Difference between `thread_execution_width` and `threads_per_simdgroup`

I have a compute kernel that makes use of simdgroup operations such as simd_shuffle_up, simd_or, etc, and I'm looking to rewrite the kernel to support older hardware. One such computation requires that I know the index of the thread in the simdgroup (thread_index_in_simdgroup). I was hoping to derive it from the thread's position in its threadgroup (thread_position_in_threadgroup) and the thread execution width (thread_execution_width), along with other knowledge about the size of the threadgroup when I noticed there was also the threads_per_simdgroup attribute. The spec describes both respectively as

thread_execution_width: The execution width of the compute unit.

threads_per_simdgroup: The thread execution width of a SIMD-group.

Under what conditions, if any, could these two values differ? If they do differ, is there a way to determine a thread's position in the simdgroup on hardware that doesn't support Metal 2.2?

Up vote post of maxwellpirtle Down vote post of maxwellpirtle
569 views
Posted by
maxwellpirtle
Reply
Add a Comment

Accepted Reply

There is no difference between thread_execution_width and threads_per_simdgroup. We've noted in Metal Shading Language 3.0 that:

thread_execution_width All OS: Since Metal 1.0. [[ Deprecated as of Metal 3.0 – use threads_per_simdgroup ]]

There's also this line:

[[threads_per_simdgroup]] and [[thread_execution_width]] are aliases of one another that reference the same concept.

Posted by
Graphics and Games Engineer
Up vote reply of Graphics and Games Engineer Down vote reply of Graphics and Games Engineer
Post marked as solved

  • Thanks for the quick reply. Looks like I was looking at the 2.4 spec 😅 I should probably remove my local copies of the old specs now

    —
    maxwellpirtle
Add a Comment

Replies

There is no difference between thread_execution_width and threads_per_simdgroup. We've noted in Metal Shading Language 3.0 that:

thread_execution_width All OS: Since Metal 1.0. [[ Deprecated as of Metal 3.0 – use threads_per_simdgroup ]]

There's also this line:

[[threads_per_simdgroup]] and [[thread_execution_width]] are aliases of one another that reference the same concept.

Posted by
Graphics and Games Engineer
Up vote reply of Graphics and Games Engineer Down vote reply of Graphics and Games Engineer
Post marked as solved

  • Thanks for the quick reply. Looks like I was looking at the 2.4 spec 😅 I should probably remove my local copies of the old specs now

    —
    maxwellpirtle
Add a Comment