[X86] Add llvm-mca tests for r328834.

Verify that the ReadAfterLd is correctly applied to FMA and 4-ops variable blend
instructions.

As Craig pointed out in D44726, some Intel models still have to be fixed.

llvm-svn: 328861

diff --git a/llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-1.s b/llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-1.s
new file mode 100644 (file)
index 0000000..6b54ee2
--- /dev/null
+++ b/llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-1.s
@@ -0,0 +1,22 @@
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=HASWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=BDWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=SKYLAKE
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ZNVER1
+
+vaddps %xmm0, %xmm0, %xmm1
+vfmadd213ps (%rdi), %xmm1, %xmm2
+
+# HASWELL:      [0,0]  DeeeER    .    ..       vaddps  %xmm0, %xmm0, %xmm1
+# HASWELL-NEXT: [0,1]  D===eeeeeeeeeeeER       vfmadd213ps     (%rdi), %xmm1, %xmm2
+
+# BDWELL:       [0,0]  DeeeER    .    .        vaddps  %xmm0, %xmm0, %xmm1
+# BDWELL-NEXT:  [0,1]  D===eeeeeeeeeeER        vfmadd213ps     (%rdi), %xmm1, %xmm2
+
+# SKYLAKE:      [0,0]  DeeeeER   .    ..       vaddps  %xmm0, %xmm0, %xmm1
+# SKYLAKE-NEXT: [0,1]  D====eeeeeeeeeeER       vfmadd213ps     (%rdi), %xmm1, %xmm2
+
+# ZNVER1:       [0,0]  DeeeER    .   .   vaddps        %xmm0, %xmm0, %xmm1
+# ZNVER1-NEXT:  [0,1]  DeeeeeeeeeeeeER   vfmadd213ps   (%rdi), %xmm1, %xmm2

diff --git a/llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-2.s b/llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-2.s
new file mode 100644 (file)
index 0000000..6b6e068
--- /dev/null
+++ b/llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-2.s
@@ -0,0 +1,22 @@
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=HASWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=BDWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=SKYLAKE
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ZNVER1
+
+vaddps %xmm0, %xmm0, %xmm2
+vfmadd213ps (%rdi), %xmm1, %xmm2
+
+# HASWELL:      [0,0]  DeeeER    .    ..       vaddps  %xmm0, %xmm0, %xmm2
+# HASWELL-NEXT: [0,1]  D===eeeeeeeeeeeER       vfmadd213ps     (%rdi), %xmm1, %xmm2
+
+# BDWELL:       [0,0]  DeeeER    .    .        vaddps  %xmm0, %xmm0, %xmm2
+# BDWELL-NEXT:  [0,1]  D===eeeeeeeeeeER        vfmadd213ps     (%rdi), %xmm1, %xmm2
+
+# SKYLAKE:      [0,0]  DeeeeER   .    ..       vaddps  %xmm0, %xmm0, %xmm2
+# SKYLAKE-NEXT: [0,1]  D====eeeeeeeeeeER       vfmadd213ps     (%rdi), %xmm1, %xmm2
+
+# ZNVER1:       [0,0]  DeeeER    .   .   vaddps        %xmm0, %xmm0, %xmm2
+# ZNVER1-NEXT:  [0,1]  DeeeeeeeeeeeeER   vfmadd213ps   (%rdi), %xmm1, %xmm2

diff --git a/llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-1.s b/llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-1.s
new file mode 100644 (file)
index 0000000..9f774a8
--- /dev/null
+++ b/llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-1.s
@@ -0,0 +1,38 @@
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=SANDY
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=ivybridge -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=IVY
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=HASWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=BDWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=SKYLAKE
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=BTVER2
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ZNVER1
+
+vaddps %xmm0, %xmm0, %xmm1
+vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+
+# SANDY:         [0,0] DeeeER    .  .  vaddps    %xmm0, %xmm0, %xmm1
+# SANDY-NEXT:    [0,1] D===eeeeeeeeER  vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# IVY:           [0,0] DeeeER    .  .  vaddps    %xmm0, %xmm0, %xmm1
+# IVY-NEXT:      [0,1] D===eeeeeeeeER  vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# HASWELL:       [0,0] DeeeER    .  .  vaddps    %xmm0, %xmm0, %xmm1
+# HASWELL-NEXT:  [0,1] D===eeeeeeeeER  vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# BDWELL:        [0,0] DeeeER    . .   vaddps    %xmm0, %xmm0, %xmm1
+# BDWELL-NEXT:   [0,1] D===eeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# SKYLAKE:       [0,0] DeeeeER   .   . vaddps    %xmm0, %xmm0, %xmm1
+# SKYLAKE-NEXT:  [0,1] D====eeeeeeeeER vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# BTVER2:        [0,0] DeeeER    .     vaddps    %xmm0, %xmm0, %xmm1
+# BTVER2-NEXT:   [0,1] .DeeeeeeeER     vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# ZNVER1:        [0,0] DeeeER    .     vaddps    %xmm0, %xmm0, %xmm1
+# ZNVER1-NEXT:   [0,1] DeeeeeeeeER     vblendvps %xmm1, (%rdi), %xmm2, %xmm3

diff --git a/llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-2.s b/llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-2.s
new file mode 100644 (file)
index 0000000..8beea9a
--- /dev/null
+++ b/llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-2.s
@@ -0,0 +1,38 @@
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=SANDY
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=ivybridge -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=IVY
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=HASWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=BDWELL
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=SKYLAKE
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=BTVER2
+
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ZNVER1
+
+vaddps %xmm0, %xmm0, %xmm2
+vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+
+# SANDY:         [0,0] DeeeER    .  .  vaddps    %xmm0, %xmm0, %xmm2
+# SANDY-NEXT:    [0,1] D===eeeeeeeeER  vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# IVY:           [0,0] DeeeER    .  .  vaddps    %xmm0, %xmm0, %xmm2
+# IVY-NEXT:      [0,1] D===eeeeeeeeER  vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# HASWELL:       [0,0] DeeeER    .  .  vaddps    %xmm0, %xmm0, %xmm2
+# HASWELL-NEXT:  [0,1] D===eeeeeeeeER  vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# BDWELL:        [0,0] DeeeER    . .   vaddps    %xmm0, %xmm0, %xmm2
+# BDWELL-NEXT:   [0,1] D===eeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# SKYLAKE:       [0,0] DeeeeER   .   . vaddps    %xmm0, %xmm0, %xmm2
+# SKYLAKE-NEXT:  [0,1] D====eeeeeeeeER vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# BTVER2:        [0,0] DeeeER    .     vaddps    %xmm0, %xmm0, %xmm2
+# BTVER2-NEXT:   [0,1] .DeeeeeeeER     vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
+# ZNVER1:        [0,0] DeeeER    .     vaddps    %xmm0, %xmm0, %xmm2
+# ZNVER1-NEXT:   [0,1] DeeeeeeeeER     vblendvps %xmm1, (%rdi), %xmm2, %xmm3