trappy.sched module

Definitions of scheduler events registered by the FTrace class

class trappy.sched.SchedContribScaleFactor

Bases: trappy.base.Base

Event to register tracing of contrib factor

name = 'sched_contrib_scale_factor'

parse_raw = False

unique_word = 'sched_contrib_scale_f:'

class trappy.sched.SchedCpuCapacity(parse_raw=False)

Bases: trappy.base.Base

Corresponds to Linux kernel trace event sched/cpu_capacity

finalize_object()

This renaming is necessary because our cpu related pivot is ‘cpu’ and not ‘cpu_id’. Otherwise you cannot ‘mix and match’ with other classes

name = 'cpu_capacity'

pivot = 'cpu'

The Pivot along which the data is orthogonal

unique_word = 'cpu_capacity:'

The unique word that will be matched in a trace line

class trappy.sched.SchedCpuFrequency(parse_raw=False)

Bases: trappy.base.Base

Corresponds to Linux kernel trace event power/cpu_frequency

finalize_object()

This renaming is necessary because our cpu related pivot is ‘cpu’ and not ‘cpu_id’. Otherwise you cannot ‘mix and match’ with other classes

name = 'cpu_frequency'

pivot = 'cpu'

The Pivot along which the data is orthogonal

unique_word = 'cpu_frequency:'

The unique word that will be matched in a trace line

class trappy.sched.SchedLoadAvgCpu

Bases: trappy.base.Base

Load and Utilization Signals for CPUs

name = 'sched_load_avg_cpu'

parse_raw = False

pivot = 'cpu'

unique_word = 'sched_load_avg_cpu:'

class trappy.sched.SchedLoadAvgSchedGroup(parse_raw=False)

Bases: trappy.base.Base

Corresponds to Linux kernel trace event sched_load_avg_sched_group

finalize_object()

This condition is necessary to force column ‘cpus’ to be printed as 8 digits w/ leading 0

name = 'sched_load_avg_sg'

pivot = 'cpus'

The Pivot along which the data is orthogonal

unique_word = 'sched_load_avg_sg:'

The unique word that will be matched in a trace line

class trappy.sched.SchedLoadAvgTask(parse_raw=False)

Bases: trappy.base.Base

Corresponds to Linux kernel trace event sched_load_avg_task

get_pids(key='')

Returns a list of (comm, pid) that contain ‘key’ in their ‘comm’.

name = 'sched_load_avg_task'

pivot = 'pid'

The Pivot along which the data is orthogonal

unique_word = 'sched_load_avg_task:'

The unique word that will be matched in a trace line

class trappy.sched.SchedSwitch

Bases: trappy.base.Base

Parse sched_switch

create_dataframe()

name = 'sched_switch'

unique_word = 'sched_switch:'

class trappy.sched.SchedWakeup

Bases: trappy.base.Base

Register SchedWakeup Event

name = 'sched_wakeup'

parse_raw = True

unique_word = 'sched_wakeup:'

class trappy.sched.SchedWakeupNew

Bases: trappy.base.Base

Register SchedWakeupNew Event

name = 'sched_wakeup_new'

parse_raw = True

unique_word = 'sched_wakeup_new:'