Hardware Counters

mohitmishra786/low-level-dev-skills/skills/profilers/hardware-counters

作者 mohitmishra786bdc58472fa9f无许可证253 个星标收录于 2026年10月9日更新于 2026年10月9日仓库3个月前更新

Hardware performance counter skill for low-level CPU analysis. Use when collecting PMU events with perf stat, using the PAPI library, measuring cache miss rates and branch misprediction ratios, computing IPC, or correlating PMU events to source lines. Activates on queries about hardware counters, PMU events, perf stat -e, PAPI, cache miss rate, branch misprediction, IPC measurement, or CPU performance events.

仅含说明Software DevelopmentData & Analytics
AI 生成的概览

指导使用 perf stat、PAPI 和 Intel PCM 进行硬件性能计数器分析，测量 IPC、缓存未命中等 CPU 指标。

功能
该技能提供在 CPU 上采集和解读硬件性能计数器（PMU）数据的指导。内容涵盖 perf stat 事件采集、原始 PMU 事件代码、使用 perf record 与 annotate 进行源码级标注、PAPI C 库以及 Intel PCM。它讲解 IPC、缓存未命中率、分支预测错误率和 MPKI 等派生指标，并给出健康与需关注数值的参考阈值。
适用场景
适用于使用硬件计数器测量 CPU 性能的场景，例如采集 PMU 事件、计算 IPC、检查缓存未命中率或分支预测错误率，以及将 PMU 事件关联到源码行。也适用于使用 perf stat、PAPI 或 Intel PCM 的情况。
运行要求
需要 Linux perf 工具（perf stat、perf record、perf annotate）；PAPI 示例还需要 PAPI 库和头文件以及 C 编译器；Intel PCM 部分需要从源码构建。根据内核设置，部分计数器采集可能需要提升权限。该技能不附带脚本，仅为说明文档。

Hardware Performance Counters

Purpose

Guide agents through hardware performance counter analysis: collecting PMU events with perf stat -e, using the PAPI library for portable counter access, interpreting cache miss rates and branch misprediction ratios, computing IPC, and correlating events to source lines with perf annotate.

Triggers

  • "How do I measure cache miss rate with perf?"
  • "How do I count branch mispredictions?"
  • "How do I compute IPC (instructions per clock) with perf?"
  • "How do I use the PAPI library for hardware counters?"
  • "How do I see which source lines cause the most cache misses?"
  • "How do I measure memory bandwidth with performance counters?"

Workflow

1. perf stat — basic counter collection

bash
# Basic hardware event summaryperf stat ./prog
# Output:#  Performance counter stats for './prog':##      1,234,567,890      instructions#        456,789,012      cycles#         12,345,678      cache-misses         #    1.23 % of all cache refs#         23,456,789      branch-misses        #    2.34 % of all branches##       0.456789012 seconds time elapsed
# Derived metrics (computed from the output)# IPC = instructions / cycles = 1,234,567,890 / 456,789,012 ≈ 2.70# CPI = cycles / instructions ≈ 0.37

2. Specifying PMU events with -e

bash
# Specific hardware eventsperf stat -e instructions,cycles,cache-misses,branch-misses ./prog
# L1/L2/L3 cache eventsperf stat -e \  L1-dcache-loads,L1-dcache-load-misses,\  L2-loads,L2-load-misses,\  LLC-loads,LLC-load-misses \  ./prog
# Memory bandwidth (Intel)perf stat -e \  uncore_imc/cas_count_read/,\  uncore_imc/cas_count_write/ \  ./prog
# TLB missesperf stat -e dTLB-loads,dTLB-load-misses,iTLB-loads,iTLB-load-misses ./prog
# Branch misprediction rateperf stat -e branches,branch-misses ./prog# Rate = branch-misses / branches × 100%
# Available events (varies by CPU)perf list hardware          # generic hardware eventsperf list cache             # cache eventsperf list pmu               # raw PMU events for your CPU

3. Key metrics and thresholds

MetricFormulaHealthyConcerning
IPCinstructions / cycles> 2.0 (modern x86)< 1.0
L1 miss rateL1-misses / L1-accesses< 1%> 5%
LLC miss rateLLC-misses / LLC-accesses< 1%> 10%
Branch miss ratebranch-misses / branches< 1%> 5%
MPKImisses per 1K instructions—L3 MPKI > 10 = memory bound
bash
# Compute MPKI (Misses Per Kilo-Instructions)perf stat -e instructions,LLC-load-misses ./prog# MPKI = LLC-load-misses / (instructions / 1000)

4. Raw PMU events (CPU-specific)

For events not in the generic aliases, use raw event codes:

bash
# Intel: use perf list or look up in Intel SDM# Format: rXXYY where XX=umask, YY=event codeperf stat -e r0124 ./prog    # example Intel raw event
# List Intel events with ocperf (OpenCL Perf Events)pip install ocperfocperf.py list | grep "mem_load"
# Use libpfm4 for event namespfm_ls | grep "MEM_LOAD"perf stat -e $(pfm_ls | grep "MEM_LOAD_RETIRED.L3_MISS") ./prog
# AMD: similar approachperf stat -e r04041 ./prog   # AMD raw event

5. Source-level annotation with perf record/annotate

bash
# Record with hardware eventsperf record -e LLC-load-misses -g ./prog
# Annotate: show source lines sorted by cache miss countperf annotate --stdio
# Interactive (requires debug symbols)perf report# Press 'a' on a function to annotate it
# Combined: record hotspot + annotateperf record -e cycles:u -g ./progperf annotate --symbol=my_function --stdio 2>/dev/null | head -40
# Example annotate output:# Percent | Source code#   45.23 |     for (int i = 0; i < N; i++)#    3.12 |         sum += data[i];   ← cache miss here (strided access)

6. PAPI — Portable API for hardware counters

PAPI provides a portable C API across different CPU architectures:

c
#include <papi.h>#include <stdio.h>
int main(void) {    int Events[] = {PAPI_TOT_INS, PAPI_TOT_CYC,                    PAPI_L2_TCM,  PAPI_BR_MSP};    long long values[4];
    if (PAPI_library_init(PAPI_VER_CURRENT) != PAPI_VER_CURRENT) {        fprintf(stderr, "PAPI init failed\n");        return 1;    }
    PAPI_start_counters(Events, 4);
    // --- Code to measure ---    do_work();    // -----------------------
    PAPI_stop_counters(values, 4);
    printf("Instructions:      %lld\n", values[0]);    printf("Cycles:            %lld\n", values[1]);    printf("IPC:               %.2f\n", (double)values[0]/values[1]);    printf("L2 cache misses:   %lld\n", values[2]);    printf("Branch mispred:    %lld\n", values[3]);
    return 0;}
bash
# Build with PAPIgcc -O2 -g -o prog prog.c -lpapi
# Available PAPI events on your systempapi_avail -a | head -30papi_native_avail | grep "L3"    # native events with "L3"

Common PAPI presets:

PresetEvent
PAPI_TOT_INSTotal instructions
PAPI_TOT_CYCTotal cycles
PAPI_L1_DCML1 data cache misses
PAPI_L2_TCML2 total cache misses
PAPI_L3_TCML3 total cache misses
PAPI_BR_MSPBranch mispredictions
PAPI_TLB_DMData TLB misses
PAPI_FP_INSFloating point instructions
PAPI_VEC_INSVector/SIMD instructions

7. Intel PCM (Performance Counter Monitor)

bash
# Intel PCM — system-wide counters, no root required on modern kernelsgit clone https://github.com/intel/pcmcd pcm && cmake -S . -B build && cmake --build build
# Measure memory bandwidth./build/bin/pcm-memory 1    # sample every 1 second
# Core utilization + IPC./build/bin/pcm 1
# Cache miss breakdown per socket./build/bin/pcm 1 -csv | head -20

Related skills

  • Use skills/profilers/intel-vtune-amd-uprof for guided microarchitecture analysis
  • Use skills/profilers/linux-perf for perf record/report and flamegraph generation
  • Use skills/low-level-programming/cpu-cache-opt for applying cache optimization patterns
  • Use skills/low-level-programming/simd-intrinsics for improving FLOPS/cycle metrics

来源与署名

来源：mohitmishra786/low-level-dev-skills位于skills/profilers/hardware-counters提交bdc5847

许可证： 无许可证

内容归原作者所有。SourceWeft 从公开仓库中收录这些内容。

举报或申请下架

更多来自 mohitmishra786/low-level-dev-skills 的技能

Zig Debugging

mohitmishra786

指导使用 GDB、LLDB 调试 Zig 程序，解读 panic、错误返回追踪、日志记录及 IDE 配置。

Software Development2533个月前更新

Simd Intrinsics

mohitmishra786

指导使用 x86 SSE/AVX 与 ARM NEON 内建函数编写和诊断 SIMD 向量化。

Software Development2533个月前更新

Zig Cross

mohitmishra786

指导 Zig 交叉编译：目标三元组、CPU 特性、用 zig cc 编译 C、嵌入式与 WASM 目标。

Software Development2533个月前更新

Rustc Basics

mohitmishra786

指导 Rust 编译器配置：RUSTFLAGS、Cargo 配置文件、构建调优，以及查看 MIR 或汇编输出。

Software Development2533个月前更新

Zig Cinterop

mohitmishra786

指导 Zig 与 C 互操作：@cImport、translate-c、extern 与 packed 结构体，以及 C/Zig 混合构建。

Software Development2533个月前更新

Zig Build System

mohitmishra786

指导编写和调试 Zig 的 build.zig 文件，涵盖目标、C 源码、选项、测试与包清单。

Software Development2533个月前更新

更多Software Development技能

Playground

anthropics

精选

构建自包含的交互式 HTML 演练场，包含控件、实时预览和可复制的提示词输出。

Software Development37K今天更新

M5 Onboard

anthropics

精选

通过 USB 检测 M5Stack ESP32 开发板，刷写 UIFlow 2.0 固件并安装 MicroPython 应用包。

Software Development37K今天更新

Cardputer Buddy

anthropics

精选

指导迭代 Cardputer-Adv 的 MicroPython 应用包：添加应用、通过 USB 串口推送文件、查看日志并运行 REPL 命令。

Software Development37K今天更新

Web Design Guidelines

vercel-labs

精选

依据从远程来源获取的 Web Interface Guidelines 审查 UI 代码，并以 file:line 形式报告问题。

Software Development32K5周前更新

Vercel React Native Skills

vercel-labs

精选

React Native 与 Expo 最佳实践规则，涵盖列表性能、动画、导航、UI 模式、状态与 monorepo 配置。

Software Development32K5周前更新

Vercel React Best Practices

vercel-labs

精选

Vercel 的 React 与 Next.js 性能指南，以 70 条规则的形式用于编写、审查和重构代码。

Software Development32K5周前更新