summaryrefslogtreecommitdiff
path: root/gpsprof
Commit message (Collapse)AuthorAgeFilesLines
* Fix an error in the whole-cycle visualizations.Eric S. Raymond2011-09-301-1/+1
|
* Avoid failure on very long gpsprof runs.Eric S. Raymond2011-09-291-2/+2
|
* Fix bug in data() method for spaceplotting.Eric S. Raymond2011-09-291-9/+8
|
* Add whole-cycle profiling to show vendor binary orotocols don't cut latency.Eric S. Raymond2011-09-241-4/+4
|
* Change the semantics of gpsprof -d so we can both dump and plot in one run.Eric S. Raymond2011-09-241-14/+18
|
* Refactor gpsprof to put most plotting machinery in a common base class.Eric S. Raymond2011-09-241-149/+142
|
* Improved captioning in the plots.Eric S. Raymond2011-09-231-22/+28
|
* Restore gpsprof's -T option.Eric S. Raymond2011-09-231-4/+9
|
* If the profiling code isn't compiled in, timing policy bit must be forced off.Eric S. Raymond2011-09-231-1/+1
|
* Dump (-d) option for gpsprof, and updated documentation.Eric S. Raymond2011-09-221-27/+35
|
* Add a useful header.Eric S. Raymond2011-09-221-3/+4
|
* Plot bars in a more useful order.Eric S. Raymond2011-09-221-6/+7
|
* Profiling can isolate RS232 transmission time.Eric S. Raymond2011-09-221-2/+4
|
* Refactor so the data method has access to device parameters.Eric S. Raymond2011-09-221-7/+7
|
* Prfile numbers now include a per-cycle chracter count.Eric S. Raymond2011-09-221-3/+4
|
* Describe the latency components.Eric S. Raymond2011-09-211-8/+15
|
* Refactor the raw plot to make adding new numbers easier.Eric S. Raymond2011-09-211-6/+6
|
* Add decode time to the profiling.Eric S. Raymond2011-09-211-5/+7
|
* Polis h the profiling code.Eric S. Raymond2011-09-211-2/+2
|
* Restore raw plot capability.Eric S. Raymond2011-09-211-2/+41
|
* More conversion to the new Python interface.Eric S. Raymond2011-09-211-2/+2
|
* Fix the gpsprof device query to work with the new Python interface.Eric S. Raymond2011-09-211-29/+18
|
* Make the 'uninstrumented' plot work again.Eric S. Raymond2011-09-211-2/+2
|
* Remove the old latency-profiling machinery.Eric S. Raymond2011-09-201-200/+2
| | | | | | It hasn't actually worked since we switched protocols to JSON. This diff puts the framework in place for new machinery using timing attributes in the TPV sentence.
* gpsprof is pychecker clean.Eric S. Raymond2011-08-231-13/+11
|
* Proposed improvement of gpsprofOliver Kurz2011-04-041-6/+13
| | | | | | | | | | | | | | | I used the "gpsprof" tool in conjunction with "gpsfake" to produce some great looking scatter plots. While working with log files including thousands of samples I recognized that the result is not looking good. The CEP circle is covered and the plot looks quite crowded. I propose a simple patch to gpsprof which adds three things: - Use "dots" instead of "points" when there are more than 1000 samples - Plot the CEP circle after the sample points, not before - Also plot 95 and 99% plot Signed-off-by: Eric S. Raymond <esr@thyrsus.com>
* Corrected bad reference to RS3232 in gpsprofJon Schlueter2011-03-051-1/+1
|
* Fix tracker bug #17832.Eric S. Raymond2011-01-251-1/+1
|
* The raw_hook mwember and set_raw_hook are gone; client API cleanup is done.Eric S. Raymond2011-01-251-1/+0
|
* Remove some unnecessary and broken code.Eric S. Raymond2010-12-161-5/+0
|
* Fixing several Python identation errors.Bernd Zeimetz2010-08-291-1/+1
|
* Carry through and document the poll() -> read() change.Eric S. Raymond2010-07-141-4/+4
|
* Revert last patch on layering of plot data.Gary E. Miller2010-04-221-3/+9
| | | | | Add EP (50%) of altitude data. My Python is pathetic, feel free to improve.
* In gpsprof, put CEP on top of the individual fixes.Gary E. Miller2010-04-221-2/+2
| | | | | That way when plotting a large number of fixes the CEP does not get covered up.
* Patch to make gpsprof use the same -D as the other gpsd helper programs. Gary E. Miller2010-04-211-4/+5
| | | | | | With man page update. Signed-off-by: Gary E. Miller <gem@rellim.com>
* Document gpsprof -v, allow > 1 level of verbosity with -vv and -vvvv. Gary E. Miller2010-04-211-3/+3
| | | | | Probably should use the -D X syntax but my Python is not that good.
* Use setup.py to handle the Python shebangs.Bernd Zeimetz2010-04-201-0/+487
|
* Substitute $PYTHON instead of assuming 'python' is python. Greg Troxel2009-12-021-481/+0
| | | | Now 'make check' runs on NetbSD, and $(prefix)/bin/gpsfake works.
* Fix typo in commentGary E. Miller2009-11-111-1/+1
|
* Add comment and remove redundant else.Gary E. Miller2009-11-111-6/+6
|
* gpsprof is fully working with the new protocol.Eric S. Raymond2009-09-291-20/+15
| | | | All regression test pass. Codebase splints clean.
* The raw plot works now.Eric S. Raymond2009-09-291-18/+17
|
* Raw plotting works.Eric S. Raymond2009-09-291-17/+16
|
* Rip out most of the old profiling support.Eric S. Raymond2009-09-291-1/+4
| | | | | Replace it with a new, lighter- weight version that can be configured out.
* Uninstrumented plot now works.Eric S. Raymond2009-09-291-1/+7
|
* Spaceplot function of gpsprof works with new protocol.Eric S. Raymond2009-09-281-43/+58
|
* More virtuous code removal.Eric S. Raymond2009-09-271-8/+4
| | | | | | There's no longer any need to collect poll-time statistics, as clients won't poll any more. All regression tests pass. Codebase splints clean.
* Better description of new Python API. Eric S. Raymond2009-09-241-4/+6
| | | | | query() is really gone; add a resulting to-do item. Codebase splints clean, all regression tests pass.
* Rebuild regressions...Eric S. Raymond2006-11-101-2/+4
| | | | | | | | ...because the way modes and error estimates are reported has legitimately changed. Also, make gpsfake and gpsprof less noisy when their output is going to a file rather than a tty.
* Make it clearer that gpsprof is importing a non-system module.Eric S. Raymond2006-10-251-1/+2
|
View Lorry Controller Status