==PROF== Kernoscope 0.1.0 - provenance-honest profile (no performance counters required)
==PROF== Target: bench/artifacts/.build/ledger_probe --kernel norm --mb 2 --target-s 1.2
==PROF== DRAM 估算不可用：显存忙碌读数 0.0% 低于 NVML utilization.memory 的分辨率下限 1 个百分点：无法区分『真的没搬』与『1.0% 被截断成 0』
==PROF== Kernel time: 630.525 ms of 12241.102 ms wall (5.2%)
==PROF== Window: mode=auto（已裁剪） 采样 7 点 / 参与积分 1.4 s / 裁掉 10.8 s / 进程 wall 12.2 s
==PROF== source[kernel activity]: nsys sqlite CUPTI_ACTIVITY_KIND_KERNEL (no counters needed)
==PROF== source[device counters]: pynvml (energy/PCIe/mem-busy are management counters, available without admin)
==PROF== source[static]: not used
==PROF== source[peaks]: measured calibration (bench/artifacts/20261008-225555/machine-profile.json)

Kernel: norm_k(const float *, const float *, float *, long)
  nsys short name: norm_k
  268502 launch(es), 630.5247 ms total, 0.0023 ms mean

  Section: Duration                           [exact]
    gpu__time_duration.sum                                630.525 ms       [exact]
    ks__launch_time_mean                                   0.0023 ms       [exact]
    ks__launch_time_min                                    0.0022 ms       [exact]
    ks__launch_time_max                                    0.0029 ms       [exact]
    ks__launch_count                                       268502          [exact]

  Section: LaunchStats                        [exact]
    launch__grid_size                                        1024          [exact]
    launch__block_size                                        256          [exact]
    launch__thread_count                                   262144          [exact]
    launch__registers_per_thread                               34          [exact]
    ks__registers_used_for_occupancy                           34          [exact]
    launch__shared_mem_per_block_static                         0 B        [exact]
    launch__shared_mem_per_block_dynamic                        0 B        [exact]
    launch__local_mem_per_thread                                0 B        [exact]
    ks__cuda_graph_capture                                    yes          [exact]

  Section: Occupancy                          [exact] (+1 n/a)
    ks__theoretical_occupancy_pct                             100 %        [exact]
    ks__blocks_per_sm                                           6          [exact]
    ks__warps_per_sm                                           48          [exact]
    ks__occupancy_limiter                          registers+threads+warps          [exact]  (并列的限制因素用 + 连接)
    sm__maximum_warps_per_active_cycle_pct                    n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: SpeedOfLight                       [unavailable] (+2 n/a)
    est__achieved_tflops                                      n/a          [unavailable]  workload not provided: pass --ks-gemm M,N,K or --ks-flops/--ks-bytes (plus --ks-calibration machine-profile.json for measured peaks)
    sm__throughput.avg.pct_of_peak_sustained_elapsed            n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: InstructionStats                   [unavailable] (+2 n/a)
    ks__static_sass_instructions                              n/a          [unavailable]  no cubin provided: pass --ks-cubin <file.so|.cubin>
    smsp__inst_executed.sum                                   n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: SchedulerStats                     [estimated] (+1 n/a)
    est__dynamic_warp_instructions                    2.44152e+11          [estimated]  (动态指令构成：alu 81.1%、control 5.4%、memory 5.4%、fp 4.5%；发射槽上限 = SM 数 × 4 调度器 × 时钟 × 时长；动态指令数按 **warp 级**计（NVBit count_warp_level=1），因此这是发射槽占用的直接计数；；指令数按 per-launch 平均对齐到本报告的 268502 次 launch（直方图那次是 1019 次）)
    est__issue_slot_utilization_pct                          84.8 %        [estimated]  (**上界口径**：每 SM 4 个调度器、每周期各发 1 条；真实可用发射槽只会更少)
    est__issued_ipc_per_sm                                  3.393 warp-instr/SM/cycle [estimated]  (同为上界口径)
    est__stall_direction                           偏发射受限：发射槽已用掉约 85%，优化方向是减少指令数（向量化、减少地址运算、合并访存）          [estimated]  (**不是** ncu 的 stall 计数器细分（smsp__warp_issue_stalled_*）：这里只回答「到底发不出去还是在等数据」，且标 estimated)
    est__warp_instructions_per_launch                   2.396e+08          [estimated]  (共 1019 次 launch)
    est__dyn_instr_share_alu                                 81.1 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_control                              5.4 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_memory                               5.4 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_fp                                   4.5 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_sync                                 3.6 %        [estimated]  (按 warp 级指令数加权)
    smsp__warp_issue_stalled_long_scoreboard_per_warp_active.pct            n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: WarpStateStats                     [estimated] (+1 n/a)
    est__stall_direction                           偏发射受限：发射槽已用掉约 85%，优化方向是减少指令数（向量化、减少地址运算、合并访存）          [estimated]  (**不是** ncu 的 stall 计数器细分（smsp__warp_issue_stalled_*）：这里只回答「到底发不出去还是在等数据」，且标 estimated)
    est__dyn_instr_share_alu                                 81.1 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_control                              5.4 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_memory                               5.4 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_fp                                   4.5 %        [estimated]  (按 warp 级指令数加权)
    est__dyn_instr_share_sync                                 3.6 %        [estimated]  (按 warp 级指令数加权)
    smsp__warp_issue_stalled_long_scoreboard_per_warp_active.pct            n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: WorkloadDistribution               [unavailable] (+1 n/a)
    ks__workloaddistribution_status                           n/a          [unavailable]  requires in-kernel instrumentation (NVBit / IKP trace) which is not enabled; install NVBit and pass an existing trace with --ks-memtrace <file> (see tools/probe

  Section: ComputeWorkloadAnalysis            [unavailable] (+1 n/a)
    ks__computeworkloadanalysis_status                        n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: MemoryWorkloadAnalysis             [unavailable] (+1 n/a)
    ks__memoryworkloadanalysis_status                         n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: MemoryWorkloadAnalysis_Tables      [unavailable] (+1 n/a)
    ks__memoryworkloadanalysis_tables_status                  n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: SchedulerStats                     [unavailable] (+1 n/a)
    ks__schedulerstats_status                                 n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: WarpStateStats                     [unavailable] (+1 n/a)
    ks__warpstatestats_status                                 n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: SourceCounters                     [unavailable] (+1 n/a)
    ks__sourcecounters_status                                 n/a          [unavailable]  requires PC sampling, which needs the profiling permission here

  Section: PmSampling                         [unavailable] (+1 n/a)
    ks__pmsampling_status                                     n/a          [unavailable]  requires GPU hardware performance counters, which are restricted here (RmProfilingAdminOnly=1 and no CAP_SYS_ADMIN/CAP_PERFMON in the container); CUPTI reports 

  Section: Nvlink_Tables                      [unavailable] (+1 n/a)
    ks__nvlink_tables_status                                  n/a          [unavailable]  requires NVLink hardware with NVML counter support (not present / not supported)

  Section: NumaAffinity                       [unavailable] (+1 n/a)
    ks__numaaffinity_status                                   n/a          [unavailable]  not applicable on this device

----------------------------------------------------------------------------------------------------
Kernel: (device)
  0 launch(es), 0.0000 ms total, 0.0000 ms mean

  Section: DeviceStatistics                   [estimated]
    meas__nvml_mem_busy_mean_pct                                0 %        [measured]  [resolution 1 %（整数百分比）]  (区间平均；是「显存子系统忙碌占比」，不是 DRAM 吞吐本身。读数 0.0% 只说明低于分辨率下限，不等于「DRAM 完全没动」；必须让 GPU 保持饱和，否则会被宿主间隙稀释（实测 96.8% vs 6.3%）)
    est__dram_busy_equivalent_gbs                               0 GB/s     [estimated]  [resolution 1 %]  (等效带宽口径；与按字节/时间的口径略有差异)
    meas__sm_clock_mean_mhz                                 891.5 MHz      [measured]  [resolution 1 MHz / 1 W（驱动报数粒度）]
    meas__sm_clock_max_mhz                                   2790 MHz      [measured]  [resolution 1 MHz / 1 W（驱动报数粒度）]
    meas__power_max_w                                       160.3 W        [measured]  [resolution 1 MHz / 1 W（驱动报数粒度）]
    meas__device_energy_j                                 542.675 J        [measured]  [resolution 1 mJ]  (ncu 不提供能耗指标)

  Section: DramTraffic                        [derived] (+4 n/a)
    est__dram_bytes_total                                     n/a          [unavailable]  显存忙碌读数 0.0% 低于 NVML utilization.memory 的分辨率下限 1 个百分点：无法区分『真的没搬』与『1.0% 被截断成 0』
    est__dram_bytes_per_launch                                n/a          [unavailable]  显存忙碌读数 0.0% 低于 NVML utilization.memory 的分辨率下限 1 个百分点：无法区分『真的没搬』与『1.0% 被截断成 0』
    est__dram_throughput_pct_of_peak                          n/a          [unavailable]  显存忙碌读数 0.0% 低于 NVML utilization.memory 的分辨率下限 1 个百分点：无法区分『真的没搬』与『1.0% 被截断成 0』
    ks__sampler_window_coverage_pct                          71.4 %        [derived]  [resolution 1 个采样点]  (前提①「窗口被被测负载主导」：<80% 不给 per-launch 归因，<50% 连总量也不给)
    est__dram_bytes_per_launch_trace                          n/a          [unavailable]  未提供访存轨迹：加 --ks-memtrace <trace> 可换用轨迹口径，在 NVML 代理口径不成立时仍能完成账本对拍

----------------------------------------------------------------------------------------------------

Provenance summary: exact=18, derived=1, measured=5, estimated=17, unavailable=21
Legend: exact=与 ncu 同源 | derived=本机实测标定/模型 | measured=我们的探针实测 | estimated=采样或代理(附误差) | unavailable=需要硬件计数器，只报 n/a
Output(json): bench/artifacts/20261008-225555/rep-issue-l2-resident.ksreport.json
Output(perfetto): bench/artifacts/20261008-225555/rep-issue-l2-resident.perfetto.json
Open the timeline at https://ui.perfetto.dev -> Load trace -> the file above
