mmhir

hercules_samples/matmul/src/matmul.hir

-fn matmul<3>(a: array(i32, #0, #1), b: array(i32, #1, #2)) -> array(i32, #0, #2)
+fn matmul<3>(a: array(i32, #0, 8), b: array(i32, 8, #2)) -> array(i32, #0, #2)
   c = constant(array(i32, #0, #2), [])
   i_j_ctrl = fork(start, #0, #2)
   i_idx = thread_id(i_j_ctrl, 0)
   j_idx = thread_id(i_j_ctrl, 1)
-  k_ctrl = fork(i_j_ctrl, #1)
+  k_ctrl = fork(i_j_ctrl, 8)
   k_idx = thread_id(k_ctrl, 0)
   k_join_ctrl = join(k_ctrl)
   i_j_join_ctrl = join(k_join_ctrl)