test/tools/llvm-mca/X86/bzhi-read-after-ld.s

   1 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
   2 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
   3 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
   4 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
   5 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
   6
   7 add     %edi, %esi
   8 bzhil   %esi, (%rdi), %eax
   9
  10 # ALL:          Iterations:        1
  11 # ALL-NEXT:     Instructions:      2
  12
  13 # BDWELL-NEXT:  Total Cycles:      9
  14 # HASWELL-NEXT: Total Cycles:      9
  15 # SKYLAKE-NEXT: Total Cycles:      9
  16 # ZNVER1-NEXT:  Total Cycles:      8
  17
  18 # ALL-NEXT:     Total uOps:        3
  19
  20 # BDWELL:       Dispatch Width:    4
  21 # BDWELL-NEXT:  uOps Per Cycle:    0.33
  22 # BDWELL-NEXT:  IPC:               0.22
  23 # BDWELL-NEXT:  Block RThroughput: 0.8
  24
  25 # HASWELL:      Dispatch Width:    4
  26 # HASWELL-NEXT: uOps Per Cycle:    0.33
  27 # HASWELL-NEXT: IPC:               0.22
  28 # HASWELL-NEXT: Block RThroughput: 0.8
  29
  30 # SKYLAKE:      Dispatch Width:    6
  31 # SKYLAKE-NEXT: uOps Per Cycle:    0.33
  32 # SKYLAKE-NEXT: IPC:               0.22
  33 # SKYLAKE-NEXT: Block RThroughput: 0.5
  34
  35 # ZNVER1:       Dispatch Width:    4
  36 # ZNVER1-NEXT:  uOps Per Cycle:    0.38
  37 # ZNVER1-NEXT:  IPC:               0.25
  38 # ZNVER1-NEXT:  Block RThroughput: 0.8
  39
  40 # ALL:          Instruction Info:
  41 # ALL-NEXT:     [1]: #uOps
  42 # ALL-NEXT:     [2]: Latency
  43 # ALL-NEXT:     [3]: RThroughput
  44 # ALL-NEXT:     [4]: MayLoad
  45 # ALL-NEXT:     [5]: MayStore
  46 # ALL-NEXT:     [6]: HasSideEffects (U)
  47
  48 # ALL:          [1]    [2]    [3]    [4]    [5]    [6]    Instructions:
  49 # ALL-NEXT:      1      1     0.25                        addl  %edi, %esi
  50
  51 # BDWELL-NEXT:   2      6     0.50    *                   bzhil %esi, (%rdi), %eax
  52 # HASWELL-NEXT:  2      6     0.50    *                   bzhil %esi, (%rdi), %eax
  53 # SKYLAKE-NEXT:  2      6     0.50    *                   bzhil %esi, (%rdi), %eax
  54 # ZNVER1-NEXT:   2      5     0.50    *                   bzhil %esi, (%rdi), %eax
  55
  56 # ALL:          Timeline view:
  57
  58 # BDWELL-NEXT:  Index     012345678
  59 # HASWELL-NEXT: Index     012345678
  60 # SKYLAKE-NEXT: Index     012345678
  61 # ZNVER1-NEXT:  Index     01234567
  62
  63 # BDWELL:       [0,0]     DeER .  .   addl      %edi, %esi
  64 # BDWELL-NEXT:  [0,1]     DeeeeeeER   bzhil     %esi, (%rdi), %eax
  65
  66 # HASWELL:      [0,0]     DeER .  .   addl      %edi, %esi
  67 # HASWELL-NEXT: [0,1]     DeeeeeeER   bzhil     %esi, (%rdi), %eax
  68
  69 # SKYLAKE:      [0,0]     DeER .  .   addl      %edi, %esi
  70 # SKYLAKE-NEXT: [0,1]     DeeeeeeER   bzhil     %esi, (%rdi), %eax
  71
  72 # ZNVER1:       [0,0]     DeER . .   addl       %edi, %esi
  73 # ZNVER1-NEXT:  [0,1]     DeeeeeER   bzhil      %esi, (%rdi), %eax
  74
  75 # ALL:          Average Wait times (based on the timeline view):
  76 # ALL-NEXT:     [0]: Executions
  77 # ALL-NEXT:     [1]: Average time spent waiting in a scheduler's queue
  78 # ALL-NEXT:     [2]: Average time spent waiting in a scheduler's queue while ready
  79 # ALL-NEXT:     [3]: Average time elapsed from WB until retire stage
  80
  81 # ALL:                [0]    [1]    [2]    [3]
  82 # ALL-NEXT:     0.     1     1.0    1.0    0.0       addl       %edi, %esi
  83 # ALL-NEXT:     1.     1     1.0    0.0    0.0       bzhil      %esi, (%rdi), %eax
  84 # ALL-NEXT:            1     1.0    0.5    0.0       <total>