cell compile fixes.

cell compile fixes.

[Dave Jones]

Missing include for __NR_syscalls, and missing sys_splice() that
causes build-time failure due to compile-time bounds check on spu_syscall_table.

Signed-off-by: Dave Jones <davej@redhat.com>

--- linux-2.6.16.noarch/arch/powerpc/platforms/cell/spufs/run.c~	2006-03-30 15:48:17.000000000 -0500
+++ linux-2.6.16.noarch/arch/powerpc/platforms/cell/spufs/run.c	2006-03-30 15:48:25.000000000 -0500
@@ -2,6 +2,7 @@
 #include <linux/ptrace.h>
 
 #include <asm/spu.h>
+#include <asm/unistd.h>
 
 #include "spufs.h"
 
--- linux-2.6.16.noarch/arch/powerpc/platforms/cell/spu_callbacks.c~	2006-03-31 21:53:04.000000000 -0500
+++ linux-2.6.16.noarch/arch/powerpc/platforms/cell/spu_callbacks.c	2006-03-31 21:53:43.000000000 -0500
@@ -316,6 +316,7 @@ void *spu_syscall_table[] = {
 	[__NR_pselect6]			sys_ni_syscall, /* sys_pselect */
 	[__NR_ppoll]			sys_ni_syscall, /* sys_ppoll */
 	[__NR_unshare]			sys_unshare,
+	[__NR_splice]			sys_splice,
 };
 
 long spu_sys_callback(struct spu_syscall_block *s)

Re: cell compile fixes.

[Arnd Bergmann]

Am Saturday 01 April 2006 06:53 schrieb Dave Jones:
> Missing include for __NR_syscalls, and missing sys_splice() that
> causes build-time failure due to compile-time bounds check on
> spu_syscall_table.
>
> Signed-off-by: Dave Jones <davej@redhat.com>

Acked-by: Arnd Bergmann <arnd@arndb.de>

It's not always obvious if a syscall can be added to that list, e.g.
pselect6 could not because the SPU must not change the signal mask,
but splice belongs in there.

Thanks,
	Arnd <><