Summary:ASTERISK-01886: Create Makefile.inc, enable more warnings, resolve warnings.
Reporter:rich (rich)Labels:
Date Opened:2004-06-24 22:48:49Date Closed:2008-01-15 15:00:42.000-0600
Versions:Frequency of
Environment:Attachments:( 0) 040624Makefiles.txt
( 1) 040624src.txt
( 2) notes.txt

Create Makefile.inc that holds:
- compiler options
- flags for features referred to in multiple Makefiles,
- location of 3rd party packages and
- installation directory.

'include' Makefile.inc in all other Makefiles
so that consistent flags are used to compile
various kinds of objects.

Enable more warnings, and more strict
compilation in general.


use g++ to generate chan_h323.so in order to
easily link in the proper version of


ast_expr.y: Use the system's definition of
QUAD_MIN on Fedora 2 and others that define
LONG_LONG_MIN in limits.h.


#ifdef 0 out unused variable.


resolve duplicate definitions of *_cb.
resolve namespace clash among include files.
resolve pointer conversion warnings from g++.


make find_tpeer() staic since almost
everything else in the files is, and it isn't
used or defined as exported anywhere else.
Otherwise the compiler complains about the
lack of a prototype.


resolve clash among include files.
wrap C language headers in extern "C" {}.


resolve clash amoung headers.


change *_cb definitions into declarations, so
that they are not multiply defined.


Use '#if 0' to comment out unused symbols
rather than '/*' to avoid compiler warnings
about nested comments.


Enable -DP_R_O_T_O_T_Y_P_E_S, which
identifies mismatch of function declarations.

Resolve mismatch of a few functions that were
declared as returning void but defined to
return int.

put declarations of f2c library functions
into f2c.h.

Move '#include "f2c.h" prior to subsequent
usage of various types defined in f2c.h in
the .c files.


tested on Fedora Core 2, Redhat 9 and FreeBSD 5.2.1.
Comments:By: Mark Spencer (markster) 2004-06-26 00:04:49

I merged the source patch, the Makefile one did not apply cleanly.  Can you please add MacOS X to your test environment and confirm it builds cleanly there, too?  I'm not familiar with these gcc macros.

By: rich (rich) 2004-06-26 21:27:20

I'm looking for Mac OS X 10.3 compatible hardware (G3 and above I believe), but it will probably take at least a month to acquire.  Meanwhile I'll find someone willing to let me test on theirs and fix OS X as soon as possible.

I'll send a revised patch as soon as I can!

By: jdturner (jdturner) 2004-06-28 20:18:03

I can test the builds on OS X now. With the latest changes the compile works:

*CLI> show version
Asterisk CVS-HEAD-06/26/04-16:54:32 built by jturner@Harp.local on a Power Macintosh running Darwin

One note about the patch I submitted. I am not sure that the fix to db1-ast/btree/bt_debug.c is the best way to get around the error from ranlib. But the make halts because ranlib on OS X doesn't like a .o file with no symbols.

By: Mark Spencer (markster) 2004-06-28 23:17:44

Does it still build for you with the attached patch?

By: Olle Johansson (oej) 2004-07-16 14:19:03

Is this patch still valid for current CVS?

By: rich (rich) 2004-07-16 16:18:10

I need to add the changes for cdr_sqlite and iax2-provision.  I'll do that tonight an d resubmit.

I hope to have access to an OS X system this weekend as well so I can test the build there.


By: Olle Johansson (oej) 2004-07-23 17:18:41

Gentle reminder from housekeeping...


By: rich (rich) 2004-07-23 17:36:37

Will do! I've got OS X access now and will get this sorted out in the next day or so.

By: rich (rich) 2004-07-25 18:34:42

I'm submitting the OSX patches, but since I'll be out of town for a week, I don't want to introduce these changes while I can't respond to any issues they raise.  I'll finish this 8/10 when I return, if that's OK.  Rich

By: Brian West (bkw918) 2004-08-17 09:50:08

Any update?

By: Brian West (bkw918) 2004-08-22 23:03:36

Please get on irc and hound kram to get this patch in or get feedback.. I can't have these patches sitting around collecting dust.. we need to MOVE MOVE MOVE on these guys.... over and out! :P

PS: I mean that with LOVE!!! :)

By: Mark Spencer (markster) 2004-09-04 11:12:14

What *is* the status of this patch anyway?

By: Mark Spencer (markster) 2004-09-07 12:56:45

I'm merging the BSD compatibility into a single bug.  Please place any comments remaining regarding BSD compatibility there.  Thanks.

By: Digium Subversion (svnbot) 2008-01-15 15:00:42.000-0600

Repository: asterisk
Revision: 3311

U   trunk/ast_expr.y
U   trunk/channels/chan_alsa.c
U   trunk/channels/chan_h323.c
U   trunk/channels/chan_iax2.c
U   trunk/channels/h323/ast_h323.cpp
U   trunk/channels/h323/ast_h323.h
U   trunk/channels/h323/chan_h323.h
U   trunk/codecs/gsm/inc/config.h
U   trunk/codecs/lpc10/analys.c
U   trunk/codecs/lpc10/bsynz.c
U   trunk/codecs/lpc10/chanwr.c
U   trunk/codecs/lpc10/dcbias.c
U   trunk/codecs/lpc10/decode.c
U   trunk/codecs/lpc10/deemp.c
U   trunk/codecs/lpc10/difmag.c
U   trunk/codecs/lpc10/dyptrk.c
U   trunk/codecs/lpc10/encode.c
U   trunk/codecs/lpc10/energy.c
U   trunk/codecs/lpc10/f2c.h
U   trunk/codecs/lpc10/ham84.c
U   trunk/codecs/lpc10/hp100.c
U   trunk/codecs/lpc10/invert.c
U   trunk/codecs/lpc10/irc2pc.c
U   trunk/codecs/lpc10/ivfilt.c
U   trunk/codecs/lpc10/lpcdec.c
U   trunk/codecs/lpc10/lpcenc.c
U   trunk/codecs/lpc10/lpfilt.c
U   trunk/codecs/lpc10/median.c
U   trunk/codecs/lpc10/mload.c
U   trunk/codecs/lpc10/onset.c
U   trunk/codecs/lpc10/pitsyn.c
U   trunk/codecs/lpc10/placea.c
U   trunk/codecs/lpc10/placev.c
U   trunk/codecs/lpc10/preemp.c
U   trunk/codecs/lpc10/prepro.c
U   trunk/codecs/lpc10/random.c
U   trunk/codecs/lpc10/rcchk.c
U   trunk/codecs/lpc10/synths.c
U   trunk/codecs/lpc10/tbdm.c
U   trunk/codecs/lpc10/voicin.c
U   trunk/codecs/lpc10/vparms.c

r3311 | markster | 2008-01-15 15:00:42 -0600 (Tue, 15 Jan 2008) | 2 lines

Merge source cleanups (bug ASTERISK-1886)