* Lossage do to #ifndef __ASSEMBLY__ in mipsregs.h
@ 2004-01-22 23:19 David Daney
2004-01-22 23:56 ` Thiemo Seufer
0 siblings, 1 reply; 3+ messages in thread
From: David Daney @ 2004-01-22 23:19 UTC (permalink / raw)
To: linux-mips
I am using gcc 3.3.1 to compile the linux_2_4 kernel obtained from cvs a
couple of days ago.
My gcc defines these assemble related macros:
[daney@dl xilleon]$ mipsel-linux-gcc -D__KERNEL__
-I/newdisk/programs/linux-mips.org/linux/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-fomit-frame-pointer -I
/newdisk/programs/linux-mips.org/linux/include/asm/gcc -G 0
-mno-abicalls -fno-pic -finline-limit=100000 -mabi=32 -march=mips32
-mips32 -Wa,-32 -Wa,-march=mips32 -Wa,-mips32 -Wa,--trap
-DLINUX_RAM_START=0x80100000 -dM -E -c xilleonIRQ.S | grep ASS
#define __ASSEMBLER__ 1
#define LANGUAGE_ASSEMBLY 1
#define __LANGUAGE_ASSEMBLY 1
#define __LANGUAGE_ASSEMBLY__ 1
#define _LANGUAGE_ASSEMBLY 1
Now in mipsregs.h we have:
.
.
.
#ifndef __ASSEMBLY__
.
.
.
So I get a ton of errors trying to assemble the file.
Now back on 2002/06/27 we have:
===================================================================
RCS file: /home/cvs/linux/include/asm-mips/mipsregs.h,v
retrieving revision 1.30.2.8
retrieving revision 1.30.2.9
diff -u -p -r1.30.2.8 -r1.30.2.9
--- linux/include/asm-mips/mipsregs.h 2002/06/27 09:53:37 1.30.2.8
+++ linux/include/asm-mips/mipsregs.h 2002/06/27 14:21:23 1.30.2.9
@@ -445,7 +445,7 @@
#define CEB_KERNEL 2 /* Count events in kernel mode EXL = ERL = 0 */
#define CEB_EXL 1 /* Count events with EXL = 1, ERL = 0 */
-#ifndef _LANGUAGE_ASSEMBLY
+#ifndef __ASSEMBLY__
/*
* Functions to access the r10k performance counter and control registers
@@ -1023,6 +1023,6 @@ do { \
__disable_fpu(); \
} while (0)
-#endif /* !defined (_LANGUAGE_ASSEMBLY) */
+#endif /* !__ASSEMBLY__ */
#endif /* _ASM_MIPSREGS_H */
Why the change?
I will fix my local mipsregs.h so that I can continue, but it seems like
the sources in CVS should be changed to allow gcc 3.3.1 to work.
David Daney.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Lossage do to #ifndef __ASSEMBLY__ in mipsregs.h
2004-01-22 23:19 Lossage do to #ifndef __ASSEMBLY__ in mipsregs.h David Daney
@ 2004-01-22 23:56 ` Thiemo Seufer
2004-01-23 0:02 ` David Daney
0 siblings, 1 reply; 3+ messages in thread
From: Thiemo Seufer @ 2004-01-22 23:56 UTC (permalink / raw)
To: David Daney; +Cc: linux-mips
David Daney wrote:
> I am using gcc 3.3.1 to compile the linux_2_4 kernel obtained from cvs a
> couple of days ago.
[snip]
> -#endif /* !defined (_LANGUAGE_ASSEMBLY) */
> +#endif /* !__ASSEMBLY__ */
>
> #endif /* _ASM_MIPSREGS_H */
>
> Why the change?
__ASSEMBLY__ is defined by the Linux build system in order to have a
compiler independent test.
> I will fix my local mipsregs.h so that I can continue, but it seems like
> the sources in CVS should be changed to allow gcc 3.3.1 to work.
It works here (for my configs). Your linux tree seems to be broken.
Thiemo
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Lossage do to #ifndef __ASSEMBLY__ in mipsregs.h
2004-01-22 23:56 ` Thiemo Seufer
@ 2004-01-23 0:02 ` David Daney
0 siblings, 0 replies; 3+ messages in thread
From: David Daney @ 2004-01-23 0:02 UTC (permalink / raw)
To: Thiemo Seufer; +Cc: linux-mips
Thiemo Seufer wrote:
>David Daney wrote:
>
>
>>I am using gcc 3.3.1 to compile the linux_2_4 kernel obtained from cvs a
>>couple of days ago.
>>
>>
>[snip]
>
>
>>-#endif /* !defined (_LANGUAGE_ASSEMBLY) */
>>+#endif /* !__ASSEMBLY__ */
>>
>>#endif /* _ASM_MIPSREGS_H */
>>
>>Why the change?
>>
>>
>
>__ASSEMBLY__ is defined by the Linux build system in order to have a
>compiler independent test.
>
I am starting to understand this.
>
>
>
>>I will fix my local mipsregs.h so that I can continue, but it seems like
>>the sources in CVS should be changed to allow gcc 3.3.1 to work.
>>
>>
>
>It works here (for my configs). Your linux tree seems to be broken.
>
>
>Thiemo
>
>
I am porting my BSP from 2.4.18 to the latest, and it seems I may have
some Makefile incompatibility. So I withdraw my criticism of
defined(__ASSEMBLY__).
David Daney
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2004-01-23 0:03 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-01-22 23:19 Lossage do to #ifndef __ASSEMBLY__ in mipsregs.h David Daney
2004-01-22 23:56 ` Thiemo Seufer
2004-01-23 0:02 ` David Daney
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox