Re: preprocessor/7558: preprocessor option -MM has change semantic

Re: preprocessor/7558: preprocessor option -MM has change semantic

[Martin v. Loewis]

The following reply was made to PR preprocessor/7558; it has been noted by GNATS.

From: martin@v.loewis.de (Martin v. Loewis)
To: Neil Booth <neil@daikokuya.co.uk>
Cc: boehme@informatik.hu-berlin.de, gcc-gnats@gcc.gnu.org,
   debian-gcc@lists.debian.org
Subject: Re: preprocessor/7558: preprocessor option -MM has change semantic
Date: 11 Aug 2002 11:02:38 +0200

 Neil Booth <neil@daikokuya.co.uk> writes:
 
 > If it's a system header, why are you lying to the compiler?
 
 I'm not lying, I use <a.h>
 
 > Maybe a real-life example and not "a.h" would help.
 
 Ok, here is the real-life example. Consider
 
 #include <jni.h>
 
 int main(){}
 
 which is compiled with
 
 g++-3.1 -MM -I/opt/JBuilder6/jdk1.3.1/include -I /opt/JBuilder6/jdk1.3.1/include/linux -c a.cc
 
 This gives as the output
 
 a.o: a.cc /opt/JBuilder6/jdk1.3.1/include/jni.h \
   /opt/JBuilder6/jdk1.3.1/include/linux/jni_md.h
 
 I don't want the dependencies on jni.h, as those are system headers.
 
 Regards,
 Martin

Re: preprocessor/7558: preprocessor option -MM has change semantic

[neil]

Synopsis: preprocessor option -MM has change semantic

State-Changed-From-To: open->closed
State-Changed-By: neil
State-Changed-When: Tue Aug 20 12:29:06 2002
State-Changed-Why:
    The new behaviour is as intended.

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7558

Re: preprocessor/7558: preprocessor option -MM has change semantic

[Neil Booth]

The following reply was made to PR preprocessor/7558; it has been noted by GNATS.

From: Neil Booth <neil@daikokuya.co.uk>
To: boehme@informatik.hu-berlin.de
Cc: gcc-gnats@gcc.gnu.org, debian-gcc@lists.debian.org,
	"Martin v. Loewis" <martin@v.loewis.de>
Subject: Re: preprocessor/7558: preprocessor option -MM has change semantic
Date: Sun, 11 Aug 2002 08:19:09 +0100

 Martin,
 
 If it's a system header, why are you lying to the compiler?
 
 Maybe a real-life example and not "a.h" would help.
 
 Neil.

Re: preprocessor/7558: preprocessor option -MM has change semantic

[Neil Booth]

The following reply was made to PR preprocessor/7558; it has been noted by GNATS.

From: Neil Booth <neil@daikokuya.co.uk>
To: "Martin v. Loewis" <martin@v.loewis.de>
Cc: boehme@informatik.hu-berlin.de, gcc-gnats@gcc.gnu.org,
	debian-gcc@lists.debian.org
Subject: Re: preprocessor/7558: preprocessor option -MM has change semantic
Date: Sun, 11 Aug 2002 10:02:20 +0100

 Martin v. Loewis wrote:-
 
 > Neil Booth <neil@daikokuya.co.uk> writes:
 > 
 > > If it's a system header, why are you lying to the compiler?
 > 
 > I'm not lying, I use <a.h>
 
 You've not told the compiler it's a system header, so it doesn't think
 it is.  Let me see what Zack thinks.
 
 Neil.

Re: preprocessor/7558: preprocessor option -MM has change semantic

[Martin v. Loewis]

The following reply was made to PR preprocessor/7558; it has been noted by GNATS.

From: martin@v.loewis.de (Martin v. Loewis)
To: Neil Booth <neil@daikokuya.co.uk>
Cc: boehme@informatik.hu-berlin.de, gcc-gnats@gcc.gnu.org,
   debian-gcc@lists.debian.org
Subject: Re: preprocessor/7558: preprocessor option -MM has change semantic
Date: 10 Aug 2002 15:45:33 +0200

 Neil Booth <neil@daikokuya.co.uk> writes:
 
 > > >Description:
 > > 	In gcc 3.1, -MM prints dependencies even to files included with
 > > 	angle brackets (<foo.h>), if those are found through -I options.
 > > 	This behaviour is unintuitive and a change from earlier versions.
 > 
 > Why are you using <> brackets?  Why is #include "foo.h" not good enough?
 
 Because it is a system header which should not be listed in the
 dependencies.
 
 Regards,
 Martin

Re: preprocessor/7558: preprocessor option -MM has change semantic

[Neil Booth]

The following reply was made to PR preprocessor/7558; it has been noted by GNATS.

From: Neil Booth <neil@daikokuya.co.uk>
To: boehme@informatik.hu-berlin.de
Cc: gcc-gnats@gcc.gnu.org, debian-gcc@lists.debian.org
Subject: Re: preprocessor/7558: preprocessor option -MM has change semantic
Date: Sat, 10 Aug 2002 10:43:41 +0100

 boehme@informatik.hu-berlin.de wrote:-
 
 > >Description:
 > 	In gcc 3.1, -MM prints dependencies even to files included with
 > 	angle brackets (<foo.h>), if those are found through -I options.
 > 	This behaviour is unintuitive and a change from earlier versions.
 
 Why are you using <> brackets?  Why is #include "foo.h" not good enough?
 
 Neil.

preprocessor/7558: preprocessor option -MM has change semantic

[boehme]

>Number:         7558
>Category:       preprocessor
>Synopsis:       preprocessor option -MM has change semantic
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          change-request
>Submitter-Id:   net
>Arrival-Date:   Fri Aug 09 09:26:01 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     
>Release:        3.1.1 (Debian testing/unstable)
>Organization:
>Environment:
System: Linux kosh 2.4.18-k7 #1 Sun Apr 14 13:19:11 EST 2002 i686 unknown unknown GNU/Linux
Architecture: i686

	
host: i386-pc-linux-gnu
build: i386-pc-linux-gnu
target: i386-pc-linux-gnu
configured with: /mnt/data/gcc-3.1/gcc-3.1-3.1.1ds3/src/configure -v --enable-languages=c,c++,java,f77,proto,objc,ada --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --with-gxx-include-dir=/usr/include/c++/-3.1 --enable-shared --with-system-zlib --enable-long-long --enable-nls --without-included-gettext --enable-clocale=gnu --enable-__cxa_atexit --enable-threads=posix --enable-java-gc=boehm --enable-objc-gc i386-linux
>Description:
	In gcc 3.1, -MM prints dependencies even to files included with
	angle brackets (<foo.h>), if those are found through -I options.
	This behaviour is unintuitive and a change from earlier versions.

	It appears that the only way to suppress these files is to use the
	-isystem directive, which is quite gcc specific and hard to use
	in a makefile that needs to work across different compilers, or
	that is generated by autoconf.
>How-To-Repeat:
	Compile the file

	#include <a.h>

	int
	main()
	{
	}

	with gcc -I. -MM; this gives

	a.o: a.c a.h

	even though a.h should not have been mentioned.

>Fix:
	No real work-around is known.
>Release-Note:
>Audit-Trail:
>Unformatted: