Hi Ian,
On 05/02/2015 19:04, Ian Campbell wrote:
On Wed, 2015-02-04 at 17:36 +, Julien Grall wrote:
I remembered to have a discussion about this change with Naresh few
month ago.
__va should only be used when the memory is direct-mapped to Xen (i.e
accessible directly). On ARM64, this only
>>> On 05.02.15 at 12:57, wrote:
> On Thu, 2015-02-05 at 11:35 +, Jan Beulich wrote:
>> >>> On 05.02.15 at 12:04, wrote:
>> > On Wed, 2015-02-04 at 17:36 +, Julien Grall wrote:
>> >> I would move the whole function (acpi_os_map_memory) per-architecture.
>> >
>> > That's an option too, si
On Thu, 2015-02-05 at 11:35 +, Jan Beulich wrote:
> >>> On 05.02.15 at 12:04, wrote:
> > On Wed, 2015-02-04 at 17:36 +, Julien Grall wrote:
> >> I would move the whole function (acpi_os_map_memory) per-architecture.
> >
> > That's an option too, since once the "/* The low first Mb is alwa
On Thu, 2015-02-05 at 11:34 +, Jan Beulich wrote:
> >> @@ -140,6 +145,7 @@ acpi_status acpi_os_write_port(acpi_io_address port,
> >> u32 value, u32 width)
> >>
> >>return AE_OK;
> >> }
> >> +#endif
> >
> > Why only x86? Linux seems to define it also for ARM64.
>
> What is a port on ARM
>>> On 05.02.15 at 12:04, wrote:
> On Wed, 2015-02-04 at 17:36 +, Julien Grall wrote:
>> I would move the whole function (acpi_os_map_memory) per-architecture.
>
> That's an option too, since once the "/* The low first Mb is always
> mapped. */" bit is removed (which it should be since it is
>>> On 04.02.15 at 18:36, wrote:
>> --- a/xen/drivers/acpi/osl.c
>> +++ b/xen/drivers/acpi/osl.c
>> @@ -96,7 +96,11 @@ acpi_os_map_memory(acpi_physical_address phys, acpi_size
>> size)
>> return __va(phys);
>> return __vmap(&pfn, PFN_UP(offs + size), 1, 1,
>> PA
On Wed, 2015-02-04 at 17:36 +, Julien Grall wrote:
> I remembered to have a discussion about this change with Naresh few
> month ago.
>
> __va should only be used when the memory is direct-mapped to Xen (i.e
> accessible directly). On ARM64, this only the case for the RAM. Can you
> confirm
Hi Parth,
On 04/02/2015 14:01, parth.di...@linaro.org wrote:
From: Naresh Bhat
xen hypervisor will use ACPI for initialisation in the same manner that
current x86/x86_64 ones do. Add the calls to initialise the ACPI tables
and load devices using the xen/drivers/acpi subsytem.
All changes in
On Wed, 4 Feb 2015, parth.di...@linaro.org wrote:
> From: Naresh Bhat
>
> xen hypervisor will use ACPI for initialisation in the same manner that
> current x86/x86_64 ones do. Add the calls to initialise the ACPI tables
> and load devices using the xen/drivers/acpi subsytem.
>
> Signed-off-by: N
From: Naresh Bhat
xen hypervisor will use ACPI for initialisation in the same manner that
current x86/x86_64 ones do. Add the calls to initialise the ACPI tables
and load devices using the xen/drivers/acpi subsytem.
Signed-off-by: Naresh Bhat
---
xen/common/sysctl.c | 2 +
10 matches
Mail list logo
