This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Patch to add snprintf/vsnprintf to libiberty [take 2]


 > From: DJ Delorie <dj@redhat.com>
 > 
 > > Well we prototype asprintf/vasprintf.  I thought these would work
 > > similarly.  I don't care, if you want I'll take them out.  Just re-ack
 > > either way.
 > 
 > I'm just asking at the moment.  It's a fairly new function, so OSs
 > implementing it are likely to have followed the spec, compared to
 > something pre-void or pre-size_t.


Ok, I don't have an opinion.  Perhaps I'll leave it the way it is
unless you specifically decide otherwise.  If we encounter problems,
we can always take them out later.


 > > I have a blanket gcc assignment on file, I assumed that covered
 > > libiberty since I thought gcc is considered the "master" for that
 > > directory.
 > 
 > It tends to, but for some things put directly into libiberty that are
 > independent of gcc, a separate libiberty assignment is useful.

Shall I wait for someone to decide whether its small enough or do I
have to initiate paperwork?


 > > WRT libstdc++-v3, I believe that GPL and/or LGPL files are not
 > > supposed to be linked into libstdc++-v3, which is why I said "public
 > > domain" cause Paolo wanted to use them.  The other libiberty files
 > > linked into libstdc++-v3 all either have public domain, the berkeley
 > > license or no license.  See the comment above the `NEEDED' variable in
 > > libiberty/Makefile.in.
 > 
 > We can do a GPL+exception license, similar to what libstdc++-v2 itself
 > uses.

I'm fine with GPL+exception.  So if I cut&paste the text at the top of
libgcc2.c, would that suffice?


 > Besides, a different type of disclaimer from your employer/school is
 > needed if you put things in the public domain, compared to assigning
 > them to the FSF.  And if you are assigning it to the FSF, shouldn't
 > the FSF (via the SC) decide the license?

Yikes! :-) I supposed you're right but we're talking about 20 lines in
the vsnprintf function and my snprintf is just a wrapper for it.
However, I'll raise it with the SC and see what they say.



 > >  > No embedded texinfo documentation?
 > > 
 > > Sorry I didn't understand it.  How does it work?
 > 
 > See libiberty/getcwd.c or libiberty/xmalloc.c
 > 
 > Configure with --enable-maintainer-mode and make will build
 > $srcdir/functions.texi from the source comments.  You'll need perl.


Got it.  I'll take a stab at this while waiting for a decision on the
license.

		--Kaveh
--
Kaveh R. Ghazi			Director of Systems Architecture
ghazi@caip.rutgers.edu		Qwest Solutions


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]