Online Developer Meeting (2020-07-28): Difference between revisions

From Octave
Jump to navigation Jump to search
(Create page.)
 
(→‎Octave 6 release status: Remove fixed bugs.)
Line 19: Line 19:
** still elusive crashes while running the test suite on the build bots (bug {{bug|57591}}) -> Maybe the broader user base of an RC can shed some light on "real-life frequency" of these crashes.
** still elusive crashes while running the test suite on the build bots (bug {{bug|57591}}) -> Maybe the broader user base of an RC can shed some light on "real-life frequency" of these crashes.
** possible background: interaction between graphics system and interpreter is different from interaction of rest of GUI with the interpreter. Possibly resolvable by a copy-on-write approach for graphics objects (postponed to Octave 7)
** possible background: interaction between graphics system and interpreter is different from interaction of rest of GUI with the interpreter. Possibly resolvable by a copy-on-write approach for graphics objects (postponed to Octave 7)
* Other important bugs
** calling script from nested functions (bug {{bug|58691}} fixed)
** empty list creation if one element is not assigned (bug {{bug|58686}} not a blocker)


* Try to tackle bugs with debuggers
* Try to tackle bugs with debuggers

Revision as of 07:28, 28 July 2020

Todays topics

  • Meet and greet 5 minutes before meeting (audio testing).

Octave 6 release status

  • We should create a first release candidate (RC) Octave 6.0.90 soon. Base for discussing and working on further bug findings.
  • Still hard to fix blocking bugs
    • Build (RC) anyway!
    • crashes during doc graphics creation seem to be fixed (bug #56952)
    • still elusive crashes while running the test suite on the build bots (bug #57591) -> Maybe the broader user base of an RC can shed some light on "real-life frequency" of these crashes.
    • possible background: interaction between graphics system and interpreter is different from interaction of rest of GUI with the interpreter. Possibly resolvable by a copy-on-write approach for graphics objects (postponed to Octave 7)
  • Try to tackle bugs with debuggers
    • valgrind slow
    • -fsanitize=thread in memory and faster

Ideas for next meeting

Topic suggestions

See also