Ben Ashbaugh

icon indicating an email [email protected]

icon location I work on GPUs for Intel, but the projects on my Github are my own. Interested in parallel and heterogeneous compute.

Results 494 comments of Ben Ashbaugh

Implement Negative Tests for clDevice Functions

I'm going to take off "focused review" until the review comments have been addressed - thanks!

Implement negative tests for cl_device_id API functions

Merging as discussed in the December 2nd teleconference.

Risky testing of versions of provisional extensions

Something we might want to consider in this area is: 1. Combine the extension check and the extension version check into a single new harness function. 2. Add a harness...

cl_khr_command_buffer access to read-only memory objects

Even though the current behavior is non-intuitive, I think it would be more confusing if the command buffer behavior (`clCommandFillImageKHR`) deviated from the command queue behavior (`clEnqueueFillImage`). Could we solve...

cl_khr_command_buffer access to read-only memory objects

Discussed in the February 11th teleconference, without a conclusion: - There is a potential issue for both read-only buffers and read-only images. - Note, there is also a request to...

Lifetime of SVM allocations in memory objects

Discussed in the October 21st teleconference. We didn't find anyplace hinting that the answer is anything other than (1), and this is what implementations are doing. Leaving this issue open...

cl_intel_subgroup_matrix_multiply_accumulate_tf32

https://opencl.gpuinfo.org/listreports.php?extension=cl_intel_subgroup_matrix_multiply_accumulate_tf32 Please note, since this is an OpenCL C-only extension, I have not modified the XML file. I still have an action item to add all of the Intel OpenCL...

cl_device_scheduling_controls_capabilities_img type missing from the XML file

Sorry for missing this previously. @paulfradgley can you please take a look?

improve error code consistency, part one

Now that v3.0.19 is published I'd like to start getting these changes over the finish line. Would it be helpful to break this PR into smaller pieces? If so, what...

improve error code consistency, part one

I'll break this PR into smaller pieces for easier review. Moving to "draft" so it doesn't get accidentally merged.