tjh.dev / kernel
Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux
1
fork

Configure Feed

Select the types of activity you want to include in your feed.

perf test: Add 'perf record cgroup' filtering test

$ sudo ./perf test filtering -vv
96: perf record sample filtering (by BPF) tests:
--- start ---
test child forked, pid 2966908
Checking BPF-filter privilege
Basic bpf-filter test
Basic bpf-filter test [Success]
Failing bpf-filter test
Failing bpf-filter test [Success]
Group bpf-filter test
Group bpf-filter test [Success]
Multiple bpf-filter test
Multiple bpf-filter test [Success]
Cgroup bpf-filter test
Cgroup bpf-filter test [Success]
---- end(0) ----
96: perf record sample filtering (by BPF) tests : Ok

Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: https://lore.kernel.org/r/20240826221045.1202305-5-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

authored by

Namhyung Kim and committed by
Arnaldo Carvalho de Melo d56a4d56 91e88437

+36 -3
+36 -3
tools/perf/tests/shell/record_bpf_filter.sh
··· 68 68 69 69 # 'cpu' requires PERF_SAMPLE_CPU flag 70 70 if ! perf record -e task-clock --filter 'cpu > 0' \ 71 - -o /dev/null true 2>&1 | grep PERF_SAMPLE_CPU 71 + -o /dev/null true 2>&1 | grep -q PERF_SAMPLE_CPU 72 72 then 73 73 echo "Failing bpf-filter test [Failed forbidden CPU]" 74 74 err=1 ··· 98 98 fi 99 99 100 100 if ! perf record -e task-clock --filter 'cpu > 0 || ip > 0' \ 101 - -o /dev/null true 2>&1 | grep PERF_SAMPLE_CPU 101 + -o /dev/null true 2>&1 | grep -q PERF_SAMPLE_CPU 102 102 then 103 103 echo "Group bpf-filter test [Failed forbidden CPU]" 104 104 err=1 ··· 106 106 fi 107 107 108 108 if ! perf record -e task-clock --filter 'period > 0 || code_pgsz > 4096' \ 109 - -o /dev/null true 2>&1 | grep PERF_SAMPLE_CODE_PAGE_SIZE 109 + -o /dev/null true 2>&1 | grep -q PERF_SAMPLE_CODE_PAGE_SIZE 110 110 then 111 111 echo "Group bpf-filter test [Failed forbidden CODE_PAGE_SIZE]" 112 112 err=1 ··· 147 147 echo "Multiple bpf-filter test [Success]" 148 148 } 149 149 150 + test_bpf_filter_cgroup() { 151 + echo "Cgroup bpf-filter test" 152 + 153 + if ! perf record -e task-clock --filter 'cgroup == /' \ 154 + -a --all-cgroups --synth=cgroup -o "${perfdata}" true 2> /dev/null 155 + then 156 + echo "Cgroup bpf-filter test [Skipped cgroup not supported]" 157 + return 158 + fi 159 + 160 + # 'cgroup' requires PERF_SAMPLE_CGROUP flag 161 + if ! perf record -e task-clock --filter 'cgroup == /' \ 162 + -o /dev/null true 2>&1 | grep -q PERF_SAMPLE_CGROUP 163 + then 164 + echo "Cgroup bpf-filter test [Failed CGROUP requires --all-cgroups]" 165 + err=1 166 + return 167 + fi 168 + 169 + if ! perf report -i "${perfdata}" -s cgroup -q | grep -q -F '100.00%' 170 + then 171 + echo "Cgroup bpf-filter test [Failed root cgroup does not have 100%]" 172 + err=1 173 + return 174 + fi 175 + 176 + echo "Cgroup bpf-filter test [Success]" 177 + } 178 + 150 179 test_bpf_filter_priv 151 180 152 181 if [ $err = 0 ]; then ··· 192 163 193 164 if [ $err = 0 ]; then 194 165 test_bpf_filter_multi 166 + fi 167 + 168 + if [ $err = 0 ]; then 169 + test_bpf_filter_cgroup 195 170 fi 196 171 197 172 cleanup