avrdude-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [avrdude-dev] AVRDUDE 6.0rc1


From: Joerg Wunsch
Subject: Re: [avrdude-dev] AVRDUDE 6.0rc1
Date: Fri, 17 May 2013 08:31:17 +0200
User-agent: Mutt/1.5.21 (2010-09-15)

As Joerg Wunsch wrote:

> I just bumped the version in configure.ac to "6.0rc1"; it has
> been at "5.11svn" for way too long now.  Yes, this is a strong
> indication we are heading for a release. ;-)

I'd like to invite everyone to browse through the bug and patch
trackers.  You might apply everything where there is clearly no doubt,
and no need for further prior discussions in the group.  Be careful to
not introduce regressions right before a release though.  When adding
new device entries, it's best if they could be derived from an
existing one using the new "parent" keyword, to avoid too much
copy&paste.  Don't forget to update documentation where needed (the
list of supported programmers and devices in the documentation is
auto-generated).

Personally, my time budget is too limited to handle all
of the existing submissions without delaying the new release
indefinitely, I'm afraid.

Upon a user request, I added the "-l logfile" option yesterday: it
calls freopen(logfile, "w", stderr), in order to redirect messages to
"logfile" which would normally go to stderr.  The user in question was
trying to invoke AVRDUDE from an environment which could not handle
shell-style stdio redirection (AVR Studio 4, in this case), but I
figured having such an option would be of general value to many users.
In particular, when recording debugging traces, we can now tell the
users to use "-l trace.txt -vvvv", and submit the resulting trace.txt
(e.g., as part of a bug tracker).

The only drawback of freopen() is that it closes the old stderr first.
If subsequently, opening the new logfile fails, there is no stderr to
complain to anymore; thus, I write the complaint to stdout then, and
exit at that point.  (I tried using fopen(), then close(stderr), and
assing the new stream to stderr.  However, this fails under MinGW as
stderr is not an lvalue there.)
-- 
cheers, Joerg               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/
Never trust an operating system you don't have sources for. ;-)



reply via email to

[Prev in Thread] Current Thread [Next in Thread]