PATCH powerpc Merge asm-ppc*/seccomp.h

PATCH powerpc Merge asm-ppc*/seccomp.h

[Jon Loeliger]

Merge asm-ppc*/seccomp.h.

Signed-off-by: Jon Loeliger <jdl@freescale.com>
Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
---


 include/asm-powerpc/seccomp.h |   21 +++++++++++++++++++++
 include/asm-ppc/seccomp.h     |   10 ----------
 include/asm-ppc64/seccomp.h   |   21 ---------------------
 3 files changed, 21 insertions(+), 31 deletions(-)


diff --git a/include/asm-powerpc/seccomp.h b/include/asm-powerpc/seccomp.h
new file mode 100644
--- /dev/null
+++ b/include/asm-powerpc/seccomp.h
@@ -0,0 +1,21 @@
+#ifndef _ASM_POWERPC_SECCOMP_H
+
+#include <linux/thread_info.h>
+
+#if defined(__powerpc64__) && !defined(TIF_32BIT)
+#error "unexpected TIF_32BIT on ppc64"
+#endif
+
+#include <linux/unistd.h>
+
+#define __NR_seccomp_read __NR_read
+#define __NR_seccomp_write __NR_write
+#define __NR_seccomp_exit __NR_exit
+#define __NR_seccomp_sigreturn __NR_rt_sigreturn
+
+#define __NR_seccomp_read_32 __NR_read
+#define __NR_seccomp_write_32 __NR_write
+#define __NR_seccomp_exit_32 __NR_exit
+#define __NR_seccomp_sigreturn_32 __NR_sigreturn
+
+#endif	/* _ASM_POWERPC_SECCOMP_H */
diff --git a/include/asm-ppc/seccomp.h b/include/asm-ppc/seccomp.h
deleted file mode 100644
--- a/include/asm-ppc/seccomp.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef _ASM_SECCOMP_H
-
-#include <linux/unistd.h>
-
-#define __NR_seccomp_read __NR_read
-#define __NR_seccomp_write __NR_write
-#define __NR_seccomp_exit __NR_exit
-#define __NR_seccomp_sigreturn __NR_rt_sigreturn
-
-#endif /* _ASM_SECCOMP_H */
diff --git a/include/asm-ppc64/seccomp.h b/include/asm-ppc64/seccomp.h
deleted file mode 100644
--- a/include/asm-ppc64/seccomp.h
+++ /dev/null
@@ -1,21 +0,0 @@
-#ifndef _ASM_SECCOMP_H
-
-#include <linux/thread_info.h> /* already defines TIF_32BIT */
-
-#ifndef TIF_32BIT
-#error "unexpected TIF_32BIT on ppc64"
-#endif
-
-#include <linux/unistd.h>
-
-#define __NR_seccomp_read __NR_read
-#define __NR_seccomp_write __NR_write
-#define __NR_seccomp_exit __NR_exit
-#define __NR_seccomp_sigreturn __NR_rt_sigreturn
-
-#define __NR_seccomp_read_32 __NR_read
-#define __NR_seccomp_write_32 __NR_write
-#define __NR_seccomp_exit_32 __NR_exit
-#define __NR_seccomp_sigreturn_32 __NR_sigreturn
-
-#endif /* _ASM_SECCOMP_H */

Re: PATCH powerpc Merge asm-ppc*/seccomp.h

[Christoph Hellwig]

> +#include <linux/thread_info.h>
> +
> +#if defined(__powerpc64__) && !defined(TIF_32BIT)
> +#error "unexpected TIF_32BIT on ppc64"
> +#endif

just kill this check, it's rather pointless

Re: PATCH powerpc Merge asm-ppc*/seccomp.h

[Jon Loeliger]

So, like, the other day Christoph Hellwig mumbled:
> > +#include <linux/thread_info.h>
> > +
> > +#if defined(__powerpc64__) && !defined(TIF_32BIT)
> > +#error "unexpected TIF_32BIT on ppc64"
> > +#endif
> 
> just kill this check, it's rather pointless

OK.

But keep the #include?  It's actually the part
that is defining TIF_32BIT for ppc64 (indirectly
through linux/thread_info.h and asm/thread_info.h).
Won't those bits be needed still?

jdl

Re: PATCH powerpc Merge asm-ppc*/seccomp.h

[Christoph Hellwig]

On Thu, Sep 22, 2005 at 01:36:52PM -0500, Jon Loeliger wrote:
> So, like, the other day Christoph Hellwig mumbled:
> > > +#include <linux/thread_info.h>
> > > +
> > > +#if defined(__powerpc64__) && !defined(TIF_32BIT)
> > > +#error "unexpected TIF_32BIT on ppc64"
> > > +#endif
> > 
> > just kill this check, it's rather pointless
> 
> OK.
> 
> But keep the #include?  It's actually the part
> that is defining TIF_32BIT for ppc64 (indirectly
> through linux/thread_info.h and asm/thread_info.h).
> Won't those bits be needed still?

I'm not sure.  This header definitly doesn't need it, but I suspect
seccompt.c expects to get it via this header.

Re: PATCH powerpc Merge asm-ppc*/seccomp.h

[Jon Loeliger]

So, like, the other day Christoph Hellwig mumbled:
> > 
> > But keep the #include?  It's actually the part
> > that is defining TIF_32BIT for ppc64 (indirectly
> > through linux/thread_info.h and asm/thread_info.h).
> > Won't those bits be needed still?
> 
> I'm not sure.  This header definitly doesn't need it, but I suspect
> seccompt.c expects to get it via this header.

I dug around in this a bit.  The nested includes here yield
a fairly wide distribution of availability of TIF_32BIT.
I'm hesitant to remove it in this "merge" patch, so I
will leave its removal to a follow on patch/effort.

New patch with #if test removed coming up.

jdl