Ben Ashbaugh
Ben Ashbaugh
Philosophically, it should be possible for a well-written OpenCL program to execute without generating any OpenCL errors. This means that no error conditions should be "unavoidable", where the only way...
From a discussion in the October 14th memory subgroup: If an application allocates SVM memory by calling **clSVMAlloc**, then wraps the SVM allocation in an OpenCL memory object by passing...
This is another PR to partially address the way we document error conditions, see https://github.com/KhronosGroup/OpenCL-Docs/issues/1320 and in particular https://github.com/KhronosGroup/OpenCL-Docs/issues/1320#issuecomment-2745919727. It contains a subset of the changes from https://github.com/KhronosGroup/OpenCL-Docs/pull/1399, specifically for...
This PR has the extension source for the cl_intel_subgroup_matrix_multiply_accumulate_tf32 extension.
I'm getting an warning building the specs with https://github.com/KhronosGroup/OpenCL-Docs/pull/1469: ``` WARNING: type: cl_device_scheduling_controls_capabilities_img IS NOT DEFINED ``` I'm not exactly sure what the type should be, but we need a...
This is another PR to partially address the way we document error conditions, see https://github.com/KhronosGroup/OpenCL-Docs/issues/1320 and in particular https://github.com/KhronosGroup/OpenCL-Docs/issues/1320#issuecomment-2745919727. It contains a subset of the changes from https://github.com/KhronosGroup/OpenCL-Docs/pull/1399, specifically for...
The description of `CL_FLT_MIN` (and `CL_DBL_MIN` / `CL_HALF_MIN`) is currently: > Minimum value of type cl_float "Minimum value" in this case could refer to the smallest floating-point value (meaning, a...
The OpenCL C section [Built-in Sub-Group Relative Shuffle Functions](https://registry.khronos.org/OpenCL/specs/3.0-unified/html/OpenCL_C.html#_built_in_sub_group_relative_shuffle_functions) is missing the note that these functions require support for the cl_khr_subgroup_shuffle_relative extension.
Fixes #1429 Removes a copy-paste error and rephrases to reduce duplication.
This is a PR to partially address the way we document error conditions, see #1320 and in particular https://github.com/KhronosGroup/OpenCL-Docs/issues/1320#issuecomment-2745919727. This doesn't modify the entire spec, but: * It's already a...