| Commit message (Collapse) | Author | Age |
| |
|
| |
|
| |
|
|
|
|
| |
Print Info out
|
| |
|
| |
|
|
|
|
| |
2) Remove pairing switch_to event from list
|
|
|
|
| |
Fix minor bugs
|
|
|
|
|
| |
2)Dump each stat to csv file
3)Print all result globally (not by cluster)
|
| |
|
|
|
|
| |
2)Prevent crush from missing Events
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Priorty Inversion.
Support New Feather Trace Event:
ST_TASKLET_RELEASE, ST_TASKLET_BEGIN, ST_TASKLET_END,
ST_WORK_RELEASE, ST_WORK_BEGIN, ST_WORK_END,
and, ST_EFF_PRIO_CHANGE
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In order to sort records from different files into a single
stream (in order of timestamp), unit-trace uses per-cpu buffers;
the length of the buffers was previously hardcoded. If that
length proved insufficient (resulting in out-of-order records),
before this commit, unit-trace produced a FATAL ERROR and
terminated.
Now, unit-trace merely keeps a record of all out-of-order
records and prints a warning at the end, listing them.
The motivation for this change was the observation that
at least some times, grossly out-of-order errors were at
the very beginning of the trace (e.g. task system release),
so they don't really matter. If we know the IDs of the records
that are unordered, we are able to check (with the -o output)
to see if their misordering actually matters or not.
Moreover, the buffer size can now be specified with -b, and
the previously hard-coded value (200) is the default. Making
this number smaller greatly improves runtime, and vice versa.
I suspect that further investigation into the problem of sorting
records will show that the current method is overkill; down the
road, we may be able to replace this method with something much
faster. (The current method has the advantage that it is extremely
scalable, but I don't think that pays off for the size of traces
we typically examine.)
|
|\
| |
| |
| |
| |
| |
| | |
- Adds time_per_maj option to visualizer
Conflicts:
unit-trace
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| | |
In the case of viz, the results in errors about not having
pycairo and pygtk potentially being printed only if the user actually
uses the -v flag
|
|/
|
|
|
|
| |
- Zapped trailing whitespace
- Bolded table headers
- Added sentence about submodule doc
|
|
|
|
| |
not require gtk/cairo, and a few other minor things.
|
| |
|
|\
| |
| |
| |
| |
| |
| |
| |
| | |
ssh://cvs.cs.unc.edu/cvs/proj/litmus/repo/unit-trace into wip-gary
Conflicts:
unit_trace/viz/draw.py
unit_trace/viz/schedule.py
unit_trace/viz/viewer.py
|
| |
| |
| |
| | |
These are outdated and I don't feel like maintaining them.
|
| |
| |
| |
| |
| | |
- Public release info
- Minor clarifications
|
| |
| |
| |
| |
| | |
- Gotchas section (including a note on time units)
- Bugs section
|
| | |
|
| |
| |
| |
| |
| | |
This much more accurately captures what is going on and should
greatly reduce confusion
|
| | |
|
| |\ |
|
| | | |
|
| | |
| | |
| | |
| | | |
Also, removed obsolete README file
|
| |/ |
|
| | |
|
| | |
|
| |
| |
| |
| | |
This allows the st_many_gsn trace to be parsed.
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| | |
This provides two advantages:
(1) Don't need root to install unit-trace
(2) Someone messing with the unit-trace code on a shared machine
won't mess up anyone else who's also using unit-trace on
that machine
|
| | |
|
| |
| |
| |
| |
| | |
A bug caused traces with < 100 records (the number of records
buffered and sorted) to yield no records.
|
| | |
|
| |
| |
| |
| |
| | |
- Check and print error if G-EDF test not enabled
- Clean up output
|
| |
| |
| |
| |
| | |
Previously this was an "intermediate" module. Makes more sense
to have it as an output module.
|
| | |
|
| | |
|