| Commit message (Collapse) | Author | Age |
|---|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
| |
Each plot processes only the records which fit within the time window displayed.
When passing records into a real-time plot, kernelshark will now decide if a
record is within the current time window using real-time timestamps, instead of
the old ftrace timestamps.
This change also enables infinite zooming into the real-time plots.
|
| | |
|
| |
|
|
| |
Currently they copy the functionality of regular task.
|
| | |
|
| |
|
|
|
| |
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
| |
The restart of drawing a cpu plot may still have a left over record.
Free it before initializing a new drawing.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
| |
When destroying a CPU plot, make sure the last record is freed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
| |
Currently when you zoom into a CPU plot, when there are no events in
the viewable area, the line turns blank (as if it were just idle).
This patch keeps the plot with a bar even if no events are visible.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Currently, if a zoomed view has events before the start of the view,
those events are not examined to determine if the bar of a graph should
be shown at the start.
This patch adds code to the cpu and task plots to look for events
that happened before the view to determine if a bar should be shown
or not.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
| |
The graph was using the "orig_pid" to determine if it should display
the box (process running) or not. But this only worked when we had
sched events available. When schedule events were enabled, this would
fail the tests.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
In the plot code, it is possible that an event exists in the
data file that was not stored in the events log. This happened
with the 2.6.35 kernel where some events had formats that were "TOO BIG"
and that event format was not saved. But the events still existed in
the trace.dat file.
The plot code assumed that events were found and when they were not
it crashed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
| |
Freeing records that were returned via "peek event" made things
a bit more complex. Since the original code was written to determine
that peek events should not be freed.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If a callback wants to hold on to a record,
it can then increase the refcount and keep
it. This is most likely to happen with the
upcoming python plugin.
Since the underlying data pages are already
refcounted, we can just add a refcount to
struct record (rather than having the error
checking with the 'locked' member).
Acked-by: Darren Hart <dvhltc@us.ibm.com>
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
|
| |
|
|
|
|
|
|
| |
Move the reading of missed comms from sched_switch events from
trace-plot-cpu.c into trace-graph.c into the sched_switch check
itself.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
| |
Added menu for adding and removing CPU plots and Tasks.
Only CPU updates have been implemented in this change set.
Task updates will follow.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
All the files that are used to create the libraries:
libparsevent and libtracecmd are converted to the LGPL
as well as the files to create the plugins.
All files now have a boilerplate header that states which
license that the file is under. A README file is created as
well as the COPYING.LIB which contains the text of the
LGPL.
All authors of the code that created these files have given
their Acks.
Acked-by: Darren Hart <dvhltc@us.ibm.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Josh Triplett <josh@joshtriplett.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
| |
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
| |
Move the code from initializing the start cursor locations of the
CPUs.
Also control when a record is passed to the plot based on its
timestamp.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
| |
Change the code to iterate once down all records to create all plots.
Use the cpu_hash, task_hash and all_recs to find what plot cares about what
data.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
| |
Add a hash in the graph info to quickly look up to see if plots
have been registered to a pid or cpus. This is currently not used
but will be in the following patches.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
| |
Added a destructor routine to the plot callbacks to be called before
the plot is destroyed. This allows the plot to clean up any resources
that it allocated before it is freed.
Also did some other minor clean ups and freeing.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
| |
Move the code to show what to display into the plot sturctures.
Have the plot structure determine what to show.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
| |
The popup needs a record to use to determin if a task should be
added to the filter or not. Add the method find_record to the
plot callback structure.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
|
| |
Move the decisions to what to plot into the cpu specific file.
Now the trace-graph.c can just manange the drawing.
The code to show the fly-over windows still needs to be made
generic.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
| |
|
|
|
|
|
|
|
| |
Make the adjustment of the graph into a plot callback.
That is, start from the first plot and work down to see
if the plot has an exact match and should be displayed in
the graph viewable area (adjust scrollbar).
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
|
|
This is a start of moving from plotting just the CPUs, but
to move the code to a plot interface. This will eventually
allow a line to just plot a task (on all CPUs).
This first patch just sets up some of the infratructure,
but still is tied to the CPU plots.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
|
