https://issues.apache.org/bugzilla/show_bug.cgi?id=55153
--- Comment #12 from Harald Oehlmann <harald.oehlm...@elmicron.de> ---
E-Mail by Jan Nijtmans on clt:
2013/7/17 Harald Oehlmann <harald.oehlm...@elmicron.de>:
> Q1) Is this use-case reasonable to cover by the tcl core?
Yes
> Q2) May a wizard please review the code
Compare your solution with the solution already present
in the OSX-version of the notifier:
<http://core.tcl.tk/tcl/artifact/8c82932603?ln=512-521>
It's very similar. Looks OK to me, even though I'm
not a wizard regarding the UNIX notifier.
> Q3) Is this a bug fix or is a TIP required ?
Give that the bug only manifests itself on UNIX
with --enable-threads, I would say it's a bug. OSX
already has a similar solution to what you propose
and Windows doesn't have fork at all. Any Tcl
application which doesn't use threads should
behave the same whether threads are enabled or
not. Failing that it's a bug, and the best way
to demonstrate that is with a test-case.
> Q4) If reasonable, we should also back-port it to tcl8.5
Bugs are normally fixed in core-8-5-branch first, then
merged to trunk. I rebased "rfe-notifier-fork" to
core-8-5-branch now and added a "testfork" test
command. Adding a test-case is all that's left to
be approved for the core.
The OSX solution using pthread_atfork() disables
the OSX notifier during the fork. I don't think that's
necessary for the UNIX notifier, but maybe someone
else can comment on that.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: rivet-dev-unsubscr...@tcl.apache.org
For additional commands, e-mail: rivet-dev-h...@tcl.apache.org
