https://crash.limetech.org/stats/dbi.mysql.ext.%25/my_real_readhttps://crash.limetech.org/stats/dbi.mysql.ext.%25/net_real_write
Both of these are caused by the VIO ptr ending up as null in the middle of reading/writing to a connection - I can't find any indication of a fix for this made to MySQL, so don't think it is a bug fix we're missing, but there are some musings around the internet that it could be caused by improper thread-safety initialisation.
`my_init` (what we had here) is called internally by `mysql_library_init` but I think would have still led to an automatic `mysql_library_init` call the first time `mysql_init` was called (which we can do on a thread in case of threaded connections), which is exactly the thread-safety issue called out by the MySQL docs, so hopefully doing things properly here will help.
PR #705 made some changes for 64-bit support that required callers to
provide a buffer for object storage, but only added the required code to
the non-virtual call builder, not the virtual call builder.
Copy it over to CreateValveVCall so that ValveParamToBinParam stops
writing to a null pointer.
Fixes#1059 (Hopefully.)
The cause of the infamous "Index error writing string table baseline"
error appears to be a timing issue between the engine's network message
processing, the stringtable encoding, and command processing when adding
stringtable entries in OnConfigsExecuted.
When the first client connects the map is re-loaded which causes a full
refresh, the game's stringtable entries are added at tick 65, the client
connection is registered at tick 66, and stringtable entries added in
OnConfigsExecuted are registered as being added in tick 67. The engine
later calls WriteBaselines with the client's signon tick, and neglects
to encode the SM added entries as it considers them from the future.
To avoid this, always pass INT_MAX as the current tick when encoding
the baseline, so all stringtable entries are included regardless of when
they were added. Tested on both Windows and Linux.
* PassType typo fixup
Just some little observations, I will look up for more in other files whenever I can
* Update natives.cpp
* Update outputnatives.cpp
* Switch CS:GO Clantag set/get to use netprops + offset over sig+offset.
* Make sure the new functions are defined before using them...
* Fix offset name
* Fix incorrectly dereferencing.
`SDKCall` has existing semantics that a `-1` retval indicates null, so use that and writes an empty string to the buffer.
Consumers can tell the difference between `""` and `NULL` by checking if the return value is `0` or `-1`.
* More TR natives
* Fix small typo
* Fix another typo
* Fix TR_AllSolid description
* Change Disp to Displacement
* Reorder params in TR_GetSurfaceName
* Fix TR_GetSurfaceName
* Another fix for TR_GetSurfaceName -_-
* Change if block to switch
* Make handle first param in TR_GetSurfaceName
* Reorder TR_GetStartPosition params
* Make TranslatePartitionFlags static
* Change tabs used for formatting to spaces
* Change inline comment style to be consistent with rest of file
This simplifies non-sdk build scripts by removing their loops over SM.archs. Instead the top-level
build script will re-evaluate them for each architecture.
* Make mac/win lookups lowercase'd
* Revert #709 & 81042cc
* Adjust HashPolicy implementation across sourcemod
Basically, in order to implement our own (actual) hash policy in
`PluginSys.h`, we needed to remove the blanket implementation of `hash`
that was used before. Now, each policy must implement `hash` along with
`matches` in order to be used with `NameHashSet`. While this does force
us to change every implementation of policies across the entirety of
sourcemod, it allows core to use flexible implementations of `hash`.
* Remove logic duplication
* Improve lowercase checks
* Create DatabaseConfBuilder & remove locking
* Remove all refcounting
This is part 1/n in regards to this PR's rework
* Move db conf lookup out of RunThreadPart
* Return default configuration for failed lookups
* RefPtr members & stop leaks
* fix uint comparison warning
Prepared statements can return multiple result sets since MySQL 5.5. That can happen when calling stored procedures using `CALL x();`.
This change removes the previous caching of result bindings, since the number of fields in a result can differ from result set to result set. This could potentially have a negative impact on prepared statements always only returning one result set, since the result binding buffers are recreated everytime the statement is executed instead of once. That difference should be negligible.
Fixes#823.
MySQL defines its own timespec_t. Official mysql builds are built using VS2013, so you still need to compile the client library yourself to be able to link it, but this change makes the dbi extension compatible with future versions already.
* Add OnPlayerRunCmdPost forward
Allow plugins to catch the values that are actually used by the game
after some eventual changes by other plugins.
* Add |const| qualifier to arrays in OnPlayerRunCmdPost forward
* Fix memory leak in Regex extension.
"x" is never deleted when compile fails.
* Handle handle allocation failures.
Handle allocation failures also result in memory leaks.
* Fix typo.
Round end reasons in CS:GO are +1 from the ones in CS:S. The
CSRoundEnd_GameStart reason for when the OnMapTimeLeftChanged forward
should be called should be one higher than in CS:S.
When the server crashed and the process got terminated, the SqDriver
instance was killed first (e.g. by atexit). SqDatabase tries to access
SqDriver in its destructor.
This patch tells SqDatabase to not use anything from SqDriver anymore
after SqDriver got destroyed.
Next to that, the clientprefs extension relied on the IDatabase pointer
being valid to get the driver pointer. Cache the pointer, so the dbi
system still knows the IDBThreadOperation belonged to the now gone
driver, even after the database object is gone.
