From ac7f3e24d3990962f7d1036e7ff2d29e17e9ef16 Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Sat, 29 Apr 2017 14:29:06 +0000
Subject: [PATCH] [X86][SSE] Add initial <2 x half> tests for PR31088

As discussed on D32391, test X86/X64 SSE2 and X64 F16C.

llvm-svn: 301744
---
 llvm/test/CodeGen/X86/pr31088.ll | 105 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 105 insertions(+)
 create mode 100644 llvm/test/CodeGen/X86/pr31088.ll

diff --git a/llvm/test/CodeGen/X86/pr31088.ll b/llvm/test/CodeGen/X86/pr31088.ll
new file mode 100644
index 0000000..ca1e08b
--- /dev/null
+++ b/llvm/test/CodeGen/X86/pr31088.ll
@@ -0,0 +1,105 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X86
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X64
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+f16c | FileCheck %s --check-prefix=F16C
+
+define <2 x half> @ir_fadd_v2f16(<2 x half> %arg0, <2 x half> %arg1) nounwind {
+; X86-LABEL: ir_fadd_v2f16:
+; X86:       # BB#0:
+; X86-NEXT:    subl $64, %esp
+; X86-NEXT:    movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; X86-NEXT:    movss %xmm0, (%esp)
+; X86-NEXT:    calll __gnu_f2h_ieee
+; X86-NEXT:    movzwl %ax, %eax
+; X86-NEXT:    movl %eax, (%esp)
+; X86-NEXT:    calll __gnu_h2f_ieee
+; X86-NEXT:    fstpt {{[0-9]+}}(%esp) # 10-byte Folded Spill
+; X86-NEXT:    movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; X86-NEXT:    movss %xmm0, (%esp)
+; X86-NEXT:    calll __gnu_f2h_ieee
+; X86-NEXT:    movzwl %ax, %eax
+; X86-NEXT:    movl %eax, (%esp)
+; X86-NEXT:    calll __gnu_h2f_ieee
+; X86-NEXT:    fstpt {{[0-9]+}}(%esp) # 10-byte Folded Spill
+; X86-NEXT:    movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; X86-NEXT:    movss %xmm0, (%esp)
+; X86-NEXT:    calll __gnu_f2h_ieee
+; X86-NEXT:    movzwl %ax, %eax
+; X86-NEXT:    movl %eax, (%esp)
+; X86-NEXT:    calll __gnu_h2f_ieee
+; X86-NEXT:    fstpt {{[0-9]+}}(%esp) # 10-byte Folded Spill
+; X86-NEXT:    movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; X86-NEXT:    movss %xmm0, (%esp)
+; X86-NEXT:    calll __gnu_f2h_ieee
+; X86-NEXT:    movzwl %ax, %eax
+; X86-NEXT:    movl %eax, (%esp)
+; X86-NEXT:    fldt {{[0-9]+}}(%esp) # 10-byte Folded Reload
+; X86-NEXT:    fstps {{[0-9]+}}(%esp)
+; X86-NEXT:    fldt {{[0-9]+}}(%esp) # 10-byte Folded Reload
+; X86-NEXT:    fstps {{[0-9]+}}(%esp)
+; X86-NEXT:    fldt {{[0-9]+}}(%esp) # 10-byte Folded Reload
+; X86-NEXT:    fstps {{[0-9]+}}(%esp)
+; X86-NEXT:    calll __gnu_h2f_ieee
+; X86-NEXT:    fstps {{[0-9]+}}(%esp)
+; X86-NEXT:    movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
+; X86-NEXT:    movss {{.*#+}} xmm1 = mem[0],zero,zero,zero
+; X86-NEXT:    addss {{[0-9]+}}(%esp), %xmm1
+; X86-NEXT:    addss {{[0-9]+}}(%esp), %xmm0
+; X86-NEXT:    movss %xmm0, {{[0-9]+}}(%esp)
+; X86-NEXT:    movss %xmm1, {{[0-9]+}}(%esp)
+; X86-NEXT:    flds {{[0-9]+}}(%esp)
+; X86-NEXT:    flds {{[0-9]+}}(%esp)
+; X86-NEXT:    addl $64, %esp
+; X86-NEXT:    retl
+;
+; X64-LABEL: ir_fadd_v2f16:
+; X64:       # BB#0:
+; X64-NEXT:    subq $24, %rsp
+; X64-NEXT:    movss %xmm2, {{[0-9]+}}(%rsp) # 4-byte Spill
+; X64-NEXT:    movss %xmm1, {{[0-9]+}}(%rsp) # 4-byte Spill
+; X64-NEXT:    movss %xmm0, {{[0-9]+}}(%rsp) # 4-byte Spill
+; X64-NEXT:    movaps %xmm3, %xmm0
+; X64-NEXT:    callq __gnu_f2h_ieee
+; X64-NEXT:    movzwl %ax, %edi
+; X64-NEXT:    callq __gnu_h2f_ieee
+; X64-NEXT:    movss %xmm0, {{[0-9]+}}(%rsp) # 4-byte Spill
+; X64-NEXT:    movss {{[0-9]+}}(%rsp), %xmm0 # 4-byte Reload
+; X64-NEXT:    # xmm0 = mem[0],zero,zero,zero
+; X64-NEXT:    callq __gnu_f2h_ieee
+; X64-NEXT:    movzwl %ax, %edi
+; X64-NEXT:    callq __gnu_h2f_ieee
+; X64-NEXT:    movss %xmm0, {{[0-9]+}}(%rsp) # 4-byte Spill
+; X64-NEXT:    movss {{[0-9]+}}(%rsp), %xmm0 # 4-byte Reload
+; X64-NEXT:    # xmm0 = mem[0],zero,zero,zero
+; X64-NEXT:    callq __gnu_f2h_ieee
+; X64-NEXT:    movzwl %ax, %edi
+; X64-NEXT:    callq __gnu_h2f_ieee
+; X64-NEXT:    movss %xmm0, {{[0-9]+}}(%rsp) # 4-byte Spill
+; X64-NEXT:    movss {{[0-9]+}}(%rsp), %xmm0 # 4-byte Reload
+; X64-NEXT:    # xmm0 = mem[0],zero,zero,zero
+; X64-NEXT:    callq __gnu_f2h_ieee
+; X64-NEXT:    movzwl %ax, %edi
+; X64-NEXT:    callq __gnu_h2f_ieee
+; X64-NEXT:    addss {{[0-9]+}}(%rsp), %xmm0 # 4-byte Folded Reload
+; X64-NEXT:    movss {{[0-9]+}}(%rsp), %xmm1 # 4-byte Reload
+; X64-NEXT:    # xmm1 = mem[0],zero,zero,zero
+; X64-NEXT:    addss {{[0-9]+}}(%rsp), %xmm1 # 4-byte Folded Reload
+; X64-NEXT:    addq $24, %rsp
+; X64-NEXT:    retq
+;
+; F16C-LABEL: ir_fadd_v2f16:
+; F16C:       # BB#0:
+; F16C-NEXT:    vcvtps2ph $4, %xmm3, %xmm3
+; F16C-NEXT:    vcvtph2ps %xmm3, %xmm3
+; F16C-NEXT:    vcvtps2ph $4, %xmm1, %xmm1
+; F16C-NEXT:    vcvtph2ps %xmm1, %xmm1
+; F16C-NEXT:    vcvtps2ph $4, %xmm2, %xmm2
+; F16C-NEXT:    vcvtph2ps %xmm2, %xmm2
+; F16C-NEXT:    vcvtps2ph $4, %xmm0, %xmm0
+; F16C-NEXT:    vcvtph2ps %xmm0, %xmm0
+; F16C-NEXT:    vaddss %xmm2, %xmm0, %xmm0
+; F16C-NEXT:    vaddss %xmm3, %xmm1, %xmm1
+; F16C-NEXT:    retq
+  %retval = fadd <2 x half> %arg0, %arg1
+  ret <2 x half> %retval
+}
-- 
2.7.4