RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-26 12:43:36 +01:00
Code Issues Projects Releases Wiki Activity
d4eedb2312
llvm-mirror/test/tools/llvm-mca/X86/Znver3/one-idioms-sse-xmm.s
Roman Lebedev 89595eb90c [X86] AMD Zen 3: same-reg SSE XMM PCMP is dep breaking one-idiom 
As measured by exegesis, and confirmed by ref docs.
Much like with MMX PCMP, it does actually have to execute, though.
2021-05-10 23:49:26 +03:00

387 lines
16 KiB
ArmAsm
Raw Blame History

# NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver3 -timeline -timeline-max-iterations=2 -register-file-stats -iterations=1000 < %s | FileCheck %s
# LLVM-MCA-BEGIN
pcmpeqb %xmm0, %xmm0
paddb %xmm0, %xmm0
# LLVM-MCA-END
# LLVM-MCA-BEGIN
pcmpeqw %xmm0, %xmm0
paddw %xmm0, %xmm0
# LLVM-MCA-END
# LLVM-MCA-BEGIN
pcmpeqd %xmm0, %xmm0
paddd %xmm0, %xmm0
# LLVM-MCA-END
# LLVM-MCA-BEGIN
pcmpeqq %xmm0, %xmm0
paddq %xmm0, %xmm0
# LLVM-MCA-END
# CHECK: [0] Code Region
# CHECK: Iterations: 1000
# CHECK-NEXT: Instructions: 2000
# CHECK-NEXT: Total Cycles: 504
# CHECK-NEXT: Total uOps: 2000
# CHECK: Dispatch Width: 6
# CHECK-NEXT: uOps Per Cycle: 3.97
# CHECK-NEXT: IPC: 3.97
# CHECK-NEXT: Block RThroughput: 0.5
# 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 1 0.25 pcmpeqb %xmm0, %xmm0
# CHECK-NEXT: 1 1 0.25 paddb %xmm0, %xmm0
# CHECK: Register File statistics:
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #1 -- Zn3FpPRF:
# CHECK-NEXT: Number of physical registers: 160
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #2 -- Zn3IntegerPRF:
# CHECK-NEXT: Number of physical registers: 192
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
# CHECK: Resources:
# CHECK-NEXT: [0] - Zn3AGU0
# CHECK-NEXT: [1] - Zn3AGU1
# CHECK-NEXT: [2] - Zn3AGU2
# CHECK-NEXT: [3] - Zn3ALU0
# CHECK-NEXT: [4] - Zn3ALU1
# CHECK-NEXT: [5] - Zn3ALU2
# CHECK-NEXT: [6] - Zn3ALU3
# CHECK-NEXT: [7] - Zn3BRU1
# CHECK-NEXT: [8] - Zn3FPP0
# CHECK-NEXT: [9] - Zn3FPP1
# CHECK-NEXT: [10] - Zn3FPP2
# CHECK-NEXT: [11] - Zn3FPP3
# CHECK-NEXT: [12.0] - Zn3FPP45
# CHECK-NEXT: [12.1] - Zn3FPP45
# CHECK-NEXT: [13] - Zn3FPSt
# CHECK-NEXT: [14.0] - Zn3LSU
# CHECK-NEXT: [14.1] - Zn3LSU
# CHECK-NEXT: [14.2] - Zn3LSU
# CHECK-NEXT: [15.0] - Zn3Load
# CHECK-NEXT: [15.1] - Zn3Load
# CHECK-NEXT: [15.2] - Zn3Load
# CHECK-NEXT: [16.0] - Zn3Store
# CHECK-NEXT: [16.1] - Zn3Store
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1]
# CHECK-NEXT: - - - - - - - - 0.50 0.50 0.50 0.50 - - - - - - - - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1] Instructions:
# CHECK-NEXT: - - - - - - - - 0.25 0.25 0.50 - - - - - - - - - - - - pcmpeqb %xmm0, %xmm0
# CHECK-NEXT: - - - - - - - - 0.25 0.25 - 0.50 - - - - - - - - - - - paddb %xmm0, %xmm0
# CHECK: Timeline view:
# CHECK-NEXT: Index 01234
# CHECK: [0,0] DeER. pcmpeqb %xmm0, %xmm0
# CHECK-NEXT: [0,1] D=eER paddb %xmm0, %xmm0
# CHECK-NEXT: [1,0] DeE-R pcmpeqb %xmm0, %xmm0
# CHECK-NEXT: [1,1] D=eER paddb %xmm0, %xmm0
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
# CHECK-NEXT: 0. 2 1.0 1.0 0.5 pcmpeqb %xmm0, %xmm0
# CHECK-NEXT: 1. 2 2.0 0.0 0.0 paddb %xmm0, %xmm0
# CHECK-NEXT: 2 1.5 0.5 0.3 <total>
# CHECK: [1] Code Region
# CHECK: Iterations: 1000
# CHECK-NEXT: Instructions: 2000
# CHECK-NEXT: Total Cycles: 504
# CHECK-NEXT: Total uOps: 2000
# CHECK: Dispatch Width: 6
# CHECK-NEXT: uOps Per Cycle: 3.97
# CHECK-NEXT: IPC: 3.97
# CHECK-NEXT: Block RThroughput: 0.5
# 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 1 0.25 pcmpeqw %xmm0, %xmm0
# CHECK-NEXT: 1 1 0.25 paddw %xmm0, %xmm0
# CHECK: Register File statistics:
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #1 -- Zn3FpPRF:
# CHECK-NEXT: Number of physical registers: 160
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #2 -- Zn3IntegerPRF:
# CHECK-NEXT: Number of physical registers: 192
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
# CHECK: Resources:
# CHECK-NEXT: [0] - Zn3AGU0
# CHECK-NEXT: [1] - Zn3AGU1
# CHECK-NEXT: [2] - Zn3AGU2
# CHECK-NEXT: [3] - Zn3ALU0
# CHECK-NEXT: [4] - Zn3ALU1
# CHECK-NEXT: [5] - Zn3ALU2
# CHECK-NEXT: [6] - Zn3ALU3
# CHECK-NEXT: [7] - Zn3BRU1
# CHECK-NEXT: [8] - Zn3FPP0
# CHECK-NEXT: [9] - Zn3FPP1
# CHECK-NEXT: [10] - Zn3FPP2
# CHECK-NEXT: [11] - Zn3FPP3
# CHECK-NEXT: [12.0] - Zn3FPP45
# CHECK-NEXT: [12.1] - Zn3FPP45
# CHECK-NEXT: [13] - Zn3FPSt
# CHECK-NEXT: [14.0] - Zn3LSU
# CHECK-NEXT: [14.1] - Zn3LSU
# CHECK-NEXT: [14.2] - Zn3LSU
# CHECK-NEXT: [15.0] - Zn3Load
# CHECK-NEXT: [15.1] - Zn3Load
# CHECK-NEXT: [15.2] - Zn3Load
# CHECK-NEXT: [16.0] - Zn3Store
# CHECK-NEXT: [16.1] - Zn3Store
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1]
# CHECK-NEXT: - - - - - - - - 0.50 0.50 0.50 0.50 - - - - - - - - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1] Instructions:
# CHECK-NEXT: - - - - - - - - 0.25 0.25 0.50 - - - - - - - - - - - - pcmpeqw %xmm0, %xmm0
# CHECK-NEXT: - - - - - - - - 0.25 0.25 - 0.50 - - - - - - - - - - - paddw %xmm0, %xmm0
# CHECK: Timeline view:
# CHECK-NEXT: Index 01234
# CHECK: [0,0] DeER. pcmpeqw %xmm0, %xmm0
# CHECK-NEXT: [0,1] D=eER paddw %xmm0, %xmm0
# CHECK-NEXT: [1,0] DeE-R pcmpeqw %xmm0, %xmm0
# CHECK-NEXT: [1,1] D=eER paddw %xmm0, %xmm0
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
# CHECK-NEXT: 0. 2 1.0 1.0 0.5 pcmpeqw %xmm0, %xmm0
# CHECK-NEXT: 1. 2 2.0 0.0 0.0 paddw %xmm0, %xmm0
# CHECK-NEXT: 2 1.5 0.5 0.3 <total>
# CHECK: [2] Code Region
# CHECK: Iterations: 1000
# CHECK-NEXT: Instructions: 2000
# CHECK-NEXT: Total Cycles: 504
# CHECK-NEXT: Total uOps: 2000
# CHECK: Dispatch Width: 6
# CHECK-NEXT: uOps Per Cycle: 3.97
# CHECK-NEXT: IPC: 3.97
# CHECK-NEXT: Block RThroughput: 0.5
# 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 1 0.25 pcmpeqd %xmm0, %xmm0
# CHECK-NEXT: 1 1 0.25 paddd %xmm0, %xmm0
# CHECK: Register File statistics:
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #1 -- Zn3FpPRF:
# CHECK-NEXT: Number of physical registers: 160
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #2 -- Zn3IntegerPRF:
# CHECK-NEXT: Number of physical registers: 192
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
# CHECK: Resources:
# CHECK-NEXT: [0] - Zn3AGU0
# CHECK-NEXT: [1] - Zn3AGU1
# CHECK-NEXT: [2] - Zn3AGU2
# CHECK-NEXT: [3] - Zn3ALU0
# CHECK-NEXT: [4] - Zn3ALU1
# CHECK-NEXT: [5] - Zn3ALU2
# CHECK-NEXT: [6] - Zn3ALU3
# CHECK-NEXT: [7] - Zn3BRU1
# CHECK-NEXT: [8] - Zn3FPP0
# CHECK-NEXT: [9] - Zn3FPP1
# CHECK-NEXT: [10] - Zn3FPP2
# CHECK-NEXT: [11] - Zn3FPP3
# CHECK-NEXT: [12.0] - Zn3FPP45
# CHECK-NEXT: [12.1] - Zn3FPP45
# CHECK-NEXT: [13] - Zn3FPSt
# CHECK-NEXT: [14.0] - Zn3LSU
# CHECK-NEXT: [14.1] - Zn3LSU
# CHECK-NEXT: [14.2] - Zn3LSU
# CHECK-NEXT: [15.0] - Zn3Load
# CHECK-NEXT: [15.1] - Zn3Load
# CHECK-NEXT: [15.2] - Zn3Load
# CHECK-NEXT: [16.0] - Zn3Store
# CHECK-NEXT: [16.1] - Zn3Store
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1]
# CHECK-NEXT: - - - - - - - - 0.50 0.50 0.50 0.50 - - - - - - - - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1] Instructions:
# CHECK-NEXT: - - - - - - - - 0.25 0.25 0.50 - - - - - - - - - - - - pcmpeqd %xmm0, %xmm0
# CHECK-NEXT: - - - - - - - - 0.25 0.25 - 0.50 - - - - - - - - - - - paddd %xmm0, %xmm0
# CHECK: Timeline view:
# CHECK-NEXT: Index 01234
# CHECK: [0,0] DeER. pcmpeqd %xmm0, %xmm0
# CHECK-NEXT: [0,1] D=eER paddd %xmm0, %xmm0
# CHECK-NEXT: [1,0] DeE-R pcmpeqd %xmm0, %xmm0
# CHECK-NEXT: [1,1] D=eER paddd %xmm0, %xmm0
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
# CHECK-NEXT: 0. 2 1.0 1.0 0.5 pcmpeqd %xmm0, %xmm0
# CHECK-NEXT: 1. 2 2.0 0.0 0.0 paddd %xmm0, %xmm0
# CHECK-NEXT: 2 1.5 0.5 0.3 <total>
# CHECK: [3] Code Region
# CHECK: Iterations: 1000
# CHECK-NEXT: Instructions: 2000
# CHECK-NEXT: Total Cycles: 504
# CHECK-NEXT: Total uOps: 2000
# CHECK: Dispatch Width: 6
# CHECK-NEXT: uOps Per Cycle: 3.97
# CHECK-NEXT: IPC: 3.97
# CHECK-NEXT: Block RThroughput: 0.5
# 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 1 0.25 pcmpeqq %xmm0, %xmm0
# CHECK-NEXT: 1 1 0.25 paddq %xmm0, %xmm0
# CHECK: Register File statistics:
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #1 -- Zn3FpPRF:
# CHECK-NEXT: Number of physical registers: 160
# CHECK-NEXT: Total number of mappings created: 2000
# CHECK-NEXT: Max number of mappings used: 74
# CHECK: * Register File #2 -- Zn3IntegerPRF:
# CHECK-NEXT: Number of physical registers: 192
# CHECK-NEXT: Total number of mappings created: 0
# CHECK-NEXT: Max number of mappings used: 0
# CHECK: Resources:
# CHECK-NEXT: [0] - Zn3AGU0
# CHECK-NEXT: [1] - Zn3AGU1
# CHECK-NEXT: [2] - Zn3AGU2
# CHECK-NEXT: [3] - Zn3ALU0
# CHECK-NEXT: [4] - Zn3ALU1
# CHECK-NEXT: [5] - Zn3ALU2
# CHECK-NEXT: [6] - Zn3ALU3
# CHECK-NEXT: [7] - Zn3BRU1
# CHECK-NEXT: [8] - Zn3FPP0
# CHECK-NEXT: [9] - Zn3FPP1
# CHECK-NEXT: [10] - Zn3FPP2
# CHECK-NEXT: [11] - Zn3FPP3
# CHECK-NEXT: [12.0] - Zn3FPP45
# CHECK-NEXT: [12.1] - Zn3FPP45
# CHECK-NEXT: [13] - Zn3FPSt
# CHECK-NEXT: [14.0] - Zn3LSU
# CHECK-NEXT: [14.1] - Zn3LSU
# CHECK-NEXT: [14.2] - Zn3LSU
# CHECK-NEXT: [15.0] - Zn3Load
# CHECK-NEXT: [15.1] - Zn3Load
# CHECK-NEXT: [15.2] - Zn3Load
# CHECK-NEXT: [16.0] - Zn3Store
# CHECK-NEXT: [16.1] - Zn3Store
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1]
# CHECK-NEXT: - - - - - - - - 0.50 0.50 0.50 0.50 - - - - - - - - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12.0] [12.1] [13] [14.0] [14.1] [14.2] [15.0] [15.1] [15.2] [16.0] [16.1] Instructions:
# CHECK-NEXT: - - - - - - - - 0.25 0.25 0.50 - - - - - - - - - - - - pcmpeqq %xmm0, %xmm0
# CHECK-NEXT: - - - - - - - - 0.25 0.25 - 0.50 - - - - - - - - - - - paddq %xmm0, %xmm0
# CHECK: Timeline view:
# CHECK-NEXT: Index 01234
# CHECK: [0,0] DeER. pcmpeqq %xmm0, %xmm0
# CHECK-NEXT: [0,1] D=eER paddq %xmm0, %xmm0
# CHECK-NEXT: [1,0] DeE-R pcmpeqq %xmm0, %xmm0
# CHECK-NEXT: [1,1] D=eER paddq %xmm0, %xmm0
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
# CHECK-NEXT: 0. 2 1.0 1.0 0.5 pcmpeqq %xmm0, %xmm0
# CHECK-NEXT: 1. 2 2.0 0.0 0.0 paddq %xmm0, %xmm0
# CHECK-NEXT: 2 1.5 0.5 0.3 <total>
Reference in New Issue View Git Blame Copy Permalink