| Summary: | undefined ATTRIBUTE_NULL_PRINTF_3 | ||
|---|---|---|---|
| Product: | gcc | Reporter: | benoit.sibaud |
| Component: | bootstrap | Assignee: | Not yet assigned to anyone <unassigned> |
| Status: | RESOLVED INVALID | ||
| Severity: | normal | CC: | bryceman, dale, dmeranda, eet, gcc-bugs |
| Priority: | P3 | Keywords: | build, documentation |
| Version: | 3.3 | ||
| Target Milestone: | --- | ||
| Host: | sparc-sun-solaris2.6 | Target: | sparc-sun-solaris2.6 |
| Build: | sparc-sun-solaris2.6 | Known to work: | |
| Known to fail: | Last reconfirmed: | 2003-05-26 14:30:33 | |
| Attachments: | dwarf2asm.out.bz2 | ||
|
Description
benoit.sibaud
2003-05-07 16:16:00 UTC
From: =?ISO-8859-1?Q?Beno=EEt_SIBAUD?= <benoit.sibaud@rd.francetelecom.com> To: gcc-prs@gcc.gnu.org, gcc-bugs@gcc.gnu.org, gcc-gnats@gcc.gnu.org Cc: Subject: Re: bootstrap/10666: undefined ATTRIBUTE_NULL_PRINTF_3 Date: Mon, 12 May 2003 11:07:46 +0200 Hi, http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=10666 just to let you know the problem is still present with the new 3.3 snapshot gcc-3.3-20030508 (tested on i686-pc-linux-gnu, sun-solaris2.6 and sun-solaris2.8). -- Benoît Sibaud From: =?ISO-8859-1?Q?Beno=EEt_SIBAUD?= <benoit.sibaud@rd.francetelecom.com> To: =?ISO-8859-1?Q?Beno=EEt_SIBAUD?= <benoit.sibaud@rd.francetelecom.com> Cc: gcc-prs@gcc.gnu.org, gcc-bugs@gcc.gnu.org, gcc-gnats@gcc.gnu.org Subject: Re: bootstrap/10666: undefined ATTRIBUTE_NULL_PRINTF_3 Date: Fri, 16 May 2003 13:37:26 +0200 Hi, http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=10666 > just to let you know the problem is still present with the new 3.3 > snapshot gcc-3.3-20030508 (tested on i686-pc-linux-gnu, sun-solaris2.6 > and sun-solaris2.8). Same problem with gcc-3.3. binutils-2.13.1/include/ansidecl.h was included, not gcc-3.3/include/ansidecl.h. Successful bootstrap on i686-pc-linux-gnu, in progress for sparc-sun-solaris2.6|7|8. -- Benoît Sibaud From: Deron Meranda <dmeranda@iac.net> To: gcc-prs@gcc.gnu.org, benoit.sibaud@rd.francetelecom.com, gcc-bugs@gcc.gnu.org, gcc-gnats@gcc.gnu.org, nobody@gcc.gnu.org Cc: Subject: Re: bootstrap/10666: undefined ATTRIBUTE_NULL_PRINTF_3 Date: Sat, 17 May 2003 04:49:13 -0400 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=10666 Also seen under hppa2.0-hp-hpux11.00 (32-bit) with gcc 3.3 final release. Using binutils 2.13.2.1. As mentioned, it is picking up the ansidecl.h under binutils rather than gcc/include. Since the only diffs seem to be these ATTRIBUTE_ macros, you can temporarily work around the bootstrap problem by replacing the ansidecl.h header in binutils with the one from gcc 3.3. This is obviously not a correct fix, but will allow the stage1 compiler to be built. For the record, here's the compile line in my setup which ends up including the wrong ansidecl.h... /usr/bin/cc -Ae -z +ESlit +DA2.0 \ -I/opt/gnu/binutils/2.13.2.1-32/include \ +Z -c -g -DIN_GCC -DHAVE_CONFIG_H \ -I. -I. -I/opt2/gnu/src/gcc/gcc-3.3/gcc \ -I/opt2/gnu/src/gcc/gcc-3.3/gcc/. \ -I/opt2/gnu/src/gcc/gcc-3.3/gcc/config \ -I/opt2/gnu/src/gcc/gcc-3.3/gcc/../include \ /opt2/gnu/src/gcc/gcc-3.3/gcc/dwarf2asm.c \ -o dwarf2asm.o The first -I directory was a direct result of me specifying it in $CC/$CFLAGS during the configure phase (needed because I have many different compilers and several different binutils versions on the system). So this may be "user error" on my part. But it seems like an easy error to make, so it may be worth a mention in the install notes until binutils catches up with gcc. A possiable fix is to add ATTRIBUTE_NULL_PRINTF_3 to binutils's ansidecl.h. *** Bug 11010 has been marked as a duplicate of this bug. *** Andrew, do you know the binutils and gcc ansidecl.h are supposed to be the same. If so, maybe binutils (or gcc) just needs to do a merge? Actually in the cvs version of binutils they are the same so the problem is that when you use a released version of binutils that was released before gcc, you might get a different version than what is in gcc. *** Bug 11255 has been marked as a duplicate of this bug. *** The only thing I can say is that is really a documentation problem with gcc that should say which version of binutils is supported in the combined build but I do not even though if the combined build is even documented. Since the only place where combined builds are documented is the cross docs, it usually reasonable that you should be using a new binutils with a new GCC so closing this as invalid. |