diff options
author | Milosz Wasilewski <milosz.wasilewski@linaro.org> | 2016-10-20 14:38:58 +0100 |
---|---|---|
committer | Milosz Wasilewski <milosz.wasilewski@linaro.org> | 2016-10-20 14:38:58 +0100 |
commit | d341c22fae0e9ebbb27cf1a1d2629c6c011f7d23 (patch) | |
tree | d828ef6c6052d0ae2d95b1743d28f556848415c3 /manual/generic/android/perf.yaml | |
parent | d1b6521e96da63cb212a2e18c4a02791bf5e07cb (diff) |
Moved manual tests to separate directory
Change-Id: I1092ce929d4d257dfd596034c605b4be5b0189a0
Signed-off-by: Milosz Wasilewski <milosz.wasilewski@linaro.org>
Diffstat (limited to 'manual/generic/android/perf.yaml')
-rw-r--r-- | manual/generic/android/perf.yaml | 42 |
1 files changed, 42 insertions, 0 deletions
diff --git a/manual/generic/android/perf.yaml b/manual/generic/android/perf.yaml new file mode 100644 index 0000000..5a6a19b --- /dev/null +++ b/manual/generic/android/perf.yaml @@ -0,0 +1,42 @@ +metadata: + name: andriod-perf + format: "Manual Test Definition 1.0" + description: "Test if perf utility works" + maintainer: + - botao.sun@linaro.org + os: + - android + scope: + - functional + devices: + - panda + - panda-es + - vexpress-a9 + - vexpress-tc2 + - arndale + environment: + - manual-test + +run: + steps: + - "Power on the board to boot into Android main screen" + - "In serial console, run: + # perf stat ls" + + expected: + - "The output of command \"perf stat ls\" should look like this: + Performance counter stats for 'ls': + + 0.659622 task-clock # 0.665 CPUs utilized + 24 context-switches # 0.036 M/sec + 0 CPU-migrations # 0.000 K/sec + 273 page-faults # 0.414 M/sec + 1'996'348 cycles # 3.027 GHz [78.98%] + 1'274'167 stalled-cycles-frontend # 63.82% frontend cycles idle + 1'005'419 stalled-cycles-backend # 50.36% backend cycles idle + 1'397'625 instructions # 0.70 insns per cycle + # 0.91 stalled cycles per insn + 277'755 branches # 421.082 M/sec + 7'689 branch-misses # 2.77% of all branches [28.76%] + + 0.000992095 seconds time elapsed" |