Add GetArrayBlockSize and GetStackBlockSize and equivalent properties to
the methodmaps to get the blocksize the array or stack was created with.
Useful when getting array handles from other plugins you didn't create
yourself.
If one of the queries fails in a transaction, the DBI system would only allocate an array of the size of the amount of all the successful queries before the failed one. It writes data for all the queries though effectively writing past the array bounds leading to heap corruption.
Create the right sized array!
Plugins were able to close the results handle in the threaded query
callback causing a double free crash. The access rights are setup right
before the handle is created, but weren't used.
SourceMod itself and thirdparty extensions call functions in plugins
without making sure it is runnable. The plugin pausing feature isn't
supported anymore and is set to get nuked #438 which will reduce the
amount of not-runnable error reports.
We can't fix every third party extension, so suppressing the error from
the logs will stop the spam until pausing is removed.
#489 can be reverted after this.
Add an ICellArray interface to expose the adt_array CellArray class from
logic to core.
Add methods to ISourceMod to create and delete ICellArray instances in
logic.
Bug 6518
Starting from no plugins loaded, after the final step of the following,
basetriggers will be marked as errored because mapchooser is missing:
Load basetriggers
Load mapchooser
Unload mapchooser
Load mapchooser
Unload mapchooser
We lose the SP_NTVFLAG_OPTIONAL when removing a WeakNative. The
__pl_mapchooser_SetNTVOptional() is only called in basetriggers when it
is loaded. It adds the OPTIONAL flag to the mapchooser natives the
basetrigger plugin uses.
When mapchooser is loaded, the native is overwritten and the OPTIONAL
flag consumed (WeakNative registered), but not preserved. Later when
mapchooser is unloaded the WeakNative is removed in
CNativeOwner::UnbindWeakRef, but the OPTIONAL flag isn't added again.
The next time mapchooser is loaded, the natives aren't optional anymore
and will cause a real dependency.
The change in ShareSys isn't needed to fix the error, but is included to
keep the SP_NTVFLAG_OPTIONAL flag consistent.
This avoids spam of "Plugin not runnable" exceptions on shutdown or
plugin unload.
When re/unloading a plugin which has other ones depending on it, like
the adminmenu, It pauses the depending plugins putting them in an
"Depends on plugin: %s" error state. ForwardSys doesn't remove them from
the forward lists on pause, specially the global forwards, and still
tries to call all the global forwards like OnPlayerRunCmd and
OnLibraryAdded etc. on the paused plugins. Executing functions in paused
runtimes has been ignored in the VM before introducing the "Exception"
mechanism, but now they're all logged.
This adds checks to make sure the plugin is runnable before calling a
function. (Stolen from #438)
When unloading a plugin no plugin is allowed to execute code. There's
been some flawed logic in the scheduling, leading it to bail out if the
plugin was marked for scheduled unloading..
Also change the "error" message of `sm plugins unload` to something
nice.
1. Fixed OnPluginUnloaded not pairing if the plugin failed.
2. Unify error message handling in the second pass.
3. Do not add libraries if a plugin failed during OnPluginStart.
Note: this also ensures that library action callbacks are balanced (i.e., we do not notify
libraries are being dropped if they were never notified as being added).
The mark-serial is a generation number to optimize dependency tracking. It did not actually get
applied correctly, meaning that in rare cases we could miss dependencies. This patch removes the
incorrect serial propagation and ensures that we don't double-count a dependent plugin.
Additionally, this patch ensures that all callers of BindNativeToPlugin() will update the mark
serial, as is required to correctly track dependencies.
This only affects plugins compiled before the param existed in the includes. NULL defaults to "GAME"
on some engine versions, but is invalid on others, causing any file to not be found.
isalpha, isdigit, isupper, and islower do not return 0/1. They return 0 or anything-else. Since the bool tag in pawn only supports exactly 0 and 1, we need to return 1 for all truthy returns in the natives that wrap these.
This has three major changes to SourcePawn. First, the API now supports the concept of "exceptions". The exception state is a global property of an instance of the SourcePawn VM. Exceptions can be caught or suppressed. Many places in SourceMod have been updated to check exceptions instead of errors.
The new API obsoletes major parts of the embedder API - all but one method of invoking functions is obsoleted, and the debug interface has been scrapped. Extensions using the native API will not be affected, however, ThrowNativeError has been deprecated in favor of ReportError.
Second, the SourcePawn concept of a "stack" has been unified at the API level. A stack frame iterator now iterates over all SourcePawn invocations, rather than the topmost plugin. This makes error handling more consistent and removes another dependency on context-per-plugin.
Finally, the implementation of stack frames has been changed dramatically. Rather than maintain a complicated and expensive return pointer stack, we now rely on the implicit one provided by the CPU. The stack frame iterator now walks the JIT stack directly. This removes many unnecessary bookkeeping instructions from the generated code, in particular making the CALL instruction 40% faster.
These changes required some fair surgery to the JIT. Its error paths are now slightly more complicated, as they have to throw an exception rather than return an error code. In addition, any path that can throw an exception is now responsible for creating an "exit frame", which exists to tell the stack frame iterator about transitions from the JIT to the VM.
This removes one the last remnants of the SourceMod 1.0 VM implementation.
The new parser introduces a number of design changes in the VM. First, the VM now takes greater responsibility for validating and sanity checking the structure of the SMX container format. Previously, malformed SMX files could easily crash SourcePawn. The loader now rejects files that have out-of-bounds offsets or incomplete sections. Complex sections, like debug info or the code stream, are verified lazily.
Internally, the sp_plugin_t structure has been removed. It has been replaced by a new LegacyImage class, designed to be independent from the SPVM API. This potentially lets us load code streams from non-.smx containers. More importantly, it removes a lot of bookkeeping and pre-computed state from PluginRuntime. The LegacyImage class is now responsible for handling debug info as well.
PluginRuntime is now intended to hold only cached or immutable data, and PluginContext holds all VM state. As such PluginContext is now responsible for allocating a plugin's runtime memory, not PluginRuntime.
Finally, some aspects of the loading process have been cleaned up. The
decompression and image handoff logic should now be easier to
understand.
