From 2f85bd146425cf16c321e023ad0331a30d0e59aa Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Tue, 2 Jul 2019 18:20:38 +0000
Subject: [PATCH] [X86] getTargetConstantBitsFromNode - remove unnecessary
 getZExtValue() (PR42486)

Don't use APInt::getZExtValue() if you can avoid it - eventually someone will call it with i128 or something that doesn't fit into 64-bits.

In this case it was completely superfluous as we'd moved the rest of the code to always use APInt.

Fixes the <1 x i128> addition bug in PR42486

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@364953 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/X86/X86ISelLowering.cpp |  3 +-
 test/CodeGen/X86/i128-add.ll       | 88 ++++++++++++++++++++++++++++++
 2 files changed, 89 insertions(+), 2 deletions(-)
 create mode 100644 test/CodeGen/X86/i128-add.ll

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 75fdd0c6f6e..33042e3900c 100644
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -5873,8 +5873,7 @@ static bool getTargetConstantBitsFromNode(SDValue Op, unsigned EltSizeInBits,
       if (UndefEltBits.getBoolValue() && !AllowPartialUndefs)
         return false;
 
-      APInt Bits = MaskBits.extractBits(EltSizeInBits, BitOffset);
-      EltBits[i] = Bits.getZExtValue();
+      EltBits[i] = MaskBits.extractBits(EltSizeInBits, BitOffset);
     }
     return true;
   };
diff --git a/test/CodeGen/X86/i128-add.ll b/test/CodeGen/X86/i128-add.ll
new file mode 100644
index 00000000000..182ada4c458
--- /dev/null
+++ b/test/CodeGen/X86/i128-add.ll
@@ -0,0 +1,88 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=X86
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=X64
+
+define i128 @add_i128(i128 %x, i128 %y) nounwind {
+; X86-LABEL: add_i128:
+; X86:       # %bb.0:
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    addl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    adcl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    adcl {{[0-9]+}}(%esp), %edx
+; X86-NEXT:    adcl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    addl $1, %esi
+; X86-NEXT:    adcl $0, %edi
+; X86-NEXT:    adcl $0, %edx
+; X86-NEXT:    adcl $0, %ecx
+; X86-NEXT:    movl %esi, (%eax)
+; X86-NEXT:    movl %edi, 4(%eax)
+; X86-NEXT:    movl %edx, 8(%eax)
+; X86-NEXT:    movl %ecx, 12(%eax)
+; X86-NEXT:    popl %esi
+; X86-NEXT:    popl %edi
+; X86-NEXT:    retl $4
+;
+; X64-LABEL: add_i128:
+; X64:       # %bb.0:
+; X64-NEXT:    movq %rdi, %rax
+; X64-NEXT:    addq %rdx, %rax
+; X64-NEXT:    adcq %rcx, %rsi
+; X64-NEXT:    addq $1, %rax
+; X64-NEXT:    adcq $0, %rsi
+; X64-NEXT:    movq %rsi, %rdx
+; X64-NEXT:    retq
+  %t0 = add i128 %x, 1
+  %t1 = add i128 %y, %t0
+  ret i128 %t1
+}
+
+; PR42486
+define <1 x i128> @add_v1i128(<1 x i128> %x, <1 x i128> %y) nounwind {
+; X86-LABEL: add_v1i128:
+; X86:       # %bb.0:
+; X86-NEXT:    pushl %edi
+; X86-NEXT:    pushl %esi
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    addl {{[0-9]+}}(%esp), %esi
+; X86-NEXT:    adcl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    adcl {{[0-9]+}}(%esp), %edx
+; X86-NEXT:    adcl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    subl $-1, %esi
+; X86-NEXT:    sbbl $-1, %edi
+; X86-NEXT:    sbbl $-1, %edx
+; X86-NEXT:    sbbl $-1, %ecx
+; X86-NEXT:    movl %esi, (%eax)
+; X86-NEXT:    movl %edi, 4(%eax)
+; X86-NEXT:    movl %edx, 8(%eax)
+; X86-NEXT:    movl %ecx, 12(%eax)
+; X86-NEXT:    popl %esi
+; X86-NEXT:    popl %edi
+; X86-NEXT:    retl $4
+;
+; X64-LABEL: add_v1i128:
+; X64:       # %bb.0:
+; X64-NEXT:    movq %rdi, %rax
+; X64-NEXT:    addq %rdx, %rax
+; X64-NEXT:    adcq %rcx, %rsi
+; X64-NEXT:    movq %rax, %xmm0
+; X64-NEXT:    movq %rsi, %xmm1
+; X64-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; X64-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; X64-NEXT:    movq %xmm0, %rdx
+; X64-NEXT:    addq $1, %rax
+; X64-NEXT:    adcq $0, %rdx
+; X64-NEXT:    retq
+  %t0 = add <1 x i128> %x, <i128 1>
+  %t1 = add <1 x i128> %y, %t0
+  ret <1 x i128> %t1
+}
-- 
2.50.1