+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -march=aarch64 -mcpu=cortex-a57 -iterations=600 -timeline -timeline-max-iterations=4 < %s | FileCheck %s
- b t
+ b t
# CHECK: Iterations: 600
# CHECK-NEXT: Instructions: 600
# CHECK-NEXT: Dispatch Width: 3
# CHECK-NEXT: IPC: 1.00
-
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: [6]: HasSideEffects
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 1 1.00 b t
-
+# CHECK-NEXT: 1 1 1.00 b t
# CHECK: Resources:
# CHECK-NEXT: [0] - A57UnitB
# CHECK-NEXT: [5] - A57UnitW
# CHECK-NEXT: [6] - A57UnitX
-
# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1.0] [1.1] [2] [3] [4] [5] [6]
-# CHECK-NEXT: 1.00 - - - - - - -
+# CHECK-NEXT: [0] [1.0] [1.1] [2] [3] [4] [5] [6]
+# CHECK-NEXT: 1.00 - - - - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1.0] [1.1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1.00 - - - - - - - b t
-
+# CHECK-NEXT: 1.00 - - - - - - - b t
# CHECK: Timeline view:
-# CHECK: Index 0123456
-# CHECK: [0,0] DeER .. b t
-# CHECK-NEXT: [1,0] D=eER.. b t
-# CHECK-NEXT: [2,0] D==eER. b t
-# CHECK-NEXT: [3,0] .D==eER b t
+# CHECK: Index 0123456
+# CHECK: [0,0] DeER .. b t
+# CHECK-NEXT: [1,0] D=eER.. b t
+# CHECK-NEXT: [2,0] D==eER. b t
+# CHECK-NEXT: [3,0] .D==eER b t
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 4 2.3 2.3 0.0 b t
+# CHECK-NEXT: 0. 4 2.3 2.3 0.0 b t
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -march=aarch64 -mcpu=exynos-m3 -iterations=300 -timeline -timeline-max-iterations=3 -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=M3
# RUN: llvm-mca -march=aarch64 -mcpu=exynos-m1 -iterations=300 -timeline -timeline-max-iterations=3 -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=M1
- b t
+ b t
-# ALL: Iterations: 300
-# ALL-NEXT: Instructions: 300
-
-# M3-NEXT: Total Cycles: 51
-# M3-NEXT: Dispatch Width: 6
-# M3-NEXT: IPC: 5.88
-
-# M1-NEXT: Total Cycles: 76
-# M1-NEXT: Dispatch Width: 4
-# M1-NEXT: IPC: 3.95
+# M3: Iterations: 300
+# M3-NEXT: Instructions: 300
+# M3-NEXT: Total Cycles: 51
+# M3-NEXT: Dispatch Width: 6
+# M3-NEXT: IPC: 5.88
+# M1: Iterations: 300
+# M1-NEXT: Instructions: 300
+# M1-NEXT: Total Cycles: 76
+# M1-NEXT: Dispatch Width: 4
+# M1-NEXT: IPC: 3.95
# ALL: Instruction Info:
# ALL-NEXT: [1]: #uOps
# ALL-NEXT: [6]: HasSideEffects
# ALL: [1] [2] [3] [4] [5] [6] Instructions:
-# ALL-NEXT: 1 0 - b t
+# ALL-NEXT: 1 0 - b t
+
+# ALL: Timeline view:
+# ALL: Index 01
+
+# ALL: [0,0] DR b t
+# ALL-NEXT: [1,0] DR b t
+# ALL-NEXT: [2,0] DR b t
# ALL: Average Wait times (based on the timeline view):
# ALL-NEXT: [0]: Executions
# ALL-NEXT: [3]: Average time elapsed from WB until retire stage
# ALL: [0] [1] [2] [3]
-# ALL-NEXT: 0. 3 0.0 0.0 0.0 b t
+# ALL-NEXT: 0. 3 0.0 0.0 0.0 b t
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -march=aarch64 -mcpu=exynos-m3 -iterations=1 -scheduler-stats -resource-pressure=false -instruction-info=false < %s | FileCheck %s -check-prefix=ALL
# RUN: llvm-mca -march=aarch64 -mcpu=exynos-m1 -iterations=1 -scheduler-stats -resource-pressure=false -instruction-info=false < %s | FileCheck %s -check-prefix=ALL
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=300 -timeline -timeline-max-iterations=3 < %s | FileCheck %s
vmulps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: Dispatch Width: 2
# CHECK-NEXT: IPC: 1.48
-
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: 1 3 1.00 vhaddps %xmm2, %xmm2, %xmm3
# CHECK-NEXT: 1 3 1.00 vhaddps %xmm3, %xmm3, %xmm4
-
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [13] - JVIMUL
# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
-# CHECK-NEXT: - - - 2.00 1.00 2.00 1.00 - - - - - - -
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - 2.00 1.00 2.00 1.00 - - - - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vhaddps %xmm2, %xmm2, %xmm3
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vhaddps %xmm3, %xmm3, %xmm4
-
# CHECK: Timeline view:
# CHECK-NEXT: 012345
# CHECK-NEXT: Index 0123456789
# CHECK-NEXT: [2,1] . D====eeeER . vhaddps %xmm2, %xmm2, %xmm3
# CHECK-NEXT: [2,2] . D======eeeER vhaddps %xmm3, %xmm3, %xmm4
-
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 3 1.0 1.0 3.3 vmulps %xmm0, %xmm1, %xmm2
-# CHECK-NEXT: 1. 3 3.3 0.7 1.0 vhaddps %xmm2, %xmm2, %xmm3
-# CHECK-NEXT: 2. 3 5.7 0.0 0.0 vhaddps %xmm3, %xmm3, %xmm4
+# CHECK-NEXT: 0. 3 1.0 1.0 3.3 vmulps %xmm0, %xmm1, %xmm2
+# CHECK-NEXT: 1. 3 3.3 0.7 1.0 vhaddps %xmm2, %xmm2, %xmm3
+# CHECK-NEXT: 2. 3 5.7 0.0 0.0 vhaddps %xmm3, %xmm3, %xmm4
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s
vshufps $0, %xmm0, %xmm1, %xmm1
vhaddps (%rdi), %xmm1, %xmm2
+# CHECK: Iterations: 1
+# CHECK-NEXT: Instructions: 2
+# CHECK-NEXT: Total Cycles: 11
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.18
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK: Timeline view:
# CHECK-NEXT: 0
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
+
# CHECK: [0,0] DeER . . vshufps $0, %xmm0, %xmm1, %xmm1
# CHECK-NEXT: [0,1] DeeeeeeeeER vhaddps (%rdi), %xmm1, %xmm2
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vshufps $0, %xmm0, %xmm1, %xmm1
+# CHECK-NEXT: 1. 1 1.0 0.0 0.0 vhaddps (%rdi), %xmm1, %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s
vshufps $0, %xmm0, %xmm1, %xmm1
vhaddps (%rdi), %ymm1, %ymm2
+# CHECK: Iterations: 1
+# CHECK-NEXT: Instructions: 2
+# CHECK-NEXT: Total Cycles: 12
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.17
+
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: 1 1 0.50 vshufps $0, %xmm0, %xmm1, %xmm1
# CHECK-NEXT: 2 8 2.00 * vhaddps (%rdi), %ymm1, %ymm2
-
# CHECK: Timeline view:
# CHECK-NEXT: 01
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
+
# CHECK: [0,0] DeER . .. vshufps $0, %xmm0, %xmm1, %xmm1
# CHECK-NEXT: [0,1] .DeeeeeeeeER vhaddps (%rdi), %ymm1, %ymm2
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vshufps $0, %xmm0, %xmm1, %xmm1
+# CHECK-NEXT: 1. 1 1.0 1.0 0.0 vhaddps (%rdi), %ymm1, %ymm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -resource-pressure=false -instruction-info=true < %s | FileCheck %s --check-prefix=ENABLED
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -resource-pressure=false -instruction-info=false < %s | FileCheck %s -check-prefix=DISABLED
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -resource-pressure=false -instruction-info < %s | FileCheck %s -check-prefix=ENABLED
# DISABLED-NOT: Instruction Info:
+# ENABLED: Iterations: 100
+# ENABLED-NEXT: Instructions: 300
+# ENABLED-NEXT: Total Cycles: 209
+# ENABLED-NEXT: Dispatch Width: 2
+# ENABLED-NEXT: IPC: 1.44
+
# ENABLED: Instruction Info:
# ENABLED-NEXT: [1]: #uOps
# ENABLED-NEXT: [2]: Latency
# ENABLED-NEXT: 1 2 1.00 vmulps %xmm0, %xmm1, %xmm2
# ENABLED-NEXT: 1 3 1.00 vhaddps %xmm2, %xmm2, %xmm3
# ENABLED-NEXT: 1 3 1.00 vhaddps %xmm3, %xmm3, %xmm4
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=100 -timeline -timeline-max-iterations=1 -noalias=false < %s | FileCheck %s
vmovaps (%rsi), %xmm0
vmovaps 48(%rsi), %xmm0
vmovaps %xmm0, 48(%rdi)
-# CHECK: Iterations: 100
+# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 800
# CHECK-NEXT: Total Cycles: 2403
# CHECK-NEXT: Dispatch Width: 2
# CHECK-NEXT: IPC: 0.33
-
-# CHECK: Instruction Info:
+# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: [3]: RThroughput
# CHECK-NEXT: [5]: MayStore
# CHECK-NEXT: [6]: HasSideEffects
-# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 5 1.00 * vmovaps (%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, (%rdi)
-# CHECK-NEXT: 1 5 1.00 * vmovaps 16(%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 16(%rdi)
-# CHECK-NEXT: 1 5 1.00 * vmovaps 32(%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 32(%rdi)
-# CHECK-NEXT: 1 5 1.00 * vmovaps 48(%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 48(%rdi)
-
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 5 1.00 * vmovaps (%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, (%rdi)
+# CHECK-NEXT: 1 5 1.00 * vmovaps 16(%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 16(%rdi)
+# CHECK-NEXT: 1 5 1.00 * vmovaps 32(%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 32(%rdi)
+# CHECK-NEXT: 1 5 1.00 * vmovaps 48(%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 48(%rdi)
-# CHECK: Resources:
+# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [2] - JDiv
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - 2.00 2.00 3.99 4.01 4.00 - 4.00 4.00 - - -
-# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
-# CHECK-NEXT: - - - 2.00 2.00 3.99 4.01 4.00 - 4.00 4.00 - - -
-
-# CHECK: Resource pressure by instruction:
+# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - - 1.00 0.99 0.01 1.00 - - - - - - vmovaps (%rsi), %xmm0
# CHECK-NEXT: - - - - - - 1.00 - - 1.00 1.00 - - - vmovaps %xmm0, (%rdi)
# CHECK-NEXT: - - - 1.00 - 1.00 - 1.00 - - - - - - vmovaps 48(%rsi), %xmm0
# CHECK-NEXT: - - - - - - 1.00 - - 1.00 1.00 - - - vmovaps %xmm0, 48(%rdi)
-
# CHECK: Timeline view:
-# CHECK-NEXT: 0123456789
+# CHECK-NEXT: 0123456789
# CHECK-NEXT: Index 0123456789 0123456
# CHECK: [0,0] DeeeeeER . . . .. vmovaps (%rsi), %xmm0
# CHECK-NEXT: [0,6] . D===============eeeeeER. vmovaps 48(%rsi), %xmm0
# CHECK-NEXT: [0,7] . D====================eER vmovaps %xmm0, 48(%rdi)
-# CHECK: Average Wait times (based on the timeline view):
+# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vmovaps (%rsi), %xmm0
-# CHECK-NEXT: 1. 1 6.0 0.0 0.0 vmovaps %xmm0, (%rdi)
-# CHECK-NEXT: 2. 1 6.0 0.0 0.0 vmovaps 16(%rsi), %xmm0
-# CHECK-NEXT: 3. 1 11.0 0.0 0.0 vmovaps %xmm0, 16(%rdi)
-# CHECK-NEXT: 4. 1 11.0 0.0 0.0 vmovaps 32(%rsi), %xmm0
-# CHECK-NEXT: 5. 1 16.0 0.0 0.0 vmovaps %xmm0, 32(%rdi)
-# CHECK-NEXT: 6. 1 16.0 0.0 0.0 vmovaps 48(%rsi), %xmm0
-# CHECK-NEXT: 7. 1 21.0 0.0 0.0 vmovaps %xmm0, 48(%rdi)
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vmovaps (%rsi), %xmm0
+# CHECK-NEXT: 1. 1 6.0 0.0 0.0 vmovaps %xmm0, (%rdi)
+# CHECK-NEXT: 2. 1 6.0 0.0 0.0 vmovaps 16(%rsi), %xmm0
+# CHECK-NEXT: 3. 1 11.0 0.0 0.0 vmovaps %xmm0, 16(%rdi)
+# CHECK-NEXT: 4. 1 11.0 0.0 0.0 vmovaps 32(%rsi), %xmm0
+# CHECK-NEXT: 5. 1 16.0 0.0 0.0 vmovaps %xmm0, 32(%rdi)
+# CHECK-NEXT: 6. 1 16.0 0.0 0.0 vmovaps 48(%rsi), %xmm0
+# CHECK-NEXT: 7. 1 21.0 0.0 0.0 vmovaps %xmm0, 48(%rdi)
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=100 -timeline -timeline-max-iterations=1 < %s | FileCheck %s
vmovaps (%rsi), %xmm0
vmovaps 48(%rsi), %xmm0
vmovaps %xmm0, 48(%rdi)
-
-# CHECK: Iterations: 100
+# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 800
# CHECK-NEXT: Total Cycles: 408
# CHECK-NEXT: Dispatch Width: 2
# CHECK-NEXT: IPC: 1.96
-
-# CHECK: Instruction Info:
+# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: [3]: RThroughput
# CHECK-NEXT: [5]: MayStore
# CHECK-NEXT: [6]: HasSideEffects
-# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 5 1.00 * vmovaps (%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, (%rdi)
-# CHECK-NEXT: 1 5 1.00 * vmovaps 16(%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 16(%rdi)
-# CHECK-NEXT: 1 5 1.00 * vmovaps 32(%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 32(%rdi)
-# CHECK-NEXT: 1 5 1.00 * vmovaps 48(%rsi), %xmm0
-# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 48(%rdi)
-
-
-# CHECK: Resources:
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 5 1.00 * vmovaps (%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, (%rdi)
+# CHECK-NEXT: 1 5 1.00 * vmovaps 16(%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 16(%rdi)
+# CHECK-NEXT: 1 5 1.00 * vmovaps 32(%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 32(%rdi)
+# CHECK-NEXT: 1 5 1.00 * vmovaps 48(%rsi), %xmm0
+# CHECK-NEXT: 1 1 1.00 * vmovaps %xmm0, 48(%rdi)
+
+# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [2] - JDiv
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - 2.00 2.00 3.97 4.03 4.00 - 4.00 4.00 - - -
-# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
-# CHECK-NEXT: - - - 2.00 2.00 3.97 4.03 4.00 - 4.00 4.00 - - -
-
-# CHECK: Resource pressure by instruction:
+# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - - 1.00 0.98 0.02 1.00 - - - - - - vmovaps (%rsi), %xmm0
# CHECK-NEXT: - - - - - - 1.00 - - 1.00 1.00 - - - vmovaps %xmm0, (%rdi)
# CHECK-NEXT: - - - 1.00 - 1.00 - 1.00 - - - - - - vmovaps 48(%rsi), %xmm0
# CHECK-NEXT: - - - - - - 1.00 - - 1.00 1.00 - - - vmovaps %xmm0, 48(%rdi)
-
-# CHECK: Timeline view:
-# CHECK-NEXT: 01
-# CHECK-NEXT: Index 0123456789
+# CHECK: Timeline view:
+# CHECK-NEXT: 01
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeER .. vmovaps (%rsi), %xmm0
# CHECK-NEXT: [0,1] D=====eER .. vmovaps %xmm0, (%rdi)
# CHECK-NEXT: [0,6] . DeeeeeER. vmovaps 48(%rsi), %xmm0
# CHECK-NEXT: [0,7] . D=====eER vmovaps %xmm0, 48(%rdi)
-
-# CHECK: Average Wait times (based on the timeline view):
+# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vmovaps (%rsi), %xmm0
-# CHECK-NEXT: 1. 1 6.0 0.0 0.0 vmovaps %xmm0, (%rdi)
-# CHECK-NEXT: 2. 1 1.0 1.0 0.0 vmovaps 16(%rsi), %xmm0
-# CHECK-NEXT: 3. 1 6.0 0.0 0.0 vmovaps %xmm0, 16(%rdi)
-# CHECK-NEXT: 4. 1 1.0 1.0 0.0 vmovaps 32(%rsi), %xmm0
-# CHECK-NEXT: 5. 1 6.0 0.0 0.0 vmovaps %xmm0, 32(%rdi)
-# CHECK-NEXT: 6. 1 1.0 1.0 0.0 vmovaps 48(%rsi), %xmm0
-# CHECK-NEXT: 7. 1 6.0 0.0 0.0 vmovaps %xmm0, 48(%rdi)
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vmovaps (%rsi), %xmm0
+# CHECK-NEXT: 1. 1 6.0 0.0 0.0 vmovaps %xmm0, (%rdi)
+# CHECK-NEXT: 2. 1 1.0 1.0 0.0 vmovaps 16(%rsi), %xmm0
+# CHECK-NEXT: 3. 1 6.0 0.0 0.0 vmovaps %xmm0, 16(%rdi)
+# CHECK-NEXT: 4. 1 1.0 1.0 0.0 vmovaps 32(%rsi), %xmm0
+# CHECK-NEXT: 5. 1 6.0 0.0 0.0 vmovaps %xmm0, 32(%rdi)
+# CHECK-NEXT: 6. 1 1.0 1.0 0.0 vmovaps 48(%rsi), %xmm0
+# CHECK-NEXT: 7. 1 6.0 0.0 0.0 vmovaps %xmm0, 48(%rdi)
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -timeline -timeline-max-iterations=2 < %s | FileCheck %s
# VALU0/VALU1
# CHECK-NEXT: [13] - JVIMUL
# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
# CHECK-NEXT: - - - 3.00 63.00 6.01 5.99 - - - 1.00 1.00 1.00 3.00
# CHECK: Resource pressure by instruction:
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -resource-pressure=false -retire-stats -iterations=1 < %s | FileCheck %s
vsqrtps %xmm0, %xmm2
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s
# The vmul can start executing 3cy in advance. That is beause the first use
# operand (i.e. %xmm1) is a ReadAfterLd. That means, the memory operand is
# evaluated before %xmm1.
-
vaddps %xmm0, %xmm0, %xmm1
vmulps (%rdi), %xmm1, %xmm2
-
# CHECK: Iterations: 1
# CHECK-NEXT: Instructions: 2
# CHECK-NEXT: Total Cycles: 10
# CHECK-NEXT: Dispatch Width: 2
-
+# CHECK-NEXT: IPC: 0.20
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: 1 3 1.00 vaddps %xmm0, %xmm0, %xmm1
# CHECK-NEXT: 1 7 1.00 * vmulps (%rdi), %xmm1, %xmm2
-
# CHECK: Timeline view:
# CHECK: Index 0123456789
+
# CHECK: [0,0] DeeeER . vaddps %xmm0, %xmm0, %xmm1
# CHECK-NEXT: [0,1] DeeeeeeeER vmulps (%rdi), %xmm1, %xmm2
-
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vaddps %xmm0, %xmm0, %xmm1
-# CHECK-NEXT: 1. 1 1.0 0.0 0.0 vmulps (%rdi), %xmm1, %xmm2
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vaddps %xmm0, %xmm0, %xmm1
+# CHECK-NEXT: 1. 1 1.0 0.0 0.0 vmulps (%rdi), %xmm1, %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -resource-pressure=0 -timeline < %s | FileCheck %s
imull %esi
imull (%rdi)
-
# The second integer multiply can start at cycle 2 because the implicit reads
# can start after the load operand is evaluated.
-# CHECK: Instruction Info:
+# CHECK: Iterations: 1
+# CHECK-NEXT: Instructions: 2
+# CHECK-NEXT: Total Cycles: 10
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.20
+
+# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: [3]: RThroughput
# CHECK-NEXT: 2 3 1.00 imull %esi
# CHECK-NEXT: 2 6 1.00 * imull (%rdi)
-
# CHECK: Timeline view:
+
# CHECK: Index 0123456789
+
# CHECK: [0,0] DeeeER . imull %esi
# CHECK-NEXT: [0,1] .DeeeeeeER imull (%rdi)
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 imull %esi
+# CHECK-NEXT: 1. 1 1.0 1.0 0.0 imull (%rdi)
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -resource-pressure=0 -timeline -dispatch=3 < %s | FileCheck %s
add %rdi, %rsi
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=5 -instruction-info=false -dispatch-stats -register-file-stats -timeline < %s | FileCheck %s
vaddps %xmm0, %xmm0, %xmm0
vmulps %xmm0, %xmm0, %xmm0
-# CHECK: Iterations: 5
+# CHECK: Iterations: 5
# CHECK-NEXT: Instructions: 10
-
+# CHECK-NEXT: Total Cycles: 28
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.36
# CHECK: Dynamic Dispatch Stall Cycles:
# CHECK-NEXT: RAT - Register unavailable: 0
# CHECK-NEXT: SQ - Store queue full: 0
# CHECK-NEXT: GROUP - Static restrictions on the dispatch group: 0
+# CHECK: Dispatch Logic - number of cycles where we saw N instructions dispatched:
+# CHECK-NEXT: [# dispatched], [# cycles]
+# CHECK-NEXT: 0, 23 (82.1%)
+# CHECK-NEXT: 2, 5 (17.9%)
# CHECK: Register File statistics:
-# CHECK-NEXT: Total number of mappings created: 10
-# CHECK-NEXT: Max number of mappings used: 10
+# CHECK-NEXT: Total number of mappings created: 10
+# CHECK-NEXT: Max number of mappings used: 10
# CHECK: * Register File #1 -- FpuPRF:
# CHECK-NEXT: Number of physical registers: 72
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
+# CHECK: Resources:
+# CHECK-NEXT: [0] - JALU0
+# CHECK-NEXT: [1] - JALU1
+# CHECK-NEXT: [2] - JDiv
+# CHECK-NEXT: [3] - JFPA
+# CHECK-NEXT: [4] - JFPM
+# CHECK-NEXT: [5] - JFPU0
+# CHECK-NEXT: [6] - JFPU1
+# CHECK-NEXT: [7] - JLAGU
+# CHECK-NEXT: [8] - JMul
+# CHECK-NEXT: [9] - JSAGU
+# CHECK-NEXT: [10] - JSTC
+# CHECK-NEXT: [11] - JVALU0
+# CHECK-NEXT: [12] - JVALU1
+# CHECK-NEXT: [13] - JVIMUL
+
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - 1.00 1.00 1.00 1.00 - - - - - - -
+
+# CHECK: Resource pressure by instruction:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
+# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vaddps %xmm0, %xmm0, %xmm0
+# CHECK-NEXT: - - - - 1.00 - 1.00 - - - - - - - vmulps %xmm0, %xmm0, %xmm0
-# CHECK: Timeline view:
-# CHECK-NEXT: 0123456789
+# CHECK: Timeline view:
+# CHECK-NEXT: 0123456789
# CHECK-NEXT: Index 0123456789 01234567
# CHECK: [0,0] DeeeER . . . . . vaddps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [3,1] . D===============eeER . . vmulps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [4,0] . D================eeeER . vaddps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [4,1] . D===================eeER vmulps %xmm0, %xmm0, %xmm0
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 5 9.0 0.2 0.0 vaddps %xmm0, %xmm0, %xmm0
+# CHECK-NEXT: 1. 5 12.0 0.0 0.0 vmulps %xmm0, %xmm0, %xmm0
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -register-file-size=5 -iterations=5 -instruction-info=false -dispatch-stats -register-file-stats -timeline < %s | FileCheck %s
vaddps %xmm0, %xmm0, %xmm0
vmulps %xmm0, %xmm0, %xmm0
-# CHECK: Iterations: 5
+# CHECK: Iterations: 5
# CHECK-NEXT: Instructions: 10
+# CHECK-NEXT: Total Cycles: 28
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.36
-
-# CHECK: Dynamic Dispatch Stall Cycles:
+# CHECK: Dynamic Dispatch Stall Cycles:
# CHECK-NEXT: RAT - Register unavailable: 13
# CHECK-NEXT: RCU - Retire tokens unavailable: 0
# CHECK-NEXT: SCHEDQ - Scheduler full: 0
# CHECK-NEXT: SQ - Store queue full: 0
# CHECK-NEXT: GROUP - Static restrictions on the dispatch group: 0
+# CHECK: Dispatch Logic - number of cycles where we saw N instructions dispatched:
+# CHECK-NEXT: [# dispatched], [# cycles]
+# CHECK-NEXT: 0, 20 (71.4%)
+# CHECK-NEXT: 2, 2 (7.1%)
+# CHECK-NEXT: 1, 6 (21.4%)
# CHECK: Register File statistics:
-# CHECK-NEXT: Total number of mappings created: 10
-# CHECK-NEXT: Max number of mappings used: 5
+# CHECK-NEXT: Total number of mappings created: 10
+# CHECK-NEXT: Max number of mappings used: 5
# CHECK: * Register File #1 -- FpuPRF:
# CHECK-NEXT: Number of physical registers: 72
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
+# CHECK: Resources:
+# CHECK-NEXT: [0] - JALU0
+# CHECK-NEXT: [1] - JALU1
+# CHECK-NEXT: [2] - JDiv
+# CHECK-NEXT: [3] - JFPA
+# CHECK-NEXT: [4] - JFPM
+# CHECK-NEXT: [5] - JFPU0
+# CHECK-NEXT: [6] - JFPU1
+# CHECK-NEXT: [7] - JLAGU
+# CHECK-NEXT: [8] - JMul
+# CHECK-NEXT: [9] - JSAGU
+# CHECK-NEXT: [10] - JSTC
+# CHECK-NEXT: [11] - JVALU0
+# CHECK-NEXT: [12] - JVALU1
+# CHECK-NEXT: [13] - JVIMUL
+
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - 1.00 1.00 1.00 1.00 - - - - - - -
+
+# CHECK: Resource pressure by instruction:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
+# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vaddps %xmm0, %xmm0, %xmm0
+# CHECK-NEXT: - - - - 1.00 - 1.00 - - - - - - - vmulps %xmm0, %xmm0, %xmm0
# CHECK: Timeline view:
-# CHECK-NEXT: 0123456789
+# CHECK-NEXT: 0123456789
# CHECK-NEXT: Index 0123456789 01234567
+
# CHECK: [0,0] DeeeER . . . . . vaddps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [0,1] D===eeER . . . . . vmulps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [1,0] .D====eeeER . . . . vaddps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [3,1] . . D========eeER . . vmulps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [4,0] . . . D========eeeER . vaddps %xmm0, %xmm0, %xmm0
# CHECK-NEXT: [4,1] . . . D========eeER vmulps %xmm0, %xmm0, %xmm0
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 5 6.6 0.2 0.0 vaddps %xmm0, %xmm0, %xmm0
+# CHECK-NEXT: 1. 5 7.8 0.0 0.0 vmulps %xmm0, %xmm0, %xmm0
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -register-file-size=5 -iterations=2 -dispatch-stats -register-file-stats -timeline < %s | FileCheck %s
idiv %eax
# CHECK: Iterations: 2
# CHECK-NEXT: Instructions: 2
+# CHECK-NEXT: Total Cycles: 55
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.04
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
# CHECK-NEXT: 2 25 25.00 * idivl %eax
-
# CHECK: Dynamic Dispatch Stall Cycles:
# CHECK-NEXT: RAT - Register unavailable: 26
# CHECK-NEXT: RCU - Retire tokens unavailable: 0
# CHECK-NEXT: SQ - Store queue full: 0
# CHECK-NEXT: GROUP - Static restrictions on the dispatch group: 0
+# CHECK: Dispatch Logic - number of cycles where we saw N instructions dispatched:
+# CHECK-NEXT: [# dispatched], [# cycles]
+# CHECK-NEXT: 0, 53 (96.4%)
+# CHECK-NEXT: 1, 2 (3.6%)
# CHECK: Register File statistics:
-# CHECK-NEXT: Total number of mappings created: 6
-# CHECK-NEXT: Max number of mappings used: 3
+# CHECK-NEXT: Total number of mappings created: 6
+# CHECK-NEXT: Max number of mappings used: 3
# CHECK: * Register File #1 -- FpuPRF:
# CHECK-NEXT: Number of physical registers: 72
# CHECK-NEXT: Total number of mappings created: 6
# CHECK-NEXT: Max number of mappings used: 3
+# CHECK: Resources:
+# CHECK-NEXT: [0] - JALU0
+# CHECK-NEXT: [1] - JALU1
+# CHECK-NEXT: [2] - JDiv
+# CHECK-NEXT: [3] - JFPA
+# CHECK-NEXT: [4] - JFPM
+# CHECK-NEXT: [5] - JFPU0
+# CHECK-NEXT: [6] - JFPU1
+# CHECK-NEXT: [7] - JLAGU
+# CHECK-NEXT: [8] - JMul
+# CHECK-NEXT: [9] - JSAGU
+# CHECK-NEXT: [10] - JSTC
+# CHECK-NEXT: [11] - JVALU0
+# CHECK-NEXT: [12] - JVALU1
+# CHECK-NEXT: [13] - JVIMUL
+
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - 1.00 25.00 - - - - - - - - - - -
+
+# CHECK: Resource pressure by instruction:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
+# CHECK-NEXT: - 1.00 25.00 - - - - - - - - - - - idivl %eax
# CHECK: Timeline view:
-# CHECK-NEXT: 0123456789 0123456789 01234
-# CHECK-NEXT: Index 0123456789 0123456789 0123456789
+# CHECK-NEXT: 0123456789 0123456789 01234
+# CHECK-NEXT: Index 0123456789 0123456789 0123456789
# CHECK: [0,0] DeeeeeeeeeeeeeeeeeeeeeeeeeER . . . . . . idivl %eax
# CHECK-NEXT: [1,0] . . . . . . DeeeeeeeeeeeeeeeeeeeeeeeeeER idivl %eax
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 2 1.0 1.0 0.0 idivl %eax
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=22 -dispatch-stats -register-file-stats -resource-pressure=false -timeline -timeline-max-iterations=3 < %s | FileCheck %s
idiv %eax
# CHECK: Iterations: 22
# CHECK-NEXT: Instructions: 22
+# CHECK-NEXT: Total Cycles: 553
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.04
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
# CHECK-NEXT: 2 25 25.00 * idivl %eax
-
# CHECK: Dynamic Dispatch Stall Cycles:
# CHECK-NEXT: RAT - Register unavailable: 6
# CHECK-NEXT: RCU - Retire tokens unavailable: 0
# CHECK-NEXT: SQ - Store queue full: 0
# CHECK-NEXT: GROUP - Static restrictions on the dispatch group: 0
+# CHECK: Dispatch Logic - number of cycles where we saw N instructions dispatched:
+# CHECK-NEXT: [# dispatched], [# cycles]
+# CHECK-NEXT: 0, 531 (96.0%)
+# CHECK-NEXT: 1, 22 (4.0%)
# CHECK: Register File statistics:
-# CHECK-NEXT: Total number of mappings created: 66
-# CHECK-NEXT: Max number of mappings used: 63
+# CHECK-NEXT: Total number of mappings created: 66
+# CHECK-NEXT: Max number of mappings used: 63
# CHECK: * Register File #1 -- FpuPRF:
# CHECK-NEXT: Number of physical registers: 72
# CHECK-NEXT: Total number of mappings created: 66
# CHECK-NEXT: Max number of mappings used: 63
-
# CHECK: Timeline view:
# CHECK-NEXT: 0123456789 0123456789 0123456789 01234567
-# CHECK-NEXT: Index 0123456789 0123456789 0123456789 0123456789
+# CHECK-NEXT: Index 0123456789 0123456789 0123456789 0123456789
+
# CHECK: [0,0] DeeeeeeeeeeeeeeeeeeeeeeeeeER . . . . . . . . . . . idivl %eax
# CHECK-NEXT: [1,0] .D========================eeeeeeeeeeeeeeeeeeeeeeeeeER . . . . . . idivl %eax
# CHECK-NEXT: [2,0] . D================================================eeeeeeeeeeeeeeeeeeeeeeeeeER idivl %eax
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 3 25.0 0.3 0.0 idivl %eax
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -resource-pressure=false -instruction-info=false -dispatch-stats -register-file-stats -timeline < %s | FileCheck %s
vdivps %ymm0, %ymm0, %ymm1
vaddps %ymm3, %ymm0, %ymm5
vaddps %ymm3, %ymm0, %ymm6
-
# CHECK: Iterations: 1
# CHECK-NEXT: Instructions: 33
# CHECK-NEXT: Total Cycles: 70
# CHECK-NEXT: Dispatch Width: 2
# CHECK-NEXT: IPC: 0.47
-
# CHECK: Dynamic Dispatch Stall Cycles:
# CHECK-NEXT: RAT - Register unavailable: 0
# CHECK-NEXT: RCU - Retire tokens unavailable: 8
# CHECK-NEXT: SQ - Store queue full: 0
# CHECK-NEXT: GROUP - Static restrictions on the dispatch group: 0
+# CHECK: Dispatch Logic - number of cycles where we saw N instructions dispatched:
+# CHECK-NEXT: [# dispatched], [# cycles]
+# CHECK-NEXT: 0, 37 (52.9%)
+# CHECK-NEXT: 1, 33 (47.1%)
# CHECK: Register File statistics:
-# CHECK-NEXT: Total number of mappings created: 66
-# CHECK-NEXT: Max number of mappings used: 64
+# CHECK-NEXT: Total number of mappings created: 66
+# CHECK-NEXT: Max number of mappings used: 64
# CHECK: * Register File #1 -- FpuPRF:
# CHECK-NEXT: Number of physical registers: 72
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
-
# CHECK: Timeline view:
-# CHECK-NEXT: 0123456789 0123456789 0123456789
+# CHECK-NEXT: 0123456789 0123456789 0123456789
# CHECK-NEXT: Index 0123456789 0123456789 0123456789 0123456789
# CHECK: [0,0] DeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeER . . . . . . vdivps %ymm0, %ymm0, %ymm1
# CHECK-NEXT: [0,30] . . . . . . D==============================eeeER . vaddps %ymm3, %ymm0, %ymm4
# CHECK-NEXT: [0,31] . . . . . . .D===============================eeeER . vaddps %ymm3, %ymm0, %ymm5
# CHECK-NEXT: [0,32] . . . . . . . . D========================eeeER vaddps %ymm3, %ymm0, %ymm6
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vdivps %ymm0, %ymm0, %ymm1
+# CHECK-NEXT: 1. 1 1.0 1.0 34.0 vaddps %ymm0, %ymm0, %ymm2
+# CHECK-NEXT: 2. 1 2.0 2.0 33.0 vaddps %ymm0, %ymm0, %ymm3
+# CHECK-NEXT: 3. 1 3.0 3.0 31.0 vaddps %ymm0, %ymm0, %ymm4
+# CHECK-NEXT: 4. 1 4.0 4.0 30.0 vaddps %ymm0, %ymm0, %ymm5
+# CHECK-NEXT: 5. 1 5.0 5.0 28.0 vaddps %ymm0, %ymm0, %ymm6
+# CHECK-NEXT: 6. 1 6.0 6.0 27.0 vaddps %ymm0, %ymm0, %ymm7
+# CHECK-NEXT: 7. 1 7.0 7.0 25.0 vaddps %ymm0, %ymm0, %ymm8
+# CHECK-NEXT: 8. 1 8.0 8.0 24.0 vaddps %ymm0, %ymm0, %ymm9
+# CHECK-NEXT: 9. 1 9.0 9.0 22.0 vaddps %ymm0, %ymm0, %ymm10
+# CHECK-NEXT: 10. 1 10.0 10.0 21.0 vaddps %ymm0, %ymm0, %ymm11
+# CHECK-NEXT: 11. 1 11.0 11.0 19.0 vaddps %ymm0, %ymm0, %ymm12
+# CHECK-NEXT: 12. 1 12.0 12.0 18.0 vaddps %ymm0, %ymm0, %ymm13
+# CHECK-NEXT: 13. 1 13.0 13.0 16.0 vaddps %ymm0, %ymm0, %ymm14
+# CHECK-NEXT: 14. 1 14.0 14.0 15.0 vaddps %ymm0, %ymm0, %ymm15
+# CHECK-NEXT: 15. 1 15.0 15.0 13.0 vaddps %ymm2, %ymm0, %ymm0
+# CHECK-NEXT: 16. 1 17.0 0.0 11.0 vaddps %ymm2, %ymm0, %ymm3
+# CHECK-NEXT: 17. 1 18.0 2.0 9.0 vaddps %ymm2, %ymm0, %ymm4
+# CHECK-NEXT: 18. 1 19.0 4.0 8.0 vaddps %ymm2, %ymm0, %ymm5
+# CHECK-NEXT: 19. 1 20.0 6.0 6.0 vaddps %ymm2, %ymm0, %ymm6
+# CHECK-NEXT: 20. 1 21.0 8.0 5.0 vaddps %ymm2, %ymm0, %ymm7
+# CHECK-NEXT: 21. 1 22.0 10.0 3.0 vaddps %ymm2, %ymm0, %ymm8
+# CHECK-NEXT: 22. 1 23.0 12.0 2.0 vaddps %ymm2, %ymm0, %ymm9
+# CHECK-NEXT: 23. 1 24.0 14.0 0.0 vaddps %ymm2, %ymm0, %ymm10
+# CHECK-NEXT: 24. 1 25.0 16.0 0.0 vaddps %ymm2, %ymm0, %ymm11
+# CHECK-NEXT: 25. 1 26.0 18.0 0.0 vaddps %ymm2, %ymm0, %ymm12
+# CHECK-NEXT: 26. 1 27.0 20.0 0.0 vaddps %ymm2, %ymm0, %ymm13
+# CHECK-NEXT: 27. 1 28.0 22.0 0.0 vaddps %ymm2, %ymm0, %ymm14
+# CHECK-NEXT: 28. 1 29.0 24.0 0.0 vaddps %ymm2, %ymm0, %ymm15
+# CHECK-NEXT: 29. 1 30.0 23.0 0.0 vaddps %ymm3, %ymm0, %ymm2
+# CHECK-NEXT: 30. 1 31.0 25.0 0.0 vaddps %ymm3, %ymm0, %ymm4
+# CHECK-NEXT: 31. 1 32.0 27.0 0.0 vaddps %ymm3, %ymm0, %ymm5
+# CHECK-NEXT: 32. 1 25.0 25.0 0.0 vaddps %ymm3, %ymm0, %ymm6
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
aesdec %xmm0, %xmm2
aeskeygenassist $22, %xmm0, %xmm2
aeskeygenassist $22, (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 2 3 1.00 aesdec %xmm0, %xmm2
+# CHECK-NEXT: 2 8 1.00 * aesdec (%rax), %xmm2
+# CHECK-NEXT: 2 3 1.00 aesdeclast %xmm0, %xmm2
+# CHECK-NEXT: 2 8 1.00 * aesdeclast (%rax), %xmm2
+# CHECK-NEXT: 2 3 1.00 aesenc %xmm0, %xmm2
+# CHECK-NEXT: 2 8 1.00 * aesenc (%rax), %xmm2
+# CHECK-NEXT: 2 3 1.00 aesenclast %xmm0, %xmm2
+# CHECK-NEXT: 2 8 1.00 * aesenclast (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 aesimc %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * aesimc (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 aeskeygenassist $22, %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * aeskeygenassist $22, (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - - - 12.00 - 6.00 - - - - - 12.00
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - - - 1.00 - - - - - - - 1.00 aesdec %xmm0, %xmm2
# CHECK-NEXT: - - - - - 1.00 - 1.00 - - - - - 1.00 aesimc (%rax), %xmm2
# CHECK-NEXT: - - - - - 1.00 - - - - - - - 1.00 aeskeygenassist $22, %xmm0, %xmm2
# CHECK-NEXT: - - - - - 1.00 - 1.00 - - - - - 1.00 aeskeygenassist $22, (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
vaddpd %xmm0, %xmm1, %xmm2
vzeroall
vzeroupper
-
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: 73 90 - * * * vzeroall
# CHECK-NEXT: 37 46 - * * * vzeroupper
-
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: 48.00 2.00 - 355.50 907.50 402.00 398.00 381.00 - 43.00 114.00 116.50 116.50 40.00
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vaddpd %xmm0, %xmm1, %xmm2
# CHECK-NEXT: - - - 1.00 1.00 1.00 1.00 2.00 - - - - - - vxorps (%rax), %ymm1, %ymm2
# CHECK-NEXT: - - - - - - - - - - - - - - vzeroall
# CHECK-NEXT: - - - - - - - - - - - - - - vzeroupper
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
pclmulqdq $11, %xmm0, %xmm2
pclmulqdq $11, (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 2 1.00 pclmulqdq $11, %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pclmulqdq $11, (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - - - 2.00 - 1.00 - - - - - 2.00
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - - - 1.00 - - - - - - - 1.00 pclmulqdq $11, %xmm0, %xmm2
# CHECK-NEXT: - - - - - 1.00 - 1.00 - - - - - 1.00 pclmulqdq $11, (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 < %s | FileCheck %s
vcvtph2ps %xmm0, %xmm2
# CHECK-NEXT: [13] - JVIMUL
# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
# CHECK-NEXT: - - - 2.00 2.00 - 12.00 2.00 - 2.00 12.00 - - -
# CHECK: Resource pressure by instruction:
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
addps %xmm0, %xmm2
xorps %xmm0, %xmm2
xorps (%rax), %xmm2
-
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: 1 1 0.50 xorps %xmm0, %xmm2
# CHECK-NEXT: 1 6 1.00 * xorps (%rax), %xmm2
-
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: 13.00 - - 44.50 183.50 37.50 47.50 42.00 - 7.00 15.00 1.00 1.00 -
-# CHECK: Resource pressure by instruction:
+# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - addps %xmm0, %xmm2
# CHECK-NEXT: - - - 1.00 - 1.00 - 1.00 - - - - - - addps (%rax), %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 1.00 - - - - - - unpcklps (%rax), %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 - - - - - - - xorps %xmm0, %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 1.00 - - - - - - xorps (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
addpd %xmm0, %xmm2
xorpd %xmm0, %xmm2
xorpd (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 3 1.00 addpd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * addpd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 addsd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * addsd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 addsubpd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * addsubpd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 andnpd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * andnpd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 andpd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * andpd (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 cmppd $0, %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * cmppd $0, (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 cmpsd $0, %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * cmpsd $0, (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 comisd %xmm0, %xmm1
+# CHECK-NEXT: 1 8 1.00 * comisd (%rax), %xmm1
+# CHECK-NEXT: 1 3 1.00 cvtdq2pd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvtdq2pd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvtdq2ps %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvtdq2ps (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvtpd2dq %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvtpd2dq (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvtpd2ps %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvtpd2ps (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvtps2dq %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvtps2dq (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvtps2pd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvtps2pd (%rax), %xmm2
+# CHECK-NEXT: 2 7 1.00 cvtsd2si %xmm0, %ecx
+# CHECK-NEXT: 2 7 1.00 cvtsd2si %xmm0, %rcx
+# CHECK-NEXT: 2 12 1.00 * cvtsd2si (%rax), %ecx
+# CHECK-NEXT: 2 12 1.00 * cvtsd2si (%rax), %rcx
+# CHECK-NEXT: 2 7 2.00 cvtsd2ss %xmm0, %xmm2
+# CHECK-NEXT: 2 12 2.00 * cvtsd2ss (%rax), %xmm2
+# CHECK-NEXT: 2 9 1.00 cvtsi2sdl %ecx, %xmm2
+# CHECK-NEXT: 2 9 1.00 cvtsi2sdq %rcx, %xmm2
+# CHECK-NEXT: 2 14 1.00 * cvtsi2sdl (%rax), %xmm2
+# CHECK-NEXT: 2 14 1.00 * cvtsi2sdl (%rax), %xmm2
+# CHECK-NEXT: 2 7 2.00 cvtss2sd %xmm0, %xmm2
+# CHECK-NEXT: 2 12 2.00 * cvtss2sd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvttpd2dq %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvttpd2dq (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 cvttps2dq %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * cvttps2dq (%rax), %xmm2
+# CHECK-NEXT: 2 7 1.00 cvttsd2si %xmm0, %ecx
+# CHECK-NEXT: 2 7 1.00 cvttsd2si %xmm0, %rcx
+# CHECK-NEXT: 2 12 1.00 * cvttsd2si (%rax), %ecx
+# CHECK-NEXT: 2 12 1.00 * cvttsd2si (%rax), %rcx
+# CHECK-NEXT: 1 19 19.00 divpd %xmm0, %xmm2
+# CHECK-NEXT: 1 24 19.00 * divpd (%rax), %xmm2
+# CHECK-NEXT: 1 19 19.00 divsd %xmm0, %xmm2
+# CHECK-NEXT: 1 24 19.00 * divsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 1.00 * * * maskmovdqu %xmm0, %xmm1
+# CHECK-NEXT: 1 2 1.00 maxpd %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * maxpd (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 maxsd %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * maxsd (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 minpd %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * minpd (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 minsd %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * minsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movapd %xmm0, %xmm2
+# CHECK-NEXT: 1 1 1.00 * movapd %xmm0, (%rax)
+# CHECK-NEXT: 1 5 1.00 * movapd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movd %eax, %xmm2
+# CHECK-NEXT: 1 5 1.00 * movd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movd %xmm0, %ecx
+# CHECK-NEXT: 1 1 1.00 * movd %xmm0, (%rax)
+# CHECK-NEXT: 1 1 0.50 movdqa %xmm0, %xmm2
+# CHECK-NEXT: 1 1 1.00 * movdqa %xmm0, (%rax)
+# CHECK-NEXT: 1 5 1.00 * movdqa (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movdqu %xmm0, %xmm2
+# CHECK-NEXT: 1 1 1.00 * movdqu %xmm0, (%rax)
+# CHECK-NEXT: 1 5 1.00 * movdqu (%rax), %xmm2
+# CHECK-NEXT: 1 1 1.00 * movhpd %xmm0, (%rax)
+# CHECK-NEXT: 1 6 1.00 * movhpd (%rax), %xmm2
+# CHECK-NEXT: 1 1 1.00 * movlpd %xmm0, (%rax)
+# CHECK-NEXT: 1 6 1.00 * movlpd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 movmskpd %xmm0, %ecx
+# CHECK-NEXT: 1 2 1.00 * movntdq %xmm0, (%rax)
+# CHECK-NEXT: 1 3 1.00 * movntpd %xmm0, (%rax)
+# CHECK-NEXT: 1 1 0.50 movq %xmm0, %xmm2
+# CHECK-NEXT: 1 1 0.50 movq %rax, %xmm2
+# CHECK-NEXT: 1 5 1.00 * movq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movq %xmm0, %rcx
+# CHECK-NEXT: 1 1 1.00 * movq %xmm0, (%rax)
+# CHECK-NEXT: 1 1 0.50 movsd %xmm0, %xmm2
+# CHECK-NEXT: 1 1 1.00 * movsd %xmm0, (%rax)
+# CHECK-NEXT: 1 5 1.00 * movsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movupd %xmm0, %xmm2
+# CHECK-NEXT: 1 1 1.00 * movupd %xmm0, (%rax)
+# CHECK-NEXT: 1 5 1.00 * movupd (%rax), %xmm2
+# CHECK-NEXT: 1 4 2.00 mulpd %xmm0, %xmm2
+# CHECK-NEXT: 1 9 2.00 * mulpd (%rax), %xmm2
+# CHECK-NEXT: 1 4 2.00 mulsd %xmm0, %xmm2
+# CHECK-NEXT: 1 9 2.00 * mulsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 orpd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * orpd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pabsb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pabsb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pabsd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pabsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pabsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pabsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 packssdw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * packssdw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 packsswb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * packsswb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 packuswb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * packuswb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddsb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddsb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddusb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddusb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddusw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddusw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 paddw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * paddw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pand %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pand (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pandn %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pandn (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pavgb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pavgb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pavgw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pavgw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpeqb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpeqb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpeqd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpeqd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpeqw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpeqw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpgtb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpgtb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpgtd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpgtd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpgtw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpgtw (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 pextrw $1, %xmm0, %ecx
+# CHECK-NEXT: 1 2 1.00 pmaddwd %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmaddwd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmaxsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmaxsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmaxub %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmaxub (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pminsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pminsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pminub %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pminub (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 pmovmskb %xmm0, %ecx
+# CHECK-NEXT: 1 2 1.00 pmulhuw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmulhuw (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 pmulhw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmulhw (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 pmullw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmullw (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 pmuludq %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmuludq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 por %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * por (%rax), %xmm2
+# CHECK-NEXT: 1 2 0.50 psadbw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * psadbw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pshufd $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pshufd $1, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pshufhw $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pshufhw $1, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pshuflw $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pshuflw $1, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pslld $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 pslld %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pslld (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pslldq $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psllq $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psllq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psllq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psllw $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psllw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psllw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psrad $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psrad %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psrad (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psraw $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psraw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psraw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psrld $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psrld %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psrld (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psrldq $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psrlq $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psrlq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psrlq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psrlw $1, %xmm2
+# CHECK-NEXT: 1 1 0.50 psrlw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psrlw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubsb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubsb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubusb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubusb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubusw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubusw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psubw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psubw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpckhbw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpckhbw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpckhdq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpckhdq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpckhqdq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpckhqdq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpckhwd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpckhwd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpcklbw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpcklbw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpckldq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpckldq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpcklqdq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpcklqdq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 punpcklwd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * punpcklwd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pxor %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pxor (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 shufpd $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * shufpd $1, (%rax), %xmm2
+# CHECK-NEXT: 1 27 27.00 sqrtpd %xmm0, %xmm2
+# CHECK-NEXT: 1 32 27.00 * sqrtpd (%rax), %xmm2
+# CHECK-NEXT: 1 27 27.00 sqrtsd %xmm0, %xmm2
+# CHECK-NEXT: 1 32 27.00 * sqrtsd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 subpd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * subpd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 subsd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * subsd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 ucomisd %xmm0, %xmm1
+# CHECK-NEXT: 1 8 1.00 * ucomisd (%rax), %xmm1
+# CHECK-NEXT: 1 1 0.50 unpckhpd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * unpckhpd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 unpcklpd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * unpcklpd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 xorpd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * xorpd (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: 17.00 2.00 - 47.50 202.50 115.00 127.00 114.00 - 12.00 43.00 65.50 65.50 10.00
-# CHECK: Resource pressure by instruction:
+# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - addpd %xmm0, %xmm2
# CHECK-NEXT: - - - 1.00 - 1.00 - 1.00 - - - - - - addpd (%rax), %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 1.00 - - - - - - unpcklpd (%rax), %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 - - - - - - - xorpd %xmm0, %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 1.00 - - - - - - xorpd (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
addsubpd %xmm0, %xmm2
movsldup %xmm0, %xmm2
movsldup (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 3 1.00 addsubpd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * addsubpd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 addsubps %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * addsubps (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 haddpd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * haddpd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 haddps %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * haddps (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 hsubpd %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * hsubpd (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 hsubps %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * hsubps (%rax), %xmm2
+# CHECK-NEXT: 1 5 1.00 * lddqu (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movddup %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * movddup (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movshdup %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * movshdup (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 movsldup %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * movsldup (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - 15.00 3.00 15.50 3.50 10.00 - - - 0.50 0.50 -
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - addsubpd %xmm0, %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 1.00 - - - - - - movshdup (%rax), %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 - - - - - - - movsldup %xmm0, %xmm2
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 1.00 - - - - - - movsldup (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
blendpd $11, %xmm0, %xmm2
roundss $1, %xmm0, %xmm2
roundss $1, (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 1 0.50 blendpd $11, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * blendpd $11, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 blendps $11, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * blendps $11, (%rax), %xmm2
+# CHECK-NEXT: 3 2 2.00 blendvpd %xmm0, %xmm0, %xmm2
+# CHECK-NEXT: 3 7 2.00 * blendvpd %xmm0, (%rax), %xmm2
+# CHECK-NEXT: 3 2 2.00 blendvps %xmm0, %xmm0, %xmm2
+# CHECK-NEXT: 3 7 2.00 * blendvps %xmm0, (%rax), %xmm2
+# CHECK-NEXT: 3 9 3.00 dppd $22, %xmm0, %xmm2
+# CHECK-NEXT: 3 14 3.00 * dppd $22, (%rax), %xmm2
+# CHECK-NEXT: 5 11 3.00 dpps $22, %xmm0, %xmm2
+# CHECK-NEXT: 5 16 3.00 * dpps $22, (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 extractps $1, %xmm0, %ecx
+# CHECK-NEXT: 1 3 1.00 * extractps $1, %xmm0, (%rax)
+# CHECK-NEXT: 1 1 0.50 insertps $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * insertps $1, (%rax), %xmm2
+# CHECK-NEXT: 1 5 1.00 * movntdqa (%rax), %xmm2
+# CHECK-NEXT: 1 3 2.00 mpsadbw $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 8 2.00 * mpsadbw $1, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 packusdw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * packusdw (%rax), %xmm2
+# CHECK-NEXT: 3 2 2.00 pblendvb %xmm0, %xmm0, %xmm2
+# CHECK-NEXT: 3 7 2.00 * pblendvb %xmm0, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pblendw $11, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pblendw $11, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpeqq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpeqq (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 pextrb $1, %xmm0, %ecx
+# CHECK-NEXT: 1 3 1.00 * pextrb $1, %xmm0, (%rax)
+# CHECK-NEXT: 1 3 1.00 pextrd $1, %xmm0, %ecx
+# CHECK-NEXT: 1 3 1.00 * pextrd $1, %xmm0, (%rax)
+# CHECK-NEXT: 1 3 1.00 pextrq $1, %xmm0, %rcx
+# CHECK-NEXT: 1 3 1.00 * pextrq $1, %xmm0, (%rax)
+# CHECK-NEXT: 1 3 1.00 * pextrw $1, %xmm0, (%rax)
+# CHECK-NEXT: 1 2 1.00 phminposuw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * phminposuw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pinsrb $1, %eax, %xmm1
+# CHECK-NEXT: 1 6 1.00 * pinsrb $1, (%rax), %xmm1
+# CHECK-NEXT: 1 1 0.50 pinsrd $1, %eax, %xmm1
+# CHECK-NEXT: 1 6 1.00 * pinsrd $1, (%rax), %xmm1
+# CHECK-NEXT: 1 1 0.50 pinsrq $1, %rax, %xmm1
+# CHECK-NEXT: 1 6 1.00 * pinsrq $1, (%rax), %xmm1
+# CHECK-NEXT: 1 1 0.50 pmaxsb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmaxsb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmaxsd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmaxsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmaxud %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmaxud (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmaxuw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmaxuw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pminsb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pminsb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pminsd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pminsd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pminud %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pminud (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pminuw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pminuw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovsxbd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovsxbd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovsxbq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovsxbq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovsxbw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovsxbw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovsxdq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovsxdq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovsxwd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovsxwd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovsxwq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovsxwq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovzxbd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovzxbd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovzxbq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovzxbq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovzxbw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovzxbw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovzxdq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovzxdq (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovzxwd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovzxwd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pmovzxwq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pmovzxwq (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 pmuldq %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmuldq (%rax), %xmm2
+# CHECK-NEXT: 3 4 2.00 pmulld %xmm0, %xmm2
+# CHECK-NEXT: 3 9 2.00 * pmulld (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 ptest %xmm0, %xmm1
+# CHECK-NEXT: 1 8 1.00 * ptest (%rax), %xmm1
+# CHECK-NEXT: 1 3 1.00 roundpd $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * roundpd $1, (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 roundps $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * roundps $1, (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 roundsd $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * roundsd $1, (%rax), %xmm2
+# CHECK-NEXT: 1 3 1.00 roundss $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 8 1.00 * roundss $1, (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: 6.00 - - 37.00 23.00 57.50 42.50 44.00 - 5.00 5.00 31.50 31.50 12.00
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - 0.50 0.50 0.50 0.50 - - - - - - - blendpd $11, %xmm0, %xmm2
# CHECK-NEXT: - - - 1.00 - 1.00 - 1.00 - - - - - - roundsd $1, (%rax), %xmm2
# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - roundss $1, %xmm0, %xmm2
# CHECK-NEXT: - - - 1.00 - 1.00 - 1.00 - - - - - - roundss $1, (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
crc32b %al, %ecx
pcmpgtq %xmm0, %xmm2
pcmpgtq (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 3 3 2.00 crc32b %al, %ecx
+# CHECK-NEXT: 3 6 2.00 * crc32b (%rax), %ecx
+# CHECK-NEXT: 3 3 2.00 crc32l %eax, %ecx
+# CHECK-NEXT: 3 6 2.00 * crc32l (%rax), %ecx
+# CHECK-NEXT: 3 3 2.00 crc32w %ax, %ecx
+# CHECK-NEXT: 3 6 2.00 * crc32w (%rax), %ecx
+# CHECK-NEXT: 3 3 2.00 crc32b %al, %rcx
+# CHECK-NEXT: 3 6 2.00 * crc32b (%rax), %rcx
+# CHECK-NEXT: 3 3 2.00 crc32q %rax, %rcx
+# CHECK-NEXT: 3 6 2.00 * crc32q (%rax), %rcx
+# CHECK-NEXT: 9 14 5.00 pcmpestri $1, %xmm0, %xmm2
+# CHECK-NEXT: 9 19 5.00 * pcmpestri $1, (%rax), %xmm2
+# CHECK-NEXT: 9 14 5.00 pcmpestrm $1, %xmm0, %xmm2
+# CHECK-NEXT: 9 19 5.00 * pcmpestrm $1, (%rax), %xmm2
+# CHECK-NEXT: 3 7 2.00 pcmpistri $1, %xmm0, %xmm2
+# CHECK-NEXT: 3 12 2.00 * pcmpistri $1, (%rax), %xmm2
+# CHECK-NEXT: 3 8 2.00 pcmpistrm $1, %xmm0, %xmm2
+# CHECK-NEXT: 3 13 2.00 * pcmpistrm $1, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 pcmpgtq %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * pcmpgtq (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: 28.00 20.00 - 8.00 - 1.00 9.00 18.00 - 8.00 - 13.00 37.00 -
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: 2.00 2.00 - - - - - - - - - - - - crc32b %al, %ecx
# CHECK-NEXT: 1.00 - - 1.00 - - 1.00 1.00 - - - - 2.00 - pcmpistrm $1, (%rax), %xmm2
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 0.50 0.50 - pcmpgtq %xmm0, %xmm2
# CHECK-NEXT: - - - - - 0.50 0.50 1.00 - - - 0.50 0.50 - pcmpgtq (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
extrq %xmm0, %xmm2
movntsd %xmm0, (%rax)
movntss %xmm0, (%rax)
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 1 0.50 extrq %xmm0, %xmm2
+# CHECK-NEXT: 1 1 0.50 extrq $22, $2, %xmm2
+# CHECK-NEXT: 1 2 2.00 insertq %xmm0, %xmm2
+# CHECK-NEXT: 1 2 2.00 insertq $22, $22, %xmm0, %xmm2
+# CHECK-NEXT: 1 3 1.00 * movntsd %xmm0, (%rax)
+# CHECK-NEXT: 1 3 1.00 * movntss %xmm0, (%rax)
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - - - 2.00 4.00 - - 2.00 2.00 5.00 5.00 -
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 0.50 0.50 - extrq %xmm0, %xmm2
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 2.00 2.00 - insertq $22, $22, %xmm0, %xmm2
# CHECK-NEXT: - - - - - - 1.00 - - 1.00 1.00 - - - movntsd %xmm0, (%rax)
# CHECK-NEXT: - - - - - - 1.00 - - 1.00 1.00 - - - movntss %xmm0, (%rax)
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -instruction-tables < %s | FileCheck %s
palignr $1, %xmm0, %xmm2
psignw %xmm0, %xmm2
psignw (%rax), %xmm2
+# CHECK: Instruction Info:
+# CHECK-NEXT: [1]: #uOps
+# CHECK-NEXT: [2]: Latency
+# CHECK-NEXT: [3]: RThroughput
+# CHECK-NEXT: [4]: MayLoad
+# CHECK-NEXT: [5]: MayStore
+# CHECK-NEXT: [6]: HasSideEffects
+
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 1 0.50 palignr $1, %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * palignr $1, (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 phaddd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * phaddd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 phaddsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * phaddsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 phaddw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * phaddw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 phsubd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * phsubd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 phsubsw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * phsubsw (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 phsubw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * phsubw (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 pmaddubsw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmaddubsw (%rax), %xmm2
+# CHECK-NEXT: 1 2 1.00 pmulhrsw %xmm0, %xmm2
+# CHECK-NEXT: 1 7 1.00 * pmulhrsw (%rax), %xmm2
+# CHECK-NEXT: 3 2 2.00 pshufb %xmm0, %xmm2
+# CHECK-NEXT: 3 7 2.00 * pshufb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psignb %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psignb (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psignd %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psignd (%rax), %xmm2
+# CHECK-NEXT: 1 1 0.50 psignw %xmm0, %xmm2
+# CHECK-NEXT: 1 6 1.00 * psignw (%rax), %xmm2
+
# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
+# CHECK: Resource pressure per iteration:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: - - - - - 15.00 11.00 13.00 - - - 14.00 14.00 4.00
+
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 0.50 0.50 - palignr $1, %xmm0, %xmm2
# CHECK-NEXT: - - - - - 0.50 0.50 1.00 - - - 0.50 0.50 - psignd (%rax), %xmm2
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 0.50 0.50 - psignw %xmm0, %xmm2
# CHECK-NEXT: - - - - - 0.50 0.50 1.00 - - - 0.50 0.50 - psignw (%rax), %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -scheduler-stats < %s | FileCheck %s
vmulps (%rsi), %xmm0, %xmm0
# CHECK-NEXT: [13] - JVIMUL
# CHECK: Resource pressure per iteration:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
# CHECK-NEXT: - 1.00 - - 1.00 - 1.00 1.00 - - - - - -
# CHECK: Resource pressure by instruction:
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=100 < %s | FileCheck %s
add %edi, %eax
-# CHECK: Iterations: 100
+# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 100
# CHECK-NEXT: Total Cycles: 103
# CHECK-NEXT: Dispatch Width: 2
# CHECK-NEXT: IPC: 0.97
-
-# CHECK: Instruction Info:
+# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: [3]: RThroughput
# CHECK-NEXT: [5]: MayStore
# CHECK-NEXT: [6]: HasSideEffects
-# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 1 0.50 addl %edi, %eax
-
+# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
+# CHECK-NEXT: 1 1 0.50 addl %edi, %eax
-# CHECK-LABEL: Resources:
+# CHECK: Resources:
# CHECK-NEXT: [0] - JALU0
# CHECK-NEXT: [1] - JALU1
# CHECK-NEXT: [2] - JDiv
# CHECK-NEXT: [12] - JVALU1
# CHECK-NEXT: [13] - JVIMUL
-
-# CHECK: Resource pressure per iteration:
+# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
# CHECK-NEXT: 0.50 0.50 - - - - - - - - - - - -
-# CHECK: Resource pressure by instruction:
-# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
-# CHECK-NEXT: 0.50 0.50 - - - - - - - - - - - - addl %edi, %eax
+# CHECK: Resource pressure by instruction:
+# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
+# CHECK-NEXT: 0.50 0.50 - - - - - - - - - - - - addl %edi, %eax
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s
vaddps %xmm0, %xmm0, %xmm1
vandps (%rdi), %xmm1, %xmm2
+# CHECK: Iterations: 1
+# CHECK-NEXT: Instructions: 2
+# CHECK-NEXT: Total Cycles: 9
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.22
+
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: 1 3 1.00 vaddps %xmm0, %xmm0, %xmm1
# CHECK-NEXT: 1 6 1.00 * vandps (%rdi), %xmm1, %xmm2
-
# CHECK: Timeline view:
+
# CHECK: Index 012345678
+
# CHECK: [0,0] DeeeER . vaddps %xmm0, %xmm0, %xmm1
# CHECK-NEXT: [0,1] DeeeeeeER vandps (%rdi), %xmm1, %xmm2
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vaddps %xmm0, %xmm0, %xmm1
+# CHECK-NEXT: 1. 1 1.0 0.0 0.0 vandps (%rdi), %xmm1, %xmm2
+
+# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s
vaddps %ymm0, %ymm0, %ymm1
vandps (%rdi), %ymm1, %ymm2
+# CHECK: Iterations: 1
+# CHECK-NEXT: Instructions: 2
+# CHECK-NEXT: Total Cycles: 10
+# CHECK-NEXT: Dispatch Width: 2
+# CHECK-NEXT: IPC: 0.20
+
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
# CHECK-NEXT: [2]: Latency
# CHECK-NEXT: 2 3 2.00 vaddps %ymm0, %ymm0, %ymm1
# CHECK-NEXT: 2 6 2.00 * vandps (%rdi), %ymm1, %ymm2
-
# CHECK: Timeline view:
+
# CHECK: Index 0123456789
+
# CHECK: [0,0] DeeeER . vaddps %ymm0, %ymm0, %ymm1
# CHECK-NEXT: [0,1] .DeeeeeeER vandps (%rdi), %ymm1, %ymm2
+
+# CHECK: Average Wait times (based on the timeline view):
+# CHECK-NEXT: [0]: Executions
+# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
+
+# CHECK: [0] [1] [2] [3]
+# CHECK-NEXT: 0. 1 1.0 1.0 0.0 vaddps %ymm0, %ymm0, %ymm1
+# CHECK-NEXT: 1. 1 1.0 1.0 0.0 vandps (%rdi), %ymm1, %ymm2
+
--- /dev/null
+#!/usr/bin/env python2.7
+
+"""A test case update script.
+
+This script is a utility to update LLVM 'llvm-mca' based test cases with new
+FileCheck patterns.
+"""
+
+import argparse
+from collections import defaultdict
+import difflib
+import glob
+import os
+import sys
+import warnings
+
+from UpdateTestChecks import common
+
+
+COMMENT_CHAR = '#'
+ADVERT_PREFIX = '{} NOTE: Assertions have been autogenerated by '.format(
+ COMMENT_CHAR)
+ADVERT = '{}utils/{}'.format(ADVERT_PREFIX, os.path.basename(__file__))
+
+
+class Error(Exception):
+ """ Generic Error to be raised without printing a traceback.
+ """
+ pass
+
+
+def _warn(msg):
+ """ Log a user warning to stderr.
+ """
+ warnings.warn(msg, Warning, stacklevel=2)
+
+
+def _configure_warnings(args):
+ warnings.resetwarnings()
+ if args.w:
+ warnings.simplefilter('ignore')
+ if args.Werror:
+ warnings.simplefilter('error')
+
+
+def _showwarning(message, category, filename, lineno, file=None, line=None):
+ """ Version of warnings.showwarning that won't attempt to print out the
+ line at the location of the warning if the line text is not explicitly
+ specified.
+ """
+ if file is None:
+ file = sys.stderr
+ if line is None:
+ line = ''
+ file.write(warnings.formatwarning(message, category, filename, lineno, line))
+
+
+def _parse_args():
+ parser = argparse.ArgumentParser(description=__doc__)
+ parser.add_argument('-v', '--verbose',
+ action='store_true',
+ help='show verbose output')
+ parser.add_argument('-w',
+ action='store_true',
+ help='suppress warnings')
+ parser.add_argument('-Werror',
+ action='store_true',
+ help='promote warnings to errors')
+ parser.add_argument('--llvm-mca-binary',
+ metavar='<path>',
+ default='llvm-mca',
+ help='the binary to use to generate the test case '
+ '(default: llvm-mca)')
+ parser.add_argument('tests',
+ metavar='<test-path>',
+ nargs='+')
+ args = parser.parse_args()
+
+ _configure_warnings(args)
+
+ if os.path.basename(args.llvm_mca_binary) != 'llvm-mca':
+ _warn('unexpected binary name: {}'.format(args.llvm_mca_binary))
+
+ return args
+
+
+def _find_run_lines(input_lines, args):
+ raw_lines = [m.group(1)
+ for m in [common.RUN_LINE_RE.match(l) for l in input_lines]
+ if m]
+ run_lines = [raw_lines[0]] if len(raw_lines) > 0 else []
+ for l in raw_lines[1:]:
+ if run_lines[-1].endswith(r'\\'):
+ run_lines[-1] = run_lines[-1].rstrip('\\') + ' ' + l
+ else:
+ run_lines.append(l)
+
+ if args.verbose:
+ sys.stderr.write('Found {} RUN line{}:\n'.format(
+ len(run_lines), '' if len(run_lines) == 1 else 's'))
+ for line in run_lines:
+ sys.stderr.write(' RUN: {}\n'.format(line))
+
+ return run_lines
+
+
+def _get_run_infos(run_lines, args):
+ run_infos = []
+ for run_line in run_lines:
+ try:
+ (tool_cmd, filecheck_cmd) = tuple([cmd.strip()
+ for cmd in run_line.split('|', 1)])
+ except ValueError:
+ _warn('could not split tool and filecheck commands: {}'.format(run_line))
+ continue
+
+ tool_basename = os.path.basename(args.llvm_mca_binary)
+
+ if not tool_cmd.startswith(tool_basename + ' '):
+ _warn('skipping non-{} RUN line: {}'.format(tool_basename, run_line))
+ continue
+
+ if not filecheck_cmd.startswith('FileCheck '):
+ _warn('skipping non-FileCheck RUN line: {}'.format(run_line))
+ continue
+
+ tool_cmd_args = tool_cmd[len(tool_basename):].strip()
+ tool_cmd_args = tool_cmd_args.replace('< %s', '').replace('%s', '').strip()
+
+ check_prefixes = [item
+ for m in common.CHECK_PREFIX_RE.finditer(filecheck_cmd)
+ for item in m.group(1).split(',')]
+ if not check_prefixes:
+ check_prefixes = ['CHECK']
+
+ run_infos.append((check_prefixes, tool_cmd_args))
+
+ return run_infos
+
+
+def _get_block_infos(run_infos, test_path, args): # noqa
+ """ For each run line, run the tool with the specified args and collect the
+ output. We use the concept of 'blocks' for uniquing, where a block is
+ a series of lines of text with no more than one newline character between
+ each one. For example:
+
+ This
+ is
+ one
+ block
+
+ This is
+ another block
+
+ This is yet another block
+
+ We then build up a 'block_infos' structure containing a dict where the
+ text of each block is the key and a list of the sets of prefixes that may
+ generate that particular block. This then goes through a series of
+ transformations to minimise the amount of CHECK lines that need to be
+ written by taking advantage of common prefixes.
+ """
+
+ def _block_key(tool_args, prefixes):
+ """ Get a hashable key based on the current tool_args and prefixes.
+ """
+ return ' '.join([tool_args] + prefixes)
+
+ all_blocks = {}
+ max_block_len = 0
+
+ # Run the tool for each run line to generate all of the blocks.
+ for prefixes, tool_args in run_infos:
+ key = _block_key(tool_args, prefixes)
+ raw_tool_output = common.invoke_tool(args.llvm_mca_binary,
+ tool_args,
+ test_path)
+
+ # Replace any lines consisting of purely whitespace with empty lines.
+ raw_tool_output = '\n'.join(line if line.strip() else ''
+ for line in raw_tool_output.splitlines())
+
+ # Split blocks, stripping all trailing whitespace, but keeping preceding
+ # whitespace except for newlines so that columns will line up visually.
+ all_blocks[key] = [b.lstrip('\n').rstrip()
+ for b in raw_tool_output.split('\n\n')]
+ max_block_len = max(max_block_len, len(all_blocks[key]))
+
+ # If necessary, pad the lists of blocks with empty blocks so that they are
+ # all the same length.
+ for key in all_blocks:
+ len_to_pad = max_block_len - len(all_blocks[key])
+ all_blocks[key] += [''] * len_to_pad
+
+ # Create the block_infos structure where it is a nested dict in the form of:
+ # block number -> block text -> list of prefix sets
+ block_infos = defaultdict(lambda: defaultdict(list))
+ for prefixes, tool_args in run_infos:
+ key = _block_key(tool_args, prefixes)
+ for block_num, block_text in enumerate(all_blocks[key]):
+ block_infos[block_num][block_text].append(set(prefixes))
+
+ # Now go through the block_infos structure and attempt to smartly prune the
+ # number of prefixes per block to the minimal set possible to output.
+ for block_num in range(len(block_infos)):
+
+ # When there are multiple block texts for a block num, remove any
+ # prefixes that are common to more than one of them.
+ # E.g. [ [{ALL,FOO}] , [{ALL,BAR}] ] -> [ [{FOO}] , [{BAR}] ]
+ all_sets = [s for s in block_infos[block_num].values()]
+ pruned_sets = []
+
+ for i, setlist in enumerate(all_sets):
+ other_set_values = set([elem for j, setlist2 in enumerate(all_sets)
+ for set_ in setlist2 for elem in set_
+ if i != j])
+ pruned_sets.append([s - other_set_values for s in setlist])
+
+ for i, block_text in enumerate(block_infos[block_num]):
+
+ # When a block text matches multiple sets of prefixes, try removing any
+ # prefixes that aren't common to all of them.
+ # E.g. [ {ALL,FOO} , {ALL,BAR} ] -> [{ALL}]
+ common_values = pruned_sets[i][0].copy()
+ for s in pruned_sets[i][1:]:
+ common_values &= s
+ if common_values:
+ pruned_sets[i] = [common_values]
+
+ # Everything should be uniqued as much as possible by now. Apply the
+ # newly pruned sets to the block_infos structure.
+ # If there are any blocks of text that still match multiple prefixes,
+ # output a warning.
+ current_set = set()
+ for s in pruned_sets[i]:
+ s = sorted(list(s))
+ if s:
+ current_set.add(s[0])
+ if len(s) > 1:
+ _warn('Multiple prefixes generating same output: {} '
+ '(discarding {})'.format(','.join(s), ','.join(s[1:])))
+
+ block_infos[block_num][block_text] = sorted(list(current_set))
+
+ return block_infos
+
+
+def _write_output(test_path, input_lines, prefix_list, block_infos, # noqa
+ args):
+ prefix_set = set([prefix for prefixes, _ in prefix_list
+ for prefix in prefixes])
+ not_prefix_set = set()
+
+ output_lines = []
+ for input_line in input_lines:
+ if input_line.startswith(ADVERT_PREFIX):
+ continue
+
+ if input_line.startswith(COMMENT_CHAR):
+ m = common.CHECK_RE.match(input_line)
+ try:
+ prefix = m.group(1)
+ except AttributeError:
+ prefix = None
+
+ if '{}-NOT:'.format(prefix) in input_line:
+ not_prefix_set.add(prefix)
+
+ if prefix not in prefix_set or prefix in not_prefix_set:
+ output_lines.append(input_line)
+ continue
+
+ if common.should_add_line_to_output(input_line, prefix_set):
+ # This input line of the function body will go as-is into the output.
+ # Except make leading whitespace uniform: 2 spaces.
+ input_line = common.SCRUB_LEADING_WHITESPACE_RE.sub(r' ', input_line)
+
+ # Skip empty lines if the previous output line is also empty.
+ if input_line or output_lines[-1]:
+ output_lines.append(input_line)
+ else:
+ continue
+
+ # Add a blank line before the new checks if required.
+ if output_lines[-1]:
+ output_lines.append('')
+
+ output_check_lines = []
+ for block_num in range(len(block_infos)):
+ for block_text in sorted(block_infos[block_num]):
+ if not block_text:
+ continue
+
+ if block_infos[block_num][block_text]:
+ lines = block_text.split('\n')
+ for prefix in block_infos[block_num][block_text]:
+ if prefix in not_prefix_set:
+ _warn('not writing for prefix {0} due to presence of "{0}-NOT:" '
+ 'in input file.'.format(prefix))
+ continue
+
+ output_check_lines.append(
+ '{} {}: {}'.format(COMMENT_CHAR, prefix, lines[0]).rstrip())
+ for line in lines[1:]:
+ output_check_lines.append(
+ '{} {}-NEXT: {}'.format(COMMENT_CHAR, prefix, line).rstrip())
+ output_check_lines.append('')
+
+ if output_check_lines:
+ output_lines.insert(0, ADVERT)
+ output_lines.extend(output_check_lines)
+
+ if input_lines == output_lines:
+ sys.stderr.write(' [unchanged]\n')
+ return
+
+ diff = list(difflib.Differ().compare(input_lines, output_lines))
+ sys.stderr.write(
+ ' [{} lines total ({} added, {} removed)]\n'.format(
+ len(output_lines),
+ len([l for l in diff if l[0] == '+']),
+ len([l for l in diff if l[0] == '-'])))
+
+ if args.verbose:
+ sys.stderr.write(
+ 'Writing {} lines to {}...\n\n'.format(len(output_lines), test_path))
+
+ with open(test_path, 'wb') as f:
+ for line in output_lines:
+ f.write('{}\n'.format(line.rstrip()).encode())
+
+
+def main():
+ args = _parse_args()
+ test_paths = [test for pattern in args.tests for test in glob.glob(pattern)]
+ for test_path in test_paths:
+ sys.stderr.write('Test: {}\n'.format(test_path))
+
+ # Call this per test. By default each warning will only be written once
+ # per source location. Reset the warning filter so that now each warning
+ # will be written once per source location per test.
+ _configure_warnings(args)
+
+ if args.verbose:
+ sys.stderr.write(
+ 'Scanning for RUN lines in test file: {}\n'.format(test_path))
+
+ if not os.path.isfile(test_path):
+ raise Error('could not find test file: {}'.format(test_path))
+
+ with open(test_path) as f:
+ input_lines = [l.rstrip() for l in f]
+
+ run_lines = _find_run_lines(input_lines, args)
+ run_infos = _get_run_infos(run_lines, args)
+ block_infos = _get_block_infos(run_infos, test_path, args)
+ _write_output(test_path, input_lines, run_infos, block_infos, args)
+
+ return 0
+
+
+if __name__ == '__main__':
+ try:
+ warnings.showwarning = _showwarning
+ sys.exit(main())
+ except Error as e:
+ sys.stdout.write('error: {}\n'.format(e))
+ sys.exit(1)
projects / platform / upstream / llvm.git / commitdiff