Perf¶

linux内核在2.6.31版本引入的性能测试工具。

实现¶

安装¶

子命令¶

命令	说明
stat
top	查看热点函数
record	对单个程序进行采样并保存
report	根据采样数据，生成报告
annotate
sched
list	列出perf支持的事件

probe¶

malloc/free跟踪¶

Bash
#添加事件
perf probe -x /lib/x86_64-linux-gnu/libc.so.6 malloc
#删除事件
perf probe -d probe_libc:malloc
#记录事件

record¶

Bash
perf record -e probe_libc:free -e probe_libc:malloc -e probe_libc:realloc -aR -p 2009 --call-graph dwarf sleep 10

perf record -e cache-misses -a -g -p <pid> sleep 10

# 内存分配相关
perf record -e kmem:mm_page_alloc,kmem:mm_page_free -a sleep 10

perf report

trace¶

Bash
sudo perf trace --no-syscalls --event 'net:*' ping 172.17.0.2 -c1 > /dev/null

sched¶

Bash
 perf sched record -C 4 -a sleep 5
 perf sched timehist
 perf sched latency

stat¶

Bash
perf stat -e cache-references,cache-misses,instructions,cycles,cpu-migrations,context-switches -p <pid> sleep 10


perf stat -e branch-misses,bus-cycles,cache-misses,cache-references,cpu-cycles,instructions,L1-dcache-load-misses,L1-dcache-loads,L1-dcache-store-misses,L1-dcache-stores,L1-icache-load-misses,L1-icache-loads,branch-load-misses,branch-loads,dTLB-load-misses,iTLB-load-misses  -a -p

# 查看内存访问模式
perf stat -e cache-references,cache-misses,L1-dcache-loads,L1-dcache-load-misses,L1-dcache-stores,L1-dcache-store-misses,LLC-loads,LLC-load-misses,LLC-stores,LLC-store-misses -a sleep 10

# 查看TLB相关
perf stat -e dTLB-loads,dTLB-load-misses,dTLB-stores,dTLB-store-misses,iTLB-loads,iTLB-load-misses -a sleep 10


# 内存带宽相关
perf stat -e uncore_imc/cas_count_read/,uncore_imc/cas_count_write/ -a sleep 10

# NUMA内存访问
perf stat -e node-loads,node-load-misses,node-stores,node-store-misses -a sleep 10

# 页面错误
perf stat -e page-faults,minor-faults,major-faults -a sleep 10

# L1/L2/L3缓存详细分析
perf stat -e L1-dcache-loads,L1-dcache-load-misses,L1-icache-loads,L1-icache-load-misses,l2_rqsts.all_demand_data_rd,l2_rqsts.demand_data_rd_miss -a sleep 10

火焰图¶

详细用法¶

perf-manpage
perf-wiki