Matthias,
Okay, I just put a bunch of stuff in the CVS:
(1) your earlier patch (pre-variable argument macro fix)
(2) some warning removals
(3) my own vararg stuff (we'll have to reconcile our versions)
(4) removed generated files from CVS
(5) fixed --disable-crypto
(6) a better undefined TUNNEWPP
against current CVS:
Index: thread.h
===
RCS file: /cvsroot/openvpn/openvpn/thread.h,v
retrieving revision 1.3
diff -u -r1.3 thread.h
--- thread.h25 Jun 2002 06:56:16 - 1.3
+++ thread.h19 Feb 2003 21:21:03 -
@@ -1
Dear Jim,
this second patch completes the varargs stuff. Tested on all machines I
listed last time, and SUNpro 6 is very happy with the ISO C99 stuff and
uses the macro (rather than the function). I tried to force the function
underneath gcc 2.95 (edited config.cache, ran ./config.status --recheck
Jim,
this is the configure.ac part of the variadic macro test patch, it seems
to work OK and cache the results.
I have added also some unrelated AC_CACHE_SAVE (supported since autoconf
2.13) to checkpoint the cache should some later test fail, so the user
can more quickly pick up when she typed "
On Wed, 19 Feb 2003, James Yonan wrote:
> > Please also note that _* names are reserved name space. Don't use them
> > for your functions if you can avoid it, for portability -- they might
> > screw some linker.
>
> Are __* names reserved? What's the recommended form for naming "private"
> var
> There were more warnings that I haven't reported because I think they
> didn't warrant fixes at this time, some "unused parameter" warnings. Try
>
> make clean
> make CFLAGS="-W -Wall"
>
> to see more warnings :-)
Yes, also ./configure --enable-strict does something like this.
Most of the war
> Please also note that _* names are reserved name space. Don't use them
> for your functions if you can avoid it, for portability -- they might
> screw some linker.
Are __* names reserved? What's the recommended form for naming "private"
variables (I'm coming from C++/Python where _* names are
On Wed, 19 Feb 2003, James Yonan wrote:
> Hey, thanks for the patch and all the testing work on different platforms.
You're welcome. I thought if I give it a whirl, I'd spin it until it was
dizzy :-)
> You raise a number of good points which I will address below:
There were more warnings that
Matthias,
Hey, thanks for the patch and all the testing work on different platforms.
You raise a number of good points which I will address below:
(1) Signed/unsigned comparison warnings in packet_id.c and socket.c -- these
are obviously just cast issues and I will clean up for a mostly warning-
OK,
here is a full patch for your convenience without the "apply on top"
junk, with this patch, the current CVS compiles on:
SuSE Linux 7.3 x86 gcc 2.95.3
SuSE Linux 8.1 x86 gcc 3.2
FreeBSD 4-STABLE x86 gcc 2.95.3+FreeBSD
Solaris 8 Sparc gcc 2.95.3
Solaris 8 Sparc SunPro Workshop 6.
The unfixed
On Wed, 19 Feb 2003, Matthias Andree wrote:
> On Sun, 16 Feb 2003, James Yonan wrote:
>
> > Beta is available on CVS as well as here:
> >
> > http://openvpn.sourceforge.net/beta/openvpn-1.3.2.9.tar.gz
>
> I tried the current CVS as of some minutes ago on Linux, Solaris and
> FreeBSD. This firs
On Sun, 16 Feb 2003, James Yonan wrote:
> 2003.02.16 -- Version 1.3.2.9
>
> * Added --replay-persist feature to allow replay
> protection across sessions.
The persist stuff seems to break when --disable-crypto is in effect:
/usr/include/netinet/in.h:69: warning: empty declaration
/home/ma/tmp/
On Sun, 16 Feb 2003, James Yonan wrote:
> Beta is available on CVS as well as here:
>
> http://openvpn.sourceforge.net/beta/openvpn-1.3.2.9.tar.gz
I tried the current CVS as of some minutes ago on Linux, Solaris and
FreeBSD. This first mail is about Linux, Solaris status is in a
separate mail.
13 matches
Mail list logo
