1a5192187SRoman Lebedev# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py 2a5baf867SRoman Lebedev# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1500 -timeline -timeline-max-iterations=3 < %s | FileCheck %s 3a5192187SRoman Lebedev 4a5192187SRoman Lebedev# perf stat reports a throughput of 0.60 IPC for this code snippet. 5a5192187SRoman Lebedev# Each lzcnt has a false dependency on %ecx; the first lzcnt has to wait on the 6a5192187SRoman Lebedev# imul. However, the folded load can start immediately. 7a5192187SRoman Lebedev# The last lzcnt has a false dependency on %cx. However, even in this case, the 8a5192187SRoman Lebedev# folded load can start immediately. 9a5192187SRoman Lebedev 10a5192187SRoman Lebedevimul %edx, %ecx 11a5192187SRoman Lebedevlzcnt (%rsp), %cx 12a5192187SRoman Lebedevlzcnt 2(%rsp), %cx 13a5192187SRoman Lebedev 14a5192187SRoman Lebedev# CHECK: Iterations: 1500 15a5192187SRoman Lebedev# CHECK-NEXT: Instructions: 4500 169db0e725SRoman Lebedev# CHECK-NEXT: Total Cycles: 12003 17a5192187SRoman Lebedev# CHECK-NEXT: Total uOps: 7500 18a5192187SRoman Lebedev 19a5192187SRoman Lebedev# CHECK: Dispatch Width: 4 209db0e725SRoman Lebedev# CHECK-NEXT: uOps Per Cycle: 0.62 219db0e725SRoman Lebedev# CHECK-NEXT: IPC: 0.37 229db0e725SRoman Lebedev# CHECK-NEXT: Block RThroughput: 4.0 23a5192187SRoman Lebedev 24a5192187SRoman Lebedev# CHECK: Instruction Info: 25a5192187SRoman Lebedev# CHECK-NEXT: [1]: #uOps 26a5192187SRoman Lebedev# CHECK-NEXT: [2]: Latency 27a5192187SRoman Lebedev# CHECK-NEXT: [3]: RThroughput 28a5192187SRoman Lebedev# CHECK-NEXT: [4]: MayLoad 29a5192187SRoman Lebedev# CHECK-NEXT: [5]: MayStore 30a5192187SRoman Lebedev# CHECK-NEXT: [6]: HasSideEffects (U) 31a5192187SRoman Lebedev 32a5192187SRoman Lebedev# CHECK: [1] [2] [3] [4] [5] [6] Instructions: 339db0e725SRoman Lebedev# CHECK-NEXT: 1 4 2.00 imull %edx, %ecx 349db0e725SRoman Lebedev# CHECK-NEXT: 2 6 2.00 * lzcntw (%rsp), %cx 359db0e725SRoman Lebedev# CHECK-NEXT: 2 6 2.00 * lzcntw 2(%rsp), %cx 36a5192187SRoman Lebedev 37a5192187SRoman Lebedev# CHECK: Resources: 38a5baf867SRoman Lebedev# CHECK-NEXT: [0.0] - PdAGLU01 39a5baf867SRoman Lebedev# CHECK-NEXT: [0.1] - PdAGLU01 40a5baf867SRoman Lebedev# CHECK-NEXT: [1] - PdBranch 41a5baf867SRoman Lebedev# CHECK-NEXT: [2] - PdCount 42a5baf867SRoman Lebedev# CHECK-NEXT: [3] - PdDiv 43a5baf867SRoman Lebedev# CHECK-NEXT: [4] - PdEX0 44a5baf867SRoman Lebedev# CHECK-NEXT: [5] - PdEX1 45a5baf867SRoman Lebedev# CHECK-NEXT: [6] - PdFPCVT 46a5baf867SRoman Lebedev# CHECK-NEXT: [7.0] - PdFPFMA 47a5baf867SRoman Lebedev# CHECK-NEXT: [7.1] - PdFPFMA 48a5baf867SRoman Lebedev# CHECK-NEXT: [8.0] - PdFPMAL 49a5baf867SRoman Lebedev# CHECK-NEXT: [8.1] - PdFPMAL 50a5baf867SRoman Lebedev# CHECK-NEXT: [9] - PdFPMMA 51a5baf867SRoman Lebedev# CHECK-NEXT: [10] - PdFPSTO 52a5baf867SRoman Lebedev# CHECK-NEXT: [11] - PdFPU0 53a5baf867SRoman Lebedev# CHECK-NEXT: [12] - PdFPU1 54a5baf867SRoman Lebedev# CHECK-NEXT: [13] - PdFPU2 55a5baf867SRoman Lebedev# CHECK-NEXT: [14] - PdFPU3 56a5baf867SRoman Lebedev# CHECK-NEXT: [15] - PdFPXBR 57b428b8b2SRoman Lebedev# CHECK-NEXT: [16.0] - PdLoad 58b428b8b2SRoman Lebedev# CHECK-NEXT: [16.1] - PdLoad 59b428b8b2SRoman Lebedev# CHECK-NEXT: [17] - PdMul 60b428b8b2SRoman Lebedev# CHECK-NEXT: [18] - PdStore 61a5192187SRoman Lebedev 62a5192187SRoman Lebedev# CHECK: Resource pressure per iteration: 63b428b8b2SRoman Lebedev# CHECK-NEXT: [0.0] [0.1] [1] [2] [3] [4] [5] [6] [7.0] [7.1] [8.0] [8.1] [9] [10] [11] [12] [13] [14] [15] [16.0] [16.1] [17] [18] 649db0e725SRoman Lebedev# CHECK-NEXT: 3.00 3.00 - - - 4.00 1.00 - - - - - - - - - - - - 3.00 3.00 2.00 - 65a5192187SRoman Lebedev 66a5192187SRoman Lebedev# CHECK: Resource pressure by instruction: 67b428b8b2SRoman Lebedev# CHECK-NEXT: [0.0] [0.1] [1] [2] [3] [4] [5] [6] [7.0] [7.1] [8.0] [8.1] [9] [10] [11] [12] [13] [14] [15] [16.0] [16.1] [17] [18] Instructions: 689db0e725SRoman Lebedev# CHECK-NEXT: - - - - - - 1.00 - - - - - - - - - - - - - - 2.00 - imull %edx, %ecx 699db0e725SRoman Lebedev# CHECK-NEXT: - 3.00 - - - 2.00 - - - - - - - - - - - - - - 3.00 - - lzcntw (%rsp), %cx 709db0e725SRoman Lebedev# CHECK-NEXT: 3.00 - - - - 2.00 - - - - - - - - - - - - - 3.00 - - - lzcntw 2(%rsp), %cx 71a5192187SRoman Lebedev 72a5192187SRoman Lebedev# CHECK: Timeline view: 73a5baf867SRoman Lebedev# CHECK-NEXT: 0123456789 749db0e725SRoman Lebedev# CHECK-NEXT: Index 0123456789 0123456 75a5192187SRoman Lebedev 769db0e725SRoman Lebedev# CHECK: [0,0] DeeeeER . . . .. imull %edx, %ecx 779db0e725SRoman Lebedev# CHECK-NEXT: [0,1] DeeeeeeER . . . .. lzcntw (%rsp), %cx 789db0e725SRoman Lebedev# CHECK-NEXT: [0,2] .D=eeeeeeER . . .. lzcntw 2(%rsp), %cx 799db0e725SRoman Lebedev# CHECK-NEXT: [1,0] .D=======eeeeER. . .. imull %edx, %ecx 809db0e725SRoman Lebedev# CHECK-NEXT: [1,1] . D======eeeeeeER . .. lzcntw (%rsp), %cx 819db0e725SRoman Lebedev# CHECK-NEXT: [1,2] . D========eeeeeeER . .. lzcntw 2(%rsp), %cx 829db0e725SRoman Lebedev# CHECK-NEXT: [2,0] . D=============eeeeER .. imull %edx, %ecx 839db0e725SRoman Lebedev# CHECK-NEXT: [2,1] . D=============eeeeeeER.. lzcntw (%rsp), %cx 849db0e725SRoman Lebedev# CHECK-NEXT: [2,2] . D==============eeeeeeER lzcntw 2(%rsp), %cx 85a5192187SRoman Lebedev 86a5192187SRoman Lebedev# CHECK: Average Wait times (based on the timeline view): 87a5192187SRoman Lebedev# CHECK-NEXT: [0]: Executions 88a5192187SRoman Lebedev# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue 89a5192187SRoman Lebedev# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready 90a5192187SRoman Lebedev# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage 91a5192187SRoman Lebedev 92a5192187SRoman Lebedev# CHECK: [0] [1] [2] [3] 939db0e725SRoman Lebedev# CHECK-NEXT: 0. 3 7.7 0.3 0.0 imull %edx, %ecx 949db0e725SRoman Lebedev# CHECK-NEXT: 1. 3 7.3 0.0 0.0 lzcntw (%rsp), %cx 959db0e725SRoman Lebedev# CHECK-NEXT: 2. 3 8.7 1.0 0.0 lzcntw 2(%rsp), %cx 96*a5e65c1cSRoman Lebedev# CHECK-NEXT: 3 7.9 0.4 0.0 <total> 97