- single slice devices need subslice remapping
- convert api thread ids to physical ids within specific device index
Related-To: NEO-6950
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
This patch uses SDI to reset all except the last event packet.
Fixes GPU address usage for copy command list.
Related-To: LOCI-3052
Signed-off-by: Joshua Santosh Ranjan <joshua.santosh.ranjan@intel.com>
- Allow for all exported symbols between L0 Dynamically linked Modules
to be accessible by adding the exported function allocations to all
linked modules unconditionally.
- This enables for L0 Function Pointers to be used to call functions
between the modules without unresolved symbols, which were a condition
to allow for exported allocations to be accessed between modules.
Signed-off-by: Neil R Spruit <neil.r.spruit@intel.com>
- new feature, enabled with PRELIM build
- implementation of debug session for linux
- move ResourceClass enum from Drm to drm_debug.h
Resolves: NEO-6814
Signed-off-by: Mateusz Hoppe <mateusz.hoppe@intel.com>
Define single .clang-tidy configuration with all used checks and use
NOLINT to selectively silence tool. That way cleanup should be easier.
third_part/ has its own configuration that disables clang-tidy for this
folder.
Signed-off-by: Artur Harasimiuk <artur.harasimiuk@intel.com>
When using implicit scaling, device allocations may have
more than one internal allocation created internally. In that case,
a separate dma-buf handle per internal allocation needs to be
exported.
So introduced two driver experimental extensions to export and
import more than one IPC handle:
- zexMemGetIpcHandles
- zexMemOpenIpcHandles
Related-To: LOCI-2919
Signed-off-by: Jaime Arteaga <jaime.a.arteaga.molina@intel.com>
Add debug variable ForceGrfNumProgrammingWithScm.
Do not update large grf value in StreamProperties when unnecessary.
Related-To: NEO-6659
Signed-off-by: Filip Hazubski <filip.hazubski@intel.com>
correct names - use xehpc instead of pvc for core tests
transform explicit xehpc tests to pvc if skip exist
Related-To: NEO-6738
Signed-off-by: Katarzyna Cencelewska <katarzyna.cencelewska@intel.com>
This change introduces checking the error code returned by
executeCommandListImmediate() inside appendMemoryRangesBarrier().
When GPU hang is encountered, the error code is propagated.
This change contains also ULTs to cover the new code and improve
coverage of that function.
Related-To: NEO-6681
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
The build of test_cmdlist_append_launch_kernel_1.cpp caused
fatal error C1128 - number of sections exceeded object file
format limit.
Signed-off-by: Patryk Wrobel <patryk.wrobel@intel.com>
This change:
* Adds support for build options section in zebinary - using
build options in binary when rebuilding.
* Appends "-cl-intel-allow-zebin" flag to build options when zebin is
used.
Resolves: NEO-6916
Signed-off-by: Krystian Chmielewski <krystian.chmielewski@intel.com>
Fix kernel args set in reverse order, add new cmd line parameter
to specify allocation size and improve input/output validation
Related-To: NEO-6839
Signed-off-by: Slawomir Milczarek <slawomir.milczarek@intel.com>
Given a following sequence of setting kernel arg svm pointers: A, nullptr, A
the kernel arg was not being set to A
Related-To: NEO-6895
Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
