On Thu, Dec 16, 2010 at 14:23, Greg Ungerer<gerg@xxxxxxxxxxxx> wrote:On 15/12/10 20:10, Al Viro wrote:On Sun, Dec 05, 2010 at 12:15:10PM +0100, Geert Uytterhoeven wrote:
Unless someone objects, I'm inclined to apply Al's patchset, as it
improves the
situation anyway.
OK?
FWIW, there's an m68knommu counterpart (ACKed by gerg a while ago). The
entire bunch is on ftp.linux.org.uk/pub/people/viro/m68k-signals - both m68k
and m68knommu sets. I don't know how you guys would prefer it done, since
Thx, good the know the nommu part is there, as I don't seem to have it
in my mailbox.
there's a trivial dependency between m68k and m68knommu parts of series -
the former has
+#ifndef __uClinux__
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
+#endif
in unistd.h and the latter does
-#ifndef __uClinux__
#define __ARCH_WANT_SYS_RT_SIGSUSPEND
-#endif
Both parts switch to generic sys_rt_sigsuspend(), for mmu and nommu resp.,
so they need __ARCH_WANT_SYS_RT_SIGSUSPEND defined as soon as conversion
is done. Other than that they are independent.
I can make sure to send to Linus after Geert. Unless Geert you
want to take and send both the m68k and m68knommu changes?
I can take the m68knommu changes with your ack, too.
I'll try to take a look at it tomorrow...