[llvm] r298045 - [X86] Use update_llc_test_checks.py to regenerate a test and add command lines to demonstrate that we don't pick EVEX encoded instruction when AVX512 and FMA3 are both enabled.
Craig Topper via llvm-commits
llvm-commits at lists.llvm.org
Thu Mar 16 23:00:01 PDT 2017
Author: ctopper
Date: Fri Mar 17 01:00:01 2017
New Revision: 298045
URL: http://llvm.org/viewvc/llvm-project?rev=298045&view=rev
Log:
[X86] Use update_llc_test_checks.py to regenerate a test and add command lines to demonstrate that we don't pick EVEX encoded instruction when AVX512 and FMA3 are both enabled.
This bug only exists on the scalar llvm.fma instrinsics. Looks like we don't test the llvm.fma intrinsics very thoroughly. In fact I don't see any tests for the vector versions.
Modified:
llvm/trunk/test/CodeGen/X86/fma.ll
Modified: llvm/trunk/test/CodeGen/X86/fma.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/fma.ll?rev=298045&r1=298044&r2=298045&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/fma.ll (original)
+++ llvm/trunk/test/CodeGen/X86/fma.ll Fri Mar 17 01:00:01 2017
@@ -1,40 +1,232 @@
-; RUN: llc < %s -mtriple=i386-apple-darwin10 -mattr=+fma,-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-INST
-; RUN: llc < %s -mtriple=i386-apple-darwin10 -mattr=-fma,-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-CALL
-; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=+fma,-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-INST
-; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=-fma,-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-CALL
-; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=+avx512f,-fma,-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-INST
-; RUN: llc < %s -march=x86 -mcpu=bdver2 -mattr=-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-INST
-; RUN: llc < %s -march=x86 -mcpu=bdver2 -mattr=-fma,-fma4 | FileCheck %s --check-prefix=CHECK --check-prefix=CHECK-FMA-CALL
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=i386-apple-darwin10 -mattr=+sse2,+fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=FMA32
+; RUN: llc < %s -mtriple=i386-apple-darwin10 -mattr=+sse2,-fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=FMACALL32
+; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=+fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=FMA64
+; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=-fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=FMACALL64
+; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=+avx512f,-fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=AVX51264NOFMA
+; RUN: llc < %s -mtriple=x86_64-apple-darwin10 -mattr=+avx512f,fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=AVX51264
+; RUN: llc < %s -mtriple=i386-apple-darwin10 -mcpu=bdver2 -mattr=-fma4 -show-mc-encoding | FileCheck %s --check-prefix=FMA32
+; RUN: llc < %s -mtriple=i386-apple-darwin10 -mcpu=bdver2 -mattr=-fma,-fma4 -show-mc-encoding | FileCheck %s --check-prefix=FMACALL32
-; CHECK-LABEL: test_f32:
-; CHECK-FMA-INST: vfmadd213ss
-; CHECK-FMA-CALL: fmaf
define float @test_f32(float %a, float %b, float %c) #0 {
+; FMA32-LABEL: test_f32:
+; FMA32: ## BB#0: ## %entry
+; FMA32-NEXT: pushl %eax ## encoding: [0x50]
+; FMA32-NEXT: vmovss {{[0-9]+}}(%esp), %xmm0 ## encoding: [0xc5,0xfa,0x10,0x44,0x24,0x08]
+; FMA32-NEXT: ## xmm0 = mem[0],zero,zero,zero
+; FMA32-NEXT: vmovss {{[0-9]+}}(%esp), %xmm1 ## encoding: [0xc5,0xfa,0x10,0x4c,0x24,0x0c]
+; FMA32-NEXT: ## xmm1 = mem[0],zero,zero,zero
+; FMA32-NEXT: vfmadd213ss {{[0-9]+}}(%esp), %xmm0, %xmm1 ## encoding: [0xc4,0xe2,0x79,0xa9,0x4c,0x24,0x10]
+; FMA32-NEXT: vmovss %xmm1, (%esp) ## encoding: [0xc5,0xfa,0x11,0x0c,0x24]
+; FMA32-NEXT: flds (%esp) ## encoding: [0xd9,0x04,0x24]
+; FMA32-NEXT: popl %eax ## encoding: [0x58]
+; FMA32-NEXT: retl ## encoding: [0xc3]
+;
+; FMACALL32-LABEL: test_f32:
+; FMACALL32: ## BB#0: ## %entry
+; FMACALL32-NEXT: jmp _fmaf ## TAILCALL
+; FMACALL32-NEXT: ## encoding: [0xeb,A]
+; FMACALL32-NEXT: ## fixup A - offset: 1, value: _fmaf-1, kind: FK_PCRel_1
+;
+; FMA64-LABEL: test_f32:
+; FMA64: ## BB#0: ## %entry
+; FMA64-NEXT: vfmadd213ss %xmm2, %xmm1, %xmm0 ## encoding: [0xc4,0xe2,0x71,0xa9,0xc2]
+; FMA64-NEXT: retq ## encoding: [0xc3]
+;
+; FMACALL64-LABEL: test_f32:
+; FMACALL64: ## BB#0: ## %entry
+; FMACALL64-NEXT: jmp _fmaf ## TAILCALL
+; FMACALL64-NEXT: ## encoding: [0xeb,A]
+; FMACALL64-NEXT: ## fixup A - offset: 1, value: _fmaf-1, kind: FK_PCRel_1
+;
+; AVX51264NOFMA-LABEL: test_f32:
+; AVX51264NOFMA: ## BB#0: ## %entry
+; AVX51264NOFMA-NEXT: vfmadd213ss %xmm2, %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc4,0xe2,0x71,0xa9,0xc2]
+; AVX51264NOFMA-NEXT: retq ## encoding: [0xc3]
+;
+; AVX51264-LABEL: test_f32:
+; AVX51264: ## BB#0: ## %entry
+; AVX51264-NEXT: vfmadd213ss %xmm2, %xmm1, %xmm0 ## encoding: [0xc4,0xe2,0x71,0xa9,0xc2]
+; AVX51264-NEXT: retq ## encoding: [0xc3]
entry:
%call = call float @llvm.fma.f32(float %a, float %b, float %c)
ret float %call
}
-; CHECK-LABEL: test_f64:
-; CHECK-FMA-INST: vfmadd213sd
-; CHECK-FMA-CALL: fma
define double @test_f64(double %a, double %b, double %c) #0 {
+; FMA32-LABEL: test_f64:
+; FMA32: ## BB#0: ## %entry
+; FMA32-NEXT: subl $12, %esp ## encoding: [0x83,0xec,0x0c]
+; FMA32-NEXT: vmovsd {{[0-9]+}}(%esp), %xmm0 ## encoding: [0xc5,0xfb,0x10,0x44,0x24,0x10]
+; FMA32-NEXT: ## xmm0 = mem[0],zero
+; FMA32-NEXT: vmovsd {{[0-9]+}}(%esp), %xmm1 ## encoding: [0xc5,0xfb,0x10,0x4c,0x24,0x18]
+; FMA32-NEXT: ## xmm1 = mem[0],zero
+; FMA32-NEXT: vfmadd213sd {{[0-9]+}}(%esp), %xmm0, %xmm1 ## encoding: [0xc4,0xe2,0xf9,0xa9,0x4c,0x24,0x20]
+; FMA32-NEXT: vmovsd %xmm1, (%esp) ## encoding: [0xc5,0xfb,0x11,0x0c,0x24]
+; FMA32-NEXT: fldl (%esp) ## encoding: [0xdd,0x04,0x24]
+; FMA32-NEXT: addl $12, %esp ## encoding: [0x83,0xc4,0x0c]
+; FMA32-NEXT: retl ## encoding: [0xc3]
+;
+; FMACALL32-LABEL: test_f64:
+; FMACALL32: ## BB#0: ## %entry
+; FMACALL32-NEXT: jmp _fma ## TAILCALL
+; FMACALL32-NEXT: ## encoding: [0xeb,A]
+; FMACALL32-NEXT: ## fixup A - offset: 1, value: _fma-1, kind: FK_PCRel_1
+;
+; FMA64-LABEL: test_f64:
+; FMA64: ## BB#0: ## %entry
+; FMA64-NEXT: vfmadd213sd %xmm2, %xmm1, %xmm0 ## encoding: [0xc4,0xe2,0xf1,0xa9,0xc2]
+; FMA64-NEXT: retq ## encoding: [0xc3]
+;
+; FMACALL64-LABEL: test_f64:
+; FMACALL64: ## BB#0: ## %entry
+; FMACALL64-NEXT: jmp _fma ## TAILCALL
+; FMACALL64-NEXT: ## encoding: [0xeb,A]
+; FMACALL64-NEXT: ## fixup A - offset: 1, value: _fma-1, kind: FK_PCRel_1
+;
+; AVX51264NOFMA-LABEL: test_f64:
+; AVX51264NOFMA: ## BB#0: ## %entry
+; AVX51264NOFMA-NEXT: vfmadd213sd %xmm2, %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc4,0xe2,0xf1,0xa9,0xc2]
+; AVX51264NOFMA-NEXT: retq ## encoding: [0xc3]
+;
+; AVX51264-LABEL: test_f64:
+; AVX51264: ## BB#0: ## %entry
+; AVX51264-NEXT: vfmadd213sd %xmm2, %xmm1, %xmm0 ## encoding: [0xc4,0xe2,0xf1,0xa9,0xc2]
+; AVX51264-NEXT: retq ## encoding: [0xc3]
entry:
%call = call double @llvm.fma.f64(double %a, double %b, double %c)
ret double %call
}
-; CHECK-LABEL: test_f80:
-; CHECK: fmal
define x86_fp80 @test_f80(x86_fp80 %a, x86_fp80 %b, x86_fp80 %c) #0 {
+; FMA32-LABEL: test_f80:
+; FMA32: ## BB#0: ## %entry
+; FMA32-NEXT: subl $60, %esp ## encoding: [0x83,0xec,0x3c]
+; FMA32-NEXT: fldt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x6c,0x24,0x40]
+; FMA32-NEXT: fldt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x6c,0x24,0x50]
+; FMA32-NEXT: fldt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x6c,0x24,0x60]
+; FMA32-NEXT: fstpt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x7c,0x24,0x20]
+; FMA32-NEXT: fstpt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x7c,0x24,0x10]
+; FMA32-NEXT: fstpt (%esp) ## encoding: [0xdb,0x3c,0x24]
+; FMA32-NEXT: calll _fmal ## encoding: [0xe8,A,A,A,A]
+; FMA32-NEXT: ## fixup A - offset: 1, value: _fmal-4, kind: FK_PCRel_4
+; FMA32-NEXT: addl $60, %esp ## encoding: [0x83,0xc4,0x3c]
+; FMA32-NEXT: retl ## encoding: [0xc3]
+;
+; FMACALL32-LABEL: test_f80:
+; FMACALL32: ## BB#0: ## %entry
+; FMACALL32-NEXT: subl $60, %esp ## encoding: [0x83,0xec,0x3c]
+; FMACALL32-NEXT: fldt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x6c,0x24,0x40]
+; FMACALL32-NEXT: fldt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x6c,0x24,0x50]
+; FMACALL32-NEXT: fldt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x6c,0x24,0x60]
+; FMACALL32-NEXT: fstpt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x7c,0x24,0x20]
+; FMACALL32-NEXT: fstpt {{[0-9]+}}(%esp) ## encoding: [0xdb,0x7c,0x24,0x10]
+; FMACALL32-NEXT: fstpt (%esp) ## encoding: [0xdb,0x3c,0x24]
+; FMACALL32-NEXT: calll _fmal ## encoding: [0xe8,A,A,A,A]
+; FMACALL32-NEXT: ## fixup A - offset: 1, value: _fmal-4, kind: FK_PCRel_4
+; FMACALL32-NEXT: addl $60, %esp ## encoding: [0x83,0xc4,0x3c]
+; FMACALL32-NEXT: retl ## encoding: [0xc3]
+;
+; FMA64-LABEL: test_f80:
+; FMA64: ## BB#0: ## %entry
+; FMA64-NEXT: subq $56, %rsp ## encoding: [0x48,0x83,0xec,0x38]
+; FMA64-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x40]
+; FMA64-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x50]
+; FMA64-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x60]
+; FMA64-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x20]
+; FMA64-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x10]
+; FMA64-NEXT: fstpt (%rsp) ## encoding: [0xdb,0x3c,0x24]
+; FMA64-NEXT: callq _fmal ## encoding: [0xe8,A,A,A,A]
+; FMA64-NEXT: ## fixup A - offset: 1, value: _fmal-4, kind: FK_PCRel_4
+; FMA64-NEXT: addq $56, %rsp ## encoding: [0x48,0x83,0xc4,0x38]
+; FMA64-NEXT: retq ## encoding: [0xc3]
+;
+; FMACALL64-LABEL: test_f80:
+; FMACALL64: ## BB#0: ## %entry
+; FMACALL64-NEXT: subq $56, %rsp ## encoding: [0x48,0x83,0xec,0x38]
+; FMACALL64-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x40]
+; FMACALL64-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x50]
+; FMACALL64-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x60]
+; FMACALL64-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x20]
+; FMACALL64-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x10]
+; FMACALL64-NEXT: fstpt (%rsp) ## encoding: [0xdb,0x3c,0x24]
+; FMACALL64-NEXT: callq _fmal ## encoding: [0xe8,A,A,A,A]
+; FMACALL64-NEXT: ## fixup A - offset: 1, value: _fmal-4, kind: FK_PCRel_4
+; FMACALL64-NEXT: addq $56, %rsp ## encoding: [0x48,0x83,0xc4,0x38]
+; FMACALL64-NEXT: retq ## encoding: [0xc3]
+;
+; AVX51264NOFMA-LABEL: test_f80:
+; AVX51264NOFMA: ## BB#0: ## %entry
+; AVX51264NOFMA-NEXT: subq $56, %rsp ## encoding: [0x48,0x83,0xec,0x38]
+; AVX51264NOFMA-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x40]
+; AVX51264NOFMA-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x50]
+; AVX51264NOFMA-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x60]
+; AVX51264NOFMA-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x20]
+; AVX51264NOFMA-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x10]
+; AVX51264NOFMA-NEXT: fstpt (%rsp) ## encoding: [0xdb,0x3c,0x24]
+; AVX51264NOFMA-NEXT: callq _fmal ## encoding: [0xe8,A,A,A,A]
+; AVX51264NOFMA-NEXT: ## fixup A - offset: 1, value: _fmal-4, kind: FK_PCRel_4
+; AVX51264NOFMA-NEXT: addq $56, %rsp ## encoding: [0x48,0x83,0xc4,0x38]
+; AVX51264NOFMA-NEXT: retq ## encoding: [0xc3]
+;
+; AVX51264-LABEL: test_f80:
+; AVX51264: ## BB#0: ## %entry
+; AVX51264-NEXT: subq $56, %rsp ## encoding: [0x48,0x83,0xec,0x38]
+; AVX51264-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x40]
+; AVX51264-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x50]
+; AVX51264-NEXT: fldt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x6c,0x24,0x60]
+; AVX51264-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x20]
+; AVX51264-NEXT: fstpt {{[0-9]+}}(%rsp) ## encoding: [0xdb,0x7c,0x24,0x10]
+; AVX51264-NEXT: fstpt (%rsp) ## encoding: [0xdb,0x3c,0x24]
+; AVX51264-NEXT: callq _fmal ## encoding: [0xe8,A,A,A,A]
+; AVX51264-NEXT: ## fixup A - offset: 1, value: _fmal-4, kind: FK_PCRel_4
+; AVX51264-NEXT: addq $56, %rsp ## encoding: [0x48,0x83,0xc4,0x38]
+; AVX51264-NEXT: retq ## encoding: [0xc3]
entry:
%call = call x86_fp80 @llvm.fma.f80(x86_fp80 %a, x86_fp80 %b, x86_fp80 %c)
ret x86_fp80 %call
}
-; CHECK-LABEL: test_f32_cst:
-; CHECK-NOT: vfmadd
define float @test_f32_cst() #0 {
+; FMA32-LABEL: test_f32_cst:
+; FMA32: ## BB#0: ## %entry
+; FMA32-NEXT: flds LCPI3_0 ## encoding: [0xd9,0x05,A,A,A,A]
+; FMA32-NEXT: ## fixup A - offset: 2, value: LCPI3_0, kind: FK_Data_4
+; FMA32-NEXT: retl ## encoding: [0xc3]
+;
+; FMACALL32-LABEL: test_f32_cst:
+; FMACALL32: ## BB#0: ## %entry
+; FMACALL32-NEXT: flds LCPI3_0 ## encoding: [0xd9,0x05,A,A,A,A]
+; FMACALL32-NEXT: ## fixup A - offset: 2, value: LCPI3_0, kind: FK_Data_4
+; FMACALL32-NEXT: retl ## encoding: [0xc3]
+;
+; FMA64-LABEL: test_f32_cst:
+; FMA64: ## BB#0: ## %entry
+; FMA64-NEXT: vmovss {{.*}}(%rip), %xmm0 ## encoding: [0xc5,0xfa,0x10,0x05,A,A,A,A]
+; FMA64-NEXT: ## fixup A - offset: 4, value: LCPI3_0-4, kind: reloc_riprel_4byte
+; FMA64-NEXT: ## xmm0 = mem[0],zero,zero,zero
+; FMA64-NEXT: retq ## encoding: [0xc3]
+;
+; FMACALL64-LABEL: test_f32_cst:
+; FMACALL64: ## BB#0: ## %entry
+; FMACALL64-NEXT: movss {{.*}}(%rip), %xmm0 ## encoding: [0xf3,0x0f,0x10,0x05,A,A,A,A]
+; FMACALL64-NEXT: ## fixup A - offset: 4, value: LCPI3_0-4, kind: reloc_riprel_4byte
+; FMACALL64-NEXT: ## xmm0 = mem[0],zero,zero,zero
+; FMACALL64-NEXT: retq ## encoding: [0xc3]
+;
+; AVX51264NOFMA-LABEL: test_f32_cst:
+; AVX51264NOFMA: ## BB#0: ## %entry
+; AVX51264NOFMA-NEXT: vmovss {{.*}}(%rip), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfa,0x10,0x05,A,A,A,A]
+; AVX51264NOFMA-NEXT: ## fixup A - offset: 4, value: LCPI3_0-4, kind: reloc_riprel_4byte
+; AVX51264NOFMA-NEXT: ## xmm0 = mem[0],zero,zero,zero
+; AVX51264NOFMA-NEXT: retq ## encoding: [0xc3]
+;
+; AVX51264-LABEL: test_f32_cst:
+; AVX51264: ## BB#0: ## %entry
+; AVX51264-NEXT: vmovss {{.*}}(%rip), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfa,0x10,0x05,A,A,A,A]
+; AVX51264-NEXT: ## fixup A - offset: 4, value: LCPI3_0-4, kind: reloc_riprel_4byte
+; AVX51264-NEXT: ## xmm0 = mem[0],zero,zero,zero
+; AVX51264-NEXT: retq ## encoding: [0xc3]
entry:
%call = call float @llvm.fma.f32(float 3.0, float 3.0, float 3.0)
ret float %call
More information about the llvm-commits
mailing list