Re: Unreviewed^3 Patches

[Rainer Orth <ro at TechFak dot Uni-Bielefeld dot DE>]

David Daney writes:

> I think as part of this you did:
> 
>     * config/mips/linux.h (FUNCTION_NAME_ALREADY_DECLARED): Define as 1.
> 
> Which causes me these warnings when configured as target=mipsisa32-linux:
> 
> <...>
> gcc -c   -g -O2 -DIN_GCC -DCROSS_COMPILE  -W -Wall -Wwrite-strings 
> -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long  
> -fno-common   -DHAVE_CONFIG_H    -I. -Icp -I../../gcccvsw/gcc/gcc 
> -I../../gcccvsw/gcc/gcc/cp -I../../gcccvsw/gcc/gcc/../include  
> ../../gcccvsw/gcc/gcc/cp/except.c -o cp/except.o
> In file included from tm.h:16,
>                  from ../../gcccvsw/gcc/gcc/cp/except.c:29:
> ../../gcccvsw/gcc/gcc/config/mips/linux.h:174:1: warning: 
> "FUNCTION_NAME_ALREADY_DECLARED" redefined
> In file included from tm.h:15,
>                  from ../../gcccvsw/gcc/gcc/cp/except.c:29:
> ../../gcccvsw/gcc/gcc/config/mips/mips.h:3294:1: warning: this is the 
> location of the previous definition
> <...>

Sorry, I hadn't seen that mips/linux.h is included after mips/mips.h.  If
the following patch fixes this for you, I'll commit as obvious since it
breaks bootstrap in a native configuration:

Fri Oct 17 18:26:57 2003  Rainer Orth  <ro@TechFak.Uni-Bielefeld.DE>

	* config/mips/linux.h (FUNCTION_NAME_ALREADY_DECLARED): Undef
	before redefinition.

Index: gcc/config/mips/linux.h
===================================================================
RCS file: /cvs/gcc/gcc/gcc/config/mips/linux.h,v
retrieving revision 1.69
diff -u -p -r1.69 linux.h
--- gcc/config/mips/linux.h	17 Oct 2003 11:44:32 -0000	1.69
+++ gcc/config/mips/linux.h	17 Oct 2003 16:28:49 -0000
@@ -171,6 +172,7 @@ Boston, MA 02111-1307, USA.  */
 
 /* Tell function_prologue in mips.c that we have already output the .ent/.end
    pseudo-ops.  */
+#undef FUNCTION_NAME_ALREADY_DECLARED
 #define FUNCTION_NAME_ALREADY_DECLARED 1
 
 #define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL)       		\


> I am also getting this again:
> <...>
> /home/daney/mipsel-gcccvs/gcc/xgcc -B/home/daney/mipsel-gcccvs/gcc/ 
> -B/home/testcvs//mipsisa32el-linux/bin/ 
> -B/home/testcvs//mipsisa32el-linux/lib/ -isystem 
> /home/testcvs//mipsisa32el-linux/include -isystem 
> /home/testcvs//mipsisa32el-linux/sys-include -DHAVE_CONFIG_H -I. 
> -I../../../gcccvsw/gcc/libffi -I. -I../../../gcccvsw/gcc/libffi/include 
> -Iinclude -I../../../gcccvsw/gcc/libffi/src -O2 -g -O2 -Wall -g 
> -fexceptions -O2 -g -O2 -c ../../../gcccvsw/gcc/libffi/src/debug.c  
> -fPIC -DPIC -o src/.libs/debug.o
> In file included from include/ffi.h:156,
>                  from ../../../gcccvsw/gcc/libffi/src/debug.c:24:
> ../../../gcccvsw/gcc/libffi/include/ffi_mips.h:39: error: parse error 
> before '--' token
> In file included from ../../../gcccvsw/gcc/libffi/src/debug.c:24:
> <...>
> 
> Which was broken and fixed.  But at some point re-broke.

No, I just had proposed a patch

	http://gcc.gnu.org/ml/gcc-patches/2003-10/msg00607.html

but there was no consensus yet.  I'll try to follow up later today on the
issues raised; time is currently limited due to start of term.

	Rainer