Kacper Nowak 26666d8c6f fix: Use getGpuAddressToPatch() of pattern allocation ...

When using FillBuffer builtin, use GPU address to patch (i.e. gpuAddress
plus allocationOffset minus gpuBaseAddress) instead of gpuAddress.
This change fixes a bug - possible abort due to huge difference between
gpuAddress() and gpuAddressToPatch() values.

Related-To: NEO-7973
Signed-off-by: Kacper Nowak <kacper.nowak@intel.com>

2023-10-19 16:23:55 +02:00

..

kernels

…

registry

…

aux_translation_builtin.h

refactor: rename hw_helper files to gfx_core_helper files

2023-02-01 19:37:51 +01:00

built_in_ops_vme.h

…

built_ins.inl

Refactor(ocl): refactor OCL Kernel and MultiDeviceKernel class 1/n

2023-04-11 11:43:26 +02:00

builtins_dispatch_builder.cpp

fix: Use getGpuAddressToPatch() of pattern allocation

2023-10-19 16:23:55 +02:00

builtins_dispatch_builder.h

performance: Bidirectional BCS split implementation for OCL

2023-05-16 13:34:18 +02:00

CMakeLists.txt

…

populate_built_ins.inl

refactor: Remove globaly enabled cl_cache

2023-04-28 23:28:49 +02:00

unknown_built_in.cpp

…

vme_builtin.cpp

refactor: Remove globaly enabled cl_cache

2023-04-28 23:28:49 +02:00

vme_builtin.h

…

vme_dispatch_builder.h

Cleanup includes 56

2023-02-16 14:42:44 +01:00