[PATCH] D17160: AMDGPU/SI: Add llvm.amdgcn.mov.dpp intrinsic

[Tom Stellard via llvm-commits]

tstellarAMD added inline comments.

================
Comment at: lib/Target/AMDGPU/VIInstructions.td:115-121
@@ +114,9 @@
+//===----------------------------------------------------------------------===//
+
+def : Pat <
+  (int_amdgcn_mov_dpp i32:$src, imm:$dpp_ctrl, imm:$bound_ctrl,
+                      imm:$bank_mask, imm:$row_mask),
+  (V_MOV_B32_dpp $src, (as_i32imm $dpp_ctrl), (as_i1imm $bound_ctrl),
+                       (as_i32imm $bank_mask), (as_i32imm $row_mask))
+>;
+
----------------
arsenm wrote:
> Is there a reason this needs to be a standalone pattern?
This is likely the only dpp pattern, so I didn't think it was worth it to modify the existing VOP1Inst class and related classes so they would be able to accept dpp patterns.


http://reviews.llvm.org/D17160