Re: [Xen-devel] [PATCH 1 of 2] Re-name xenstore key used to save VM generation ID buffer address

[Paul Durrant <Paul.Durrant@xxxxxxxxxx>]

> -----Original Message-----
> From: Ian Campbell
> Sent: 14 December 2011 11:39
> To: Paul Durrant
> Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
> Subject: RE: [Xen-devel] [PATCH 1 of 2] Re-name xenstore key used to
> save VM generation ID buffer address
> 
> On Wed, 2011-12-14 at 11:17 +0000, Paul Durrant wrote:
> >
> > > -----Original Message-----
> > > From: Ian Campbell
> > > Sent: 14 December 2011 10:12
> > > To: Paul Durrant
> > > Cc: xen-devel@xxxxxxxxxxxxxxxxxxx
> > > Subject: Re: [Xen-devel] [PATCH 1 of 2] Re-name xenstore key
> used to
> > > save VM generation ID buffer address
> > >
> > > On Wed, 2011-12-14 at 09:31 +0000, Paul Durrant wrote:
> > > > # HG changeset patch
> > > > # User Paul Durrant <paul.durrant@xxxxxxxxxx> # Date
> 1323854952 0
> > > #
> > > > Node ID fded65be5d82461e87de54960db14ce8feb4625f
> > > > # Parent  03138a08366b895d79e143119d4c9c72833cdbcd
> > > > Re-name xenstore key used to save VM generation ID buffer
> address.
> > > >
> > > > Signed-off-by: Paul Durrant <paul.durrant@xxxxxxxxxx>
> > > >
> > > > diff -r 03138a08366b -r fded65be5d82
> > > tools/firmware/hvmloader/acpi/build.c
> > > > --- a/tools/firmware/hvmloader/acpi/build.c     Fri Dec 09
> 16:19:36
> > > 2011 +0000
> > > > +++ b/tools/firmware/hvmloader/acpi/build.c     Wed Dec 14
> 09:29:12
> > > 2011 +0000
> > > > @@ -309,7 +309,7 @@ unsigned long new_vm_gid(void)
> > > >      if ( snprintf(addr, sizeof(addr), "0x%lx",
> virt_to_phys(buf))
> > > >           >= sizeof(addr) )
> > > >          return 0;
> > > > -    xenstore_write("data/generation-id", addr);
> > > > +    xenstore_write("data/generation-id-address", addr);
> > >
> > > data/ seems like an odd home for this, isn't that the area where
> > > guests can expect to store their own bits and bobs, agent stuff
> etc?
> > >
> >
> > No what data is 'for' as such. It seemed like a reasonable place
> to
> > put something created by the guest and for the tools' consumption.
> 
> Although it runs in guest context hvmloader should really be
> considered an extension of the tools and not part of the guest.
> 
> > I could move it under hvmloader, but is that the right place?
> > hvmloader/bios is *read* by hvmloader, not written by it.
> 
> I think it's a good as anywhere else.
> 

Ok, does that mean that we should create hmvloader as writable by the guest, or 
should we special-case the generation-id-address key?

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