# *** WARNING: *** # Configuration listed in this file is NOT COMPLETE. This file sets the default # configuration for WA and gives EXAMPLES of other configuration available. It # is not supposed to be an exhaustive list. # PLEASE REFER TO WA DOCUMENTATION FOR THE COMPLETE LIST OF AVAILABLE # EXTENSIONS AND THEIR CONFIGURATION. # This defines when the device will be rebooted during Workload Automation execution. # # # # Valid policies are: # # never: The device will never be rebooted. # # as_needed: The device will only be rebooted if the need arises (e.g. if it # # becomes unresponsive # # initial: The device will be rebooted when the execution first starts, just before executing # # the first workload spec. # # each_spec: The device will be rebooted before running a new workload spec. # # each_iteration: The device will be rebooted before each new iteration. # # # reboot_policy = 'as_needed' # Defines the order in which the agenda spec will be executed. At the moment, # # the following execution orders are supported: # # # # by_iteration: The first iteration of each workload spec is executed one ofter the other, # # so all workloads are executed before proceeding on to the second iteration. # # This is the default if no order is explicitly specified. # # If multiple sections were specified, this will also split them up, so that specs # # in the same section are further apart in the execution order. # # by_section: Same as "by_iteration", but runn specs from the same section one after the other # # by_spec: All iterations of the first spec are executed before moving on to the next # # spec. This may also be specified as ``"classic"``, as this was the way # # workloads were executed in earlier versions of WA. # # random: Randomisizes the order in which specs run. # execution_order = 'by_iteration' # This indicates when a job will be re-run. # Possible values: # OK: This iteration has completed and no errors have been detected # PARTIAL: One or more instruments have failed (the iteration may still be running). # FAILED: The workload itself has failed. # ABORTED: The user interupted the workload # # If set to an empty list, a job will not be re-run ever. retry_on_status = ['FAILED', 'PARTIAL'] # How many times a job will be re-run before giving up max_retries = 3 # If WA should delete its files from the device after the run is completed clean_up = False #################################################################################################### ######################################### Device Settings ########################################## #################################################################################################### # Specify the device you want to run workload automation on. This must be a # # string with the ID of the device. At the moment, only 'TC2' is supported. # # # device = 'generic_android' # Configuration options that will be passed onto the device. These are obviously device-specific, # # so check the documentation for the particular device to find out which options and values are # # valid. The settings listed below are common to all devices # # # device_config = dict( # The name used by adb to identify the device. Use "adb devices" in bash to list # the devices currently seen by adb. #adb_name='10.109.173.2:5555', # The directory on the device that WA will use to push files to #working_directory='/sdcard/wa-working', # This specifies the device's CPU cores. The order must match how they # appear in cpufreq. The example below is for TC2. # core_names = ['a7', 'a7', 'a7', 'a15', 'a15'] # Specifies cluster mapping for the device's cores. # core_clusters = [0, 0, 0, 1, 1] ) #################################################################################################### ################################### Instrumention Configuration #################################### #################################################################################################### # This defines the additionnal instrumentation that will be enabled during workload execution, # # which in turn determines what additional data (such as /proc/interrupts content or Streamline # # traces) will be available in the results directory. # # # instrumentation = [ # Records the time it took to run the workload 'execution_time', # Collects /proc/interrupts before and after execution and does a diff. 'interrupts', # Collects the contents of/sys/devices/system/cpu before and after execution and does a diff. 'cpufreq', # Gets energy usage from the workload form HWMON devices # NOTE: the hardware needs to have the right sensors in order for this to work #'hwmon', # Run perf in the background during workload execution and then collect the results. perf is a # standard Linux performance analysis tool. #'perf', # Collect Streamline traces during workload execution. Streamline is part of DS-5 #'streamline', # Collects traces by interacting with Ftrace Linux kernel internal tracer #'trace-cmd', # Obtains the power consumption of the target device's core measured by National Instruments # Data Acquisition(DAQ) device. #'daq', # Collects CCI counter data. #'cci_pmu_logger', # Collects FPS (Frames Per Second) and related metrics (such as jank) from # the View of the workload (Note: only a single View per workload is # supported at the moment, so this is mainly useful for games). #'fps', ] #################################################################################################### ################################# Result Processors Configuration ################################## #################################################################################################### # Specifies how results will be processed and presented. # # # result_processors = [ # Creates a status.txt that provides a summary status for the run 'status', # Creates a results.txt file for each iteration that lists all collected metrics # in "name = value (units)" format 'standard', # Creates a results.csv that contains metrics for all iterations of all workloads # in the .csv format. 'csv', # Creates a summary.csv that contains summary metrics for all iterations of all # all in the .csv format. Summary metrics are defined on per-worklod basis # are typically things like overall scores. The contents of summary.csv are # always a subset of the contents of results.csv (if it is generated). #'summary_csv', # Creates a results.csv that contains metrics for all iterations of all workloads # in the JSON format #'json', # Write results to an sqlite3 database. By default, a new database will be # generated for each run, however it is possible to specify a path to an # existing DB file (see result processor configuration below), in which # case results from multiple runs may be stored in the one file. #'sqlite', ] #################################################################################################### ################################### Logging output Configuration ################################### #################################################################################################### # Specify the format of logging messages. The format uses the old formatting syntax: # # # # http://docs.python.org/2/library/stdtypes.html#string-formatting-operations # # # # The attributes that can be used in formats are listested here: # # # # http://docs.python.org/2/library/logging.html#logrecord-attributes # # # logging = { # Log file format 'file format': '%(asctime)s %(levelname)-8s %(name)s: %(message)s', # Verbose console output format 'verbose format': '%(asctime)s %(levelname)-8s %(name)s: %(message)s', # Regular console output format 'regular format': '%(levelname)-8s %(message)s', # Colouring the console output 'colour_enabled': True, } #################################################################################################### #################################### Instruments Configuration ##################################### #################################################################################################### # Instrumention Configuration is related to specific instrument's settings. Some of the # # instrumentations require specific settings in order for them to work. These settings are # # specified here. # # Note that these settings only take effect if the corresponding instrument is # enabled above. #################################################################################################### ######################################## perf configuration ######################################## # The hardware events such as instructions executed, cache-misses suffered, or branches # mispredicted to be reported by perf. Events can be obtained from the device by tpying # 'perf list'. #perf_events = ['migrations', 'cs'] # The perf options which can be obtained from man page for perf-record #perf_options = '-a -i' #################################################################################################### ####################################### hwmon configuration ######################################## # The kinds of sensors hwmon instrument will look for #hwmon_sensors = ['energy', 'temp'] #################################################################################################### ###################################### trace-cmd configuration ##################################### # trace-cmd events to be traced. The events can be found by rooting on the device then type # 'trace-cmd list -e' #trace_events = ['power*'] #################################################################################################### ######################################### DAQ configuration ######################################## # The host address of the machine that runs the daq Server which the instrument communicates with #daq_server_host = '10.1.17.56' # The port number for daq Server in which daq instrument communicates with #daq_server_port = 56788 # The values of resistors 1 and 2 (in Ohms) across which the voltages are measured #daq_resistor_values = [0.002, 0.002] #################################################################################################### ################################### cci_pmu_logger configuration ################################### # The events to be counted by PMU # NOTE: The number of events must not exceed the number of counters available (which is 4 for CCI-400) #cci_pmu_events = ['0x63', '0x83'] # The name of the events which will be used when reporting PMU counts #cci_pmu_event_labels = ['event_0x63', 'event_0x83'] # The period (in jiffies) between counter reads #cci_pmu_period = 15 #################################################################################################### ################################### fps configuration ############################################## # Data points below this FPS will dropped as not constituting "real" gameplay. The assumption # being that while actually running, the FPS in the game will not drop below X frames per second, # except on loading screens, menus, etc, which should not contribute to FPS calculation. #fps_drop_threshold=5 # If set to True, this will keep the raw dumpsys output in the results directory (this is maily # used for debugging). Note: frames.csv with collected frames data will always be generated # regardless of this setting. #fps_keep_raw=False #################################################################################################### ################################# Result Processor Configuration ################################### #################################################################################################### # Specifies an alternative database to store results in. If the file does not # exist, it will be created (the directiory of the file must exist however). If # the file does exist, the results will be added to the existing data set (each # run as a UUID, so results won't clash even if identical agendas were used). # Note that in order for this to work, the version of the schema used to generate # the DB file must match that of the schema used for the current run. Please # see "What's new" secition in WA docs to check if the schema has changed in # recent releases of WA. #sqlite_database = '/work/results/myresults.sqlite' # If the file specified by sqlite_database exists, setting this to True will # cause that file to be overwritten rather than updated -- existing results in # the file will be lost. #sqlite_overwrite = False # distribution: internal #################################################################################################### #################################### Resource Getter configuration ################################# #################################################################################################### # The location on your system where /arm/scratch is mounted. Used by # Scratch resource getter. #scratch_mount_point = '/arm/scratch' # end distribution