This maps a monotonic timestamp to the
corresponding real-time timestamp, which
can be used to match up the traces with
other logs that use real-time.
Also write clock_sync records first instead of at
the end, to avoid not being to write it due to the
buffer being full.
Bug: 23668823
Change-Id: I644aeea496197e194ec30f808f754e3e043d905f
Cherry-pick of c2c6ecd119 from AOSP
Tracing the state of the voltage and current regulators enables a developer
to attribute power consumption measurements to specific perhipherals and
cores. Unlike other indirect methods of tracing when peripherals and cores
are toggled, the regulator ftrace event indicates the exact point in time
when the kernel switches on or off the device. For example, in the following
trace, a developer can see exactly when the krait2 core starts receiving
power, and and when the voltage is set for voltage scaling.
mpdecision-2172 [000] ...1 1566.665481: regulator_enable: name=krait2
mpdecision-2172 [000] ...1 1566.665493: regulator_enable_delay: name=krait2
mpdecision-2172 [000] ...1 1566.665495: regulator_enable_complete: name=krait2
<...>-4133 [002] ...1 1566.666891: regulator_set_voltage: name=krait2 (1075000-1100000)
Change-Id: Ia322206ef74496daf1d9baa03545d8e89e398487
Signed-off-by: Aaron Schulman <aschulman@google.com>
The availability of these sysfs entries will vary based on the kernel's
supported filesystems, so don't block the disk category if some of them
are missing.
Bug: 18467498
Change-Id: I96200f1b5cf3f1a8fa8186602552adc96ddffc04
Signed-off-by: Greg Hackmann <ghackmann@google.com>
Both Sync Manager tracing and kernel sync tracing were using "sync" to
enable tracing and since Sync Manager was found first, this prevented
enabling kernel sync tracing.
To fix this, use "syncman" to enable Sync Manager tracing.
Change-Id: Id3b799e01e5041c582cd752c8c40d3e36954f821
Signed-off-by: Iliyan Malchev <malchev@google.com>
Writing to trace_clock erases the trace buffer, even if the value
hasn't changed. This prevents use of --async_start and --async_dump
to leave background tracing running and dump after an even that
needs debugging, because --async_dump writes to trace_clock and
resets the buffer before it can read it.
Read and parse the current value from trace_clock before writing,
and skip the write if the value isn't changing.
Change-Id: Ia2ec5bb654fb0bd179771b511ff261731ba47dca
This change adds support for the -a command line argument to atrace. This
argument sets the comma separated list of application names (i.e.
/proc/self/cmdline values) for which to enable application-level tracing.
Change-Id: I812d57c4cdc581943247229413f09b7e1adc10d0
This change adds support for tracing specific kernel function calls using the
function_graph tracer. It adds a '-k' option to atrace that accepts a comma
separated list of kernel function names for which tracing will be enabled.
Change-Id: I872b2f1d474b8ebb904053853fc8cf8c0a98089c
This change switches atrace to use a table-driven approach to enabling and
disabling options. It unifies how framework and kernel tracing are enabled,
and causes userland tracing options to be picked up by currently running
processes.
Change-Id: Iba2a3012ca0a67c4defdd076bce597db26e9f539
Martijn Coenen