aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* Fixes.Jonathan Herman2013-04-27
|
* Bug fixes from testing.Jonathan Herman2013-04-25
|
* Merge branch 'master' into wip-color-mcJonathan Herman2013-04-23
|\ | | | | | | | | | | | | Conflicts: gen/generator.py parse/sched.py parse_exps.py
| * Improved accuracy of sched_trace measurement parsing.Jonathan Herman2013-04-23
| | | | | | | | | | | | | | | | | | | | | | | | * Measurements from tasks missing > 20% of their scheduling records are ignored. This is configurable in config/config.py. * Measurements which only have zero values are ignored. * If either of these 2 situations are encountered print out a message the first time using the common.log_once() method. See parse_exps.py for how this is used with multiple threads. * Measurements from a task's last job are ignored. * Miss ratio is calculated only as a fraction of the number of jobs whose matching release and completion records were found, not just release.
| * Improved robustness of run_exps.py execution.Jonathan Herman2013-04-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Thanks to bcw and gelliott for debugging and ideas. * Print out experiment number and total experiments when starting experiments. * Only sleep and re-release tasks if tasks are waiting to release. * Fail experiment with verbose messages if any tasks fail before becoming ready to release. * When waiting for tasks to become ready for release, reset the waiting time whenever a new task (or task(s)) become ready. * Start regular tracers BEFORE the plugin switch to log data from the switch. * Check the number of running tasks AFTER trying to switch the linux scheduler. This gives plugin deactivate code the opportunity to kill these tasks. * If an invalid executable is specified in the schedule file, fail before attempting to run the experiment and print out the problem. * Propogate exceptions up from experiment failures instead of creating ExperimentFailed exceptions. This commit also made clock-frequency automatically ignored by parse_exps.py. The value of this would change by +- a Mhz between experiments, ruining graphs.
| * Scripts read directories created by other scripts if no arguments.Jonathan Herman2013-04-21
| | | | | | | | | | | | | | | | | | | | With no arguments, all scripts first try to load the current directory. If the current directory has no data, the scripts search for the output of the previous scripts in the toolchain, e.g. parse_exps.py loads run-data/*, created by run_exps.py. This commit also switched messages to stderr where they belong, and adds in missing lock and unlock overheads.
| * Made default generator behavior more intuitive.Jonathan Herman2013-04-19
| | | | | | | | | | | | GenOption defaults can be specified as a single value in addition to a list of values. PartitionedEdfGenerator's now use worst-fit partitioning for the most even distribution of tasks.
| * Switched sched_trace data to verbose ctypes structs.Jonathan Herman2013-04-18
| |
| * Merge branch 'master' of github.com:hermanjl/experiment-scriptsBryan Ward2013-04-17
| |\
| | * Fixed calculation of tardiness.Jonathan Herman2013-04-17
| | |
| | * Cleaned up sched_trace output and code.Jonathan Herman2013-04-17
| | |
| * | which -> thatBryan Ward2013-04-17
| |/
* | Created infrastructure for calculating scaling factors.Jonathan Herman2013-04-17
| |
* | Added ColorMcGenerator.Jonathan Herman2013-04-15
| |
* | Added mixed-criticality generator.Jonathan Herman2013-04-15
|/
* Updated readme to reflect new multiprocessing options.Jonathan Herman2013-04-12
|
* Formatting error in readme.Jonathan Herman2013-04-12
|
* Improved error handling in parse_ and plot_exps.py.Jonathan Herman2013-04-12
|
* Improved error handling in run_exps.py.Jonathan Herman2013-04-12
|
* Added option to send email when experiments complete.Jonathan Herman2013-04-10
|
* run_exps will run any command whose last argument is the duration.Jonathan Herman2013-04-10
|
* Updated README.md to reflect recent changes.Jonathan Herman2013-04-08
|
* Create generator map in gen/__init__.py.Jonathan Herman2013-04-08
|
* Replaced num_tasks with tasks for consistency.Jonathan Herman2013-04-08
|
* Added parameter file options for tracers and pre/post experiment commands.Jonathan Herman2013-04-08
|
* Added -j option to send jabber update messages.Jonathan Herman2013-04-08
|
* Removed poorly cherry-picked code.Jonathan Herman2013-04-02
|
* Minor bug fixes from end-to-end testing.Jonathan Herman2013-04-02
|
* Added option to ignore environment issues and rewrote part of the generation ↵Jonathan Herman2013-04-02
| | | | scripts for easier generator creation.
* Formatting edits.Jonathan Herman2013-03-19
|
* Formatted rest of README.Jonathan Herman2013-03-19
|
* Formatted first half of README.Jonathan Herman2013-03-19
|
* Moved README to README.md.Jonathan Herman2013-03-19
|
* Added README.Jonathan Herman2013-03-19
|
* Allow csvs in plot_exps which don't have configurations in their name.Jonathan Herman2013-03-18
|
* Seperate directory output in plot_exps.py.Jonathan Herman2013-03-18
|
* Allow users to specify required kernel configuration options.Jonathan Herman2013-03-18
|
* Remove unused 'clean' option.Jonathan Herman2013-03-18
|
* Use smarter defaults which can handle data from Bjorns scripts.Jonathan Herman2013-03-18
|
* Made trace-cmd support optional.Jonathan Herman2013-03-18
|
* Automatically merge experiments with the trial parameter.Jonathan Herman2013-03-18
|
* Saved trial number in experiment parameters.Jonathan Herman2013-03-07
|
* Moved num_tasks option into base Generator class.Jonathan Herman2013-03-07
|
* Seperated BaseGenerator into EdfGenerator and its superclass Generator.Jonathan Herman2013-03-07
|
* Seperated BaseGenerator into EdfGenerator and its superclass Generator.Jonathan Herman2013-03-07
|
* Added num-trials option to gen_exps.pyJonathan Herman2013-03-07
|
* Many bugfixes motivated by some end-to-end testing.Jonathan Herman2013-02-21
|
* Shortened experiment log lines.Jonathan Herman2013-02-17
|
* Save feather-trace timer frequency, not CPU frequency, with each experiment.Jonathan Herman2013-02-17
|
* Replaced fixed sleep times in experiment setup with loops which sleep until ↵Jonathan Herman2013-02-14
| | | | events occur.