Sleeping in tests and attempting to rely on fine-grained measurements to
check code validity is intrinsically brittle. Wall-clock time is
unreliable in an environment where tests are run under a hypervisor that
may choose to suspend your VM at any point, or in situations where the
OS cannot schedule your thread in time. The correct way to reintroduce
these tests in the future is provide an override for the timestamp
queries that can be injected in the test environment to control the wall
time deterministically.
Signed-off-by: Jeremy Ong <jcong@amazon.com>
In some cases, the interface being queried may be persistently nullptr
(e.g. a gem isn't active, or a feature was disabled by a cvar). In this
case, querying the interface would always enter the writer-lock branch
with existing code because nullptr was synonmous with "we haven't
queried for this variable yet."
This PR adds an additional state variable that is set when s_instance is
assigned to, so that after the variable is queried and found to be null,
future queries can enter the read-lock branch.
Signed-off-by: Jeremy Ong <jcong@amazon.com>
- Use a vector for shared string storage (to avoid the double heap allocation for AZStd::string)
- Use a shared heap allocated any for opaque types (instead of an unsafe ref)
- Add a string comparison key lookup benchmark to measure the impact of Name
Signed-off-by: Nicholas Van Sickle <nvsickle@amazon.com>
Direct conversion helpers for quaternion to the scaled axis-angle representation and back without the need to convert them first to the axis-angle format and manually scale (or normalize on the way back). This also avoids having to deal with the special case of an identity representation which is 0,0,0 in the scaled axis-angle format while our convention for axis-angle is 0,1,0 for the axis and 0 for the angle.
Added unit tests that check the conversion round-trips from quaternion -> (scaled) axis-angle -> quaternion as well as comparing the scaled axis-angle representations from the direct helper functions as well as the axis-angle while manually scaling/normalizing.
Signed-off-by: Benjamin Jillich <jillich@amazon.com>
* Prefab hotloading - load and maintain asset references
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Working hotloading
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Clean up code, add comments
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Add missing includes
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Add event for template removal.
Clear ProceduralPrefabSystemComponent lookup when templates are removed.
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Add unit test.
Fix missing include, make sure source name is set
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix missing include
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix missing include
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Update comments to make ebus event behavior more clear
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix missing include
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix missing includes
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix missing include
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix temp directory usage which doesn't work on linux
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix temp directory usage which doesn't work on linux
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Use AZ::IO::Path for comparison. Print paths on failure
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Fix unit test - register asset with catalog so it can be looked up later
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Preapre codebase for FileRequest compiletime improvements
This is preparing grounds for the next PR that will contain the 'meat'
of the changes.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove spurious newline.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
Small change in behavior context attribute for a few
classes so they become visible when extracting Lua symbols
from the ScriptContextDebug() API.
From:
Attribute(AZ::Script::Attributes::ExcludeFrom,
AZ::Script::Attributes::ExcludeFlags::All)
To:
Attribute(AZ::Script::Attributes::ExcludeFrom,
AZ::Script::Attributes::ExcludeFlags::ListOnly)
It seems the previous code was simply trying to hide those symbols from
the Script Canvas node palette. But it was also, seemingly unintentional,
hiding those symbol from Lua Documentation.
Signed-off-by: galibzon <66021303+galibzon@users.noreply.github.com>
longer read form the users `~/.o3de/Registry` directory in non
debug/profile configurations.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
This was part of another instrumentation framework that was never fully integrated
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
The Legacy Cry XConsole code validates that CVars only contain
underscore and alphanumeric characters via an Assert in `debug`
configurations.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
- Use ref types
- Support using rapidjson::Value in lieu of rapidjson::Document
- Use the existing JSON comparison util function in tests
Signed-off-by: Nicholas Van Sickle <nvsickle@amazon.com>
* Adds crc validation checks
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Fixes invalid CRCs
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Changes test to smoke suite
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* excludes some test data from the validator
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* uses pathlib instead of os.path
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* fixes wrong path to test scripts
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Escape not needed
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
The SettingsRegistryMergeUtils.cpp now supports specifying a path to a JSON Merge Patch formated settings registry file via the --regset-file option.
The option supports specify an anchor key to merge the settings underneath that is separated from the filepath via "::"
Ex. `--regset-file="Registry/custom.setreg::/Custom/Anchor"`
An AZ::Console command of "sr_regset-file" has also been added to allow merging of a setting registry file as well.
closes#5767
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
Fixed incorrect frame boundary guess when loading a profile capture
Added some missing default initializers
Removed implicit dependency on Atom timing marker constants
Fixed issue with small visualizer viewport bounds when loading a saved capture
Signed-off-by: AMZN-ScottR 24445312+AMZN-ScottR@users.noreply.github.com
* Add a way to set the render vsync from code
Signed-off-by: rgba16f <82187279+rgba16f@users.noreply.github.com>
* Update change with PR feedback. Missing whitespace plus a comment on why the assignment is outside the value changed check
Signed-off-by: rgba16f <82187279+rgba16f@users.noreply.github.com>
GHI bug https://github.com/o3de/o3de/issues/5866 describes a scenario in which the Editor freezes on 6-thread CPUs. This occurs because the Job Manager ends up with only 2 threads. The freeze occurs when the main thread spawns a culling job and blocks on it, a Mesh Feature Processor job spawns processing jobs and blocks on them, and a Material/ShaderVariant load queues an asset load and blocks on it. No new jobs can get started, and the main thread can't continue to process the asset load, so it freezes.
This workaround boosts the minimum number of threads from 2 to 3 by default, which fixes this specific scenario. The longer-term fix would be to fix any jobs that block on other jobs to no longer block, so that even 1 job thread would work correctly.
Signed-off-by: Mike Balfour <82224783+mbalfour-amzn@users.noreply.github.com>
* Added SerializeContext Reflection of FixedMaxPath class
Added UnitTest for validating path clas reflection
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Updated the paths loaded during the serialization to use test path
separator
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Generic Cleanup
Removals:
* Unused cvars,member variables and defines
* Windows media center edition support routines
* CSystem - removed debug_GetCallStackRaw/GetRootWindowMessageHandler/
UnloadDLL/ShutdownModuleLibraries
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Fix CryMessageBox return values in Windows build
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove named selection group and a few smaller unused functionalities
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove ObjectManager export functionality + 2 deprecated functions
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove object legacy freeze/hide support from ObjectManager
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove ObjectManager duplicate name dection, as well as object renaming
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove ObjectManager serialization and selection callbacks
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* ObjectManager - remove MoveObjects/HitTestObject/EndEditParams and related members/functions
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove a bunch of unused clone related functionality + misc
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* ObjectManager - misc removals
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* ObjectManager - more removals
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove unused object selection/retrieval in ObjectManager and LoadObjects in ObjectLoader
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* CBaseObject - remove unused material layers mask support
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* More CBaseObject cleanups.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* CBaseObject - remove SubObj functions and IMouseCreateCallback
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove unused procedural floor management, helper scale and tags.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* CBaseObject - remove more unused methods.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Removals in multiple places.
CBaseObject - unused `OBJTYPE_DUMMY` flag removed, member `IsSameClass`
CObjectArchive removed unused methods/members
CObjectManager removed unused `Update` method
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* More removals
IDisplayViewport - HitTestLine/GetGridStep/setHitcontext/
GetConstructionPlane
Unused Cry_Matrix44 template specializations.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* O3DE.exe Project-Centric "Open Editor" fix (#5852)
* The O3DE.exe Open Editor button now attempts to open the Editor in the
build directory of the project being opened.
If their is no Editor within the build directory of the Project, it uses
the Editor.exe in the current O3DE.exe executable directory if it exists
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Engine .gitignore now ignores the build directory if placed in the
AutomatedTesting project
Previously it was just ignoring a `[Bb]uild` directory if it was
directly within the engine root.
This change matches the behavior of the project templates.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Renamed the ProjectUtils GetEditorDirectory function to
GetEditorExecutablePath
Added a platform specific implementation for retrieving the path to the
Editor executable in the GetEditorExectuablePath function.
It first attempts to locate the Editor via checking the project build
directory for an Editor executable before falling back to checking the
binary directory of the currently running O3DE executable.
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* Correct the MacOS GetEditorExecutablePath to return the Editor path
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Adding missing C++20 std::erase implementations (#5735)
There were already implementations for std::erase_if.
This adds the counterpart AZStd::erase versions
resolves#5734
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* More removals and refactors
`ShowMessage` and `CryMessageBox` return void now
Simplify code in `CSystem::WarningV`
Remove unused `CryGetTicksPerSec`
Remove unused WinBase functionality
Replace `nanosleep` call in WinBase with `std::this_thread::sleep_for`
Remove unused Win32Wrapper routines
Remove unused IFunctorBase.h and IEntityObjectListener.h
Fix VectorAndArray.cpp compilation
Use QMessageBox instead of CryMessageBox in the editor.
Remove empty ArchiveVars platform specific files
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Fix test code.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove an unused function
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* CBaseObject and undo description removals.
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* A bunch of removals
* Remove IRenderNode
* Remove editor's KDTree
* Remove StatObjBus, InstanceStatObjEventBus::ReleaseData is redundant (
same functionality in the only available handler is triggered by
`OnCryEditorCloseScene`)
* Remove CExportManager::AddStatObj/AddMeshes/AddMesh,
IExportManager::ExportSingleStatObj
* Remove CIconManager/IIconManager::GetObject
* Remove CBaseObject::IntersectRayMesh
* Remove IIndexedMesh and related structs.
* Unused IUndoObject::GetDescription and all derived implementations.
* Unused CUndoBaseLibrary/CUndoBaseLibraryManager
* Unused Matrix34_tpl typedefs
* Legacy Xml classes cleanup
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* Remove unused `Vec3ToVector3D` helper
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
* fix - remove unused static variable
Signed-off-by: nemerle <96597+nemerle@users.noreply.github.com>
Co-authored-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
There were already implementations for std::erase_if.
This adds the counterpart AZStd::erase versions
resolves#5734
Signed-off-by: lumberyard-employee-dm <56135373+lumberyard-employee-dm@users.noreply.github.com>
* update return type for viewport screen functions
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
* add tests for AZ::Matrix3x4::CreateFromMatrix4x4 and add TransformPoint to Matrix3x4
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
* update NDC -> Ndc
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
* updates following review feedback
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
* updates and improvements following PR feedback
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
* add forward declaration of Matrix3x4 type
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
* update where forward declarations are defined for Matrix3x4
Signed-off-by: Tom Hulton-Harrop <82228511+hultonha@users.noreply.github.com>
AZStd::fixed_vector had all its functions marked with constexpr, but this requires all member variables to be fully initialized. This meant that the internal array used to store elements always has to be fully initialized. This was done for trivial classes but not for non-trivial classes. As a result trivial classes always did a memset (or more optimized versions for smaller buffers) while the non-trivial version couldn't actually be stored in a constexpr variable. Since AZStd::fixed_vector is meant to be dynamic the choice was made to remove the constexpr from all non-static member functions in favor of avoiding the overhead of memset, which profiling showed was a considerable overhead depending on the reserved size. If a truly constexpr array is needed than AZStd::array is a better choice as that's designed to not by dynamic.
Signed-off-by: AMZN-koppersr <82230785+AMZN-koppersr@users.noreply.github.com>
* Cherry-pick 49e8f35858
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Merging differences from development of other changes that need to be there for deb packaging
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Picks a needed change for the installer
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Fixes warning in mac
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Takes version from environment if defined
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Do not pick up version if it is empty string since that will also break version comparison
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* creating temp directories if they dont exist
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* removing a dependency to itself (Multiplayer.Builders is an alias of Multiplayer.Editor)
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* Filters which runtime dependencies are passed from private build dependencies to only those that are actual targets.
This avoids something like a "d3d12" private build dependency from being passed to the runtime dependencies
Signed-off-by: Esteban Papp <81431996+amznestebanpapp@users.noreply.github.com>
* [SPEC-7644] ParallelDeepAssetReferences is failing intermittently (#5721)
* Fixed race condition caused by trying to handle asset ready event before asset container has finished filling out all the data structures.
Added check to only handle asset ready once init is complete
Added unit test to verify fix
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Re-enable test
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Add missing space to error message
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Add comment on sleep
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Collapse nested namespace
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
* Collapse nested namespace
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>
(cherry picked from commit 56900484fc)
# Conflicts:
# Code/Framework/AzCore/AzCore/Asset/AssetContainer.cpp
* Fix indentation
Signed-off-by: amzn-mike <80125227+amzn-mike@users.noreply.github.com>