[Intel-gfx] [PATCH 02/22] drm/i915/icl: Enable Sampler DFR

Oscar Mateo oscar.mateo at intel.com
Fri Apr 20 20:52:24 UTC 2018



On 04/20/2018 01:48 PM, Rodrigo Vivi wrote:
> On Fri, Apr 20, 2018 at 01:33:46PM -0700, Oscar Mateo wrote:
>> Sampler Dynamic Frequency Rebalancing (DFR) aims to reduce Sampler
>> power by dynamically changing its clock frequency in low-throughput
>> conditions. This patches enables it by default on Gen11.
>>
>> v2: Wrong operation to clear the bit (Praveen)
>> v3: Rebased on top of the WA refactoring
>>
>> Cc: Praveen Paneri <praveen.paneri at intel.com>
>> Cc: Mika Kuoppala <mika.kuoppala at linux.intel.com>
>> Signed-off-by: Oscar Mateo <oscar.mateo at intel.com>
>> Reviewed-by: Sagar Arun Kamble <sagar.a.kamble at intel.com>
>> ---
>>   drivers/gpu/drm/i915/i915_reg.h          | 3 +++
>>   drivers/gpu/drm/i915/intel_workarounds.c | 4 ++++
>>   2 files changed, 7 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
>> index f2ee225..4b7e6bc 100644
>> --- a/drivers/gpu/drm/i915/i915_reg.h
>> +++ b/drivers/gpu/drm/i915/i915_reg.h
>> @@ -8218,6 +8218,9 @@ enum {
>>   #define GEN8_GARBCNTL                   _MMIO(0xB004)
>>   #define   GEN9_GAPS_TSV_CREDIT_DISABLE  (1<<7)
>>   
>> +#define GEN10_DFR_RATIO_EN_AND_CHICKEN	_MMIO(0x9550)
>> +#define   DFR_DISABLE			(1 << 9)
>> +
>>   /* IVYBRIDGE DPF */
>>   #define GEN7_L3CDERRST1(slice)		_MMIO(0xB008 + (slice) * 0x200) /* L3CD Error Status 1 */
>>   #define   GEN7_L3CDERRST1_ROW_MASK	(0x7ff<<14)
>> diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c
>> index 3f00623..60a5b1d 100644
>> --- a/drivers/gpu/drm/i915/intel_workarounds.c
>> +++ b/drivers/gpu/drm/i915/intel_workarounds.c
>> @@ -692,6 +692,10 @@ static void icl_gt_workarounds_apply(struct drm_i915_private *dev_priv)
>>   	I915_WRITE(_3D_CHICKEN3,
>>   		   _MASKED_BIT_ENABLE(_3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE));
>>   
>> +	/* This is not an Wa. Enable to reduce Sampler power */
> First of all it is strange that a feature is under a chicken bit,

There is a precedent: _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE

> but if it is not an wa, but a PM feature, shouldn't we add it to some sort of
> init clock gating function, or a new specific function called when it makes sense?

I'm open to suggestions

>> +	I915_WRITE(GEN10_DFR_RATIO_EN_AND_CHICKEN,
>> +		   (I915_READ(GEN10_DFR_RATIO_EN_AND_CHICKEN) & ~DFR_DISABLE));
>> +
>>   	/* WaInPlaceDecompressionHang:icl */
>>   	I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, (I915_READ(GEN9_GAMT_ECO_REG_RW_IA) |
>>   					     GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS));
>> -- 
>> 1.9.1
>>
>> _______________________________________________
>> Intel-gfx mailing list
>> Intel-gfx at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/intel-gfx



More information about the Intel-gfx mailing list