You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
CL_INVALID_IMAGE_DESCRIPTOR if values specified in image_desc are not valid or if image_desc is NULL.
CL_INVALID_IMAGE_SIZE if image dimensions specified in image_desc exceed the maximum image dimensions described in the Device Queries table for all devices in context.
However for the legacy clCreateImage3D, spec says:
CL_INVALID_IMAGE_SIZE if image_width or image_height are 0 or if image_depth ≤ 1, or if they exceed the maximum values specified in CL_DEVICE_IMAGE3D_MAX_WIDTH, CL_DEVICE_IMAGE3D_MAX_HEIGHT or CL_DEVICE_IMAGE3D_MAX_DEPTH respectively for all devices in context, or if values specified by image_row_pitch and image_slice_pitch do not follow rules described in the argument description above.
This means, for e.g., that when trying to create a 3D image with depth 0 (and hence invalid), the new APIs are required to return CL_INVALID_IMAGE_DESCRIPTOR whereas the legacy API would return CL_INVALID_IMAGE_SIZE. It would be ideal to reconcile the error descriptions so we use/expand CL_INVALID_IMAGE_SIZE for this case as well rather than restrict it to exceeding the maximum dimension supported.
The text was updated successfully, but these errors were encountered:
Do you mind if I transfer this issue to OpenCL-Docs? It seems like more of a spec issue than a test issue, unless there is also a test for the current error behavior.
Under error codes for clCreateImage/clCreateImageWithProperties, the API spec says:
CL_INVALID_IMAGE_DESCRIPTOR if values specified in image_desc are not valid or if image_desc is NULL.
CL_INVALID_IMAGE_SIZE if image dimensions specified in image_desc exceed the maximum image dimensions described in the Device Queries table for all devices in context.
However for the legacy clCreateImage3D, spec says:
CL_INVALID_IMAGE_SIZE if image_width or image_height are 0 or if image_depth ≤ 1, or if they exceed the maximum values specified in CL_DEVICE_IMAGE3D_MAX_WIDTH, CL_DEVICE_IMAGE3D_MAX_HEIGHT or CL_DEVICE_IMAGE3D_MAX_DEPTH respectively for all devices in context, or if values specified by image_row_pitch and image_slice_pitch do not follow rules described in the argument description above.
This means, for e.g., that when trying to create a 3D image with depth 0 (and hence invalid), the new APIs are required to return CL_INVALID_IMAGE_DESCRIPTOR whereas the legacy API would return CL_INVALID_IMAGE_SIZE. It would be ideal to reconcile the error descriptions so we use/expand CL_INVALID_IMAGE_SIZE for this case as well rather than restrict it to exceeding the maximum dimension supported.
The text was updated successfully, but these errors were encountered: