1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-25 04:02:41 +01:00
llvm-mirror/test/tools/llvm-mca/X86/BdVer2/rcu-statistics.s
Roman Lebedev 49a3c934c1 [X86] AMD Piledriver (BdVer2): major cleanup (mainly inverse throughput)
I've started this cleanup more several times now, but got sidetracked
elsewhere, e.g. by llvm-exegesis problems. Not this time, finally!

This is mainly cleaning up the inverse throughput values,
and a few latencies/uops, based on the llvm-exegesis measured values.

Though this is not complete by any means,
there's certainly more cleanup to be done.

The performance numbers (i've only checked by RawSpeed benchmark) aren't
really surprising - overall this *slightly* (< -1%) improves perf.

llvm-svn: 360341
2019-05-09 13:54:51 +00:00

67 lines
3.0 KiB
ArmAsm

# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -resource-pressure=false -retire-stats -iterations=1 < %s | FileCheck %s
vsqrtps %xmm0, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
vaddps %xmm0, %xmm1, %xmm2
# CHECK: Iterations: 1
# CHECK-NEXT: Instructions: 16
# CHECK-NEXT: Total Cycles: 22
# CHECK-NEXT: Total uOps: 16
# CHECK: Dispatch Width: 4
# CHECK-NEXT: uOps Per Cycle: 0.73
# CHECK-NEXT: IPC: 0.73
# CHECK-NEXT: Block RThroughput: 15.0
# 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 (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
# CHECK-NEXT: 1 9 4.50 vsqrtps %xmm0, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK-NEXT: 1 5 1.00 vaddps %xmm0, %xmm1, %xmm2
# CHECK: Retire Control Unit - number of cycles where we saw N instructions retired:
# CHECK-NEXT: [# retired], [# cycles]
# CHECK-NEXT: 0, 11 (50.0%)
# CHECK-NEXT: 1, 9 (40.9%)
# CHECK-NEXT: 3, 1 (4.5%)
# CHECK-NEXT: 4, 1 (4.5%)
# CHECK: Total ROB Entries: 128
# CHECK-NEXT: Max Used ROB Entries: 16 ( 12.5% )
# CHECK-NEXT: Average Used ROB Entries per cy: 9 ( 7.0% )