Re: [Xen-devel] [PATCH v4 14/24] arm/acpi: Prepare EFI system table for Dom0

[Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>]

On Fri, 4 Mar 2016, Shannon Zhao wrote:
> On 2016/3/1 20:49, Stefano Stabellini wrote:
> > On Tue, 1 Mar 2016, Jan Beulich wrote:
> >>>>> > >>> On 01.03.16 at 03:35, <zhaoshenglong@xxxxxxxxxx> wrote:
> >> > 
> >>> > > 
> >>> > > On 2016/2/29 22:36, Jan Beulich wrote:
> >>>>>>> > >>>>> On 29.02.16 at 15:25, <stefano.stabellini@xxxxxxxxxxxxx> 
> >>>>>>> > >>>>> wrote:
> >>>>>> > >>> > On Mon, 29 Feb 2016, Jan Beulich wrote:
> >>>>>>>> > >>>> >> Also it doesn't look very nice to me to (ab)use xz's CRC32 
> >>>>>>>> > >>>> >> code
> >>>>>>>> > >>>> >> here; I don't know who has suggested doing so.
> >>>>>> > >>> >  
> >>>>>> > >>> > It was suggested by Julien.
> >>>>>> > >>> > 
> >>>>>> > >>> > I agree that including ../../../common/xz/crc32.c seems a bit 
> >>>>>> > >>> > fragile
> >>>>>> > >>> > but introducing another copy of xz_crc32 seems even worse to 
> >>>>>> > >>> > me (see
> >>>>>> > >>> > http://marc.info/?l=xen-devel&m=144775375427921&w=2). Maybe 
> >>>>>> > >>> > you have a
> >>>>>> > >>> > better suggestion?
> >>>> > >> Well, at some point there was talk of ARM not wanting to
> >>>> > >> ExitBootServices() as early as x86 does. In that case there
> >>>> > >> would be a CRC32 function among the various boot services
> >>>> > >> ones.
> >>>> > >> 
> >>> > > At this point, I think it already ExitBootServices() while it's 
> >>> > > creating
> >>> > > Dom0.
> >> > 
> >> > I understand that's the case today, hence my saying "at some
> >> > point there was talk of ...".
> >> > 
> >>>> > >> The other option would be to have a generic crc32() function,
> >>>> > >> and maybe make xz use it.
> >>> > > Ok, I'll go for this way.
> >> > 
> >> > Well, for the avoidance of doubt: With the code moving into an
> >> > ARM specific file, if Stefano is fine with the xz hack, I certainly
> >> > won't insist on you adding a separate crc32().
> > Having a generic crc32() function like you suggested would be nicer than
> > the xz hack. If Shannon is OK with doing that, it would be best.
> Since this will introduce more changes, I want to stay with current way.
 
OK

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