[x86] add tests for potentially miscompiling cvttp2si (PR37751); NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@334367 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/test/CodeGen/X86/avx-cvttp2si.ll b/test/CodeGen/X86/avx-cvttp2si.ll
new file mode 100644 (file)
index 0000000..ab8a0d9
--- /dev/null
+++ b/test/CodeGen/X86/avx-cvttp2si.ll
@@ -0,0 +1,64 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-- -mattr=avx              | FileCheck %s --check-prefixes=AVX,AVX1
+; RUN: llc < %s -mtriple=x86_64-- -mattr=avx512f,avx512vl | FileCheck %s --check-prefixes=AVX,AVX512
+
+; PR37551 - https://bugs.llvm.org/show_bug.cgi?id=37751
+; We can't combine into 'round' instructions because the behavior is different for out-of-range values.
+
+declare <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float>)
+declare <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double>)
+
+define <8 x float> @float_to_int_to_float_mem_v8f32(<8 x float>* %p) {
+; AVX1-LABEL: float_to_int_to_float_mem_v8f32:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vroundps $11, (%rdi), %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX512-LABEL: float_to_int_to_float_mem_v8f32:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vmovups (%rdi), %ymm0
+; AVX512-NEXT:    vroundps $11, %ymm0, %ymm0
+; AVX512-NEXT:    retq
+  %x = load <8 x float>, <8 x float>* %p, align 16
+  %fptosi = tail call <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float> %x)
+  %sitofp = sitofp <8 x i32> %fptosi to <8 x float>
+  ret <8 x float> %sitofp
+}
+
+define <8 x float> @float_to_int_to_float_reg_v8f32(<8 x float> %x) {
+; AVX-LABEL: float_to_int_to_float_reg_v8f32:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vroundps $11, %ymm0, %ymm0
+; AVX-NEXT:    retq
+  %fptosi = tail call <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float> %x)
+  %sitofp = sitofp <8 x i32> %fptosi to <8 x float>
+  ret <8 x float> %sitofp
+}
+
+define <4 x double> @float_to_int_to_float_mem_v4f64(<4 x double>* %p) {
+; AVX1-LABEL: float_to_int_to_float_mem_v4f64:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vroundpd $11, (%rdi), %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX512-LABEL: float_to_int_to_float_mem_v4f64:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vmovupd (%rdi), %ymm0
+; AVX512-NEXT:    vroundpd $11, %ymm0, %ymm0
+; AVX512-NEXT:    retq
+  %x = load <4 x double>, <4 x double>* %p, align 16
+  %fptosi = tail call <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double> %x)
+  %sitofp = sitofp <4 x i32> %fptosi to <4 x double>
+  ret <4 x double> %sitofp
+}
+
+define <4 x double> @float_to_int_to_float_reg_v4f64(<4 x double> %x) {
+; AVX-LABEL: float_to_int_to_float_reg_v4f64:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vroundpd $11, %ymm0, %ymm0
+; AVX-NEXT:    retq
+  %fptosi = tail call <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double> %x)
+  %sitofp = sitofp <4 x i32> %fptosi to <4 x double>
+  ret <4 x double> %sitofp
+}
+

diff --git a/test/CodeGen/X86/sse-cvttp2si.ll b/test/CodeGen/X86/sse-cvttp2si.ll
new file mode 100644 (file)
index 0000000..a8b52a7
--- /dev/null
+++ b/test/CodeGen/X86/sse-cvttp2si.ll
@@ -0,0 +1,85 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-- -mattr=sse4.1           | FileCheck %s --check-prefixes=SSE
+; RUN: llc < %s -mtriple=x86_64-- -mattr=avx              | FileCheck %s --check-prefixes=AVX,AVX1
+; RUN: llc < %s -mtriple=x86_64-- -mattr=avx512f,avx512vl | FileCheck %s --check-prefixes=AVX,AVX512
+
+; PR37551 - https://bugs.llvm.org/show_bug.cgi?id=37751
+; We can't combine into 'round' instructions because the behavior is different for out-of-range values.
+
+declare <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float>)
+declare <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double>)
+
+define <4 x float> @float_to_int_to_float_mem_v4f32(<4 x float>* %p) {
+; SSE-LABEL: float_to_int_to_float_mem_v4f32:
+; SSE:       # %bb.0:
+; SSE-NEXT:    roundps $11, (%rdi), %xmm0
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: float_to_int_to_float_mem_v4f32:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vroundps $11, (%rdi), %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX512-LABEL: float_to_int_to_float_mem_v4f32:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vmovaps (%rdi), %xmm0
+; AVX512-NEXT:    vroundps $11, %xmm0, %xmm0
+; AVX512-NEXT:    retq
+  %x = load <4 x float>, <4 x float>* %p, align 16
+  %fptosi = tail call <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float> %x)
+  %sitofp = sitofp <4 x i32> %fptosi to <4 x float>
+  ret <4 x float> %sitofp
+}
+
+define <4 x float> @float_to_int_to_float_reg_v4f32(<4 x float> %x) {
+; SSE-LABEL: float_to_int_to_float_reg_v4f32:
+; SSE:       # %bb.0:
+; SSE-NEXT:    roundps $11, %xmm0, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: float_to_int_to_float_reg_v4f32:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vroundps $11, %xmm0, %xmm0
+; AVX-NEXT:    retq
+  %fptosi = tail call <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float> %x)
+  %sitofp = sitofp <4 x i32> %fptosi to <4 x float>
+  ret <4 x float> %sitofp
+}
+
+define <2 x double> @float_to_int_to_float_mem_v2f64(<2 x double>* %p) {
+; SSE-LABEL: float_to_int_to_float_mem_v2f64:
+; SSE:       # %bb.0:
+; SSE-NEXT:    cvttpd2dq (%rdi), %xmm0
+; SSE-NEXT:    cvtdq2pd %xmm0, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: float_to_int_to_float_mem_v2f64:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vcvttpd2dqx (%rdi), %xmm0
+; AVX-NEXT:    vcvtdq2pd %xmm0, %xmm0
+; AVX-NEXT:    retq
+  %x = load <2 x double>, <2 x double>* %p, align 16
+  %fptosi = tail call <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double> %x)
+  %concat = shufflevector <4 x i32> %fptosi, <4 x i32> undef, <2 x i32> <i32 0, i32 1>
+  %sitofp = sitofp <2 x i32> %concat to <2 x double>
+  ret <2 x double> %sitofp
+}
+
+define <2 x double> @float_to_int_to_float_reg_v2f64(<2 x double> %x) {
+; SSE-LABEL: float_to_int_to_float_reg_v2f64:
+; SSE:       # %bb.0:
+; SSE-NEXT:    cvttpd2dq %xmm0, %xmm0
+; SSE-NEXT:    cvtdq2pd %xmm0, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: float_to_int_to_float_reg_v2f64:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vcvttpd2dq %xmm0, %xmm0
+; AVX-NEXT:    vcvtdq2pd %xmm0, %xmm0
+; AVX-NEXT:    retq
+  %fptosi = tail call <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double> %x)
+  %concat = shufflevector <4 x i32> %fptosi, <4 x i32> undef, <2 x i32> <i32 0, i32 1>
+  %sitofp = sitofp <2 x i32> %concat to <2 x double>
+  ret <2 x double> %sitofp
+}
+