114 lines · plain
1# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py2# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -timeline -timeline-max-iterations=2 < %s | FileCheck %s3 4# VALU0/VALU15vpmulld %xmm0, %xmm1, %xmm26vpand %xmm0, %xmm1, %xmm27 8# VIMUL/STC9vcvttps2dq %xmm0, %xmm210vpclmulqdq $0, %xmm0, %xmm1, %xmm211 12# FPA/FPM13vaddps %xmm0, %xmm1, %xmm214vsqrtps %xmm0, %xmm215 16# FPA/FPM YMM17vaddps %ymm0, %ymm1, %ymm218vsqrtps %ymm0, %ymm219 20# CHECK: Iterations: 10021# CHECK-NEXT: Instructions: 80022# CHECK-NEXT: Total Cycles: 630623# CHECK-NEXT: Total uOps: 120024 25# CHECK: Dispatch Width: 226# CHECK-NEXT: uOps Per Cycle: 0.1927# CHECK-NEXT: IPC: 0.1328# CHECK-NEXT: Block RThroughput: 63.029 30# CHECK: Instruction Info:31# CHECK-NEXT: [1]: #uOps32# CHECK-NEXT: [2]: Latency33# CHECK-NEXT: [3]: RThroughput34# CHECK-NEXT: [4]: MayLoad35# CHECK-NEXT: [5]: MayStore36# CHECK-NEXT: [6]: HasSideEffects (U)37 38# CHECK: [1] [2] [3] [4] [5] [6] Instructions:39# CHECK-NEXT: 3 4 2.00 vpmulld %xmm0, %xmm1, %xmm240# CHECK-NEXT: 1 1 0.50 vpand %xmm0, %xmm1, %xmm241# CHECK-NEXT: 1 3 1.00 vcvttps2dq %xmm0, %xmm242# CHECK-NEXT: 1 2 1.00 vpclmulqdq $0, %xmm0, %xmm1, %xmm243# CHECK-NEXT: 1 3 1.00 vaddps %xmm0, %xmm1, %xmm244# CHECK-NEXT: 1 21 21.00 vsqrtps %xmm0, %xmm245# CHECK-NEXT: 2 3 2.00 vaddps %ymm0, %ymm1, %ymm246# CHECK-NEXT: 2 42 42.00 vsqrtps %ymm0, %ymm247 48# CHECK: Resources:49# CHECK-NEXT: [0] - JALU050# CHECK-NEXT: [1] - JALU151# CHECK-NEXT: [2] - JDiv52# CHECK-NEXT: [3] - JFPA53# CHECK-NEXT: [4] - JFPM54# CHECK-NEXT: [5] - JFPU055# CHECK-NEXT: [6] - JFPU156# CHECK-NEXT: [7] - JLAGU57# CHECK-NEXT: [8] - JMul58# CHECK-NEXT: [9] - JSAGU59# CHECK-NEXT: [10] - JSTC60# CHECK-NEXT: [11] - JVALU061# CHECK-NEXT: [12] - JVALU162# CHECK-NEXT: [13] - JVIMUL63 64# CHECK: Resource pressure per iteration:65# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]66# CHECK-NEXT: - - - 3.00 63.00 6.01 5.99 - - - 1.00 1.00 1.00 3.0067 68# CHECK: Resource pressure by instruction:69# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:70# CHECK-NEXT: - - - - - 2.00 1.00 - - - - 0.02 0.98 2.00 vpmulld %xmm0, %xmm1, %xmm271# CHECK-NEXT: - - - - - 0.01 0.99 - - - - 0.98 0.02 - vpand %xmm0, %xmm1, %xmm272# CHECK-NEXT: - - - - - - 1.00 - - - 1.00 - - - vcvttps2dq %xmm0, %xmm273# CHECK-NEXT: - - - - - 1.00 - - - - - - - 1.00 vpclmulqdq $0, %xmm0, %xmm1, %xmm274# CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vaddps %xmm0, %xmm1, %xmm275# CHECK-NEXT: - - - - 21.00 - 1.00 - - - - - - - vsqrtps %xmm0, %xmm276# CHECK-NEXT: - - - 2.00 - 2.00 - - - - - - - - vaddps %ymm0, %ymm1, %ymm277# CHECK-NEXT: - - - - 42.00 - 2.00 - - - - - - - vsqrtps %ymm0, %ymm278 79# CHECK: Timeline view:80# CHECK-NEXT: 0123456789 0123456789 0123456789 081# CHECK-NEXT: Index 0123456789 0123456789 0123456789 012345678982 83# CHECK: [0,0] DeeeeER . . . . . . . . . . . . . vpmulld %xmm0, %xmm1, %xmm284# CHECK-NEXT: [0,1] .DeE--R . . . . . . . . . . . . . vpand %xmm0, %xmm1, %xmm285# CHECK-NEXT: [0,2] . DeeeER . . . . . . . . . . . . . vcvttps2dq %xmm0, %xmm286# CHECK-NEXT: [0,3] . DeeE-R . . . . . . . . . . . . . vpclmulqdq $0, %xmm0, %xmm1, %xmm287# CHECK-NEXT: [0,4] . DeeeER . . . . . . . . . . . . . vaddps %xmm0, %xmm1, %xmm288# CHECK-NEXT: [0,5] . DeeeeeeeeeeeeeeeeeeeeeER . . . . . . . . . vsqrtps %xmm0, %xmm289# CHECK-NEXT: [0,6] . DeeeE-----------------R . . . . . . . . . vaddps %ymm0, %ymm1, %ymm290# CHECK-NEXT: [0,7] . D===================eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeER . vsqrtps %ymm0, %ymm291# CHECK-NEXT: [1,0] . .DeeeeE--------------------------------------------------------R . vpmulld %xmm0, %xmm1, %xmm292# CHECK-NEXT: [1,1] . . DeE-----------------------------------------------------------R. vpand %xmm0, %xmm1, %xmm293# CHECK-NEXT: [1,2] . . DeeeE--------------------------------------------------------R. vcvttps2dq %xmm0, %xmm294# CHECK-NEXT: [1,3] . . DeeE----------------------------------------------------------R vpclmulqdq $0, %xmm0, %xmm1, %xmm295# CHECK-NEXT: [1,4] . . DeeeE--------------------------------------------------------R vaddps %xmm0, %xmm1, %xmm296# CHECK-NEXT: Truncated display due to cycle limit97 98# CHECK: Average Wait times (based on the timeline view):99# CHECK-NEXT: [0]: Executions100# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue101# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready102# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage103 104# CHECK: [0] [1] [2] [3]105# CHECK-NEXT: 0. 2 1.0 1.0 28.0 vpmulld %xmm0, %xmm1, %xmm2106# CHECK-NEXT: 1. 2 1.0 1.0 30.5 vpand %xmm0, %xmm1, %xmm2107# CHECK-NEXT: 2. 2 1.0 1.0 28.0 vcvttps2dq %xmm0, %xmm2108# CHECK-NEXT: 3. 2 1.0 1.0 29.5 vpclmulqdq $0, %xmm0, %xmm1, %xmm2109# CHECK-NEXT: 4. 2 1.0 1.0 28.0 vaddps %xmm0, %xmm1, %xmm2110# CHECK-NEXT: 5. 2 29.5 29.5 0.0 vsqrtps %xmm0, %xmm2111# CHECK-NEXT: 6. 2 1.0 1.0 45.5 vaddps %ymm0, %ymm1, %ymm2112# CHECK-NEXT: 7. 2 48.5 48.5 0.0 vsqrtps %ymm0, %ymm2113# CHECK-NEXT: 2 10.5 10.5 23.7 <total>114