[X86] Keep EXTRACT_VECTOR_ELT result type as f128 for Android x86_64.

Android x86_64 target uses f128 type and stores f128 values in %xmm* registers.
SoftenFloatRes_EXTRACT_VECTOR_ELT should not convert result value
from f128 to i128.

Differential Revision: http://reviews.llvm.org/D32102

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@300583 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp b/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp
index 6f2b1b94ce465069e8cbb2477be2c47055792171..c1cb5d9b5235e7cdc6da521e14890a004c51af6c 100644 (file)
--- a/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp
+++ b/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp
@@ -72,7 +72,7 @@ bool DAGTypeLegalizer::SoftenFloatResult(SDNode *N, unsigned ResNo) {
     case ISD::BUILD_PAIR:  R = SoftenFloatRes_BUILD_PAIR(N); break;
     case ISD::ConstantFP:  R = SoftenFloatRes_ConstantFP(N, ResNo); break;
     case ISD::EXTRACT_VECTOR_ELT:
-      R = SoftenFloatRes_EXTRACT_VECTOR_ELT(N); break;
+      R = SoftenFloatRes_EXTRACT_VECTOR_ELT(N, ResNo); break;
     case ISD::FABS:        R = SoftenFloatRes_FABS(N, ResNo); break;
     case ISD::FMINNUM:     R = SoftenFloatRes_FMINNUM(N); break;
     case ISD::FMAXNUM:     R = SoftenFloatRes_FMAXNUM(N); break;
@@ -171,7 +171,10 @@ SDValue DAGTypeLegalizer::SoftenFloatRes_ConstantFP(SDNode *N, unsigned ResNo) {
   }
 }
 
-SDValue DAGTypeLegalizer::SoftenFloatRes_EXTRACT_VECTOR_ELT(SDNode *N) {
+SDValue DAGTypeLegalizer::SoftenFloatRes_EXTRACT_VECTOR_ELT(SDNode *N, unsigned ResNo) {
+  // When LegalInHWReg, keep the extracted value in register.
+  if (isLegalInHWReg(N->getValueType(ResNo)))
+    return SDValue(N, ResNo);
   SDValue NewOp = BitConvertVectorToIntegerVector(N->getOperand(0));
   return DAG.getNode(ISD::EXTRACT_VECTOR_ELT, SDLoc(N),
                      NewOp.getValueType().getVectorElementType(),

diff --git a/lib/CodeGen/SelectionDAG/LegalizeTypes.h b/lib/CodeGen/SelectionDAG/LegalizeTypes.h
index 80c939700518f7dbee3262fbad87580afee45627..af55a22972a6106c2f0b6b9906d8d3d5635c1ceb 100644 (file)
--- a/lib/CodeGen/SelectionDAG/LegalizeTypes.h
+++ b/lib/CodeGen/SelectionDAG/LegalizeTypes.h
@@ -428,7 +428,7 @@ private:
   SDValue SoftenFloatRes_BITCAST(SDNode *N, unsigned ResNo);
   SDValue SoftenFloatRes_BUILD_PAIR(SDNode *N);
   SDValue SoftenFloatRes_ConstantFP(SDNode *N, unsigned ResNo);
-  SDValue SoftenFloatRes_EXTRACT_VECTOR_ELT(SDNode *N);
+  SDValue SoftenFloatRes_EXTRACT_VECTOR_ELT(SDNode *N, unsigned ResNo);
   SDValue SoftenFloatRes_FABS(SDNode *N, unsigned ResNo);
   SDValue SoftenFloatRes_FMINNUM(SDNode *N);
   SDValue SoftenFloatRes_FMAXNUM(SDNode *N);

diff --git a/test/CodeGen/X86/extract-store.ll b/test/CodeGen/X86/extract-store.ll
index 1751f03731d3aaa21b2d42e64cf0bfad359744a4..5286a1b635d1ad57fe94b24122638f9fb0b66e75 100644 (file)
--- a/test/CodeGen/X86/extract-store.ll
+++ b/test/CodeGen/X86/extract-store.ll
@@ -5,6 +5,10 @@
 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=X64 --check-prefix=SSE-X64 --check-prefix=SSE41-X64
 ; RUN: llc < %s -mtriple=i686-unknown -mattr=+avx      | FileCheck %s --check-prefix=X32 --check-prefix=AVX-X32
 ; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx    | FileCheck %s --check-prefix=X64 --check-prefix=AVX-X64
+; RUN: llc < %s -O2 -mtriple=x86_64-linux-android -mattr=+mmx -enable-legalize-types-checking \
+; RUN:   | FileCheck %s --check-prefix=X64 --check-prefix=SSE-X64 --check-prefix=SSE-F128
+; RUN: llc < %s -O2 -mtriple=x86_64-linux-gnu -mattr=+mmx -enable-legalize-types-checking \
+; RUN:   | FileCheck %s --check-prefix=X64 --check-prefix=SSE-X64 --check-prefix=SSE-F128
 
 define void @extract_i8_0(i8* nocapture %dst, <16 x i8> %foo) nounwind {
 ; SSE2-X32-LABEL: extract_i8_0:
@@ -458,6 +462,26 @@ define void @extract_f64_1(double* nocapture %dst, <2 x double> %foo) nounwind {
   ret void
 }
 
+define void @extract_f128_0(fp128* nocapture %dst, <2 x fp128> %foo) nounwind {
+; SSE-F128-LABEL: extract_f128_0:
+; SSE-F128:       # BB#0:
+; SSE-F128-NEXT:    movaps     %xmm0, (%rdi)
+; SSE-F128-NEXT:    retq
+  %vecext = extractelement <2 x fp128> %foo, i32 0
+  store fp128 %vecext, fp128* %dst, align 1
+  ret void
+}
+
+define void @extract_f128_1(fp128* nocapture %dst, <2 x fp128> %foo) nounwind {
+; SSE-F128-LABEL: extract_f128_1:
+; SSE-F128:       # BB#0:
+; SSE-F128-NEXT:    movaps     %xmm1, (%rdi)
+; SSE-F128-NEXT:    retq
+  %vecext = extractelement <2 x fp128> %foo, i32 1
+  store fp128 %vecext, fp128* %dst, align 1
+  ret void
+}
+
 define void @extract_i8_undef(i8* nocapture %dst, <16 x i8> %foo) nounwind {
 ; X32-LABEL: extract_i8_undef:
 ; X32:       # BB#0:
@@ -535,3 +559,16 @@ define void @extract_f64_undef(double* nocapture %dst, <2 x double> %foo) nounwi
   store double %vecext, double* %dst, align 1
   ret void
 }
+
+define void @extract_f128_undef(fp128* nocapture %dst, <2 x fp128> %foo) nounwind {
+; X32-LABEL: extract_f128_undef:
+; X32:       # BB#0:
+; X32-NEXT:    retl
+;
+; X64-LABEL: extract_f128_undef:
+; X64:       # BB#0:
+; X64-NEXT:    retq
+  %vecext = extractelement <2 x fp128> %foo, i32 2 ; undef
+  store fp128 %vecext, fp128* %dst, align 1
+  ret void
+}

diff --git a/test/CodeGen/X86/fp128-extract.ll b/test/CodeGen/X86/fp128-extract.ll
new file mode 100644 (file)
index 0000000..5006ac8
--- /dev/null
+++ b/test/CodeGen/X86/fp128-extract.ll
@@ -0,0 +1,22 @@
+; RUN: llc < %s -O2 -mtriple=x86_64-linux-android -mattr=+mmx \
+; RUN:     -enable-legalize-types-checking | FileCheck %s
+; RUN: llc < %s -O2 -mtriple=x86_64-linux-gnu -mattr=+mmx \
+; RUN:     -enable-legalize-types-checking | FileCheck %s
+
+; Test the softened result of extractelement op code.
+define fp128 @TestExtract(<2 x double> %x) {
+entry:
+  ; Simplified instruction pattern from the output of llvm before r289042,
+  ; for a boost function ...::insert<...>::traverse<...>().
+  %a = fpext <2 x double> %x to <2 x fp128>
+  %0 = extractelement <2 x fp128> %a, i32 0
+  %1 = extractelement <2 x fp128> %a, i32 1
+  %2 = fmul fp128 %0, %1
+  ret fp128 %2
+; CHECK-LABEL: TestExtract:
+; CHECK:       movaps  %xmm0, (%rsp)
+; CHECK:       callq   __extenddftf2
+; CHECK:       callq   __extenddftf2
+; CHECK:       callq    __multf3
+; CHECK:       retq
+}