Re: [Xen-devel] [PATCH RFC 15/20] acpi: Move ACPI code to xen/common/libacpi

[Roger Pau Monné <roger.pau@xxxxxxxxxx>]

On Mon, Jun 06, 2016 at 12:09:21PM -0400, Boris Ostrovsky wrote:
> On 06/06/2016 09:05 AM, Jan Beulich wrote:
> >>>> On 06.04.16 at 03:25, <boris.ostrovsky@xxxxxxxxxx> wrote:
> >>  .gitignore                                                        | 8 
> >> ++++----
> >>  tools/firmware/hvmloader/Makefile                                 | 3 +--
> >>  tools/firmware/hvmloader/smbios.c                                 | 1 +
> >>  tools/firmware/rombios/32bit/Makefile                             | 2 +-
> >>  tools/firmware/rombios/32bit/tcgbios/Makefile                     | 2 +-
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/Makefile    | 6 
> >> +++---
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/README      | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/acpi2_0.h   | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/build.c     | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/dsdt.asl    | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/mk_dsdt.c   | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/ssdt_pm.asl | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/ssdt_s3.asl | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/ssdt_s4.asl | 0
> >>  .../firmware/hvmloader/acpi => xen/common/libacpi}/ssdt_tpm.asl   | 0
> >>  .../hvmloader/acpi => xen/common/libacpi}/static_tables.c         | 0
> >>  {tools/firmware/hvmloader/acpi => xen/common/libacpi}/x86.h       | 0
> >>  17 files changed, 11 insertions(+), 11 deletions(-)
> > As mentioned before, new placement subject to determination
> > whether this is to eventually be used by the hypervisor.
> 
> 
> Roger, when do you think you'll be able to see whether dom0 builder will
> use this? I can start working on v1 with assumption that this will be
> used by hypervisor. We can drop the last patch (and modify this one to
> move acpi to somewhere in tools) if it becomes clear that hypervisor
> does not want it.

What we need to do is quite similar to what ARM does, and they seem to 
prefer to have different ACPI code for Dom0/DomU, so I'm quite sure we could 
do the same for x86 also. FWIW, when Boris and I discussed this we though it 
would be better to have a single place where all the ACPI code resides, both 
to create tables or to modify them.

Would it be a lot of work to leave it in tools/ for the time being and move 
it later if we see that we could benefit from merging both? Or maybe leave 
it in xen/common without linking it with the hypervisor just yet (but that 
would need sorting out later anyway).

Roger.

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