[PATCH] D122241: [X86] Fold MMX_MOVD64from64rr + store to MMX_MOVQ64mr instead of MMX_MOVD64from64mr.

[Craig Topper via Phabricator via llvm-commits]

craig.topper created this revision.
craig.topper added reviewers: skan, pengfei, RKSimon.
Herald added subscribers: StephenFan, hiraditya.
Herald added a project: All.
craig.topper requested review of this revision.
Herald added a project: LLVM.

MMX_MOVD64from64rr moves an MMX register to a 64-bit GPR.

MMX_MOVD64from64mr is the memory version of moving a MMX register to a
64-bit GPR. It requires the REX.W bit to be set. There are no isel
patterns that use this instruction.

MMX_MOVQ64mr is the MMX register store instruction. It doesn't
require a REX.W prefix. This makes it one byte shorter to encode
than MMX_MOVD64from64mr in many cases.

Both store instructions output the same mnemonic string. The assembler
would choose MMX_MOVQ64mr if it was to parse the output. Which is
another reason using it is the correct thing to do.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D122241

Files:
  llvm/lib/Target/X86/X86InstrFoldTables.cpp
  llvm/test/CodeGen/X86/stack-folding-mmx.ll


Index: llvm/test/CodeGen/X86/stack-folding-mmx.ll
===================================================================
--- llvm/test/CodeGen/X86/stack-folding-mmx.ll
+++ llvm/test/CodeGen/X86/stack-folding-mmx.ll
@@ -140,7 +140,7 @@
 ; CHECK-NEXT:    pushq %r12
 ; CHECK-NEXT:    pushq %rbx
 ; CHECK-NEXT:    paddb %mm0, %mm0
-; CHECK-NEXT:    movq %mm0, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Folded Spill
+; CHECK-NEXT:    movq %mm0, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; CHECK-NEXT:    #APP
 ; CHECK-NEXT:    nop
 ; CHECK-NEXT:    #NO_APP
Index: llvm/lib/Target/X86/X86InstrFoldTables.cpp
===================================================================
--- llvm/lib/Target/X86/X86InstrFoldTables.cpp
+++ llvm/lib/Target/X86/X86InstrFoldTables.cpp
@@ -292,7 +292,7 @@
   { X86::JMP32r_NT,           X86::JMP32m_NT,           TB_FOLDED_LOAD },
   { X86::JMP64r,              X86::JMP64m,              TB_FOLDED_LOAD },
   { X86::JMP64r_NT,           X86::JMP64m_NT,           TB_FOLDED_LOAD },
-  { X86::MMX_MOVD64from64rr,  X86::MMX_MOVD64from64mr,  TB_FOLDED_STORE | TB_NO_REVERSE },
+  { X86::MMX_MOVD64from64rr,  X86::MMX_MOVQ64mr,        TB_FOLDED_STORE | TB_NO_REVERSE },
   { X86::MMX_MOVD64grr,       X86::MMX_MOVD64mr,        TB_FOLDED_STORE | TB_NO_REVERSE },
   { X86::MOV16ri,             X86::MOV16mi,             TB_FOLDED_STORE },
   { X86::MOV16rr,             X86::MOV16mr,             TB_FOLDED_STORE },


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D122241.417337.patch
Type: text/x-patch
Size: 1426 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20220322/475dc846/attachment.bin>