The fix for issue #624 added an error condition for the slice pitch for clEnqueueReadImage and clEnqueueWriteImage but a few related parts of the spec could still use attention:
Should there be a requirement that the row_pitch or input_row_pitch must be zero?
The currently descriptions are slightly contradictory and say that e.g. the value "must be greater than or equal to the element size in bytes × width" but then that if "row_pitch is set to 0, the appropriate row pitch is calculated". It would be better to say that the value must be zero or greater than the proper value in these cases.
I initially thought similar clarifications could be needed for the row pitch and slice pitch arguments to clEnqueueReadBufferRect, clEnqueueWriteBufferRect, and clEnqueueCopyBufferRect would be needed, but after looking more closely I think they're OK.
The text was updated successfully, but these errors were encountered:
The fix for issue #624 added an error condition for the slice pitch for clEnqueueReadImage and clEnqueueWriteImage but a few related parts of the spec could still use attention:
row_pitch
orinput_row_pitch
must be zero?I initially thought similar clarifications could be needed for the row pitch and slice pitch arguments to
clEnqueueReadBufferRect
,clEnqueueWriteBufferRect
, andclEnqueueCopyBufferRect
would be needed, but after looking more closely I think they're OK.The text was updated successfully, but these errors were encountered: