Re: [Xen-devel] [PATCH, RFC] Re: x86: gnttab_clear_flag() abusing clear_bit()

[Keir Fraser <keir.xen@xxxxxxxxx>]

On 09/02/2012 01:01, "Jan Beulich" <JBeulich@xxxxxxxx> wrote:

>> Looks fine to me, in principle. I would add a comment to the x86
>> gnttab_clear_flag() explaining why we have to open code something that looks
>> a lot like clear_bit().
> 
> That one I already did, will submit soon (desiring clarification on the
> below).
> 
> As to the "+m" constraint - I'm being told that "+m" (var) is equivalent
> to "=m" (var) : "m" (var), no matter what the documentation says
> regarding '+' (but they're also not seeing a need to adjust the docs
> accordingly).
> 
> The question is whether we should go with the (documentation-wise
> correct) form, or the shorter one (which they're unlikely to change
> the meaning of, given in how many places "+m" is used in e.g. Linux).

You could switch us to "+m" and see how we get on.

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel