Ok to push this? Sorry, I haven't yet explored how valgrind tests are
supposed to be added to gnulib.
>From 165ffc9aaa9bfe41a9977a80edf80e830d765ab9 Mon Sep 17 00:00:00 2001
From: Joel E. Denny
Date: Tue, 15 Dec 2009 02:43:07 -0500
Subject: [PATCH] fprintf-posix: fix memory leak
* lib/fprintf.
Hi Bruno,
I'd like to relax the areadlink's license to LGPLv2+,
since I'm about to use it from libvirt.
Is that ok with you?
Ralf Wildenhues wrote:
> * Jim Meyering wrote on Sun, Dec 13, 2009 at 11:44:15AM CET:
>> Paolo Bonzini wrote:
>> > This rule could actually be moved to Automake's distcheck target.
>>
>> Good idea.
>> FYI, here's the definition I have now:
>> (added GPERF and LIBTOOL, and made it overridable)
Hi R
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Bruno Haible on 12/11/2009 8:38 AM:
>> + switch (action)
>> +{
>> +case F_DUPFD_CLOEXEC:
>> + {
>> +int target = va_arg (arg, int);
>
> Is the arg of type 'int' or 'long'? POSIX says it's an 'int'. But
> the Linux ma
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
According to Eric Blake on 11/25/2009 6:41 AM:
> As discussed on coreutils; some Linux kernels support utimensat but not
> symlink timestamps. Now pushed.
This still has the potential for declaring an unused variable. Fixed as
follows:
- --
Don't w
