* Re-enabled linux tests physics
* Fix for python load errors on Linux (#5627)
* Explicitly load libpython on Linux
Downstream loads of python modules that weren't linked to libpython would
fail to load because libraries were loaded using the RTLD_LOCAL flag. This
adds a function that will explicitly load libpython on Linux using the
RTLD_GLOBAL flag.
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Fix misspelled function name
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Addressing PR feedback
- Updates naming and location of things.
- Adds load code to a Gem template.
- Updates error checking.
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Address further feedback
Removes the api function in favor of just having modules inherit off a
PythonLoader class, that way we get RAAI behavior and lifetime management
for free.
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
Signed-off-by: aljanru <aljanru@amazon.co.uk>
Co-authored-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Explicitly load libpython on Linux
Downstream loads of python modules that weren't linked to libpython would
fail to load because libraries were loaded using the RTLD_LOCAL flag. This
adds a function that will explicitly load libpython on Linux using the
RTLD_GLOBAL flag.
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Fix misspelled function name
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Addressing PR feedback
- Updates naming and location of things.
- Adds load code to a Gem template.
- Updates error checking.
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
* Address further feedback
Removes the api function in favor of just having modules inherit off a
PythonLoader class, that way we get RAAI behavior and lifetime management
for free.
Signed-off-by: amzn-phist <52085794+amzn-phist@users.noreply.github.com>
Replaced and removed the CryTimer (gEnv->pTimer). The new TimeSystem is a merger of the current time functionality found in the engine.
* Rename TimeSystemComponent.h/.cpp to TimeSystem.h/.cpp
* Adding New TimeSystem
* remove old timer cvars
* small improvements to the time system.
- updated parts to use the time conversion functions.
- in AdvanceTickDeltaTimes applying t_simulationTickScale is now uses doubles instead of floats.
* Replace gEnv->pTimer / ITimer usages with TimeSystem
* Updating usages of AZ::TimeMs{ 0 } and AZ::TimeUs{ 0 } to AZ::Time::ZeroTimeMs and AZ::Time::ZeroTimeUs
* red code the CryTimer
* using TimeUs instead of TimeMs is some cases + updating usages of old cvars to new
Signed-off-by: amzn-sean <75276488+amzn-sean@users.noreply.github.com>
* Enable relocation of the Project Game Release Layout
Relocating the Project Game Release Layout to another directory on the file system failed due to the querying of the engine root failing due to the ComponentApplication::m_engineRoot not using the project path stored in the SettingsRegisry if the engine root cannot be detected
Removed the ApplicationRequestBus GetEngineRoot function.
The ComponentApplicationRequestBus has a function of the same name that returns the same path.
Removed the deprecated GetAppRoot function.
The path it returns has no defined value. It was not the engine root or the project root.
Removed unused CFileUtil and CFileUtil_impl functions that were invoking the ApplicationREquestBus GetEngineRoot function.
On the way to update the functions it was discovered that they aren't called
Added a CalculateBranchToken overload that can populate a fixed_string to avoid heap allocations
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Protect against an empty list of artifacts to remove when generating the
engine.pak
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
Changed material document to load a source data are for the parent material as well. It was also a previously loading the parent material products asset which would be out of date compared to the source data.
Changed material document to track source file dependency changes instead of product asset changes.
Fixed a bug or copy paste error in the document manager that was using the same container to track documents the modified externally and from other dependency changes.
Returning source data dependencies when creating a material asset from source.
Signed-off-by: Guthrie Adams <guthadam@amazon.com>
* ATOM-16747 RPISystemInterface::GetDefaultScene returns the scene created by PreviewRenderer but not the Main Scene
Deprecate GetDefaultScene() function.
Update all the places which use GetDefaultScene to use Scene::GetFeatureProcessorFromEntityId or GetMainScene.
Tested with Editor, UI Editor, Material Editor, game launcher.
Signed-off-by: Qing Tao <55564570+VickyAtAZ@users.noreply.github.com>
(cherry picked from commit 8da6bea073)
* Fixed some files missed when groundplane_521 was renamed to 512 (#4958)
* Fixed references to 521x521 to reference the correct 512x512 FBX file
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Fixed asset hints
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Moved the Asset Catalog loading from LmbrCentral to the AzFramework::Application (#4568)
* Moved the loading of the AssetCatalog from LmbrCentralSystemComponent to AzFramework Application
Modified the AssetCatalog::InitializeCatalog function to no longer rely on the TickBus to send out the `AssetCatalogEventBus::OnCatalogLoaded` event.
It now queues a function on the AssetCatalogRequestBus to send the OnCatalogLoaded event as soon as the dispatching for the AssetCatalogRequestBus has completed on the current thread.
This is done by updating the AssetCatalogRequestBus to use EBus ThreadDispatchPolicy to add a callback to invoke any queued function has soon a thread has finished dispatching and has released its DispatchMutex
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AssetCatalogRequestBus to add a custom DispatchLockGuard
The AssetCatalogRequestBus uses the custom lock guard to dispatch queued
events after it has unlocked it's context mutex for the current thread.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed GetContext call from the
AssetCatalogRequests::PostThreadDispatchInvoker
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the definition of FileTagQueryManager::GetDefaultFileTagFilePath
function to return a path
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AZ_CONSOLEFREEFUNC macro to actually use the _NAME
The _NAME parameter was not being used before, resulting in the Console
stringified name of the function being used.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed CrySystem dependencies from the BundlingSystemComponent
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Moved the loading of the AssetCatalog from LmbrCentralSystemComponent to AzFramework Application
Modified the AssetCatalog::InitializeCatalog function to no longer rely on the TickBus to send out the `AssetCatalogEventBus::OnCatalogLoaded` event.
It now queues a function on the AssetCatalogRequestBus to send the OnCatalogLoaded event as soon as the dispatching for the AssetCatalogRequestBus has completed on the current thread.
This is done by updating the AssetCatalogRequestBus to use EBus ThreadDispatchPolicy to add a callback to invoke any queued function has soon a thread has finished dispatching and has released its DispatchMutex
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AssetCatalogRequestBus to add a custom DispatchLockGuard
The AssetCatalogRequestBus uses the custom lock guard to dispatch queued
events after it has unlocked it's context mutex for the current thread.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed GetContext call from the
AssetCatalogRequests::PostThreadDispatchInvoker
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the definition of FileTagQueryManager::GetDefaultFileTagFilePath
function to return a path
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AZ_CONSOLEFREEFUNC macro to actually use the _NAME
The _NAME parameter was not being used before, resulting in the Console
stringified name of the function being used.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed CrySystem dependencies from the BundlingSystemComponent
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Addded missing template parameter to AssetCatalogRequests
The fixes the compile error.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Adding AssetBus::MultiHandler::BusDisconnect call
The BlastSystemComponent was connecting to the Bus, but not
disconnecting from it, causing an assert to fire to it being a
multi-thread bus
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Added support for DataDrive lifecycle events to the ComponentApplication
The events are using the SettingsRegistry NotifyEvent to track when
certain keys are modified to signal handlers.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Corrected invalid JSON creation in ModuleManager::DeactivateEntities
Resolved clang warning about used type alias
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Fix for dangling reference in lambda registered to the SettingsRegistry
Notifier event
This was causing the EditorPythonBinding tests to crash due to the
following circumstances.
First Python has created an instance of a SettingsRegistryProxy
Second the SettingsRegistry sends an event during the time when the
SettingsRegistryProxy exists.
This issue was exposed due to the ComponentApplication Lifecycle events
using the SettingsRegistry to dispatch during various times of the
application workflow.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Added the generated cmake_dependencies.*.setreg files to engine.pak (#5073)
* Copied the generated cmake_dependencies.*.setreg file to the Cache
directory
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed the platform name from the bootstrap.game.*.setreg
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Fixes for release builds with DCO fix (#5164)
* This set of changes is work toward allowing release builds to work with asset bundler generated bundles and legacy, non-prefab levels. This requires some other in-flight changes before this work is complete.
Updated engine seed list + fixed automated test
ComponentApplicationLifecycle has the ability to automatically register events if asked to register a handler and the event doesn't exist. This is only intended for cases where you need to register a handler early in startup before the settings registry file is loaded.
Added two new lifecycle events: One after the system entity has been activated, and one after the system interface has been created.
If you load an archive before the system entity has been activated, archive.cpp caches information about those archives until that time, so it can finish registration. This is because the serialization system and BundlingSystemComponent both need to be available to do this registration, but the bundles have to be loaded before those are initialized so that the settings registry file can be loaded.
Fixed an error were mounted pak files were searching for levels.pak and not level.pak, and not finding them. I'm pretty sure this logic doesn't do anything functional either way, but I've been testing legacy levels with this change and they work now.
Moved wildcard pak loading to where engine.pak is loaded. This is because the settings registry file that defines the IO stack to spin up must be available early in application startup, and this file must be within a mounted pak file. If you're using asset bundler generated bundles, they need to be loaded at this time so that file can be loaded.
Atom's BootstrapSystemComponent.cpp no longer initializes on AssetCatalogLoaded, and instead initializes on the ApplicationLifecycle event SystemInterfaceCreated. This is because the base assetcatalog.xml file is really just a development time concept, this file should not be used in packaged release builds, because those builds will make use of delta catalogs in each bundle loaded. The asset catalog contains the list of all assets that were in the cache at development time, and this contains content that developers don't want to ship, and they may want to specifically hide from their customers, so data miners don't find secrets about upcoming game content.
Recovering from a branch that had incorrect DCO
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Fixed an incorrect ebus disconnect and removed an include that's no longer needed
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Fixed a copy and paste typo from trying to recover the previous pull request
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Updated product IDs for the settings registry builder to no longer collide with the JSON builder. Now they are based on a hash of the configuration.
Updated the engine default seed list to include the new asset ID info for the renamed bootstrap file
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Updated the path to the application lifecycle events, because runtime settings aren't included in the merged bootstrap file.
Addressed some feedback on printing out a string view on an error
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Removed a test that uses old assets that aren't relevant. We may not need this test anymore, but if we do we've backlogged a task to create a new test to cover this behavior without using old assets.
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Renamed SystemInterfaceCreated event to LegacySystemInterfaceCreated
Removed SystemEntityActivated event. Now that I have the rest of the fixes in this pull request, this new event wasn't needed, the already existing SystemComponentsActivated event does what I need.
Changed list to vector
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
Co-authored-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* This set of changes is work toward allowing release builds to work with asset bundler generated bundles and legacy, non-prefab levels. This requires some other in-flight changes before this work is complete.
Updated engine seed list + fixed automated test
ComponentApplicationLifecycle has the ability to automatically register events if asked to register a handler and the event doesn't exist. This is only intended for cases where you need to register a handler early in startup before the settings registry file is loaded.
Added two new lifecycle events: One after the system entity has been activated, and one after the system interface has been created.
If you load an archive before the system entity has been activated, archive.cpp caches information about those archives until that time, so it can finish registration. This is because the serialization system and BundlingSystemComponent both need to be available to do this registration, but the bundles have to be loaded before those are initialized so that the settings registry file can be loaded.
Fixed an error were mounted pak files were searching for levels.pak and not level.pak, and not finding them. I'm pretty sure this logic doesn't do anything functional either way, but I've been testing legacy levels with this change and they work now.
Moved wildcard pak loading to where engine.pak is loaded. This is because the settings registry file that defines the IO stack to spin up must be available early in application startup, and this file must be within a mounted pak file. If you're using asset bundler generated bundles, they need to be loaded at this time so that file can be loaded.
Atom's BootstrapSystemComponent.cpp no longer initializes on AssetCatalogLoaded, and instead initializes on the ApplicationLifecycle event SystemInterfaceCreated. This is because the base assetcatalog.xml file is really just a development time concept, this file should not be used in packaged release builds, because those builds will make use of delta catalogs in each bundle loaded. The asset catalog contains the list of all assets that were in the cache at development time, and this contains content that developers don't want to ship, and they may want to specifically hide from their customers, so data miners don't find secrets about upcoming game content.
Recovering from a branch that had incorrect DCO
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Fixed an incorrect ebus disconnect and removed an include that's no longer needed
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Fixed a copy and paste typo from trying to recover the previous pull request
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Updated product IDs for the settings registry builder to no longer collide with the JSON builder. Now they are based on a hash of the configuration.
Updated the engine default seed list to include the new asset ID info for the renamed bootstrap file
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Updated the path to the application lifecycle events, because runtime settings aren't included in the merged bootstrap file.
Addressed some feedback on printing out a string view on an error
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Removed a test that uses old assets that aren't relevant. We may not need this test anymore, but if we do we've backlogged a task to create a new test to cover this behavior without using old assets.
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* Renamed SystemInterfaceCreated event to LegacySystemInterfaceCreated
Removed SystemEntityActivated event. Now that I have the rest of the fixes in this pull request, this new event wasn't needed, the already existing SystemComponentsActivated event does what I need.
Changed list to vector
Signed-off-by: stankowi <4838196+AMZN-stankowi@users.noreply.github.com>
* ATOM-16747 RPISystemInterface::GetDefaultScene returns the scene created by PreviewRenderer but not the Main Scene
Deprecate GetDefaultScene() function.
Update all the places which use GetDefaultScene to use Scene::GetFeatureProcessorFromEntityId or GetMainScene.
Tested with Editor, UI Editor, Material Editor, game launcher.
Signed-off-by: Qing Tao <55564570+VickyAtAZ@users.noreply.github.com>
* Moved the loading of the AssetCatalog from LmbrCentralSystemComponent to AzFramework Application
Modified the AssetCatalog::InitializeCatalog function to no longer rely on the TickBus to send out the `AssetCatalogEventBus::OnCatalogLoaded` event.
It now queues a function on the AssetCatalogRequestBus to send the OnCatalogLoaded event as soon as the dispatching for the AssetCatalogRequestBus has completed on the current thread.
This is done by updating the AssetCatalogRequestBus to use EBus ThreadDispatchPolicy to add a callback to invoke any queued function has soon a thread has finished dispatching and has released its DispatchMutex
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AssetCatalogRequestBus to add a custom DispatchLockGuard
The AssetCatalogRequestBus uses the custom lock guard to dispatch queued
events after it has unlocked it's context mutex for the current thread.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed GetContext call from the
AssetCatalogRequests::PostThreadDispatchInvoker
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the definition of FileTagQueryManager::GetDefaultFileTagFilePath
function to return a path
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AZ_CONSOLEFREEFUNC macro to actually use the _NAME
The _NAME parameter was not being used before, resulting in the Console
stringified name of the function being used.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed CrySystem dependencies from the BundlingSystemComponent
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Moved the loading of the AssetCatalog from LmbrCentralSystemComponent to AzFramework Application
Modified the AssetCatalog::InitializeCatalog function to no longer rely on the TickBus to send out the `AssetCatalogEventBus::OnCatalogLoaded` event.
It now queues a function on the AssetCatalogRequestBus to send the OnCatalogLoaded event as soon as the dispatching for the AssetCatalogRequestBus has completed on the current thread.
This is done by updating the AssetCatalogRequestBus to use EBus ThreadDispatchPolicy to add a callback to invoke any queued function has soon a thread has finished dispatching and has released its DispatchMutex
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AssetCatalogRequestBus to add a custom DispatchLockGuard
The AssetCatalogRequestBus uses the custom lock guard to dispatch queued
events after it has unlocked it's context mutex for the current thread.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed GetContext call from the
AssetCatalogRequests::PostThreadDispatchInvoker
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the definition of FileTagQueryManager::GetDefaultFileTagFilePath
function to return a path
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the AZ_CONSOLEFREEFUNC macro to actually use the _NAME
The _NAME parameter was not being used before, resulting in the Console
stringified name of the function being used.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Removed CrySystem dependencies from the BundlingSystemComponent
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Addded missing template parameter to AssetCatalogRequests
The fixes the compile error.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Adding AssetBus::MultiHandler::BusDisconnect call
The BlastSystemComponent was connecting to the Bus, but not
disconnecting from it, causing an assert to fire to it being a
multi-thread bus
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Added support for DataDrive lifecycle events to the ComponentApplication
The events are using the SettingsRegistry NotifyEvent to track when
certain keys are modified to signal handlers.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Corrected invalid JSON creation in ModuleManager::DeactivateEntities
Resolved clang warning about used type alias
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Fix for dangling reference in lambda registered to the SettingsRegistry
Notifier event
This was causing the EditorPythonBinding tests to crash due to the
following circumstances.
First Python has created an instance of a SettingsRegistryProxy
Second the SettingsRegistry sends an event during the time when the
SettingsRegistryProxy exists.
This issue was exposed due to the ComponentApplication Lifecycle events
using the SettingsRegistry to dispatch during various times of the
application workflow.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Refactor Vulkan swapchain so it can recreate when error occurs
Signed-off-by: jiaweig <jiaweig@amazon.com>
* revert the workaround
Signed-off-by: jiaweig <jiaweig@amazon.com>
* Move semaphore. Revert some viewport changes.
Signed-off-by: jiaweig <jiaweig@amazon.com>
* Added comments. Moved recreation out of AcquireNewImage.
Signed-off-by: jiaweig <jiaweig@amazon.com>
- Fixed fallback connections for hair pipeline to allow disabling the parent pass hierarchy when not required
- Renamed the Thumbnail pipeline to be used as generic minimal tools pipeline
- Reused the Tools pipeline for the preview renderer - minimal FPs and passes
Remark:
- The tools pipeline should have folloup submits for reducing passes to minimal required render passes
Signed-off-by: Adi-Amazon <Adi Bar-Lev barlev@amazon.com>
Co-authored-by: Adi-Amazon <Adi Bar-Lev barlev@amazon.com>
- Prevent P4 thread to run if we cannot detect the P4 command to begin with
- Add a trait to disable calling the parent ComponentApplication::Destroy(), instead calling _exit() to skip the module unloading on exit
Signed-off-by: Steve Pham <spham@amazon.com>
Preventing multiple preview renderers from being created with multiple catalog loaded notifications
Purging material preview images reaching a maximum number of stored images
Purging material previews if entity gets destroyed
Signed-off-by: Guthrie Adams <guthadam@amazon.com>
- Updated the RenderViewportWidget::event method to call 'SendWindowResizeEvent' on specific events
- Moved the onResize to one of the events in RenderViewportWidget::event
Signed-off-by: Steve Pham <spham@amazon.com>
A request was captured and held onto to by a lambda that was not getting cleared before the system was destroyed
This caused an entity to be destroyed late, accessing the culling system that was already torn down
Solution is to only store the success and failure callbacks inside of the attachment pass read back callback lambda so the rest of the content can be released
Signed-off-by: Guthrie Adams <guthadam@amazon.com>
puvvadar
Mike Balfour
Tommy Walton
AMZN-AlexOteiza
Tom Hulton-Harrop
amzn-phist
Guthrie Adams
amzn-sean
Nicholas Van Sickle
Guthrie Adams
Roman
lumberyard-employee-dm
Qing Tao
AMZN-stankowi
jiaweig
Steve Pham
Adi Bar-Lev
jonawals
hultonha