[PATCH] xen/ACPI: don't upload Px/Cx data for disabled processors

[PATCH] xen/ACPI: don't upload Px/Cx data for disabled processors

[Jan Beulich]

This is unnecessary and triggers a warning in the hypervisor.

Often systems have more processor entries in their ACPI tables than are
actually installed/active. The ACPI_STA_DEVICE_PRESENT bit cannot be
reliably used, but the ACPI_MADT_ENABLED bit can. In order to not
introduce new functions in the main ACPI processor driver code, simply
use acpi_get_phys_id(), which does more than we need, but which checks
the MADT enabled bit in the process. Any CPU for which we can't
determine the APIC ID is unlikely to work properly anyway, so the extra
checks done by acpi_get_phys_id() should do no harm.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
---
 drivers/acpi/processor_core.c    |    1 +
 drivers/xen/xen-acpi-processor.c |    6 ++++++
 2 files changed, 7 insertions(+)

--- 4.18-rc2/drivers/acpi/processor_core.c
+++ 4.18-rc2-xen-ACPI-processor-skip-disabled/drivers/acpi/processor_core.c
@@ -205,6 +205,7 @@ phys_cpuid_t acpi_get_phys_id(acpi_handl
 
 	return phys_id;
 }
+EXPORT_SYMBOL_GPL(acpi_get_phys_id);
 
 int acpi_map_cpuid(phys_cpuid_t phys_id, u32 acpi_id)
 {
--- 4.18-rc2/drivers/xen/xen-acpi-processor.c
+++ 4.18-rc2-xen-ACPI-processor-skip-disabled/drivers/xen/xen-acpi-processor.c
@@ -362,6 +362,12 @@ read_acpi_id(acpi_handle handle, u32 lvl
 	default:
 		return AE_OK;
 	}
+	if (invalid_phys_cpuid(acpi_get_phys_id(handle,
+						acpi_type == ACPI_TYPE_DEVICE,
+						acpi_id))) {
+		pr_debug("CPU with ACPI ID %u is unavailable\n", acpi_id);
+		return AE_OK;
+	}
 	/* There are more ACPI Processor objects than in x2APIC or MADT.
 	 * This can happen with incorrect ACPI SSDT declerations. */
 	if (acpi_id >= nr_acpi_bits) {





_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Ping: [PATCH] xen/ACPI: don't upload Px/Cx data for disabled processors

[Jan Beulich]

>>> On 25.06.18 at 12:17,  wrote:
> This is unnecessary and triggers a warning in the hypervisor.
> 
> Often systems have more processor entries in their ACPI tables than are
> actually installed/active. The ACPI_STA_DEVICE_PRESENT bit cannot be
> reliably used, but the ACPI_MADT_ENABLED bit can. In order to not
> introduce new functions in the main ACPI processor driver code, simply
> use acpi_get_phys_id(), which does more than we need, but which checks
> the MADT enabled bit in the process. Any CPU for which we can't
> determine the APIC ID is unlikely to work properly anyway, so the extra
> checks done by acpi_get_phys_id() should do no harm.
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>
> ---
>  drivers/acpi/processor_core.c    |    1 +
>  drivers/xen/xen-acpi-processor.c |    6 ++++++
>  2 files changed, 7 insertions(+)

With Jürgen's R-b in place, may I ask for an ack for the processor_core.c
change, or - in case you dislike the new export - an alternative suggestion?

Thanks, Jan

> --- 4.18-rc2/drivers/acpi/processor_core.c
> +++ 4.18-rc2-xen-ACPI-processor-skip-disabled/drivers/acpi/processor_core.c
> @@ -205,6 +205,7 @@ phys_cpuid_t acpi_get_phys_id(acpi_handl
>  
>  	return phys_id;
>  }
> +EXPORT_SYMBOL_GPL(acpi_get_phys_id);
>  
>  int acpi_map_cpuid(phys_cpuid_t phys_id, u32 acpi_id)
>  {
> --- 4.18-rc2/drivers/xen/xen-acpi-processor.c
> +++ 
> 4.18-rc2-xen-ACPI-processor-skip-disabled/drivers/xen/xen-acpi-processor.c
> @@ -362,6 +362,12 @@ read_acpi_id(acpi_handle handle, u32 lvl
>  	default:
>  		return AE_OK;
>  	}
> +	if (invalid_phys_cpuid(acpi_get_phys_id(handle,
> +						acpi_type == ACPI_TYPE_DEVICE,
> +						acpi_id))) {
> +		pr_debug("CPU with ACPI ID %u is unavailable\n", acpi_id);
> +		return AE_OK;
> +	}
>  	/* There are more ACPI Processor objects than in x2APIC or MADT.
>  	 * This can happen with incorrect ACPI SSDT declerations. */
>  	if (acpi_id >= nr_acpi_bits) {
> 
> 
> 
> 




_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: Ping: [PATCH] xen/ACPI: don't upload Px/Cx data for disabled processors

[Rafael J. Wysocki]

On Wed, Aug 15, 2018 at 8:44 AM Jan Beulich <JBeulich@suse.com> wrote:
>
> >>> On 25.06.18 at 12:17,  wrote:
> > This is unnecessary and triggers a warning in the hypervisor.
> >
> > Often systems have more processor entries in their ACPI tables than are
> > actually installed/active. The ACPI_STA_DEVICE_PRESENT bit cannot be
> > reliably used, but the ACPI_MADT_ENABLED bit can. In order to not
> > introduce new functions in the main ACPI processor driver code, simply
> > use acpi_get_phys_id(), which does more than we need, but which checks
> > the MADT enabled bit in the process. Any CPU for which we can't
> > determine the APIC ID is unlikely to work properly anyway, so the extra
> > checks done by acpi_get_phys_id() should do no harm.
> >
> > Signed-off-by: Jan Beulich <jbeulich@suse.com>
> > ---
> >  drivers/acpi/processor_core.c    |    1 +
> >  drivers/xen/xen-acpi-processor.c |    6 ++++++
> >  2 files changed, 7 insertions(+)
>
> With Jürgen's R-b in place, may I ask for an ack for the processor_core.c
> change, or - in case you dislike the new export - an alternative suggestion?

It is fine to export it IMO.  If you need my ACK for that, please feel
free to add it to the patch.

Thanks,
Rafael

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: Ping: [PATCH] xen/ACPI: don't upload Px/Cx data for disabled processors

[Boris Ostrovsky]

On 08/15/2018 03:55 AM, Rafael J. Wysocki wrote:
> On Wed, Aug 15, 2018 at 8:44 AM Jan Beulich <JBeulich@suse.com> wrote:
>>>>> On 25.06.18 at 12:17,  wrote:
>>> This is unnecessary and triggers a warning in the hypervisor.
>>>
>>> Often systems have more processor entries in their ACPI tables than are
>>> actually installed/active. The ACPI_STA_DEVICE_PRESENT bit cannot be
>>> reliably used, but the ACPI_MADT_ENABLED bit can. In order to not
>>> introduce new functions in the main ACPI processor driver code, simply
>>> use acpi_get_phys_id(), which does more than we need, but which checks
>>> the MADT enabled bit in the process. Any CPU for which we can't
>>> determine the APIC ID is unlikely to work properly anyway, so the extra
>>> checks done by acpi_get_phys_id() should do no harm.
>>>
>>> Signed-off-by: Jan Beulich <jbeulich@suse.com>
>>> ---
>>>  drivers/acpi/processor_core.c    |    1 +
>>>  drivers/xen/xen-acpi-processor.c |    6 ++++++
>>>  2 files changed, 7 insertions(+)
>> With Jürgen's R-b in place, may I ask for an ack for the processor_core.c
>> change, or - in case you dislike the new export - an alternative suggestion?
> It is fine to export it IMO.  If you need my ACK for that, please feel
> free to add it to the patch.


Applied to for-linus-4.19.

-boris

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel