1 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py 2 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -timeline -timeline-max-iterations=2 < %s | FileCheck %s 3 4 # VALU0/VALU1 5 vpmulld %xmm0, %xmm1, %xmm2 6 vpand %xmm0, %xmm1, %xmm2 7 8 # VIMUL/STC 9 vcvttps2dq %xmm0, %xmm2 10 vpclmulqdq $0, %xmm0, %xmm1, %xmm2 11 12 # FPA/FPM 13 vaddps %xmm0, %xmm1, %xmm2 14 vsqrtps %xmm0, %xmm2 15 16 # FPA/FPM YMM 17 vaddps %ymm0, %ymm1, %ymm2 18 vsqrtps %ymm0, %ymm2 19 20 # CHECK: Iterations: 100 21 # CHECK-NEXT: Instructions: 800 22 # CHECK-NEXT: Total Cycles: 6306 23 # CHECK-NEXT: Dispatch Width: 2 24 # CHECK-NEXT: IPC: 0.13 25 # CHECK-NEXT: Block RThroughput: 63.0 26 27 # CHECK: Instruction Info: 28 # CHECK-NEXT: [1]: #uOps 29 # CHECK-NEXT: [2]: Latency 30 # CHECK-NEXT: [3]: RThroughput 31 # CHECK-NEXT: [4]: MayLoad 32 # CHECK-NEXT: [5]: MayStore 33 # CHECK-NEXT: [6]: HasSideEffects (U) 34 35 # CHECK: [1] [2] [3] [4] [5] [6] Instructions: 36 # CHECK-NEXT: 3 4 2.00 vpmulld %xmm0, %xmm1, %xmm2 37 # CHECK-NEXT: 1 1 0.50 vpand %xmm0, %xmm1, %xmm2 38 # CHECK-NEXT: 1 3 1.00 vcvttps2dq %xmm0, %xmm2 39 # CHECK-NEXT: 1 2 1.00 vpclmulqdq $0, %xmm0, %xmm1, %xmm2 40 # CHECK-NEXT: 1 3 1.00 vaddps %xmm0, %xmm1, %xmm2 41 # CHECK-NEXT: 1 21 21.00 vsqrtps %xmm0, %xmm2 42 # CHECK-NEXT: 2 3 2.00 vaddps %ymm0, %ymm1, %ymm2 43 # CHECK-NEXT: 2 42 42.00 vsqrtps %ymm0, %ymm2 44 45 # CHECK: Resources: 46 # CHECK-NEXT: [0] - JALU0 47 # CHECK-NEXT: [1] - JALU1 48 # CHECK-NEXT: [2] - JDiv 49 # CHECK-NEXT: [3] - JFPA 50 # CHECK-NEXT: [4] - JFPM 51 # CHECK-NEXT: [5] - JFPU0 52 # CHECK-NEXT: [6] - JFPU1 53 # CHECK-NEXT: [7] - JLAGU 54 # CHECK-NEXT: [8] - JMul 55 # CHECK-NEXT: [9] - JSAGU 56 # CHECK-NEXT: [10] - JSTC 57 # CHECK-NEXT: [11] - JVALU0 58 # CHECK-NEXT: [12] - JVALU1 59 # CHECK-NEXT: [13] - JVIMUL 60 61 # CHECK: Resource pressure per iteration: 62 # CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] 63 # CHECK-NEXT: - - - 3.00 63.00 6.01 5.99 - - - 1.00 1.00 1.00 3.00 64 65 # CHECK: Resource pressure by instruction: 66 # CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions: 67 # CHECK-NEXT: - - - - - 2.00 1.00 - - - - 0.02 0.98 2.00 vpmulld %xmm0, %xmm1, %xmm2 68 # CHECK-NEXT: - - - - - 0.01 0.99 - - - - 0.98 0.02 - vpand %xmm0, %xmm1, %xmm2 69 # CHECK-NEXT: - - - - - - 1.00 - - - 1.00 - - - vcvttps2dq %xmm0, %xmm2 70 # CHECK-NEXT: - - - - - 1.00 - - - - - - - 1.00 vpclmulqdq $0, %xmm0, %xmm1, %xmm2 71 # CHECK-NEXT: - - - 1.00 - 1.00 - - - - - - - - vaddps %xmm0, %xmm1, %xmm2 72 # CHECK-NEXT: - - - - 21.00 - 1.00 - - - - - - - vsqrtps %xmm0, %xmm2 73 # CHECK-NEXT: - - - 2.00 - 2.00 - - - - - - - - vaddps %ymm0, %ymm1, %ymm2 74 # CHECK-NEXT: - - - - 42.00 - 2.00 - - - - - - - vsqrtps %ymm0, %ymm2 75 76 # CHECK: Timeline view: 77 # CHECK-NEXT: 0123456789 0123456789 0123456789 0 78 # CHECK-NEXT: Index 0123456789 0123456789 0123456789 0123456789 79 80 # CHECK: [0,0] DeeeeER . . . . . . . . . . . . . vpmulld %xmm0, %xmm1, %xmm2 81 # CHECK-NEXT: [0,1] .DeE--R . . . . . . . . . . . . . vpand %xmm0, %xmm1, %xmm2 82 # CHECK-NEXT: [0,2] . DeeeER . . . . . . . . . . . . . vcvttps2dq %xmm0, %xmm2 83 # CHECK-NEXT: [0,3] . DeeE-R . . . . . . . . . . . . . vpclmulqdq $0, %xmm0, %xmm1, %xmm2 84 # CHECK-NEXT: [0,4] . DeeeER . . . . . . . . . . . . . vaddps %xmm0, %xmm1, %xmm2 85 # CHECK-NEXT: [0,5] . DeeeeeeeeeeeeeeeeeeeeeER . . . . . . . . . vsqrtps %xmm0, %xmm2 86 # CHECK-NEXT: [0,6] . DeeeE-----------------R . . . . . . . . . vaddps %ymm0, %ymm1, %ymm2 87 # CHECK-NEXT: [0,7] . D===================eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeER . vsqrtps %ymm0, %ymm2 88 # CHECK-NEXT: [1,0] . .DeeeeE--------------------------------------------------------R . vpmulld %xmm0, %xmm1, %xmm2 89 # CHECK-NEXT: [1,1] . . DeE-----------------------------------------------------------R. vpand %xmm0, %xmm1, %xmm2 90 # CHECK-NEXT: [1,2] . . DeeeE--------------------------------------------------------R. vcvttps2dq %xmm0, %xmm2 91 # CHECK-NEXT: [1,3] . . DeeE----------------------------------------------------------R vpclmulqdq $0, %xmm0, %xmm1, %xmm2 92 # CHECK-NEXT: [1,4] . . DeeeE--------------------------------------------------------R vaddps %xmm0, %xmm1, %xmm2 93 94 # CHECK: Average Wait times (based on the timeline view): 95 # CHECK-NEXT: [0]: Executions 96 # CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue 97 # CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready 98 # CHECK-NEXT: [3]: Average time elapsed from WB until retire stage 99 100 # CHECK: [0] [1] [2] [3] 101 # CHECK-NEXT: 0. 2 1.0 1.0 28.0 vpmulld %xmm0, %xmm1, %xmm2 102 # CHECK-NEXT: 1. 2 1.0 1.0 30.5 vpand %xmm0, %xmm1, %xmm2 103 # CHECK-NEXT: 2. 2 1.0 1.0 28.0 vcvttps2dq %xmm0, %xmm2 104 # CHECK-NEXT: 3. 2 1.0 1.0 29.5 vpclmulqdq $0, %xmm0, %xmm1, %xmm2 105 # CHECK-NEXT: 4. 2 1.0 1.0 28.0 vaddps %xmm0, %xmm1, %xmm2 106 # CHECK-NEXT: 5. 1 1.0 1.0 0.0 vsqrtps %xmm0, %xmm2 107 # CHECK-NEXT: 6. 1 1.0 1.0 17.0 vaddps %ymm0, %ymm1, %ymm2 108 # CHECK-NEXT: 7. 1 20.0 20.0 0.0 vsqrtps %ymm0, %ymm2 109