This class of tests can fail randomnly since they assume that time to
create first set of stream should be greater than second set. This is
incorrect since there can be multiple reason why the time for second
batch can be greater than the first batch.
Change-Id: I5f91c2bab7119608014dead25fdeb38523c61682
https://gerrit-git.amd.com/c/compute/ec/clr/+/849086 changed stream sync behaviour and accordingly updated Unit_hipStreamCreateWithFlags_DefaultStreamInteraction.
Update error code to match for hipStreamWaitEvent.
Thread detach makes executions all async and while that is happening, It is unexpected effect for device reset so join thread is better.
Change-Id: I1affa84089626dee478d8bcc5aaa318e320fd6b0
* SWDEV-385161 - Deprecate usage of env vars in HIP samples and tests env vars such as HIP_PATH, ROCM_PATH, HIP_COMPILER, HIP_RUNTIME, HSA_PATH etc are removed.
Change-Id: I280e20b291cfb14675846aae72d97fa5b3ad567d
- remove null stream from operation with other created streams on multithread test
- some general test fix
Change-Id: Icec7436f92a2d90dcee93ed5cdc4c8934d803fde
* SWDEV-359379 - catch2: Standalone single exe per file
-workaround for rsp file issue.
-Creates single exe per file
-tests detection is still during execution time and NOT compile time
Change-Id: Iddfb83d57b2d767212f3d9307a276b7d572da6cd
* SWDEV-359379 - Update CMakeList
* Update Catch.cmake
* Temporarily disable failing tests
- Negative parameter tests
- Validate basic behavior
- Validate the behavior when pageable memory access is supported
- Validate the behavior for hipMemAttachGlobal
- Validate the behavior for hipMemAttachHost
- Validate the behavior for hipMemAttachSingle
- Add simple test that waits for an event - Unit_hipStreamWaitEvent_Default
- Add negative test when stream is uninitialized - Unit_hipStreamWaitEvent_WithUninitializedStream
- Minor modification of Unit_hipStreamWaitEvent_DifferentStreamstxt
- Modify uninitialized stream error code
- Implement negative test for hipStream<Write/Wait>Value when stream is uninitialized
- Implement negative test for hipStreamSynchronize when stream is uninitialized
- Invalid flag negative test for hipStreamWriteValue API removed as flag parameter is not used
- Fix bug in calculation of normal priority in Unit_hipStreamGetPriority_StreamsWithCUMask tests
- Delete duplicate test Unit_hipStreamGetPriority_InvalidPriorityPointer
- Separate Unit_hipStreamAddCallback_ParamTst into positive and negative test
- Remove Unit_hipStreamDestroy_Negative_DoubleDestroy test for Nvidia devices due to error
- For gfx >= 10, one work group processor encompasses 2 CUs & hence the CUs need to be enabled in pair
Change-Id: I359df1b221b4400b260b02201b7b0385054784f3
* AHTS-90 - Add missing test case for hipStreamCreate
* Update hipStreamGetFlags testing
* Added testing for hipStreamSynchronize
* Added macro to test a particular error code is returned from an expression
* Update hip_test_common.hh
* Correcting checkers to properly list the test count
* fix copy paste error in HIP_CHECK_ERROR
* Add support for hipStreamWaitValue
* Remove changes unrelated to this PR
* Revert "Added testing for hipStreamSynchronize"
* Remove changes unrelated to this PR
* Added skip test for failure on AMD devices and removed changed to hip_test_common.hh
* Fixed build issues on Nvidia platforms by disabled incompatible tests; Updated negative tests to check the correct return values
Co-authored-by: Jatin Chaudhary <jatin.chaudhary@codeplay.com>
Co-authored-by: Fábio Mestre <fabio.mestre@codeplay.com>
Co-authored-by: Finlay Marno <finlay.marno@codeplay.com>
Co-authored-by: Fábio <fabio.m.mestre@gmail.com>