Re: Getting ready for 1.0.6
- From: Owen Taylor <otaylor redhat com>
- To: gtk-devel-list redhat com
- Subject: Re: Getting ready for 1.0.6
- Date: 15 Sep 1998 13:24:39 -0400
Aaron Digulla <digulla@wi-pc44.fh-konstanz.de> writes:
> Quoting Owen Taylor (otaylor@redhat.com):
>
> > I've been working recently on putting together a release 1.0.6 of
> > GTK+. It is my hope that this will be the last 1.0.x release of GTK+,
> > so I'd like to make sure it is good enough to last to GTK+-1.2.
>
> Ok, here is the compile report on Linux/i386 with egcs-1.1a:
>
> gdkimage.c: In function `gdk_image_new_bitmap':
> gdkimage.c:87: warning: implicit declaration of function `malloc'
> gdkimage.c:87: warning: assignment makes pointer from integer without a cast
> ggdkimage.c: In function `gdk_image_new':
> gdkimage.c:168: warning: assignment makes pointer from integer without a cast
> gdkimage.c:190: warning: assignment makes pointer from integer without a cast
> gdkimage.c:199: warning: implicit declaration of function `free'
> gdkimage.c:282: warning: assignment makes pointer from integer without a cast
>
> Why is malloc used here ? It's not used anywhere else, right ?
It was used there because some pointers (but not actually the those
ones...) were being passed to X which freed them with free().
For some configurations of GLIB, it is a bad idea to mix
calls to g_malloc() and free() or vice-versa.
I changed the ones that didn't need to be malloc() back to
g_new() and added #include <stdlib.h>, so I things should be OK now.
Thanks for the report.
But actually, I don't think those changes were in the 1.0.x
branch... So I should mention that what I was asking people to test
above was GTK+ checked out as:
cvs get -r gtk-1-0 -d gtk-1-0 gtk+
which is the maintenance version of GTK+, instead of the unstable
GTK+ 1.1 version.
Regards,
Owen
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]