Commit Graph

65 Commits

Author SHA1 Message Date
not-a-robot[bot]
74be165c84 Merge #524 #525
524: Don't pass a negative value to __redisAsyncCommand if redisFormatSdsCommandArgv fails r=badboy



525: Fix compilation on FreeBSD 10.3 with default compiler r=badboy

Given FreeBSD 10.3 with default compiler:

> $ cc -v
> FreeBSD clang version 3.4.1 (tags/RELEASE_34/dot1-final 208032) 20140512
> Target: x86_64-unknown-freebsd10.3

Defining _XOPEN_SOURCE to 600 on the fixed line leads to the following errors:

> cc -std=c99 -pedantic -c -O3 -fPIC  -Wall -W -Wstrict-prototypes -Wwrite-strings -g -ggdb  net.c
> net.c:435:29: error: use of undeclared identifier 'AF_LOCAL'
>     if (redisCreateSocket(c,AF_LOCAL) < 0)
>                             ^
> net.c:460:21: error: use of undeclared identifier 'AF_LOCAL'
>     sa.sun_family = AF_LOCAL;
>                     ^
> 2 errors generated.
> 

AF_LOCAL is defined in sys/socket.h within ifdef __BSD_VISIBLE.

__BSD_VISIBLE could be defined in sys/cdefs.h, but it is never done if _XOPEN_SOURCE is defined.

So on FreeBSD _XOPEN_SOURCE shouldn't be defined.
2017-07-15 14:17:07 +00:00
amallia
b46413647f Assert statement calls a function which has side effects 2017-06-15 20:33:23 +01:00
Frederik Deweerdt
706129a161 Don't pass a negative value to __redisAsyncCommand if redisFormatSdsCommandArgv fails 2017-05-15 09:20:03 -07:00
charsyam
e2f6ee239c fixing typos 2015-10-28 02:19:24 +09:00
Jan-Erik Rediger
36a9802c54 Prevent crash on pending replies in async code
Fixes #335.
2015-07-27 23:35:02 +02:00
antirez
2fc39eb4c3 Fix memory leak in async spontaneous reply handling
When an asynchronous hiredis connection subscribes to a Pub/Sub channel
and gets an error, and in other related conditions, the function
redisProcessCallbacks() enters a code path where the link is
disconnected, however the function returns before freeing the allocated
reply object. This causes a memory leak. The memory leak was trivial to
trigger in Redis Sentinel, which uses hiredis, every time we tried to
subscribe to an instance that required a password, in case the Sentinel
was configured either with the wrong password or without password at
all. In this case, the -AUTH error caused the leaking code path to be
executed.
2015-04-28 22:00:48 +02:00
Matt Stancliff
6a00a4643b Fix redisAppendCommand error result
Previously, redisvAppendCommand() would return OOM even with formatting
errors.  Now we use OTHER with an error string telling the user the
error was formatting related, not memory related.

This also fixes a potentially worse bug where were would pass error result
of -1 as an actual length to another function taking an unsigned length,
which would result in crash/overallocation/errors.  Now for that case,
we just return an error immediately and stop processing the command.

Fixes #177
2015-01-05 16:52:08 -05:00
Matt Stancliff
9b83ddc2d9 Fix clang analyzer warning
redisAsyncInitialize() can return NULL, but then we pass
the return value from redisAsyncInitialize() into something
dereferencing the return value, which can cause crashies.
2015-01-05 16:39:30 -05:00
mike
7c4d2557c4 Add support for SO_REUSEADDR
[This introduces some new API functions.]

* Adds new flag to the connection context indicating SO_REUSEADDR
  should be set.
* Adds max number of retries constant for when connect() hits
  EADDRNOTAVAIL.
* Adds new function, redisAsyncConnectBindWithReuse(), letting
  clients enable this functionality.

[Removed trailing whitespace in new header lines.]

Closes #264
2015-01-05 16:39:30 -05:00
michael-grunder
40f7035ba4 Improve redisAppendCommandArgv performance
OK, perhaps the second time is a charm.  I forgot that I had
hiredis forked from a long time ago, so the initial pull
request was hosed.  :)

* Pulled in sdscatfmt() from Redis, and modified it to accept a
  size_t (%T) style format specifier.

* Pulled in sdsll2str() and sdsull2str() from Redis (needed by
  sdscatfmt).

* Added a new method, redisFormatSdsCommandArgv() which takes
  and sds* as the target, rather than char* (and uses sdscatfmt
  instead of sprintf for the construction).

I get roughly the following improvement:

Old: 1.044806
New: 0.481620

The benchmark code itself can be found here:
https://gist.github.com/michael-grunder/c92ef31bb632b3d0ad81

Closes #260
2015-01-05 11:21:38 -05:00
Hang Su
3315c09839 Use stricter function argument types
'const' where we can.

Closes #268
2015-01-05 11:21:38 -05:00
Jan-Erik Rediger
2a6dbdd225 Free string if it is unused
If the string is already in the dict, we need to free this.

Original fix by @tt33415366.

Fixes #256
Closes #286
2015-01-05 11:21:38 -05:00
Pieter Noordhuis
a9c21e4d48 Fix const correctness 2014-04-09 22:30:04 -07:00
Matt Stancliff
37d25a392c Add ability to bind source address on connect
Some environments require binding to specific source addresses instead
of letting the system determine which IP a connection should originate
from.

Closes #233
2014-04-08 19:37:54 -04:00
Matt Stancliff
61eeedbe77 Fix potential undefined struct read
All the assignments to cb are inside conditionals, so it's
vaguely possible it never gets initialized before we try
to read from it with (cb.fn == NULL).

Condition discovered with scan-build.

Closes #229
2014-04-08 19:37:45 -04:00
Matt Stancliff
7f0c7a29fd Remove possiblity of multiple close on same fd
With all the async connects and disconnects and error handling
going on in hiredis, we need to centralize how we close our fd
and set it so it doesn't get re-closed.  Prior to this commit,
sometimes we'd close(fd), run an async error handler, then
call close(fd) again.

To stop multiple closes, we now set fd to -1 after we free it,
but that requires not passing fd as an independent argument to
functions.

This commit moves all fd usage to c->fd.  Since the context
has a fd field and all functions receive the context, it makes
more sense to use the fd inside of c instead of passing along fd
as an independent argument.

Also, by only using c->fd, we can set c->fd after we close it to
signify we shouldn't re-close the same fd again.

This does change one semi-public interface function redisCheckSocketError()
to only take (context) instead of (context, fd).  A search on github
returned zero occasions of people using redisCheckSocketError()
outside of net.{c,h} in hiredis itself.

Commit inspired by the bug report at:
https://groups.google.com/forum/#!topic/redis-db/mQm46XkIPOY

Thanks go out to Thijs for trying high-frequency reconnects on
a host that isn't there.

Closes #230
2014-04-08 19:37:29 -04:00
Charlie Somerville
070da21623 async.c: avoid a NULL deref when redisAsyncInitialize returns NULL 2014-01-15 11:38:45 +11:00
Pieter Noordhuis
c532edc27f Fix const-related compiler errors 2013-07-10 23:31:30 -07:00
Pieter Noordhuis
049346aa33 Merge pull request #132 from nwmcsween/constify
constify: constify some variables / functions
2013-07-10 23:13:36 -07:00
Henri Doreau
d7e3268f48 Prevent AsyncConnect from crashing on memory allocation failures. 2013-01-22 15:53:17 +01:00
Henri Doreau
814be4f5bd Made connect functions return NULL on alloc failures.
Updated documentation and examples accordingly.
2013-01-22 10:19:46 +01:00
Nathan McSween
ba6c3c152c constify: constify some variables / functions 2012-11-12 19:48:21 +00:00
Pieter Noordhuis
7ec4df9403 Spontaneous error reply can always happen 2012-07-31 17:40:46 -07:00
Alex Leverington
01cce89d9d async: support for determining monitor mode, if so, repush replies callback in expectation of another reply. 2012-07-11 02:53:58 -05:00
R. Tyler Croy
e6cf455be5 Include stdlib.h in async.c for free(3), realloc(3) and strtol(3) support 2011-07-10 21:04:03 -07:00
Pieter Noordhuis
3313bcd191 Change prototype of connect callback
This commit adds a status argument to the connect callback. It will be
called in the event of an unsuccessful connection as well, where the
status argument is set to REDIS_ERR. It is set to REDIS_OK otherwise.
2011-06-27 23:52:29 +02:00
Pieter Noordhuis
026d5ae750 Use macro's for event loop hooks 2011-06-27 22:49:25 +02:00
Pieter Noordhuis
367ac44ac1 Remove private feature macros for Solaris from compiler flags 2011-06-18 16:28:57 +02:00
Pieter Noordhuis
1c29aafd47 Clarify rationale behind issue #43 2011-05-29 09:52:18 -07:00
Blake Matheny
63dcf9b741 Make sure subscribe is still respected 2011-05-24 12:59:02 -04:00
Blake Matheny
7268857254 Fix the case where an error reply is received before any callbacks are registered 2011-05-24 11:46:19 -04:00
Pieter Noordhuis
0cb7c27d9c Update license 2011-04-21 22:46:23 +02:00
Pieter Noordhuis
f3f02b6419 Let the async API segfault on OOM for now 2011-04-21 22:40:36 +02:00
Pieter Noordhuis
6d207ea98e Create protocol reader when creating context 2011-04-21 15:04:42 +02:00
Pieter Noordhuis
c6b8bd77c0 Make dictionary functions static and include the .c file 2011-01-14 12:07:29 +01:00
Pierre Riteau
4def8569b8 strcasecmp and strncasecmp are defined in strings.h 2011-01-10 22:44:37 +01:00
Pieter Noordhuis
ec922cd007 Unsubscribe messages always have 3 elements 2010-12-31 15:49:26 +01:00
Pieter Noordhuis
ed97945c27 Clean up dictionaries on free 2010-12-31 15:49:26 +01:00
Pieter Noordhuis
a0ebc5417f Delegate pub/sub replies to the right callback 2010-12-31 15:49:26 +01:00
Pieter Noordhuis
3ac8ef927d Add callbacks to channel/pattern dictionaries 2010-12-31 15:49:26 +01:00
Pieter Noordhuis
e6621d05b4 Add fields for subscribed channel/pattern names 2010-12-31 15:49:26 +01:00
Pieter Noordhuis
3ce8d5b08b Change reply processing code to prepare for pub/sub 2010-12-31 15:49:21 +01:00
Pieter Noordhuis
2d53a6a711 Copy entire callback in one call 2010-12-31 11:22:13 +01:00
Pieter Noordhuis
18c55a8f1e Scope event library related data and hooks to a struct 2010-12-29 15:41:03 +01:00
Pieter Noordhuis
8cb4d52cd2 Run pending callbacks with a NULL reply on redisAsyncFree() 2010-12-28 20:49:18 +01:00
Pieter Noordhuis
29ea901b24 Fix the async free() and disconnect() functions
To make sure that these functions can also be called from functions
other than command callbacks, the flag IN_CALLBACK is introduced that
holds whether the context is currently executing a callback. If so,
redisAsyncFree() and redisAsyncDisconnect() should delegate their task
to the reply processor to avoid segfaults.
2010-12-28 20:29:29 +01:00
Pieter Noordhuis
c882a3621a Only check REDIS_FREEING when a callback was executed 2010-12-28 19:35:26 +01:00
Pieter Noordhuis
e3776bfaa6 Add function to explicitly free an async context 2010-12-28 19:19:25 +01:00
Pieter Noordhuis
3d76f3fe02 Add write event after setting connect callback 2010-12-28 17:59:26 +01:00
Pieter Noordhuis
a1e2c6dfed Add myself to license in some files 2010-12-16 22:08:46 +01:00