Re: [Intel-gfx] [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-25 Thread Ingo Molnar 

* Jani Nikula  wrote:

> On Wed, 25 Jan 2017, Jani Nikula  wrote:
> > On Wed, 25 Jan 2017, Ingo Molnar  wrote:
> >> * Paulo Zanoni  wrote:
> >>
> >>> So don't forget to reserve its stolen memory bits.
> >>> 
> >>> Cc: Ingo Molnar 
> >>> Cc: H. Peter Anvin 
> >>> Cc: Ander Conselvan de Oliveira 
> >>> Cc: x...@kernel.org
> >>> Reviewed-by: Ander Conselvan de Oliveira 
> >>> 
> >>> Signed-off-by: Paulo Zanoni 
> >>> ---
> >>>  arch/x86/kernel/early-quirks.c | 1 +
> >>>  1 file changed, 1 insertion(+)
> >>> 
> >>> Found by code inspection. This is completely untested since I don't have
> >>> GLK hardware.
> >>> 
> >>> diff --git a/arch/x86/kernel/early-quirks.c 
> >>> b/arch/x86/kernel/early-quirks.c
> >>> index 6a08e25..23c4f1c 100644
> >>> --- a/arch/x86/kernel/early-quirks.c
> >>> +++ b/arch/x86/kernel/early-quirks.c
> >>> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
> >>> __initconst = {
> >>>   INTEL_SKL_IDS(_early_ops),
> >>>   INTEL_BXT_IDS(_early_ops),
> >>>   INTEL_KBL_IDS(_early_ops),
> >>> + INTEL_GLK_IDS(_early_ops),
> >>>  };
> >>
> >> There's no INTEL_GLK_IDS() upstream - is there any dependency here on 
> >> other 
> >> changes to the i915 GPU driver?
> >
> > It's in our -next. Ack for merging this through drm-intel along with the
> > deps?

Yeah, LGTM:

  Acked-by: Ingo Molnar 

Thanks,

Ingo

Re: [Intel-gfx] [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-25 Thread Ingo Molnar 

* Jani Nikula  wrote:

> On Wed, 25 Jan 2017, Jani Nikula  wrote:
> > On Wed, 25 Jan 2017, Ingo Molnar  wrote:
> >> * Paulo Zanoni  wrote:
> >>
> >>> So don't forget to reserve its stolen memory bits.
> >>> 
> >>> Cc: Ingo Molnar 
> >>> Cc: H. Peter Anvin 
> >>> Cc: Ander Conselvan de Oliveira 
> >>> Cc: x...@kernel.org
> >>> Reviewed-by: Ander Conselvan de Oliveira 
> >>> 
> >>> Signed-off-by: Paulo Zanoni 
> >>> ---
> >>>  arch/x86/kernel/early-quirks.c | 1 +
> >>>  1 file changed, 1 insertion(+)
> >>> 
> >>> Found by code inspection. This is completely untested since I don't have
> >>> GLK hardware.
> >>> 
> >>> diff --git a/arch/x86/kernel/early-quirks.c 
> >>> b/arch/x86/kernel/early-quirks.c
> >>> index 6a08e25..23c4f1c 100644
> >>> --- a/arch/x86/kernel/early-quirks.c
> >>> +++ b/arch/x86/kernel/early-quirks.c
> >>> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
> >>> __initconst = {
> >>>   INTEL_SKL_IDS(_early_ops),
> >>>   INTEL_BXT_IDS(_early_ops),
> >>>   INTEL_KBL_IDS(_early_ops),
> >>> + INTEL_GLK_IDS(_early_ops),
> >>>  };
> >>
> >> There's no INTEL_GLK_IDS() upstream - is there any dependency here on 
> >> other 
> >> changes to the i915 GPU driver?
> >
> > It's in our -next. Ack for merging this through drm-intel along with the
> > deps?

Yeah, LGTM:

  Acked-by: Ingo Molnar 

Thanks,

Ingo

Re: [Intel-gfx] [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-25 Thread Jani Nikula 

(Now without the typo in Daniel's address; apologies for the extra
noise.)

On Wed, 25 Jan 2017, Jani Nikula  wrote:
> On Wed, 25 Jan 2017, Ingo Molnar  wrote:
>> * Paulo Zanoni  wrote:
>>
>>> So don't forget to reserve its stolen memory bits.
>>> 
>>> Cc: Ingo Molnar 
>>> Cc: H. Peter Anvin 
>>> Cc: Ander Conselvan de Oliveira 
>>> Cc: x...@kernel.org
>>> Reviewed-by: Ander Conselvan de Oliveira 
>>> 
>>> Signed-off-by: Paulo Zanoni 
>>> ---
>>>  arch/x86/kernel/early-quirks.c | 1 +
>>>  1 file changed, 1 insertion(+)
>>> 
>>> Found by code inspection. This is completely untested since I don't have
>>> GLK hardware.
>>> 
>>> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
>>> index 6a08e25..23c4f1c 100644
>>> --- a/arch/x86/kernel/early-quirks.c
>>> +++ b/arch/x86/kernel/early-quirks.c
>>> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
>>> __initconst = {
>>> INTEL_SKL_IDS(_early_ops),
>>> INTEL_BXT_IDS(_early_ops),
>>> INTEL_KBL_IDS(_early_ops),
>>> +   INTEL_GLK_IDS(_early_ops),
>>>  };
>>
>> There's no INTEL_GLK_IDS() upstream - is there any dependency here on other 
>> changes to the i915 GPU driver?
>
> It's in our -next. Ack for merging this through drm-intel along with the
> deps?
>
> BR,
> Jani.
>
>>
>> Thanks,
>>
>>  Ingo
>> ___
>> Intel-gfx mailing list
>> intel-...@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Jani Nikula, Intel Open Source Technology Center

Re: [Intel-gfx] [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-25 Thread Jani Nikula 

(Now without the typo in Daniel's address; apologies for the extra
noise.)

On Wed, 25 Jan 2017, Jani Nikula  wrote:
> On Wed, 25 Jan 2017, Ingo Molnar  wrote:
>> * Paulo Zanoni  wrote:
>>
>>> So don't forget to reserve its stolen memory bits.
>>> 
>>> Cc: Ingo Molnar 
>>> Cc: H. Peter Anvin 
>>> Cc: Ander Conselvan de Oliveira 
>>> Cc: x...@kernel.org
>>> Reviewed-by: Ander Conselvan de Oliveira 
>>> 
>>> Signed-off-by: Paulo Zanoni 
>>> ---
>>>  arch/x86/kernel/early-quirks.c | 1 +
>>>  1 file changed, 1 insertion(+)
>>> 
>>> Found by code inspection. This is completely untested since I don't have
>>> GLK hardware.
>>> 
>>> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
>>> index 6a08e25..23c4f1c 100644
>>> --- a/arch/x86/kernel/early-quirks.c
>>> +++ b/arch/x86/kernel/early-quirks.c
>>> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
>>> __initconst = {
>>> INTEL_SKL_IDS(_early_ops),
>>> INTEL_BXT_IDS(_early_ops),
>>> INTEL_KBL_IDS(_early_ops),
>>> +   INTEL_GLK_IDS(_early_ops),
>>>  };
>>
>> There's no INTEL_GLK_IDS() upstream - is there any dependency here on other 
>> changes to the i915 GPU driver?
>
> It's in our -next. Ack for merging this through drm-intel along with the
> deps?
>
> BR,
> Jani.
>
>>
>> Thanks,
>>
>>  Ingo
>> ___
>> Intel-gfx mailing list
>> intel-...@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Jani Nikula, Intel Open Source Technology Center

Re: [Intel-gfx] [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-25 Thread Jani Nikula 
On Wed, 25 Jan 2017, Ingo Molnar  wrote:
> * Paulo Zanoni  wrote:
>
>> So don't forget to reserve its stolen memory bits.
>> 
>> Cc: Ingo Molnar 
>> Cc: H. Peter Anvin 
>> Cc: Ander Conselvan de Oliveira 
>> Cc: x...@kernel.org
>> Reviewed-by: Ander Conselvan de Oliveira 
>> 
>> Signed-off-by: Paulo Zanoni 
>> ---
>>  arch/x86/kernel/early-quirks.c | 1 +
>>  1 file changed, 1 insertion(+)
>> 
>> Found by code inspection. This is completely untested since I don't have
>> GLK hardware.
>> 
>> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
>> index 6a08e25..23c4f1c 100644
>> --- a/arch/x86/kernel/early-quirks.c
>> +++ b/arch/x86/kernel/early-quirks.c
>> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
>> __initconst = {
>>  INTEL_SKL_IDS(_early_ops),
>>  INTEL_BXT_IDS(_early_ops),
>>  INTEL_KBL_IDS(_early_ops),
>> +INTEL_GLK_IDS(_early_ops),
>>  };
>
> There's no INTEL_GLK_IDS() upstream - is there any dependency here on other 
> changes to the i915 GPU driver?

It's in our -next. Ack for merging this through drm-intel along with the
deps?

BR,
Jani.

>
> Thanks,
>
>   Ingo
> ___
> Intel-gfx mailing list
> intel-...@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Jani Nikula, Intel Open Source Technology Center

Re: [Intel-gfx] [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-25 Thread Jani Nikula 
On Wed, 25 Jan 2017, Ingo Molnar  wrote:
> * Paulo Zanoni  wrote:
>
>> So don't forget to reserve its stolen memory bits.
>> 
>> Cc: Ingo Molnar 
>> Cc: H. Peter Anvin 
>> Cc: Ander Conselvan de Oliveira 
>> Cc: x...@kernel.org
>> Reviewed-by: Ander Conselvan de Oliveira 
>> 
>> Signed-off-by: Paulo Zanoni 
>> ---
>>  arch/x86/kernel/early-quirks.c | 1 +
>>  1 file changed, 1 insertion(+)
>> 
>> Found by code inspection. This is completely untested since I don't have
>> GLK hardware.
>> 
>> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
>> index 6a08e25..23c4f1c 100644
>> --- a/arch/x86/kernel/early-quirks.c
>> +++ b/arch/x86/kernel/early-quirks.c
>> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
>> __initconst = {
>>  INTEL_SKL_IDS(_early_ops),
>>  INTEL_BXT_IDS(_early_ops),
>>  INTEL_KBL_IDS(_early_ops),
>> +INTEL_GLK_IDS(_early_ops),
>>  };
>
> There's no INTEL_GLK_IDS() upstream - is there any dependency here on other 
> changes to the i915 GPU driver?

It's in our -next. Ack for merging this through drm-intel along with the
deps?

BR,
Jani.

>
> Thanks,
>
>   Ingo
> ___
> Intel-gfx mailing list
> intel-...@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Jani Nikula, Intel Open Source Technology Center

Re: [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-24 Thread Ingo Molnar 

* Paulo Zanoni  wrote:

> So don't forget to reserve its stolen memory bits.
> 
> Cc: Ingo Molnar 
> Cc: H. Peter Anvin 
> Cc: Ander Conselvan de Oliveira 
> Cc: x...@kernel.org
> Reviewed-by: Ander Conselvan de Oliveira 
> 
> Signed-off-by: Paulo Zanoni 
> ---
>  arch/x86/kernel/early-quirks.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> Found by code inspection. This is completely untested since I don't have
> GLK hardware.
> 
> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
> index 6a08e25..23c4f1c 100644
> --- a/arch/x86/kernel/early-quirks.c
> +++ b/arch/x86/kernel/early-quirks.c
> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
> __initconst = {
>   INTEL_SKL_IDS(_early_ops),
>   INTEL_BXT_IDS(_early_ops),
>   INTEL_KBL_IDS(_early_ops),
> + INTEL_GLK_IDS(_early_ops),
>  };

There's no INTEL_GLK_IDS() upstream - is there any dependency here on other 
changes to the i915 GPU driver?

Thanks,

Ingo

Re: [PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-24 Thread Ingo Molnar 

* Paulo Zanoni  wrote:

> So don't forget to reserve its stolen memory bits.
> 
> Cc: Ingo Molnar 
> Cc: H. Peter Anvin 
> Cc: Ander Conselvan de Oliveira 
> Cc: x...@kernel.org
> Reviewed-by: Ander Conselvan de Oliveira 
> 
> Signed-off-by: Paulo Zanoni 
> ---
>  arch/x86/kernel/early-quirks.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> Found by code inspection. This is completely untested since I don't have
> GLK hardware.
> 
> diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
> index 6a08e25..23c4f1c 100644
> --- a/arch/x86/kernel/early-quirks.c
> +++ b/arch/x86/kernel/early-quirks.c
> @@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
> __initconst = {
>   INTEL_SKL_IDS(_early_ops),
>   INTEL_BXT_IDS(_early_ops),
>   INTEL_KBL_IDS(_early_ops),
> + INTEL_GLK_IDS(_early_ops),
>  };

There's no INTEL_GLK_IDS() upstream - is there any dependency here on other 
changes to the i915 GPU driver?

Thanks,

Ingo

[PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-24 Thread Paulo Zanoni 
So don't forget to reserve its stolen memory bits.

Cc: Ingo Molnar 
Cc: H. Peter Anvin 
Cc: Ander Conselvan de Oliveira 
Cc: x...@kernel.org
Reviewed-by: Ander Conselvan de Oliveira 
Signed-off-by: Paulo Zanoni 
---
 arch/x86/kernel/early-quirks.c | 1 +
 1 file changed, 1 insertion(+)

Found by code inspection. This is completely untested since I don't have
GLK hardware.

diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 6a08e25..23c4f1c 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
__initconst = {
INTEL_SKL_IDS(_early_ops),
INTEL_BXT_IDS(_early_ops),
INTEL_KBL_IDS(_early_ops),
+   INTEL_GLK_IDS(_early_ops),
 };
 
 static void __init
-- 
2.7.4

[PATCH] x86/gpu: GLK uses the same GMS values as SKL

2017-01-24 Thread Paulo Zanoni 
So don't forget to reserve its stolen memory bits.

Cc: Ingo Molnar 
Cc: H. Peter Anvin 
Cc: Ander Conselvan de Oliveira 
Cc: x...@kernel.org
Reviewed-by: Ander Conselvan de Oliveira 
Signed-off-by: Paulo Zanoni 
---
 arch/x86/kernel/early-quirks.c | 1 +
 1 file changed, 1 insertion(+)

Found by code inspection. This is completely untested since I don't have
GLK hardware.

diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c
index 6a08e25..23c4f1c 100644
--- a/arch/x86/kernel/early-quirks.c
+++ b/arch/x86/kernel/early-quirks.c
@@ -526,6 +526,7 @@ static const struct pci_device_id intel_early_ids[] 
__initconst = {
INTEL_SKL_IDS(_early_ops),
INTEL_BXT_IDS(_early_ops),
INTEL_KBL_IDS(_early_ops),
+   INTEL_GLK_IDS(_early_ops),
 };
 
 static void __init
-- 
2.7.4