Re: [Xen-devel] Should we revert "mm: New XENMEM space, XENMAPSPACE_gmfn_range"?

[Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>]

On Thu, 2 Aug 2012, Jan Beulich wrote:
> >>> On 02.08.12 at 15:13, Stefano Stabellini 
> >>> <stefano.stabellini@xxxxxxxxxxxxx>
> wrote:
> > On Thu, 2 Aug 2012, Jan Beulich wrote:
> >> >>> On 01.08.12 at 19:55, Stefano Stabellini 
> >> >>> <stefano.stabellini@xxxxxxxxxxxxx> 
> > wrote:
> >> > I was reading more about this commit because this patch breaks the ABI
> >> > on ARM, when I realized that on x86 there is no standard that specifies
> >> > the alignment of fields in a struct.
> >> 
> >> There is - the psABI supplements to the SVR4 ABI.
> > 
> > Thank you very much, that document was exactly what I was looking for.
> > 
> > Also it explains where my confusion was coming from: Jean's patch doesn't
> > break the ABI on ARM or x86, but I am carrying a patch in my patch queue
> > that does (unless Jean's patch is applied):
> > 
> > http://marc.info/?l=xen-devel&m=134305777903771 
> > 
> > As you can see this patch splits .space into two shorts, and as a side
> > effect changes the offset of .space, removing the padding.
> > Thus it led me to think that Jean's patch was breaking the ABI when actually
> > with "arm: initial XENMAPSPACE_gmfn_foreign" applied, it becomes
> > required to keep the binary interface compatible.
> 
> And then you wouldn't need to split 'space' and break the ABI
> at all, you could simply put 'size' and 'foreign_domid' into a union.

Yes, that's a good suggestion.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel