Trouble bisecting

Hello,
I haven't bisected the Python source tree in a long time and it seems our current way of making releases is messing it up.
# Start on tip of branch 3.9 (3.9)$ git bisect start
# 3.9 tip is good (3.9|BISECTING)$ git bisect good
# Switch to release 3.9.1, where the bug wasn't fixed yet (3.9|BISECTING)$ git checkout v3.9.1 [...] ((v3.9.1)|BISECTING)$ LANG=C git bisect bad Some good revs are not ancestors of the bad rev. git bisect cannot work properly in this case. Maybe you mistook good and bad revs?
How can this be worked around? Not being able to bisect a fix or regression is really annoying.
Regards
Antoine.

Oh, my bad. I finally understood my mistake. "git bisect" always expects you to search for a regression, not the other way round. So "bad" revisions are supposed to be more recent than "good" revisions.
Regards
Antoine.
Le 26/06/2021 à 17:00, Antoine Pitrou a écrit :
Hello,
I haven't bisected the Python source tree in a long time and it seems our current way of making releases is messing it up.
# Start on tip of branch 3.9
(3.9)$ git bisect start
# 3.9 tip is good
(3.9|BISECTING)$ git bisect good
# Switch to release 3.9.1, where the bug wasn't fixed yet
(3.9|BISECTING)$ git checkout v3.9.1 [...] ((v3.9.1)|BISECTING)$ LANG=C git bisect bad Some good revs are not ancestors of the bad rev. git bisect cannot work properly in this case. Maybe you mistook good and bad revs?
How can this be worked around? Not being able to bisect a fix or regression is really annoying.
Regards
Antoine.
python-committers mailing list -- python-committers@python.org To unsubscribe send an email to python-committers-leave@python.org https://mail.python.org/mailman3/lists/python-committers.python.org/ Message archived at https://mail.python.org/archives/list/python-committers@python.org/message/T... Code of Conduct: https://www.python.org/psf/codeofconduct/

Hi
I'm always confused by expressions "good" and "bad". Previously, I added shell comments:
$ git bisect bad # slow $ git bisect good # fast
or:
$ git bisect bad # no leak $ git bisect good # leak
where good = leak is counter-intuitive...
Since it confused too many people (including myself), Git now allows to use a different terms than "good" and "bad" (or "old" and "new"). Example:
git bisect start --term-new=leak --term-bad=ok ...
Then you can type directly:
git bisect leak git bisect ok
See also "git bisect terms" to remind you which terms you picked ;-)
Victor
On Sat, Jun 26, 2021 at 5:33 PM Antoine Pitrou <antoine@python.org> wrote:
Oh, my bad. I finally understood my mistake. "git bisect" always expects you to search for a regression, not the other way round. So "bad" revisions are supposed to be more recent than "good" revisions.
Regards
Antoine.
Le 26/06/2021 à 17:00, Antoine Pitrou a écrit :
Hello,
I haven't bisected the Python source tree in a long time and it seems our current way of making releases is messing it up.
# Start on tip of branch 3.9
(3.9)$ git bisect start
# 3.9 tip is good
(3.9|BISECTING)$ git bisect good
# Switch to release 3.9.1, where the bug wasn't fixed yet
(3.9|BISECTING)$ git checkout v3.9.1 [...] ((v3.9.1)|BISECTING)$ LANG=C git bisect bad Some good revs are not ancestors of the bad rev. git bisect cannot work properly in this case. Maybe you mistook good and bad revs?
How can this be worked around? Not being able to bisect a fix or regression is really annoying.
Regards
Antoine.
python-committers mailing list -- python-committers@python.org To unsubscribe send an email to python-committers-leave@python.org https://mail.python.org/mailman3/lists/python-committers.python.org/ Message archived at https://mail.python.org/archives/list/python-committers@python.org/message/T... Code of Conduct: https://www.python.org/psf/codeofconduct/
python-committers mailing list -- python-committers@python.org To unsubscribe send an email to python-committers-leave@python.org https://mail.python.org/mailman3/lists/python-committers.python.org/ Message archived at https://mail.python.org/archives/list/python-committers@python.org/message/7... Code of Conduct: https://www.python.org/psf/codeofconduct/
-- Night gathers, and now my watch begins. It shall not end until my death.
participants (2)
-
Antoine Pitrou
-
Victor Stinner