[PATCH] D155995: [AMDGPU]: Allow combining into v_dot4

[Jeffrey Byrnes via Phabricator via llvm-commits]

jrbyrnes added inline comments.


================
Comment at: llvm/lib/Target/AMDGPU/SIISelLowering.cpp:12397
+
+    if (Elt->second == 0x3020100)
+      return EltVal;
----------------
arsenm wrote:
> maybe comment skipping bswap?
It's the original SDValue, not bswap. I'll add comment.


```
def : GCNPat <
  (i32 (bswap i32:$a)),
  (V_PERM_B32_e64 (i32 0), VSrc_b32:$a, (S_MOV_B32 (i32 0x00010203)))
>;
```


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155995/new/

https://reviews.llvm.org/D155995