RE: CVS Update@oss.sgi.com: linux

[Harald Koerfgen <Harald.Koerfgen@home.ivm.de>]

On 18-Nov-00 Ralf Baechle wrote:
> CVSROOT:      /home/pub/cvs
> Module name:  linux
> Changes by:   ralf@oss.sgi.com        00/11/17 17:19:36
> 
> Modified files:
>       arch/mips/kernel: sysmips.c 
> 
> Log message:
>       Hopefully R3k-proof version of sysmips(MIPS_ATOMIC_SET, ...).

Compiles but will never work on non-ll/sc CPUs, obviously. Wouldn't it be
better to let the compilation fail than to generate code which is known not to
work for R3000?

-- 
Regards,
Harald

--- snip here ---
diff -ruN /nfs/cvs/linux-2.3/linux/arch/mips/kernel/sysmips.c
linux/arch/mips/kernel/sysmips.c
--- /nfs/cvs/linux-2.3/linux/arch/mips/kernel/sysmips.c Sat Nov 18 14:31:06 2000
+++ linux/arch/mips/kernel/sysmips.c    Sat Nov 18 17:56:01 2000
@@ -83,6 +83,7 @@
                        return errno;
                errno = 0;
 
+#ifdef CONFIG_CPU_HAS_LLSC
                __asm__(".set\tpush\t\t\t# sysmips(MIPS_ATOMIC, ...)\n\t"
                        ".set\tmips2\n\t"
                        ".set\tnoat\n\t"
@@ -101,6 +102,9 @@
                        : "=&r" (tmp), "=o" (* (u32 *) p), "=r" (errno)
                        : "r" (arg2), "o" (* (u32 *) p), "2" (errno)
                        : "$1");
+#else
+#error "sys_sysmips(MIPS_ATOMIC_SET, ...): Fixme, I'm broken"
+#endif
 
                if (errno)
                        return -EFAULT;