Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1669560 > unrolled thread

Re: [PATCH v3 1/2] acpica: iort: Update SMMUv3 header for proximity domain mapping

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v3 1/2] acpica: iort: Update SMMUv3 header for proximity  domain mapping Robin Murphy <robin.murphy@arm.com> - 2017-06-19 20:00 +0200
    Re: [PATCH v3 1/2] acpica: iort: Update SMMUv3 header for proximity  domain mapping Ganapatrao Kulkarni <gpkulkarni@gmail.com> - 2017-06-20 04:30 +0200

#1669560 — Re: [PATCH v3 1/2] acpica: iort: Update SMMUv3 header for proximity domain mapping

On 08/06/17 05:44, Ganapatrao Kulkarni wrote:
> ARM IORT specification (rev. C) has added two new fields to define
> proximity domain for the SMMUv3 node in the IORT table.
> 
> Proximity Domain Valid:
> 	Set to 1 if the value provided in the Proximity Domain field is
> 	valid. Set to 0 otherwise.
> 
> Proximity domain:
> 	If the Proximity Domain Valid flag is set to 1, this entry
> 	provides the proximity domain to which this SMMU
> 	instance belongs.
> 
> Update header file to reflect this.
> 
> Signed-off-by: Ganapatrao Kulkarni <ganapatrao.kulkarni@cavium.com>
> ---
>  include/acpi/actbl2.h | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/include/acpi/actbl2.h b/include/acpi/actbl2.h
> index 4b306a6..389e91f 100644
> --- a/include/acpi/actbl2.h
> +++ b/include/acpi/actbl2.h
> @@ -805,6 +805,9 @@ struct acpi_iort_smmu_v3 {
>  	u32 pri_gsiv;
>  	u32 gerr_gsiv;
>  	u32 sync_gsiv;
> +	u8 pxm;
> +	u8 reserved1;
> +	u16 reserved2;

Why add the padding if you're not also adding the following field that
it aligns? (side note: could it not just be a u8[3]?)

If the actual structure definition is changing, do there not need to be
corresponding IASL changes too?

>  };
>  
>  /* Values for Model field above */
> @@ -817,6 +820,7 @@ struct acpi_iort_smmu_v3 {
>  
>  #define ACPI_IORT_SMMU_V3_COHACC_OVERRIDE   (1)
>  #define ACPI_IORT_SMMU_V3_HTTU_OVERRIDE     (1<<1)
> +#define ACPI_IORT_SMMU_V3_PXM_VALID         (1<<3)

Maybe it's just me, but "PXM" seems incredibly cryptic.

Robin.

>  
>  /*******************************************************************************
>   *
> 

[toc] | [next] | [standalone]

#1670205

On Mon, Jun 19, 2017 at 11:24 PM, Robin Murphy <robin.murphy@arm.com> wrote:
> On 08/06/17 05:44, Ganapatrao Kulkarni wrote:
>> ARM IORT specification (rev. C) has added two new fields to define
>> proximity domain for the SMMUv3 node in the IORT table.
>>
>> Proximity Domain Valid:
>>       Set to 1 if the value provided in the Proximity Domain field is
>>       valid. Set to 0 otherwise.
>>
>> Proximity domain:
>>       If the Proximity Domain Valid flag is set to 1, this entry
>>       provides the proximity domain to which this SMMU
>>       instance belongs.
>>
>> Update header file to reflect this.
>>
>> Signed-off-by: Ganapatrao Kulkarni <ganapatrao.kulkarni@cavium.com>
>> ---
>>  include/acpi/actbl2.h | 4 ++++
>>  1 file changed, 4 insertions(+)
>>
>> diff --git a/include/acpi/actbl2.h b/include/acpi/actbl2.h
>> index 4b306a6..389e91f 100644
>> --- a/include/acpi/actbl2.h
>> +++ b/include/acpi/actbl2.h
>> @@ -805,6 +805,9 @@ struct acpi_iort_smmu_v3 {
>>       u32 pri_gsiv;
>>       u32 gerr_gsiv;
>>       u32 sync_gsiv;
>> +     u8 pxm;
>> +     u8 reserved1;
>> +     u16 reserved2;
>
> Why add the padding if you're not also adding the following field that
> it aligns? (side note: could it not just be a u8[3]?)

added to align as well to comply with  the spec.

>
> If the actual structure definition is changing, do there not need to be
> corresponding IASL changes too?

yes, it is been added and sent to acpica devel ML.
https://lists.acpica.org/pipermail/devel/2017-June/001243.html

>
>>  };
>>
>>  /* Values for Model field above */
>> @@ -817,6 +820,7 @@ struct acpi_iort_smmu_v3 {
>>
>>  #define ACPI_IORT_SMMU_V3_COHACC_OVERRIDE   (1)
>>  #define ACPI_IORT_SMMU_V3_HTTU_OVERRIDE     (1<<1)
>> +#define ACPI_IORT_SMMU_V3_PXM_VALID         (1<<3)
>
> Maybe it's just me, but "PXM" seems incredibly cryptic.

PXM stands for proximity(refer ACPI spec)
it is widely used in spec and also in Linux kernel.

>
> Robin.
>
>>
>>  /*******************************************************************************
>>   *
>>
>

thanks
Ganapat

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web