_WIDTH 2
#define EXTENT_BITS_STATE_SHIFT (EXTENT_BITS_ZEROED_WIDTH + EXTENT_BITS_ZEROED_SHIFT)
#define EXTENT_BITS_STATE_MASK MASK(EXTENT_BITS_STATE_WIDTH, EXTENT_BITS_STATE_SHIFT)
#define EXTENT_BITS_SZIND_WIDTH LG_CEIL(SC_NSIZES)
#define EXTENT_BITS_SZIND_SHIFT (EXTENT_BITS_STATE_WIDTH + EXTENT_BITS_STATE_SHIFT)
#define EXTENT_BITS_SZIND_MASK MASK(EXTENT_BITS_SZIND_WIDTH, EXTENT_BITS_SZIND_SHIFT)
#define EXTENT_BITS_NFREE_WIDTH (LG_SLAB_MAXREGS + 1)
#define EXTENT_BITS_NFREE_SHIFT (EXTENT_BITS_SZIND_WIDTH + EXTENT_BITS_SZIND_SHIFT)
#define EXTENT_BITS_NFREE_MASK MASK(EXTENT_BITS_NFREE_WIDTH, EXTENT_BITS_NFREE_SHIFT)
#define EXTENT_BITS_BINSHARD_WIDTH 6
#define EXTENT_BITS_BINSHARD_SHIFT (EXTENT_BITS_NFREE_WIDTH + EXTENT_BITS_NFREE_SHIFT)
#define EXTENT_BITS_BINSHARD_MASK MASK(EXTENT_BITS_BINSHARD_WIDTH, EXTENT_BITS_BINSHARD_SHIFT)
#define EXTENT_BITS_IS_HEAD_WIDTH 1
#define EXTENT_BITS_IS_HEAD_SHIFT (EXTENT_BITS_BINSHARD_WIDTH + EXTENT_BITS_BINSHARD_SHIFT)
#define EXTENT_BITS_IS_HEAD_MASK MASK(EXTENT_BITS_IS_HEAD_WIDTH, EXTENT_BITS_IS_HEAD_SHIFT)
#define EXTENT_BITS_SN_SHIFT (EXTENT_BITS_IS_HEAD_WIDTH + EXTENT_BITS_IS_HEAD_SHIFT)
#define EXTENT_BITS_SN_MASK (UINT64_MAX << EXTENT_BITS_SN_SHIFT)
/* Pointer to the extent that this structure is responsible for. */
void *e_addr;
union {
/*
* Extent size and serial number associated with the extent
* structure (different than the serial number for the extent at
* e_addr).
*
* ssssssss [...] ssssssss ssssnnnn nnnnnnnn
*/
size_t e_size_esn;
#define EXTENT_SIZE_MASK ((size_t)~(PAGE-1))
#define EXTENT_ESN_MASK ((size_t)PAGE-1)
/* Base extent size, which may not be a multiple of PAGE. */
size_t e_bsize;
};
/*
* List linkage, used by a variety of lists:
* - bin_t's slabs_full
* - extents_t's LRU
* - stashed dirty extents
* - arena's large allocations
*/
ql_elm(extent_t) ql_link;
/*
* Linkage for per size class sn/address-ordered heaps, and
* for extent_avail
*/
phn(extent_t) ph_link;
union {
/* Small region slab metadata. */
arena_slab_data_t e_slab_data;
/* Profiling data, used for large objects. */
struct {
/* Time when this was allocated. */
nstime_t e_alloc_time;
/* Points to a prof_tctx_t. */
atomic_p_t e_prof_tctx;
};
};
};
typedef ql_head(extent_t) extent_list_t;
typedef ph(extent_t) extent_tree_t;
typedef ph(extent_t) extent_heap_t;
/* Quantized collection of extents, with built-in LRU queue. */
struct extents_s {
malloc_mutex_t mtx;
/*
* Quantized per size class heaps of extents.
*
* Synchronization: mtx.
*/
extent_heap_t heaps[SC_NPSIZES + 1];
atomic_zu_t nextents[SC_NPSIZES + 1];
atomic_zu_t nbytes[SC_NPSIZES + 1];
/*
* Bitmap for which set bits correspond to non-empty heaps.
*
* Synchronization: mtx.
*/
bitmap_t bitmap[BITMAP_GROUPS(SC_NPSIZES + 1)];
/*
* LRU of all extents in heaps.
*
* Synchronization: mtx.
*/
extent_list_t lru;
/*
* Page sum for all extents in heaps.
*
* The synchronization here is a little tricky. Modifications to npages
* must hold mtx, but reads need not (though, a reader who sees npages
* without holding the mutex can't assume anything about the rest of the
* state of the extents_t).
*/
atomic_zu_t npages;
/* All stored extents must be in the same state. */
extent_state_t state;
/*
* If true, delay coalescing until eviction; otherwise coalesce during
* deallocation.
*/
bool delay_coalesce;
};
/*
* The following two structs are for experimental purposes. See
* experimental_utilization_query_ctl and
* experimental_utilization_batch_query_ctl in src/ctl.c.
*/
struct extent_util_stats_s {
size_t nfree;
size_t nregs;
size_t size;
};
struct extent_util_stats_verbose_s {
void *slabcur_addr;
size_t nfree;
size_t nregs;
size_t size;
size_t bin_nfree;
size_t bin_nregs;
};
#endif /* JEMALLOC_INTERNAL_EXTENT_STRUCTS_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/extent_types.h
Vendored
-23
View File
@@ -1,23 +0,0 @@
#ifndef JEMALLOC_INTERNAL_EXTENT_TYPES_H
#define JEMALLOC_INTERNAL_EXTENT_TYPES_H
typedef struct extent_s extent_t;
typedef struct extents_s extents_t;
typedef struct extent_util_stats_s extent_util_stats_t;
typedef struct extent_util_stats_verbose_s extent_util_stats_verbose_t;
#define EXTENT_HOOKS_INITIALIZER NULL
/*
* When reuse (and split) an active extent, (1U << opt_lg_extent_max_active_fit)
* is the max ratio between the size of the active extent and the new extent.
*/
#define LG_EXTENT_MAX_ACTIVE_FIT_DEFAULT 6
typedef enum {
EXTENT_NOT_HEAD,
EXTENT_IS_HEAD /* Only relevant for Windows && opt.retain. */
} extent_head_state_t;
#endif /* JEMALLOC_INTERNAL_EXTENT_TYPES_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/hook.h
Vendored
-163
View File
@@ -1,163 +0,0 @@
#ifndef JEMALLOC_INTERNAL_HOOK_H
#define JEMALLOC_INTERNAL_HOOK_H
#include "jemalloc/internal/tsd.h"
/*
* This API is *extremely* experimental, and may get ripped out, changed in API-
* and ABI-incompatible ways, be insufficiently or incorrectly documented, etc.
*
* It allows hooking the stateful parts of the API to see changes as they
* happen.
*
* Allocation hooks are called after the allocation is done, free hooks are
* called before the free is done, and expand hooks are called after the
* allocation is expanded.
*
* For realloc and rallocx, if the expansion happens in place, the expansion
* hook is called. If it is moved, then the alloc hook is called on the new
* location, and then the free hook is called on the old location (i.e. both
* hooks are invoked in between the alloc and the dalloc).
*
* If we return NULL from OOM, then usize might not be trustworthy. Calling
* realloc(NULL, size) only calls the alloc hook, and calling realloc(ptr, 0)
* only calls the free hook. (Calling realloc(NULL, 0) is treated as malloc(0),
* and only calls the alloc hook).
*
* Reentrancy:
* Reentrancy is guarded against from within the hook implementation. If you
* call allocator functions from within a hook, the hooks will not be invoked
* again.
* Threading:
* The installation of a hook synchronizes with all its uses. If you can
* prove the installation of a hook happens-before a jemalloc entry point,
* then the hook will get invoked (unless there's a racing removal).
*
* Hook insertion appears to be atomic at a per-thread level (i.e. if a thread
* allocates and has the alloc hook invoked, then a subsequent free on the
* same thread will also have the free hook invoked).
*
* The *removal* of a hook does *not* block until all threads are done with
* the hook. Hook authors have to be resilient to this, and need some
* out-of-band mechanism for cleaning up any dynamically allocated memory
* associated with their hook.
* Ordering:
* Order of hook execution is unspecified, and may be different than insertion
* order.
*/
#define HOOK_MAX 4
enum hook_alloc_e {
hook_alloc_malloc,
hook_alloc_posix_memalign,
hook_alloc_aligned_alloc,
hook_alloc_calloc,
hook_alloc_memalign,
hook_alloc_valloc,
hook_alloc_mallocx,
/* The reallocating functions have both alloc and dalloc variants */
hook_alloc_realloc,
hook_alloc_rallocx,
};
/*
* We put the enum typedef after the enum, since this file may get included by
* jemalloc_cpp.cpp, and C++ disallows enum forward declarations.
*/
typedef enum hook_alloc_e hook_alloc_t;
enum hook_dalloc_e {
hook_dalloc_free,
hook_dalloc_dallocx,
hook_dalloc_sdallocx,
/*
* The dalloc halves of reallocation (not called if in-place expansion
* happens).
*/
hook_dalloc_realloc,
hook_dalloc_rallocx,
};
typedef enum hook_dalloc_e hook_dalloc_t;
enum hook_expand_e {
hook_expand_realloc,
hook_expand_rallocx,
hook_expand_xallocx,
};
typedef enum hook_expand_e hook_expand_t;
typedef void (*hook_alloc)(
void *extra, hook_alloc_t type, void *result, uintptr_t result_raw,
uintptr_t args_raw[3]);
typedef void (*hook_dalloc)(
void *extra, hook_dalloc_t type, void *address, uintptr_t args_raw[3]);
typedef void (*hook_expand)(
void *extra, hook_expand_t type, void *address, size_t old_usize,
size_t new_usize, uintptr_t result_raw, uintptr_t args_raw[4]);
typedef struct hooks_s hooks_t;
struct hooks_s {
hook_alloc alloc_hook;
hook_dalloc dalloc_hook;
hook_expand expand_hook;
void *extra;
};
/*
* Begin implementation details; everything above this point might one day live
* in a public API. Everything below this point never will.
*/
/*
* The realloc pathways haven't gotten any refactoring love in a while, and it's
* fairly difficult to pass information from the entry point to the hooks. We
* put the informaiton the hooks will need into a struct to encapsulate
* everything.
*
* Much of these pathways are force-inlined, so that the compiler can avoid
* materializing this struct until we hit an extern arena function. For fairly
* goofy reasons, *many* of the realloc paths hit an extern arena function.
* These paths are cold enough that it doesn't matter; eventually, we should
* rewrite the realloc code to make the expand-in-place and the
* free-then-realloc paths more orthogonal, at which point we don't need to
* spread the hook logic all over the place.
*/
typedef struct hook_ralloc_args_s hook_ralloc_args_t;
struct hook_ralloc_args_s {
/* I.e. as opposed to rallocx. */
bool is_realloc;
/*
* The expand hook takes 4 arguments, even if only 3 are actually used;
* we add an extra one in case the user decides to memcpy without
* looking too closely at the hooked function.
*/
uintptr_t args[4];
};
/*
* Returns an opaque handle to be used when removing the hook. NULL means that
* we couldn't install the hook.
*/
bool hook_boot();
void *hook_install(tsdn_t *tsdn, hooks_t *hooks);
/* Uninstalls the hook with the handle previously returned from hook_install. */
void hook_remove(tsdn_t *tsdn, void *opaque);
/* Hooks */
void hook_invoke_alloc(hook_alloc_t type, void *result, uintptr_t result_raw,
uintptr_t args_raw[3]);
void hook_invoke_dalloc(hook_dalloc_t type, void *address,
uintptr_t args_raw[3]);
void hook_invoke_expand(hook_expand_t type, void *address, size_t old_usize,
size_t new_usize, uintptr_t result_raw, uintptr_t args_raw[4]);
#endif /* JEMALLOC_INTERNAL_HOOK_H */
deps/jemalloc-5.2.1/include/jemalloc/internal/jemalloc_internal.h
Vendored
-1134
View File
File diff suppressed because it is too large Load Diff
deps/jemalloc-5.2.1/include/jemalloc/internal/jemalloc_internal_decls.h
Vendored
-94
View File
@@ -1,94 +0,0 @@
#ifndef JEMALLOC_INTERNAL_DECLS_H
#define JEMALLOC_INTERNAL_DECLS_H
#include <math.h>
#ifdef _WIN32
# include <windows.h>
# include "msvc_compat/windows_extra.h"
# ifdef _WIN64
# if LG_VADDR <= 32
# error Generate the headers using x64 vcargs
# endif
# else
# if LG_VADDR > 32
# undef LG_VADDR
# define LG_VADDR 32
# endif
# endif
#else
# include <sys/param.h>
# include <sys/mman.h>
# if !defined(__pnacl__) && !defined(__native_client__)
# include <sys/syscall.h>
# if !defined(SYS_write) && defined(__NR_write)
# define SYS_write __NR_write
# endif
# if defined(SYS_open) && defined(__aarch64__)
/* Android headers may define SYS_open to __NR_open even though
* __NR_open may not exist on AArch64 (superseded by __NR_openat). */
# undef SYS_open
# endif
# include <sys/uio.h>
# endif
# include <pthread.h>
# ifdef __FreeBSD__
# include <pthread_np.h>
# endif
# include <signal.h>
# ifdef JEMALLOC_OS_UNFAIR_LOCK
# include <os/lock.h>
# endif
# ifdef JEMALLOC_GLIBC_MALLOC_HOOK
# include <sched.h>
# endif
# include <errno.h>
# include <sys/time.h>
# include <time.h>
# ifdef JEMALLOC_HAVE_MACH_ABSOLUTE_TIME
# include <mach/mach_time.h>
# endif
#endif
#include <sys/types.h>
#include <limits.h>
#ifndef SIZE_T_MAX
# define SIZE_T_MAX SIZE_MAX
#endif
#ifndef SSIZE_MAX
# define SSIZE_MAX ((ssize_t)(SIZE_T_MAX >> 1))
#endif
#include <stdarg.h>
#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include <stddef.h>
#ifndef offsetof
# define offsetof(type, member) ((size_t)&(((type *)NULL)->member))
#endif
#include <string.h>
#include <strings.h>
#include <ctype.h>
#ifdef _MSC_VER
# include <io.h>
typedef intptr_t ssize_t;
# define PATH_MAX 1024
# define STDERR_FILENO 2
# define __func__ __FUNCTION__
# ifdef JEMALLOC_HAS_RESTRICT
# define restrict __restrict
# endif
/* Disable warnings about deprecated system functions. */
# pragma warning(disable: 4996)
#if _MSC_VER < 1800
static int
isblank(int c) {
return (c == '\t' || c == ' ');
}
#endif
#else
# include <unistd.h>
#endif
#include <fcntl.h>
#endif /* JEMALLOC_INTERNAL_H */

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