For now, the budget of the hash table that is used for fprobe_ip_table is fixed, which is 256, and can cause huge overhead when the hooked functions is a huge quantity. In this series, we use rhashtable for fprobe_ip_table to reduce the overhead. Meanwhile, we also add the benchmark testcase "kprobe-multi-all", which will hook all the kernel functions during the testing. Before this series, the performance is: usermode-count : 875.380 ± 0.366M/s kernel-count : 435.924 ± 0.461M/s syscall-count : 31.004 ± 0.017M/s fentry : 134.076 ± 1.752M/s fexit : 68.319 ± 0.055M/s fmodret : 71.530 ± 0.032M/s rawtp : 202.751 ± 0.138M/s tp : 79.562 ± 0.084M/s kprobe : 55.587 ± 0.028M/s kprobe-multi : 56.481 ± 0.043M/s kprobe-multi-all: 6.283 ± 0.005M/s << look this kretprobe : 22.378 ± 0.028M/s kretprobe-multi: 28.205 ± 0.025M/s With this series, the performance is: usermode-count : 897.083 ± 5.347M/s kernel-count : 431.638 ± 1.781M/s syscall-count : 30.807 ± 0.057M/s fentry : 134.803 ± 1.045M/s fexit : 68.763 ± 0.018M/s fmodret : 71.444 ± 0.052M/s rawtp : 202.344 ± 0.149M/s tp : 79.644 ± 0.376M/s kprobe : 55.480 ± 0.108M/s kprobe-multi : 57.302 ± 0.119M/s kprobe-multi-all: 57.855 ± 0.144M/s << look this kretprobe : 22.265 ± 0.023M/s kretprobe-multi: 27.740 ± 0.023M/s The benchmark of "kprobe-multi-all" increase from 6.283M/s to 57.855M/s. Menglong Dong (4): fprobe: use rhashtable selftests/bpf: move get_ksyms and get_addrs to trace_helpers.c selftests/bpf: add benchmark testing for kprobe-multi-all selftests/bpf: skip recursive functions for kprobe_multi include/linux/fprobe.h | 2 +- kernel/trace/fprobe.c | 144 ++++++----- tools/testing/selftests/bpf/bench.c | 2 + .../selftests/bpf/benchs/bench_trigger.c | 30 +++ .../selftests/bpf/benchs/run_bench_trigger.sh | 2 +- .../bpf/prog_tests/kprobe_multi_test.c | 220 +---------------- tools/testing/selftests/bpf/trace_helpers.c | 230 ++++++++++++++++++ tools/testing/selftests/bpf/trace_helpers.h | 3 + 8 files changed, 351 insertions(+), 282 deletions(-) -- 2.50.1
