[PATCH 2/2] drm/i915/gt: Set default CCS mode '1'

John Harrison john.c.harrison at intel.com
Thu Feb 15 22:55:25 UTC 2024


On 2/15/2024 14:34, Andi Shyti wrote:
> Hi John,
>
> On Thu, Feb 15, 2024 at 01:23:24PM -0800, John Harrison wrote:
>> On 2/15/2024 05:59, Andi Shyti wrote:
>>> Since CCS automatic load balancing is disabled, we will impose a
>>> fixed balancing policy that involves setting all the CCS engines
>>> to work together on the same load.
>>>
>>> Simultaneously, the user will see only 1 CCS rather than the
>>> actual number. As of now, this change affects only DG2.
>> These two paragraphs are mutually exclusive. You can't have four CCS engines
>> 'working together' if only one engine exists. I think you are meaning that
>> we only export 1 CCS engine and that single engine is configured to control
>> all the EUs. As opposed to running in 4 CCS engine mode where the EUs are
>> (dynamically or statically) divided amongst those four engines.
> The balancing is done statically. The dynamic balancing is
> disabled in patch 1.
>
> The 2 or 4 CCS engines will share the same workload.
But they don't.

In i915, we use 'engine' to refer to a command streamer and all the 
associated hardware. This is distinct from the EUs which sit behind and 
can be driven by one or more command streamers. Saying that multiple 
engines are sharing a workload implies that you are submitting the 
context to multiple command streamers in parallel. I.e. a similar 
process to media frame split where they have a set of LRCA contexts 
bound together which are submitted in parallel to two or more video 
decode engines (VCS0, VCS1, etc.). That is not what is happening here.

Here, you are submitting a single context with a singe ring buffer to a 
single engine - CCS0. That engine is configured to own all EUs. Which 
actually means that submitting a compute task to another CCS engine will 
achieve nothing because there are no EUs available to those other 
engines. They will simply hang when waiting for the walker instruction 
to complete.

>
> Because the user won't be able anymore to select the CCS engine
> he wants to use, he will see only one CCS.
>
> I think we are saying the same thing using different words :)
But words are important.

John.

> I can try in v2 to reword the commit better.
>
> Thanks for looking into this.
> Andi
>
>> John.
>>
>>> Fixes: d2eae8e98d59 ("drm/i915/dg2: Drop force_probe requirement")
>>> Signed-off-by: Andi Shyti <andi.shyti at linux.intel.com>
>>> Cc: Chris Wilson <chris.p.wilson at linux.intel.com>
>>> Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
>>> Cc: Matt Roper <matthew.d.roper at intel.com>
>>> Cc: <stable at vger.kernel.org> # v6.2+
>>> ---
>>>    drivers/gpu/drm/i915/gt/intel_gt.c      | 11 +++++++++++
>>>    drivers/gpu/drm/i915/gt/intel_gt_regs.h |  2 ++
>>>    drivers/gpu/drm/i915/i915_drv.h         | 17 +++++++++++++++++
>>>    drivers/gpu/drm/i915/i915_query.c       |  5 +++--
>>>    4 files changed, 33 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c
>>> index a425db5ed3a2..e19df4ef47f6 100644
>>> --- a/drivers/gpu/drm/i915/gt/intel_gt.c
>>> +++ b/drivers/gpu/drm/i915/gt/intel_gt.c
>>> @@ -168,6 +168,14 @@ static void init_unused_rings(struct intel_gt *gt)
>>>    	}
>>>    }
>>> +static void intel_gt_apply_ccs_mode(struct intel_gt *gt)
>>> +{
>>> +	if (!IS_DG2(gt->i915))
>>> +		return;
>>> +
>>> +	intel_uncore_write(gt->uncore, XEHP_CCS_MODE, 0);
>>> +}
>>> +
>>>    int intel_gt_init_hw(struct intel_gt *gt)
>>>    {
>>>    	struct drm_i915_private *i915 = gt->i915;
>>> @@ -195,6 +203,9 @@ int intel_gt_init_hw(struct intel_gt *gt)
>>>    	intel_gt_init_swizzling(gt);
>>> +	/* Configure CCS mode */
>>> +	intel_gt_apply_ccs_mode(gt);
>>> +
>>>    	/*
>>>    	 * At least 830 can leave some of the unused rings
>>>    	 * "active" (ie. head != tail) after resume which
>>> diff --git a/drivers/gpu/drm/i915/gt/intel_gt_regs.h b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
>>> index cf709f6c05ae..c148113770ea 100644
>>> --- a/drivers/gpu/drm/i915/gt/intel_gt_regs.h
>>> +++ b/drivers/gpu/drm/i915/gt/intel_gt_regs.h
>>> @@ -1605,6 +1605,8 @@
>>>    #define   GEN12_VOLTAGE_MASK			REG_GENMASK(10, 0)
>>>    #define   GEN12_CAGF_MASK			REG_GENMASK(19, 11)
>>> +#define XEHP_CCS_MODE                          _MMIO(0x14804)
>>> +
>>>    #define GEN11_GT_INTR_DW(x)			_MMIO(0x190018 + ((x) * 4))
>>>    #define   GEN11_CSME				(31)
>>>    #define   GEN12_HECI_2				(30)
>>> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
>>> index e81b3b2858ac..0853ffd3cb8d 100644
>>> --- a/drivers/gpu/drm/i915/i915_drv.h
>>> +++ b/drivers/gpu/drm/i915/i915_drv.h
>>> @@ -396,6 +396,23 @@ static inline struct intel_gt *to_gt(const struct drm_i915_private *i915)
>>>    	     (engine__); \
>>>    	     (engine__) = rb_to_uabi_engine(rb_next(&(engine__)->uabi_node)))
>>> +/*
>>> + * Exclude unavailable engines.
>>> + *
>>> + * Only the first CCS engine is utilized due to the disabling of CCS auto load
>>> + * balancing. As a result, all CCS engines operate collectively, functioning
>>> + * essentially as a single CCS engine, hence the count of active CCS engines is
>>> + * considered '1'.
>>> + * Currently, this applies to platforms with more than one CCS engine,
>>> + * specifically DG2.
>>> + */
>>> +#define for_each_available_uabi_engine(engine__, i915__) \
>>> +	for_each_uabi_engine(engine__, i915__) \
>>> +		if ((IS_DG2(i915__)) && \
>>> +		    ((engine__)->uabi_class == I915_ENGINE_CLASS_COMPUTE) && \
>>> +		    ((engine__)->uabi_instance)) { } \
>>> +		else
>>> +
>>>    #define INTEL_INFO(i915)	((i915)->__info)
>>>    #define RUNTIME_INFO(i915)	(&(i915)->__runtime)
>>>    #define DRIVER_CAPS(i915)	(&(i915)->caps)
>>> diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c
>>> index fa3e937ed3f5..2d41bda626a6 100644
>>> --- a/drivers/gpu/drm/i915/i915_query.c
>>> +++ b/drivers/gpu/drm/i915/i915_query.c
>>> @@ -124,6 +124,7 @@ static int query_geometry_subslices(struct drm_i915_private *i915,
>>>    	return fill_topology_info(sseu, query_item, sseu->geometry_subslice_mask);
>>>    }
>>> +
>>>    static int
>>>    query_engine_info(struct drm_i915_private *i915,
>>>    		  struct drm_i915_query_item *query_item)
>>> @@ -140,7 +141,7 @@ query_engine_info(struct drm_i915_private *i915,
>>>    	if (query_item->flags)
>>>    		return -EINVAL;
>>> -	for_each_uabi_engine(engine, i915)
>>> +	for_each_available_uabi_engine(engine, i915)
>>>    		num_uabi_engines++;
>>>    	len = struct_size(query_ptr, engines, num_uabi_engines);
>>> @@ -155,7 +156,7 @@ query_engine_info(struct drm_i915_private *i915,
>>>    	info_ptr = &query_ptr->engines[0];
>>> -	for_each_uabi_engine(engine, i915) {
>>> +	for_each_available_uabi_engine(engine, i915) {
>>>    		info.engine.engine_class = engine->uabi_class;
>>>    		info.engine.engine_instance = engine->uabi_instance;
>>>    		info.flags = I915_ENGINE_INFO_HAS_LOGICAL_INSTANCE;



More information about the dri-devel mailing list