37c49e22b3
chromeos_test_image: added a runtime parameter to disable ui
2015-09-08 17:43:10 +01:00
9e12930a43
dhrystone: kill any running instances during setup (also handle CTRL-C).
2015-09-08 17:42:18 +01:00
d5f4457701
cpufreq: refined availability check not to rely on the top-level cpu/cpufreq directory
2015-09-08 17:41:25 +01:00
c9f86b05dd
flashing: fixing vexpress flashing
2015-09-07 17:56:05 +01:00
5eb7ca07fe
energy_model: fix cluster power estimation based on voltage.
2015-09-07 11:52:31 +01:00
6d6cddff58
Added 'negative_values' to daq instrument which can be used to specify how negative values in the samples should be handled.
2015-09-04 17:22:17 +01:00
94cc17271e
sysfs_extractor: fixed pulled files verification.
...
When no files are listed for one the specified directoris, the
instrument checks whether there are any files in that directory on the
device. If that directory itself does not exist; that will result in an
error. This is now handled correctly.
2015-09-04 08:56:19 +01:00
9aadb9087a
http_assets: fixing that which was broken
2015-09-04 08:56:15 +01:00
2e35d4003f
nenamark: made duration configurable.
2015-09-03 17:29:42 +01:00
0d076fe8ba
andebench: added parameter to only run native portion of the benchmark.
2015-09-03 17:21:45 +01:00
bfed59a7cf
Adding an HTTP-based resource getter.
2015-09-03 13:55:44 +01:00
047308a904
ipython: pylint fixes
2015-09-03 12:07:41 +01:00
0179d45b5b
energy_model: minor fix for previous adjustment.
2015-09-03 11:06:26 +01:00
c04b98c75a
ipython: updated version check.
...
Version 4.0.0 changes API and breaks WA's usage of it. Updated version
check to only initialize ipython utils if version is < 4.0.0.
2015-09-02 17:39:11 +01:00
3501eccb8e
energy_model: yet another adjustment to leakage compensation.
2015-09-02 15:31:13 +01:00
a80780b9ed
energy_model: idle state fix
2015-09-01 16:19:53 +01:00
1b6b0907f9
energy_model: further fixes to idle measurement.
...
- Fix to core frequency to min
- Only disable idle states that are deeper than the measured state.
Keep shallower states enabed.
2015-08-28 17:12:24 +01:00
227c39d95d
trace: fix data file trace name in warning
...
The warning refers to trace.bin, which is not the extension that we use
for it. Instead, use the variable with the default trace
name (trace.dat) for the warning.
2015-08-28 15:52:43 +01:00
25b9350fea
energy_model: fixing idle state handling to encompass all active cores on a cluster.
2015-08-26 12:48:22 +01:00
9b7b57a4d4
androbench: removed trailing spaces from metric names.
2015-08-21 13:58:00 +01:00
5aae705172
androbench: added early check for root on device; updated description with link to website.
2015-08-21 09:37:53 +01:00
b653beacd3
Merge pull request #45 from bathepawan/master
...
Androbench Storage Benchmark
2015-08-21 09:16:28 +01:00
88f57e5251
Agenda: default config to dict
2015-08-21 08:34:49 +01:00
f44fd9df7a
Change device pull to handle root,and renamed local file as well history.db from results.db
2015-08-20 21:25:55 +05:30
8513304aeb
python sqlite3 to remove host/DUT dependencies + other changes
2015-08-19 21:53:50 +05:30
e38c87f258
Fix pylint errors
2015-08-18 01:28:09 +05:30
03a9470007
Fixed issue reported by pep8 checkers
2015-08-18 01:09:45 +05:30
2d8b8ba799
moved UiAutomator code to java, and removed not required python code for device interfacing
2015-08-18 00:46:48 +05:30
e4ee496bc9
Use sqllite3 instead of sqllite to remove host dependencies and result unit in proper format
2015-08-17 23:20:30 +05:30
205934d55b
juno: use bootargs on hard_reset with u-boot
2015-08-17 14:36:28 +01:00
95c3f049fb
trace_cmd: handle trace headers and thread names with spaces
2015-08-17 12:37:58 +01:00
25c0fd7b8b
Allow setting classifiers via agenda.
2015-08-17 10:37:40 +01:00
1cf60c2615
Results format
2015-08-15 02:39:45 +05:30
c3d8128ff3
Androbench Storage Benchmark Workload Addition
2015-08-15 02:10:08 +05:30
aa2ae03ce6
daq: daq_power.csv now matches the order of labels (if specified).
2015-08-13 16:10:58 +01:00
6137d5650f
delay: added fixed_before_start configuration; updated all to be very_slow
...
- fixed_before_start is a fixed time period equivalent to
temperature_before_start
- Changed existing *_between_specs and *_between_iterations callbacks to
be very_slow, as they should always run after everything else.
2015-08-12 16:12:38 +01:00
0703db05cf
rt-app: relaxing timeouts.
2015-08-12 11:53:29 +01:00
0e1990a2bb
ssh: fixed error reporting on failed exit code extraction.
2015-08-12 11:28:38 +01:00
14f5858e3d
pep8
2015-08-11 17:02:38 +01:00
a6d374bcff
wa create workload: better error reporting
...
Give more informative errors if Android SDK installed but no
platform has been downloaded.
2015-08-11 16:59:25 +01:00
85eba9c37a
Better error reporting for subprocess.CalledProcessError
2015-08-11 16:51:34 +01:00
1d67dd3b99
energy_model: adjusting to compensate for leakage when evaluating cluster power
2015-08-10 10:48:26 +01:00
ab45c4499f
utils.misc.normalize: only normalize string keys.
2015-08-10 10:42:20 +01:00
75e4b7d2ae
fps: filter out bogus actual-present times and ignore janks above PAUSE_LATENCY
...
The PAUSE_LATENCY thing is in line with what Google do inside
SurfaceFlingerHelper.
2015-08-06 11:20:58 +01:00
5e92728d77
pylint fixes
2015-07-27 10:11:59 +01:00
b5d879a90b
APK workloads: added an option to skip host-side APK check entirely.
2015-07-27 09:55:23 +01:00
507efaec48
perf: reverting type change for optionstring
2015-07-24 16:20:14 +01:00
0db14e2466
Updating energy_model to mitigate cluster power issue
...
A quadratic is now fitted to single and two-core power measured across
frequencies. This quadratic is then used in projection of cluster power.
This mitigates issues with cluster powers going negative or "crossing
over".
2015-07-24 08:11:23 +01:00
6ced04daf0
Fix for the previous fix...
2015-07-23 10:25:13 +01:00
b6064fa348
perf: fix for Android: killall sleep as root.
2015-07-23 10:17:01 +01:00
