Re: [Sofia-sip-devel] [PATCH] Fix configuration under MinGW

2010-07-08 Thread Stefano Sabatini 
On date Tuesday 2010-07-06 19:35:38 +0300, Pekka Pessi phoned this:
 2010/7/6 Stefano Sabatini ssabat...@reilabs.com:
  this patch fixes the configuration problem as reported here:
  http://thread.gmane.org/gmane.comp.telephony.sofia-sip.devel/3853
 
 Thanks for the patch.
 
 -CFLAGS=$CFLAGS -I\$(top_srcdir)/win32/pthread -DWINVER=0x0501 \
 +CFLAGS=$CFLAGS -I${srcdir}/win32/pthread -DWINVER=0x0501 \
 
 Can you actually compile sofia-sip with this? It seems to me that the
 relative srcdir does dot work when make descends deeper into source
 tree.

Mmh no, indeed compilation was failing with a libtool error (I had to
hack the generated libtool script to fix it, as reported in another
thread).

So it seems that the patch defined in sac-general.m4 needs to be
*absolute*. I tried to use $abs_srcdir but that is expanded to the
empty string, then I hacked this:

abs_srcdir=`(cd $srcdir  pwd)`
CFLAGS=$CFLAGS -I${abs_srcdir}/win32/pthread -DWINVER=0x0501 \ ...

and now it is working (configure and libtool and all).

Now I need to understand why it isn't expanding $abs_srcdir...

Regards.

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel

[Sofia-sip-devel] [PATCH] Fix configuration under MinGW

2010-07-06 Thread Stefano Sabatini 
Hi all,

this patch fixes the configuration problem as reported here:
http://thread.gmane.org/gmane.comp.telephony.sofia-sip.devel/3853

$LIBS and $CFLAGS need to be expanded at the configuration level,
otherwise when using them during configuration they will be wrong,
resulting in many failing tests and in a fatal error which happens
in the code from configure.ac:

if test $ac_cv_c_ll_format = yes; then
  AC_DEFINE([LLU], [%llu], [Format (%llu) for unsigned long long])dnl
  AC_DEFINE([LLI], [%lli], [Format (%lli) for long long])dnl
  AC_DEFINE([LLX], [%llx], [Format (%llx) for long long hex])dnl
else
  AC_MSG_ERROR(printf cannot handle 64-bit integers)
fi

This is what you can find in config.log:

configure:24311: gcc -o conftest.exe -g -O2 -I./win32/pthread -DWINVER=0x0501
-D_WIN32_WINNT=0x0501 -DIN_LIBSOFIA_SIP_UA -DIN_LIBSOFIA_SRES   -mms-bitfields
-pipe -mno-cygwin -mwindows -mconsole -Wall -g -O0   -Wl,--enable-auto-image-bas
e conftest.c -L$(top_srcdir)/win32/pthread -lpthreadVC2 -lws2_32-lwsock3
2 5
c:/mingw/bin/../lib/gcc/mingw32/4.5.0/../../../../mingw32/bin/ld.exe: cannot fin
d -lpthreadVC2
collect2: ld returned 1 exit status
configure:24311: $? = 1
configure: program exited with status 1
configure: failed program was:

Patch attached, regards.
From 7d74c8b50b33469667006c50e902e690620ce5dd Mon Sep 17 00:00:00 2001
Message-Id: 7d74c8b50b33469667006c50e902e690620ce5dd.1278413394.git.stefano.sabatini-l...@poste.it
From: Stefano Sabatini stefano.sabatini-l...@poste.it
Date: Tue, 6 Jul 2010 12:28:11 +0200
Subject: [fix-mingw-compilation PATCH] Fix configuration in MinGW.

CFLAGS and LIBS need to be expanded at configuration time, otherwise
many configure test will fail and the configure will abort with the
message:
configure: error: printf cannot handle 64-bit integers
---
 m4/sac-general.m4 |4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/m4/sac-general.m4 b/m4/sac-general.m4
index b494de1..de0ba9a 100644
--- a/m4/sac-general.m4
+++ b/m4/sac-general.m4
@@ -359,12 +359,12 @@ fi
 ])
 
 if test $ac_cc_environment = mingw32 ; then
-CFLAGS=$CFLAGS -I\$(top_srcdir)/win32/pthread -DWINVER=0x0501 \
+CFLAGS=$CFLAGS -I${srcdir}/win32/pthread -DWINVER=0x0501 \
 	-D_WIN32_WINNT=0x0501 -DIN_LIBSOFIA_SIP_UA -DIN_LIBSOFIA_SRES \
 	-mms-bitfields \
 	-pipe -mno-cygwin -mwindows -mconsole -Wall -g -O0
 LDFLAGS=$LDFLAGS -Wl,--enable-auto-image-base
-LIBS=-L\$(top_srcdir)/win32/pthread -lpthreadVC2 -lws2_32 \
+LIBS=-L${srcdir}/win32/pthread -lpthreadVC2 -lws2_32 \
 	-lwsock32
 MINGW_ENVIRONMENT=1
 AC_SUBST(MINGW_ENVIRONMENT)
-- 
1.7.1

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first___
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel

Re: [Sofia-sip-devel] [PATCH] Fix configuration under MinGW

2010-07-06 Thread Michael Jerris 
Do these CFLAGS get into the build as well?  a bit worried that we need to do 
both this for the tests, and using top_ for the build.

Mike

On Jul 6, 2010, at 6:53 AM, Stefano Sabatini wrote:

 Hi all,
 
 this patch fixes the configuration problem as reported here:
 http://thread.gmane.org/gmane.comp.telephony.sofia-sip.devel/3853
 
 $LIBS and $CFLAGS need to be expanded at the configuration level,
 otherwise when using them during configuration they will be wrong,
 resulting in many failing tests and in a fatal error which happens
 in the code from configure.ac:
 
 if test $ac_cv_c_ll_format = yes; then
  AC_DEFINE([LLU], [%llu], [Format (%llu) for unsigned long long])dnl
  AC_DEFINE([LLI], [%lli], [Format (%lli) for long long])dnl
  AC_DEFINE([LLX], [%llx], [Format (%llx) for long long hex])dnl
 else
  AC_MSG_ERROR(printf cannot handle 64-bit integers)
 fi
 
 This is what you can find in config.log:
 
 configure:24311: gcc -o conftest.exe -g -O2 -I./win32/pthread -DWINVER=0x0501
 -D_WIN32_WINNT=0x0501 -DIN_LIBSOFIA_SIP_UA -DIN_LIBSOFIA_SRES   -mms-bitfields
 -pipe -mno-cygwin -mwindows -mconsole -Wall -g -O0   
 -Wl,--enable-auto-image-bas
 e conftest.c -L$(top_srcdir)/win32/pthread -lpthreadVC2 -lws2_32
 -lwsock3
 2 5
 c:/mingw/bin/../lib/gcc/mingw32/4.5.0/../../../../mingw32/bin/ld.exe: cannot 
 fin
 d -lpthreadVC2
 collect2: ld returned 1 exit status
 configure:24311: $? = 1
 configure: program exited with status 1
 configure: failed program was:
 
 Patch attached, regards.
 0001-Fix-configuration-in-MinGW.patch--
 This SF.net email is sponsored by Sprint
 What will you do first with EVO, the first 4G phone?
 Visit sprint.com/first -- 
 http://p.sf.net/sfu/sprint-com-first___
 Sofia-sip-devel mailing list
 Sofia-sip-devel@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel


--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel

Re: [Sofia-sip-devel] [PATCH] Fix configuration under MinGW

2010-07-06 Thread Michael Jerris 
I'll take a look at this issue.

Mike

On Jul 6, 2010, at 12:20 PM, Stefano Sabatini wrote:

 On date Tuesday 2010-07-06 11:55:57 -0400, Michael Jerris phoned:
 Do these CFLAGS get into the build as well?  a bit worried that we
 need to do both this for the tests, and using top_ for the build.
 
 Indeed I'm not sure this fix is right at all, at least it is working
 for me but it may badly fails in other scenarios. I have no special
 expertise with auto* tools, and I still can't grasp how it solves the
 problem of defining the same flags for both the configure file and the
 Makefile.
 
 Hope someone with expertise in that area (Pekka?) can shed some light.
 
 [...]
 
 Regards.


--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel

Re: [Sofia-sip-devel] [PATCH] Fix configuration under MinGW

2010-07-06 Thread Pekka Pessi 
2010/7/6 Stefano Sabatini ssabat...@reilabs.com:
 this patch fixes the configuration problem as reported here:
 http://thread.gmane.org/gmane.comp.telephony.sofia-sip.devel/3853

Thanks for the patch.

-CFLAGS=$CFLAGS -I\$(top_srcdir)/win32/pthread -DWINVER=0x0501 \
+CFLAGS=$CFLAGS -I${srcdir}/win32/pthread -DWINVER=0x0501 \

Can you actually compile sofia-sip with this? It seems to me that the
relative srcdir does dot work when make descends deeper into source
tree.

--Pekka

-- 
Pekka.Pessi mail at nokia.com

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Sofia-sip-devel mailing list
Sofia-sip-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sofia-sip-devel