This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: GDB 7.6.1 release Thu Aug 29?
- From: Joel Brobecker <brobecker at adacore dot com>
- To: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Thu, 22 Aug 2013 07:11:15 -0700
- Subject: Re: GDB 7.6.1 release Thu Aug 29?
- References: <20130821150945 dot GD5266 at adacore dot com> <20130822132552 dot GA23257 at host2 dot jankratochvil dot net>
> I believe the Todo: list should have been fixed in 7.6.1, at least these list
> regressions of 7.6 against previous FSF GDB releases:
Argh! I completely missed the fact that the TODO list is not empty.
> [pedro/tom] Fix CTRL-C 7.6 regression
> - Fixed but not pushed to 7.6:
> http://sourceware.org/bugzilla/show_bug.cgi?id=15603
A fairly big patch, touching sensitive areas (Eg: handle_inferior_event).
Fortunately, it's been in HEAD for 2 months now, so we have a bit of
testing behind our belt. I'd like both Pedro and Tom's approval
for pushing the the 7.6 branch.
> [dje] Fix gdbserver socket leak 7.5 regression
> - Simple but not yet done by anyone, a bit annoying in practice:
> http://sourceware.org/bugzilla/show_bug.cgi?id=15604
Not critical, IMO, and not a regression in the strict sense. I think
it can wait for 7.7.
> [jan] Fix argv[0] symlink regression
> - Not investigated yet, to be done these days:
> http://sourceware.org/bugzilla/show_bug.cgi?id=15415
An annoying regression indeed. Let's try to get a fix within
the next week or so - do you think that would be possible?
> I have no idea about these items:
>
> [dje] The mi fullname change is causing issues for Eclipse users
> that may not have been anticipated.
In the PR, Doug says he does not know how to handle this, since
he got the OK from MarcK re: Eclipse. And we don't know for sure
what's causing the issues in Eclipse either. I suggest we keep it
out of 7.6.1. We can produce a 7.6.2 with a fix afterwards if
we understand the problem better and deem it to be critical in
nature.
> [keiths/dje] Revert
> http://sourceware.org/ml/gdb-patches/2013-03/msg00251.html for
> msymbol_hash_iw [or use strspn?]. Doug to test on his
> mega-application.
This was on the 7.6 TODO, IIRC, and we decided to ignore it for
the first release. The URL points to:
[RFA] cleanup: while (isspace) -> skip_spaces{,_const}
Not sure what the problem is. I suggest the same as above:
Ignore for 7.6.1 unless we have more data, and possibly keep
the option of a 7.6.2 if we have a critical problem to fix.
Note that the tentative 7.7 branch is getting closer, and I would
delay it a little bit, but not too much in order to not disrupt
the transition to git too much.
Thanks Jan!
--
Joel