In this mode AUB csr will be created, however, no aub file will be created
Related-To: NEO-11097
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
Move ulls controller thread stopping to execution environment
destructor. This removes the vptr race from ulls controller destructor.
Put tests that are actually running the ulls controller into multi thread
tests.
That way they are compiled with thread sanitizer and removes the
variable test time from traditional ULTS.
Related-To: NEO-10942
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
- samplers using bindless adressing require patching bindless offsets to
sampler states on kernel's cross thread data
Related-To: NEO-10505
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
- Move logic from product helper to compiler product helper
- Add method for adjusting fp16 and extra capabilities using release
helper (if present).
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>
aub tests compilation can be enabled with cmake flag NEO_SKIP_AUB_TESTS=0
aub tests execution can be enabled with cmake flag NEO_SKIP_AUB_TEST_RUN=0
enabled aub tests execution forces compilation of aub tests targets
Signed-off-by: Mateusz Jablonski <mateusz.jablonski@intel.com>
ULLS controller timeout settings will be adjusted based on ac line
status and lowest queue throttle from submissions.
Lowest queue throttle is reset when controller stops ULLS.
Related-To: NEO-10800
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
- report Bindless_images extension in driver only when at least one
device has global bindless heaps enabled
Related-To: NEO-10352
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
Make bindless heaps resident right after heap allocation.
Motivation is that SYCL bindless image can be passed as a value argument
or through memory. Therefore, we're not able to make its bindless heap
resident during kernel initialization or setting kernel arguments.
This fixes SYCL bindless image read_write_*D.cpp tests on DG2.
Related-To: NEO-7063
Signed-off-by: Wenju He <wenju.he@intel.com>
When calculating the size of the indirect object heap,
the local work group size from kernel implicit args is taken into account.
If the LWS is not set before this calculation,
it can lead to insufficient ioh allocation size.
Such a problem is seen when local ids are generated by the runtime
and then written to ioh. The write fails due to lack of space in the allocation.
Related-To: IGC-7708
Signed-off-by: Fabian Zwolinski <fabian.zwolinski@intel.com>
- use virtualFileSystem in ULTs, writeDataToFile() does not create file
but instead saves filename. Call to fileExists() checks if file was
previously written
Related-To: NEO-7006
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
explicit test for xe hpc
add missing compression flag to compressed buffer performance hint test
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
Ensure that device with CSR_AUB is created
after the aub name override is properly initialized.
For devices with RCS disabled and single CCS enabled,
allow dual context cases to utilize the same CCS engine twice.
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
Related-To: NEO-9151
With this change OpenGL should be able to use
cl_gl_private_intel_structures.h header
directly from openCL repository.
Signed-off-by: Maciej Plewka <maciej.plewka@intel.com>
