94 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=3 -iterations=1500 < %s | FileCheck %s3 4# perf stat reports an IPC of 2.00 for this block of code.5 6# All of the vector packed compares from this test are dependency breaking7# instructions. That means, there is no RAW dependency between any of the8# instructions, and the code can be fully parallelized in hardware.9 10vpcmpeqb %xmm0, %xmm0, %xmm111vpcmpeqw %xmm1, %xmm1, %xmm212vpcmpeqd %xmm2, %xmm2, %xmm313vpcmpeqq %xmm3, %xmm3, %xmm014 15# CHECK: Iterations: 150016# CHECK-NEXT: Instructions: 600017# CHECK-NEXT: Total Cycles: 300318# CHECK-NEXT: Total uOps: 600019 20# CHECK: Dispatch Width: 221# CHECK-NEXT: uOps Per Cycle: 2.0022# CHECK-NEXT: IPC: 2.0023# CHECK-NEXT: Block RThroughput: 2.024 25# CHECK: Instruction Info:26# CHECK-NEXT: [1]: #uOps27# CHECK-NEXT: [2]: Latency28# CHECK-NEXT: [3]: RThroughput29# CHECK-NEXT: [4]: MayLoad30# CHECK-NEXT: [5]: MayStore31# CHECK-NEXT: [6]: HasSideEffects (U)32 33# CHECK: [1] [2] [3] [4] [5] [6] Instructions:34# CHECK-NEXT: 1 1 0.50 vpcmpeqb %xmm0, %xmm0, %xmm135# CHECK-NEXT: 1 1 0.50 vpcmpeqw %xmm1, %xmm1, %xmm236# CHECK-NEXT: 1 1 0.50 vpcmpeqd %xmm2, %xmm2, %xmm337# CHECK-NEXT: 1 1 0.50 vpcmpeqq %xmm3, %xmm3, %xmm038 39# CHECK: Resources:40# CHECK-NEXT: [0] - JALU041# CHECK-NEXT: [1] - JALU142# CHECK-NEXT: [2] - JDiv43# CHECK-NEXT: [3] - JFPA44# CHECK-NEXT: [4] - JFPM45# CHECK-NEXT: [5] - JFPU046# CHECK-NEXT: [6] - JFPU147# CHECK-NEXT: [7] - JLAGU48# CHECK-NEXT: [8] - JMul49# CHECK-NEXT: [9] - JSAGU50# CHECK-NEXT: [10] - JSTC51# CHECK-NEXT: [11] - JVALU052# CHECK-NEXT: [12] - JVALU153# CHECK-NEXT: [13] - JVIMUL54 55# CHECK: Resource pressure per iteration:56# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]57# CHECK-NEXT: - - - - - 2.00 2.00 - - - - 2.00 2.00 -58 59# CHECK: Resource pressure by instruction:60# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:61# CHECK-NEXT: - - - - - - 1.00 - - - - - 1.00 - vpcmpeqb %xmm0, %xmm0, %xmm162# CHECK-NEXT: - - - - - 1.00 - - - - - 1.00 - - vpcmpeqw %xmm1, %xmm1, %xmm263# CHECK-NEXT: - - - - - - 1.00 - - - - - 1.00 - vpcmpeqd %xmm2, %xmm2, %xmm364# CHECK-NEXT: - - - - - 1.00 - - - - - 1.00 - - vpcmpeqq %xmm3, %xmm3, %xmm065 66# CHECK: Timeline view:67# CHECK-NEXT: Index 01234567868 69# CHECK: [0,0] DeER . . vpcmpeqb %xmm0, %xmm0, %xmm170# CHECK-NEXT: [0,1] DeER . . vpcmpeqw %xmm1, %xmm1, %xmm271# CHECK-NEXT: [0,2] .DeER. . vpcmpeqd %xmm2, %xmm2, %xmm372# CHECK-NEXT: [0,3] .DeER. . vpcmpeqq %xmm3, %xmm3, %xmm073# CHECK-NEXT: [1,0] . DeER . vpcmpeqb %xmm0, %xmm0, %xmm174# CHECK-NEXT: [1,1] . DeER . vpcmpeqw %xmm1, %xmm1, %xmm275# CHECK-NEXT: [1,2] . DeER . vpcmpeqd %xmm2, %xmm2, %xmm376# CHECK-NEXT: [1,3] . DeER . vpcmpeqq %xmm3, %xmm3, %xmm077# CHECK-NEXT: [2,0] . DeER. vpcmpeqb %xmm0, %xmm0, %xmm178# CHECK-NEXT: [2,1] . DeER. vpcmpeqw %xmm1, %xmm1, %xmm279# CHECK-NEXT: [2,2] . DeER vpcmpeqd %xmm2, %xmm2, %xmm380# CHECK-NEXT: [2,3] . DeER vpcmpeqq %xmm3, %xmm3, %xmm081 82# CHECK: Average Wait times (based on the timeline view):83# CHECK-NEXT: [0]: Executions84# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue85# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready86# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage87 88# CHECK: [0] [1] [2] [3]89# CHECK-NEXT: 0. 3 1.0 1.0 0.0 vpcmpeqb %xmm0, %xmm0, %xmm190# CHECK-NEXT: 1. 3 1.0 1.0 0.0 vpcmpeqw %xmm1, %xmm1, %xmm291# CHECK-NEXT: 2. 3 1.0 1.0 0.0 vpcmpeqd %xmm2, %xmm2, %xmm392# CHECK-NEXT: 3. 3 1.0 1.0 0.0 vpcmpeqq %xmm3, %xmm3, %xmm093# CHECK-NEXT: 3 1.0 1.0 0.0 <total>94