[PATCH XEN] xen: event channel arrays are xen_ulong_t and not unsigned long

[Jan Beulich]

>>> On 22.02.13 at 09:55, Ian Campbell <Ian.Campbell at citrix.com> wrote:
> On Fri, 2013-02-22 at 08:48 +0000, Jan Beulich wrote:
>> >>> On 22.02.13 at 09:28, Ian Campbell <Ian.Campbell at citrix.com> wrote:
>> > On Fri, 2013-02-22 at 08:12 +0000, Jan Beulich wrote:
>> >> >>> On 21.02.13 at 18:16, Ian Campbell <Ian.Campbell at citrix.com> wrote:
>> >> > On Tue, 2013-02-19 at 14:49 +0000, Ian Campbell wrote:
>> >> >> On ARM we want these to be the same size on 32- and 64-bit.
>> >> >> 
>> >> >> This is an ABI change on ARM. X86 does not change.
>> >> >> 
>> >> >> Signed-off-by: Ian Campbell <ian.campbell at citrix.com>
>> >> >> Cc: Jan Beulich <JBeulich at suse.com>
>> >> >> Cc: Keir (Xen.org) <keir at xen.org>
>> >> > 
>> >> > Are you guys (un)happy with this change from the Xen & x86 side?
>> >> 
>> >> I don't see any problem with it.
>> > 
>> > I'll take this as an Acked-by if that's ok with you.
>> 
>> Well, I specifically didn't say "ack": I don't really mind the change,
>> but I'm also not eager see it go in. Not seeing a problem with it
>> doesn't really mean there's none lurking - type changes in public
>> interfaces are always an at least slightly risky business.
> 
> For x86 there is no change here since:
> xen/include/public/arch-x86/xen.h:typedef unsigned long xen_ulong_t;
> 
> Also the tools/include/xen-foreign checks show no change in the size of
> things on x86.

Which is why I said I see no problem with it.

> On ARM the change is deliberate, if there is any fallout we can fix it.

I understand that.

Jan