[PATCH] D147775: [X86] Fix problem in D147541

[Phoebe Wang via Phabricator via llvm-commits]

This revision was not accepted when it landed; it landed in state "Needs Review".
This revision was automatically updated to reflect the committed changes.
Closed by commit rG2ffdfb5f9dff: [X86] Fix problem in D147541 (authored by pengfei).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D147775/new/

https://reviews.llvm.org/D147775

Files:
  llvm/lib/Target/X86/X86FixupInstTuning.cpp


Index: llvm/lib/Target/X86/X86FixupInstTuning.cpp
===================================================================
--- llvm/lib/Target/X86/X86FixupInstTuning.cpp
+++ llvm/lib/Target/X86/X86FixupInstTuning.cpp
@@ -154,8 +154,8 @@
     return true;
   };
 
-  // `vunpcklpd/vmovlhps r, r` -> `vshufps r, r, 0x44`
-  // `vunpckhpd/vmovlhps r, r` -> `vshufps r, r, 0xee`
+  // `vunpcklpd/vmovlhps r, r` -> `vshufpd r, r, 0x00`
+  // `vunpckhpd/vmovlhps r, r` -> `vshufpd r, r, 0xff`
   // `vunpcklpd r, r, k` -> `vshufpd r, r, 0x00`
   // `vunpckhpd r, r, k` -> `vshufpd r, r, 0xff`
   // iff `vshufps` is faster than `vunpck{l|h}pd`. Otherwise stick with
@@ -234,12 +234,12 @@
     // processors where its safe.
   case X86::MOVLHPSrr:
   case X86::UNPCKLPDrr:
-    return ProcessUNPCKLPDrr(X86::SHUFPSrri);
+    return ProcessUNPCKLPDrr(X86::SHUFPDrri);
   case X86::VMOVLHPSrr:
   case X86::VUNPCKLPDrr:
-    return ProcessUNPCKLPDrr(X86::VSHUFPSrri);
+    return ProcessUNPCKLPDrr(X86::VSHUFPDrri);
   case X86::VUNPCKLPDYrr:
-    return ProcessUNPCKLPDrr(X86::VSHUFPSYrri);
+    return ProcessUNPCKLPDrr(X86::VSHUFPDYrri);
     // VMOVLHPS is always 128 bits.
   case X86::VMOVLHPSZrr:
   case X86::VUNPCKLPDZ128rr:


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D147775.511684.patch
Type: text/x-patch
Size: 1220 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20230407/2508009c/attachment.bin>