[X86][MMX] Added custom lowering action for MMX SELECT (PR30418)

Fix for pr30418 - error in backend: Cannot select: t17: x86mmx = select_cc t2, Constant:i64<0>, t7, t8, seteq:ch
Differential Revision: https://reviews.llvm.org/D34661

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@309614 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 081b711a145c63082d1eef18ec7ce4858e2cb3b0..e1dd6d03550bf0a11221da675bb1599452f64e28 100644 (file)
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -419,6 +419,11 @@ X86TargetLowering::X86TargetLowering(const X86TargetMachine &TM,
     setOperationAction(ISD::SELECT, VT, Custom);
     setOperationAction(ISD::SETCC,  VT, Custom);
   }
+
+  // Custom action for SELECT MMX and expand action for SELECT_CC MMX
+  setOperationAction(ISD::SELECT, MVT::x86mmx, Custom);
+  setOperationAction(ISD::SELECT_CC, MVT::x86mmx, Expand);
+
   setOperationAction(ISD::EH_RETURN       , MVT::Other, Custom);
   // NOTE: EH_SJLJ_SETJMP/_LONGJMP supported here is NOT intended to support
   // SjLj exception handling but a light-weight setjmp/longjmp replacement to
@@ -30631,6 +30636,14 @@ static SDValue combineSelect(SDNode *N, SelectionDAG &DAG,
       return SDValue(N, 0);
   }
 
+  // Custom action for SELECT MMX
+  if (VT == MVT::x86mmx) {
+    LHS = DAG.getBitcast(MVT::i64, LHS);
+    RHS = DAG.getBitcast(MVT::i64, RHS);
+    SDValue newSelect = DAG.getNode(ISD::SELECT, DL, MVT::i64, Cond, LHS, RHS);
+    return DAG.getBitcast(VT, newSelect);
+  }
+
   return SDValue();
 }
 

diff --git a/test/CodeGen/X86/select-mmx.ll b/test/CodeGen/X86/select-mmx.ll
new file mode 100644 (file)
index 0000000..9e6382f
--- /dev/null
+++ b/test/CodeGen/X86/select-mmx.ll
@@ -0,0 +1,120 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=x86_64-unknown-unknown -mattr=+mmx < %s | FileCheck %s --check-prefix=X64
+; RUN: llc -mtriple=i686-unknown-unknown -mattr=+mmx < %s | FileCheck %s --check-prefix=I32
+
+
+; From source: clang -02
+;__m64 test47(int a)
+;{
+;    __m64 x = (a)? (__m64)(7): (__m64)(0);
+; return __builtin_ia32_psllw(x, x);
+;}
+
+define i64 @test47(i64 %arg)  {
+;
+; X64-LABEL: test47:
+; X64:       # BB#0:
+; X64-NEXT:    xorl %eax, %eax
+; X64-NEXT:    testq %rdi, %rdi
+; X64-NEXT:    movl $7, %ecx
+; X64-NEXT:    cmoveq %rcx, %rax
+; X64-NEXT:    movd %rax, %mm0
+; X64-NEXT:    psllw %mm0, %mm0
+; X64-NEXT:    movd %mm0, %rax
+; X64-NEXT:    retq
+;
+; I32-LABEL: test47:
+; I32:       # BB#0:
+; I32-NEXT:    pushl %ebp
+; I32-NEXT:  .Lcfi0:
+; I32-NEXT:    .cfi_def_cfa_offset 8
+; I32-NEXT:  .Lcfi1:
+; I32-NEXT:    .cfi_offset %ebp, -8
+; I32-NEXT:    movl %esp, %ebp
+; I32-NEXT:  .Lcfi2:
+; I32-NEXT:    .cfi_def_cfa_register %ebp
+; I32-NEXT:    andl $-8, %esp
+; I32-NEXT:    subl $16, %esp
+; I32-NEXT:    movl 8(%ebp), %eax
+; I32-NEXT:    orl 12(%ebp), %eax
+; I32-NEXT:    movl $7, %eax
+; I32-NEXT:    je .LBB0_2
+; I32-NEXT:  # BB#1:
+; I32-NEXT:    xorl %eax, %eax
+; I32-NEXT:  .LBB0_2:
+; I32-NEXT:    movl %eax, {{[0-9]+}}(%esp)
+; I32-NEXT:    movl $0, {{[0-9]+}}(%esp)
+; I32-NEXT:    movq {{[0-9]+}}(%esp), %mm0
+; I32-NEXT:    psllw %mm0, %mm0
+; I32-NEXT:    movq %mm0, (%esp)
+; I32-NEXT:    movl (%esp), %eax
+; I32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; I32-NEXT:    movl %ebp, %esp
+; I32-NEXT:    popl %ebp
+; I32-NEXT:    retl
+  %cond = icmp eq i64 %arg, 0
+  %slct = select i1 %cond, x86_mmx bitcast (i64 7 to x86_mmx), x86_mmx bitcast (i64 0 to x86_mmx)
+  %psll = tail call x86_mmx @llvm.x86.mmx.psll.w(x86_mmx %slct, x86_mmx %slct)
+  %retc = bitcast x86_mmx %psll to i64
+  ret i64 %retc
+}
+
+
+; From source: clang -O2
+;__m64 test49(int a, long long n, long long m)
+;{
+;    __m64 x = (a)? (__m64)(n): (__m64)(m);
+; return __builtin_ia32_psllw(x, x);
+;}
+
+define i64 @test49(i64 %arg, i64 %x, i64 %y) {
+;
+; X64-LABEL: test49:
+; X64:       # BB#0:
+; X64-NEXT:    testq %rdi, %rdi
+; X64-NEXT:    cmovneq %rdx, %rsi
+; X64-NEXT:    movd %rsi, %mm0
+; X64-NEXT:    psllw %mm0, %mm0
+; X64-NEXT:    movd %mm0, %rax
+; X64-NEXT:    retq
+;
+; I32-LABEL: test49:
+; I32:       # BB#0:
+; I32-NEXT:    pushl %ebp
+; I32-NEXT:  .Lcfi3:
+; I32-NEXT:    .cfi_def_cfa_offset 8
+; I32-NEXT:  .Lcfi4:
+; I32-NEXT:    .cfi_offset %ebp, -8
+; I32-NEXT:    movl %esp, %ebp
+; I32-NEXT:  .Lcfi5:
+; I32-NEXT:    .cfi_def_cfa_register %ebp
+; I32-NEXT:    andl $-8, %esp
+; I32-NEXT:    subl $8, %esp
+; I32-NEXT:    movl 8(%ebp), %eax
+; I32-NEXT:    orl 12(%ebp), %eax
+; I32-NEXT:    je .LBB1_1
+; I32-NEXT:  # BB#2:
+; I32-NEXT:    leal 24(%ebp), %eax
+; I32-NEXT:    jmp .LBB1_3
+; I32-NEXT:  .LBB1_1:
+; I32-NEXT:    leal 16(%ebp), %eax
+; I32-NEXT:  .LBB1_3:
+; I32-NEXT:    movq (%eax), %mm0
+; I32-NEXT:    psllw %mm0, %mm0
+; I32-NEXT:    movq %mm0, (%esp)
+; I32-NEXT:    movl (%esp), %eax
+; I32-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; I32-NEXT:    movl %ebp, %esp
+; I32-NEXT:    popl %ebp
+; I32-NEXT:    retl
+  %cond = icmp eq i64 %arg, 0
+  %xmmx = bitcast i64 %x to x86_mmx
+  %ymmx = bitcast i64 %y to x86_mmx
+  %slct = select i1 %cond, x86_mmx %xmmx, x86_mmx %ymmx
+  %psll = tail call x86_mmx @llvm.x86.mmx.psll.w(x86_mmx %slct, x86_mmx %slct)
+  %retc = bitcast x86_mmx %psll to i64
+  ret i64 %retc
+}
+
+declare x86_mmx @llvm.x86.mmx.psll.w(x86_mmx, x86_mmx)
+