[RFC] drm/i915: store all mmio bases in intel_engines
diff mbox

Message ID 20180307194515.17135-1-daniele.ceraolospurio@intel.com
State New
Headers show

Commit Message

Daniele Ceraolo Spurio March 7, 2018, 7:45 p.m. UTC
The mmio bases we're currently storing in the intel_engines array are
only valid for a subset of gens, so we need to ignore them and use
different values in some cases. Instead of doing that, we can have a
table of [starting gen, mmio base] pairs for each engine in
intel_engines and select the correct one based on the gen we're running
on in a consistent way.

Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
---
 drivers/gpu/drm/i915/intel_engine_cs.c  | 77 +++++++++++++++++++++------------
 drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
 2 files changed, 49 insertions(+), 29 deletions(-)

Comments

Chris Wilson March 7, 2018, 7:56 p.m. UTC | #1
Quoting Daniele Ceraolo Spurio (2018-03-07 19:45:15)
> The mmio bases we're currently storing in the intel_engines array are
> only valid for a subset of gens, so we need to ignore them and use
> different values in some cases. Instead of doing that, we can have a
> table of [starting gen, mmio base] pairs for each engine in
> intel_engines and select the correct one based on the gen we're running
> on in a consistent way.
> 
> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
> ---
>  drivers/gpu/drm/i915/intel_engine_cs.c  | 77 +++++++++++++++++++++------------
>  drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>  2 files changed, 49 insertions(+), 29 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
> index 4ba139c27fba..1dd92cac8d67 100644
> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
> @@ -81,12 +81,16 @@ static const struct engine_class_info intel_engine_classes[] = {
>         },
>  };
>  
> +#define MAX_MMIO_BASES 3
>  struct engine_info {
>         unsigned int hw_id;
>         unsigned int uabi_id;
>         u8 class;
>         u8 instance;
> -       u32 mmio_base;
> +       struct engine_mmio_base {
> +               u32 gen : 8;
> +               u32 base : 24;
> +       } mmio_bases[MAX_MMIO_BASES];

Needs a note to mention the array must be in reverse gen order.

I would even add a selftest just to verify the arrays.

>         unsigned irq_shift;
>  };
>  
> @@ -96,7 +100,9 @@ static const struct engine_info intel_engines[] = {
>                 .uabi_id = I915_EXEC_RENDER,
>                 .class = RENDER_CLASS,
>                 .instance = 0,
> -               .mmio_base = RENDER_RING_BASE,
> +               .mmio_bases = {
> +                       { .gen = 1, .base = RENDER_RING_BASE }

Even gen0 (i740) has the render ring :)

> +static u32 __engine_mmio_base(struct drm_i915_private *i915,
> +                             const struct engine_mmio_base* bases)
> +{
> +       int i;
> +
> +       for (i = 0; i < MAX_MMIO_BASES; i++)
> +               if (INTEL_GEN(i915) >= bases[i].gen)
> +                       break;
> +
> +       GEM_BUG_ON(i == MAX_MMIO_BASES);
> +       GEM_BUG_ON(!bases[i].base);
> +
> +       return bases[i].base;
> +}
> +
>  static int
>  intel_engine_setup(struct drm_i915_private *dev_priv,
>                    enum intel_engine_id id)
> @@ -257,25 +296,7 @@ intel_engine_setup(struct drm_i915_private *dev_priv,
>                          class_info->name, info->instance) >=
>                 sizeof(engine->name));
>         engine->hw_id = engine->guc_id = info->hw_id;
> -       if (INTEL_GEN(dev_priv) >= 11) {
> -               switch (engine->id) {
> -               case VCS:
> -                       engine->mmio_base = GEN11_BSD_RING_BASE;
> -                       break;
> -               case VCS2:
> -                       engine->mmio_base = GEN11_BSD2_RING_BASE;
> -                       break;
> -               case VECS:
> -                       engine->mmio_base = GEN11_VEBOX_RING_BASE;
> -                       break;
> -               default:
> -                       /* take the original value for all other engines  */
> -                       engine->mmio_base = info->mmio_base;
> -                       break;
> -               }
> -       } else {
> -               engine->mmio_base = info->mmio_base;
> -       }
> +       engine->mmio_base = __engine_mmio_base(dev_priv, info->mmio_bases);
>         engine->irq_shift = info->irq_shift;
>         engine->class = info->class;
>         engine->instance = info->instance;

Very neat.
-Chris
Tvrtko Ursulin March 8, 2018, 9:31 a.m. UTC | #2
On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
> The mmio bases we're currently storing in the intel_engines array are
> only valid for a subset of gens, so we need to ignore them and use
> different values in some cases. Instead of doing that, we can have a
> table of [starting gen, mmio base] pairs for each engine in
> intel_engines and select the correct one based on the gen we're running
> on in a consistent way.
> 
> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
> ---
>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 +++++++++++++++++++++------------
>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>   2 files changed, 49 insertions(+), 29 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
> index 4ba139c27fba..1dd92cac8d67 100644
> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
> @@ -81,12 +81,16 @@ static const struct engine_class_info intel_engine_classes[] = {
>   	},
>   };
>   
> +#define MAX_MMIO_BASES 3
>   struct engine_info {
>   	unsigned int hw_id;
>   	unsigned int uabi_id;
>   	u8 class;
>   	u8 instance;
> -	u32 mmio_base;
> +	struct engine_mmio_base {
> +		u32 gen : 8;
> +		u32 base : 24;
> +	} mmio_bases[MAX_MMIO_BASES];
>   	unsigned irq_shift;
>   };

It's not bad, but I am just wondering if it is too complicated versus 
simply duplicating the tables.

Duplicated tables would certainly be much less code and complexity, but 
what about the size of pure data?

In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
NUM_ENGINES equals 240 bytes for intel_engines[] array with this scheme.

Separate tables per gens would be:

sizeof(struct engine_info) is 18 bytes.

For < gen6 we'd need 3 engines * 18 = 54
< gen11 = 5 engines * 18 = 90
 >= gen11 = 8 engines * 18 = 144

54 + 90 + 144 = 288 bytes

So a little bit bigger. Hm. Plus we would need to refactor so 
intel_engines[] is not indexed by engine->id but is contiguous array 
with engine->id in the elements. Code to lookup the compact array should 
be simpler than combined new code from this patch, especially if you add 
the test as Chris suggested. So separate engine info arrays would win I 
think overall - when considering size of text + size of data + size of 
source code.

But on the other hand your solution might be more future proof. So I 
don't know. Use the crystal ball to decide? :)

Regards,

Tvrtko


>   
> @@ -96,7 +100,9 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_RENDER,
>   		.class = RENDER_CLASS,
>   		.instance = 0,
> -		.mmio_base = RENDER_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 1, .base = RENDER_RING_BASE }
> +		},
>   		.irq_shift = GEN8_RCS_IRQ_SHIFT,
>   	},
>   	[BCS] = {
> @@ -104,7 +110,9 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_BLT,
>   		.class = COPY_ENGINE_CLASS,
>   		.instance = 0,
> -		.mmio_base = BLT_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 6, .base = BLT_RING_BASE }
> +		},
>   		.irq_shift = GEN8_BCS_IRQ_SHIFT,
>   	},
>   	[VCS] = {
> @@ -112,7 +120,11 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_BSD,
>   		.class = VIDEO_DECODE_CLASS,
>   		.instance = 0,
> -		.mmio_base = GEN6_BSD_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 11, .base = GEN11_BSD_RING_BASE },
> +			{ .gen = 6, .base = GEN6_BSD_RING_BASE },
> +			{ .gen = 4, .base = BSD_RING_BASE }
> +		},
>   		.irq_shift = GEN8_VCS1_IRQ_SHIFT,
>   	},
>   	[VCS2] = {
> @@ -120,7 +132,10 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_BSD,
>   		.class = VIDEO_DECODE_CLASS,
>   		.instance = 1,
> -		.mmio_base = GEN8_BSD2_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 11, .base = GEN11_BSD2_RING_BASE },
> +			{ .gen = 8, .base = GEN8_BSD2_RING_BASE }
> +		},
>   		.irq_shift = GEN8_VCS2_IRQ_SHIFT,
>   	},
>   	[VCS3] = {
> @@ -128,7 +143,9 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_BSD,
>   		.class = VIDEO_DECODE_CLASS,
>   		.instance = 2,
> -		.mmio_base = GEN11_BSD3_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 11, .base = GEN11_BSD3_RING_BASE }
> +		},
>   		.irq_shift = 0, /* not used */
>   	},
>   	[VCS4] = {
> @@ -136,7 +153,9 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_BSD,
>   		.class = VIDEO_DECODE_CLASS,
>   		.instance = 3,
> -		.mmio_base = GEN11_BSD4_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 11, .base = GEN11_BSD4_RING_BASE }
> +		},
>   		.irq_shift = 0, /* not used */
>   	},
>   	[VECS] = {
> @@ -144,7 +163,10 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_VEBOX,
>   		.class = VIDEO_ENHANCEMENT_CLASS,
>   		.instance = 0,
> -		.mmio_base = VEBOX_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 11, .base = GEN11_VEBOX_RING_BASE },
> +			{ .gen = 7, .base = VEBOX_RING_BASE }
> +		},
>   		.irq_shift = GEN8_VECS_IRQ_SHIFT,
>   	},
>   	[VECS2] = {
> @@ -152,7 +174,9 @@ static const struct engine_info intel_engines[] = {
>   		.uabi_id = I915_EXEC_VEBOX,
>   		.class = VIDEO_ENHANCEMENT_CLASS,
>   		.instance = 1,
> -		.mmio_base = GEN11_VEBOX2_RING_BASE,
> +		.mmio_bases = {
> +			{ .gen = 11, .base = GEN11_VEBOX2_RING_BASE }
> +		},
>   		.irq_shift = 0, /* not used */
>   	},
>   };
> @@ -223,6 +247,21 @@ __intel_engine_context_size(struct drm_i915_private *dev_priv, u8 class)
>   	}
>   }
>   
> +static u32 __engine_mmio_base(struct drm_i915_private *i915,
> +			      const struct engine_mmio_base* bases)
> +{
> +	int i;
> +
> +	for (i = 0; i < MAX_MMIO_BASES; i++)
> +		if (INTEL_GEN(i915) >= bases[i].gen)
> +			break;
> +
> +	GEM_BUG_ON(i == MAX_MMIO_BASES);
> +	GEM_BUG_ON(!bases[i].base);
> +
> +	return bases[i].base;
> +}
> +
>   static int
>   intel_engine_setup(struct drm_i915_private *dev_priv,
>   		   enum intel_engine_id id)
> @@ -257,25 +296,7 @@ intel_engine_setup(struct drm_i915_private *dev_priv,
>   			 class_info->name, info->instance) >=
>   		sizeof(engine->name));
>   	engine->hw_id = engine->guc_id = info->hw_id;
> -	if (INTEL_GEN(dev_priv) >= 11) {
> -		switch (engine->id) {
> -		case VCS:
> -			engine->mmio_base = GEN11_BSD_RING_BASE;
> -			break;
> -		case VCS2:
> -			engine->mmio_base = GEN11_BSD2_RING_BASE;
> -			break;
> -		case VECS:
> -			engine->mmio_base = GEN11_VEBOX_RING_BASE;
> -			break;
> -		default:
> -			/* take the original value for all other engines  */
> -			engine->mmio_base = info->mmio_base;
> -			break;
> -		}
> -	} else {
> -		engine->mmio_base = info->mmio_base;
> -	}
> +	engine->mmio_base = __engine_mmio_base(dev_priv, info->mmio_bases);
>   	engine->irq_shift = info->irq_shift;
>   	engine->class = info->class;
>   	engine->instance = info->instance;
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 1d599524a759..2e4408477ab5 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -2079,7 +2079,6 @@ int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine)
>   		engine->emit_flush = gen6_bsd_ring_flush;
>   		engine->irq_enable_mask = GT_BSD_USER_INTERRUPT;
>   	} else {
> -		engine->mmio_base = BSD_RING_BASE;
>   		engine->emit_flush = bsd_ring_flush;
>   		if (IS_GEN5(dev_priv))
>   			engine->irq_enable_mask = ILK_BSD_USER_INTERRUPT;
>
Daniele Ceraolo Spurio March 8, 2018, 6:46 p.m. UTC | #3
On 08/03/18 01:31, Tvrtko Ursulin wrote:
> 
> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>> The mmio bases we're currently storing in the intel_engines array are
>> only valid for a subset of gens, so we need to ignore them and use
>> different values in some cases. Instead of doing that, we can have a
>> table of [starting gen, mmio base] pairs for each engine in
>> intel_engines and select the correct one based on the gen we're running
>> on in a consistent way.
>>
>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
>> ---
>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>> +++++++++++++++++++++------------
>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>> index 4ba139c27fba..1dd92cac8d67 100644
>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>> intel_engine_classes[] = {
>>       },
>>   };
>> +#define MAX_MMIO_BASES 3
>>   struct engine_info {
>>       unsigned int hw_id;
>>       unsigned int uabi_id;
>>       u8 class;
>>       u8 instance;
>> -    u32 mmio_base;
>> +    struct engine_mmio_base {
>> +        u32 gen : 8;
>> +        u32 base : 24;
>> +    } mmio_bases[MAX_MMIO_BASES];
>>       unsigned irq_shift;
>>   };
> 
> It's not bad, but I am just wondering if it is too complicated versus 
> simply duplicating the tables.
> 
> Duplicated tables would certainly be much less code and complexity, but 
> what about the size of pure data?
> 
> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
> sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
> NUM_ENGINES equals 240 bytes for intel_engines[] array with this scheme.
>

we remove a u32 (the old mmio base) so we only grow 8 per engine, but 
the compiler rounds up to a multiple of u32 so 28 per engine, for a 
total of 224.

> Separate tables per gens would be:
> 
> sizeof(struct engine_info) is 18 bytes.
> 
> For < gen6 we'd need 3 engines * 18 = 54
> < gen11 = 5 engines * 18 = 90
>  >= gen11 = 8 engines * 18 = 144
> 
> 54 + 90 + 144 = 288 bytes
> 
> So a little bit bigger. Hm. Plus we would need to refactor so 
> intel_engines[] is not indexed by engine->id but is contiguous array 
> with engine->id in the elements. Code to lookup the compact array should 
> be simpler than combined new code from this patch, especially if you add 
> the test as Chris suggested. So separate engine info arrays would win I 
> think overall - when considering size of text + size of data + size of 
> source code.
> 

Not sure. I would exclude the selftest, which is usually not compiled 
for released kernels, which leads to:

add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
Function                                     old     new   delta
intel_engines                                160     224     +64
__func__                                   13891   13910     +19
intel_engines_init_mmio                     1247    1264     +17
intel_init_bsd_ring_buffer                   142     135      -7
Total: Before=1479626, After=1479719, chg +0.01%

Total growth is 93, which is less then your estimation for the growth 
introduced by replicating the table.

> But on the other hand your solution might be more future proof. So I 
> don't know. Use the crystal ball to decide? :)
> 

I think we should expect that the total engine count could grow with 
future gens. In this case to me adding a new entry to a unified table 
seems much cleaner (and uses less data) than adding a completely new 
table each time.

Daniele

> Regards,
> 
> Tvrtko
> 
> 
>> @@ -96,7 +100,9 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_RENDER,
>>           .class = RENDER_CLASS,
>>           .instance = 0,
>> -        .mmio_base = RENDER_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 1, .base = RENDER_RING_BASE }
>> +        },
>>           .irq_shift = GEN8_RCS_IRQ_SHIFT,
>>       },
>>       [BCS] = {
>> @@ -104,7 +110,9 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_BLT,
>>           .class = COPY_ENGINE_CLASS,
>>           .instance = 0,
>> -        .mmio_base = BLT_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 6, .base = BLT_RING_BASE }
>> +        },
>>           .irq_shift = GEN8_BCS_IRQ_SHIFT,
>>       },
>>       [VCS] = {
>> @@ -112,7 +120,11 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_BSD,
>>           .class = VIDEO_DECODE_CLASS,
>>           .instance = 0,
>> -        .mmio_base = GEN6_BSD_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 11, .base = GEN11_BSD_RING_BASE },
>> +            { .gen = 6, .base = GEN6_BSD_RING_BASE },
>> +            { .gen = 4, .base = BSD_RING_BASE }
>> +        },
>>           .irq_shift = GEN8_VCS1_IRQ_SHIFT,
>>       },
>>       [VCS2] = {
>> @@ -120,7 +132,10 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_BSD,
>>           .class = VIDEO_DECODE_CLASS,
>>           .instance = 1,
>> -        .mmio_base = GEN8_BSD2_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 11, .base = GEN11_BSD2_RING_BASE },
>> +            { .gen = 8, .base = GEN8_BSD2_RING_BASE }
>> +        },
>>           .irq_shift = GEN8_VCS2_IRQ_SHIFT,
>>       },
>>       [VCS3] = {
>> @@ -128,7 +143,9 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_BSD,
>>           .class = VIDEO_DECODE_CLASS,
>>           .instance = 2,
>> -        .mmio_base = GEN11_BSD3_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 11, .base = GEN11_BSD3_RING_BASE }
>> +        },
>>           .irq_shift = 0, /* not used */
>>       },
>>       [VCS4] = {
>> @@ -136,7 +153,9 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_BSD,
>>           .class = VIDEO_DECODE_CLASS,
>>           .instance = 3,
>> -        .mmio_base = GEN11_BSD4_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 11, .base = GEN11_BSD4_RING_BASE }
>> +        },
>>           .irq_shift = 0, /* not used */
>>       },
>>       [VECS] = {
>> @@ -144,7 +163,10 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_VEBOX,
>>           .class = VIDEO_ENHANCEMENT_CLASS,
>>           .instance = 0,
>> -        .mmio_base = VEBOX_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 11, .base = GEN11_VEBOX_RING_BASE },
>> +            { .gen = 7, .base = VEBOX_RING_BASE }
>> +        },
>>           .irq_shift = GEN8_VECS_IRQ_SHIFT,
>>       },
>>       [VECS2] = {
>> @@ -152,7 +174,9 @@ static const struct engine_info intel_engines[] = {
>>           .uabi_id = I915_EXEC_VEBOX,
>>           .class = VIDEO_ENHANCEMENT_CLASS,
>>           .instance = 1,
>> -        .mmio_base = GEN11_VEBOX2_RING_BASE,
>> +        .mmio_bases = {
>> +            { .gen = 11, .base = GEN11_VEBOX2_RING_BASE }
>> +        },
>>           .irq_shift = 0, /* not used */
>>       },
>>   };
>> @@ -223,6 +247,21 @@ __intel_engine_context_size(struct 
>> drm_i915_private *dev_priv, u8 class)
>>       }
>>   }
>> +static u32 __engine_mmio_base(struct drm_i915_private *i915,
>> +                  const struct engine_mmio_base* bases)
>> +{
>> +    int i;
>> +
>> +    for (i = 0; i < MAX_MMIO_BASES; i++)
>> +        if (INTEL_GEN(i915) >= bases[i].gen)
>> +            break;
>> +
>> +    GEM_BUG_ON(i == MAX_MMIO_BASES);
>> +    GEM_BUG_ON(!bases[i].base);
>> +
>> +    return bases[i].base;
>> +}
>> +
>>   static int
>>   intel_engine_setup(struct drm_i915_private *dev_priv,
>>              enum intel_engine_id id)
>> @@ -257,25 +296,7 @@ intel_engine_setup(struct drm_i915_private 
>> *dev_priv,
>>                class_info->name, info->instance) >=
>>           sizeof(engine->name));
>>       engine->hw_id = engine->guc_id = info->hw_id;
>> -    if (INTEL_GEN(dev_priv) >= 11) {
>> -        switch (engine->id) {
>> -        case VCS:
>> -            engine->mmio_base = GEN11_BSD_RING_BASE;
>> -            break;
>> -        case VCS2:
>> -            engine->mmio_base = GEN11_BSD2_RING_BASE;
>> -            break;
>> -        case VECS:
>> -            engine->mmio_base = GEN11_VEBOX_RING_BASE;
>> -            break;
>> -        default:
>> -            /* take the original value for all other engines  */
>> -            engine->mmio_base = info->mmio_base;
>> -            break;
>> -        }
>> -    } else {
>> -        engine->mmio_base = info->mmio_base;
>> -    }
>> +    engine->mmio_base = __engine_mmio_base(dev_priv, info->mmio_bases);
>>       engine->irq_shift = info->irq_shift;
>>       engine->class = info->class;
>>       engine->instance = info->instance;
>> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c 
>> b/drivers/gpu/drm/i915/intel_ringbuffer.c
>> index 1d599524a759..2e4408477ab5 100644
>> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
>> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
>> @@ -2079,7 +2079,6 @@ int intel_init_bsd_ring_buffer(struct 
>> intel_engine_cs *engine)
>>           engine->emit_flush = gen6_bsd_ring_flush;
>>           engine->irq_enable_mask = GT_BSD_USER_INTERRUPT;
>>       } else {
>> -        engine->mmio_base = BSD_RING_BASE;
>>           engine->emit_flush = bsd_ring_flush;
>>           if (IS_GEN5(dev_priv))
>>               engine->irq_enable_mask = ILK_BSD_USER_INTERRUPT;
>>
Tvrtko Ursulin March 9, 2018, 9:53 a.m. UTC | #4
On 08/03/2018 18:46, Daniele Ceraolo Spurio wrote:
> On 08/03/18 01:31, Tvrtko Ursulin wrote:
>>
>> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>>> The mmio bases we're currently storing in the intel_engines array are
>>> only valid for a subset of gens, so we need to ignore them and use
>>> different values in some cases. Instead of doing that, we can have a
>>> table of [starting gen, mmio base] pairs for each engine in
>>> intel_engines and select the correct one based on the gen we're running
>>> on in a consistent way.
>>>
>>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
>>> ---
>>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>>> +++++++++++++++++++++------------
>>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>>> index 4ba139c27fba..1dd92cac8d67 100644
>>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>>> intel_engine_classes[] = {
>>>       },
>>>   };
>>> +#define MAX_MMIO_BASES 3
>>>   struct engine_info {
>>>       unsigned int hw_id;
>>>       unsigned int uabi_id;
>>>       u8 class;
>>>       u8 instance;
>>> -    u32 mmio_base;
>>> +    struct engine_mmio_base {
>>> +        u32 gen : 8;
>>> +        u32 base : 24;
>>> +    } mmio_bases[MAX_MMIO_BASES];
>>>       unsigned irq_shift;
>>>   };
>>
>> It's not bad, but I am just wondering if it is too complicated versus 
>> simply duplicating the tables.
>>
>> Duplicated tables would certainly be much less code and complexity, 
>> but what about the size of pure data?
>>
>> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
>> sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
>> NUM_ENGINES equals 240 bytes for intel_engines[] array with this scheme.
>>
> 
> we remove a u32 (the old mmio base) so we only grow 8 per engine, but 
> the compiler rounds up to a multiple of u32 so 28 per engine, for a 
> total of 224.
> 
>> Separate tables per gens would be:
>>
>> sizeof(struct engine_info) is 18 bytes.
>>
>> For < gen6 we'd need 3 engines * 18 = 54
>> < gen11 = 5 engines * 18 = 90
>>  >= gen11 = 8 engines * 18 = 144
>>
>> 54 + 90 + 144 = 288 bytes
>>
>> So a little bit bigger. Hm. Plus we would need to refactor so 
>> intel_engines[] is not indexed by engine->id but is contiguous array 
>> with engine->id in the elements. Code to lookup the compact array 
>> should be simpler than combined new code from this patch, especially 
>> if you add the test as Chris suggested. So separate engine info arrays 
>> would win I think overall - when considering size of text + size of 
>> data + size of source code.
>>
> 
> Not sure. I would exclude the selftest, which is usually not compiled 
> for released kernels, which leads to:

Yes, but we cannot exclude its source since selftests for separate 
tables wouldn't be needed.

> add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
> Function                                     old     new   delta
> intel_engines                                160     224     +64
> __func__                                   13891   13910     +19
> intel_engines_init_mmio                     1247    1264     +17
> intel_init_bsd_ring_buffer                   142     135      -7
> Total: Before=1479626, After=1479719, chg +0.01%
> 
> Total growth is 93, which is less then your estimation for the growth 
> introduced by replicating the table.
> 
>> But on the other hand your solution might be more future proof. So I 
>> don't know. Use the crystal ball to decide? :)
>>
> 
> I think we should expect that the total engine count could grow with 
> future gens. In this case to me adding a new entry to a unified table 
> seems much cleaner (and uses less data) than adding a completely new 
> table each time.

Okay I was off in my estimates. But in general I was coming from the 
angle of thinking that every new mmio base difference in this scheme 
grows the size for all engines. So if just VCS grows one new base, 
impact is multiplied by NUM_ENGINES. But maybe separate tables are also 
not a solution. Perhaps pulling out mmio_base arrays to outside struct 
engine_info in another set of static tables, so they could be null 
terminated would be best of both worlds?

struct engine_mmio_base {
	u32 gen : 8;
	u32 base : 24;
};

static const struct engine_mmio_base vcs0_mmio_bases[] = {
	{ .gen = 11, .base = GEN11_BSD_RING_BASE },
	{ .gen = 6, .base = GEN6_BSD_RING_BASE },
	{ .gen = 4, .base = BSD_RING_BASE },
	{ },
};

And then in intel_engines array, for BSD:

    {
	...
	.mmio_bases = &vcs0_mmio_bases;
	...
    },

This way we limit data growth only to engines which change their mmio 
bases frequently.

Just an idea.

Regards,

Tvrtko
Daniele Ceraolo Spurio March 9, 2018, 6:47 p.m. UTC | #5
On 09/03/18 01:53, Tvrtko Ursulin wrote:
> 
> On 08/03/2018 18:46, Daniele Ceraolo Spurio wrote:
>> On 08/03/18 01:31, Tvrtko Ursulin wrote:
>>>
>>> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>>>> The mmio bases we're currently storing in the intel_engines array are
>>>> only valid for a subset of gens, so we need to ignore them and use
>>>> different values in some cases. Instead of doing that, we can have a
>>>> table of [starting gen, mmio base] pairs for each engine in
>>>> intel_engines and select the correct one based on the gen we're running
>>>> on in a consistent way.
>>>>
>>>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
>>>> ---
>>>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>>>> +++++++++++++++++++++------------
>>>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>>>
>>>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>>>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>> index 4ba139c27fba..1dd92cac8d67 100644
>>>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>>>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>>>> intel_engine_classes[] = {
>>>>       },
>>>>   };
>>>> +#define MAX_MMIO_BASES 3
>>>>   struct engine_info {
>>>>       unsigned int hw_id;
>>>>       unsigned int uabi_id;
>>>>       u8 class;
>>>>       u8 instance;
>>>> -    u32 mmio_base;
>>>> +    struct engine_mmio_base {
>>>> +        u32 gen : 8;
>>>> +        u32 base : 24;
>>>> +    } mmio_bases[MAX_MMIO_BASES];
>>>>       unsigned irq_shift;
>>>>   };
>>>
>>> It's not bad, but I am just wondering if it is too complicated versus 
>>> simply duplicating the tables.
>>>
>>> Duplicated tables would certainly be much less code and complexity, 
>>> but what about the size of pure data?
>>>
>>> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
>>> sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
>>> NUM_ENGINES equals 240 bytes for intel_engines[] array with this scheme.
>>>
>>
>> we remove a u32 (the old mmio base) so we only grow 8 per engine, but 
>> the compiler rounds up to a multiple of u32 so 28 per engine, for a 
>> total of 224.
>>
>>> Separate tables per gens would be:
>>>
>>> sizeof(struct engine_info) is 18 bytes.
>>>
>>> For < gen6 we'd need 3 engines * 18 = 54
>>> < gen11 = 5 engines * 18 = 90
>>>  >= gen11 = 8 engines * 18 = 144
>>>
>>> 54 + 90 + 144 = 288 bytes
>>>
>>> So a little bit bigger. Hm. Plus we would need to refactor so 
>>> intel_engines[] is not indexed by engine->id but is contiguous array 
>>> with engine->id in the elements. Code to lookup the compact array 
>>> should be simpler than combined new code from this patch, especially 
>>> if you add the test as Chris suggested. So separate engine info 
>>> arrays would win I think overall - when considering size of text + 
>>> size of data + size of source code.
>>>
>>
>> Not sure. I would exclude the selftest, which is usually not compiled 
>> for released kernels, which leads to:
> 
> Yes, but we cannot exclude its source since selftests for separate 
> tables wouldn't be needed.
> 
>> add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
>> Function                                     old     new   delta
>> intel_engines                                160     224     +64
>> __func__                                   13891   13910     +19
>> intel_engines_init_mmio                     1247    1264     +17
>> intel_init_bsd_ring_buffer                   142     135      -7
>> Total: Before=1479626, After=1479719, chg +0.01%
>>
>> Total growth is 93, which is less then your estimation for the growth 
>> introduced by replicating the table.
>>
>>> But on the other hand your solution might be more future proof. So I 
>>> don't know. Use the crystal ball to decide? :)
>>>
>>
>> I think we should expect that the total engine count could grow with 
>> future gens. In this case to me adding a new entry to a unified table 
>> seems much cleaner (and uses less data) than adding a completely new 
>> table each time.
> 
> Okay I was off in my estimates. But in general I was coming from the 
> angle of thinking that every new mmio base difference in this scheme 
> grows the size for all engines. So if just VCS grows one new base, 
> impact is multiplied by NUM_ENGINES. But maybe separate tables are also 
> not a solution. Perhaps pulling out mmio_base arrays to outside struct 
> engine_info in another set of static tables, so they could be null 
> terminated would be best of both worlds?
> 
> struct engine_mmio_base {
>      u32 gen : 8;
>      u32 base : 24;
> };
> 
> static const struct engine_mmio_base vcs0_mmio_bases[] = {
>      { .gen = 11, .base = GEN11_BSD_RING_BASE },
>      { .gen = 6, .base = GEN6_BSD_RING_BASE },
>      { .gen = 4, .base = BSD_RING_BASE },
>      { },
> };
> 
> And then in intel_engines array, for BSD:
> 
>     {
>      ...
>      .mmio_bases = &vcs0_mmio_bases;
>      ...
>     },
> 
> This way we limit data growth only to engines which change their mmio 
> bases frequently.
> 
> Just an idea.
> 

I gave this a try and the code actually grows:

add/remove: 8/0 grow/shrink: 2/0 up/down: 120/0 (120)
Function                                     old     new   delta
intel_engines                                224     256     +32
vcs0_mmio_bases                                -      16     +16
vecs1_mmio_bases                               -      12     +12
vecs0_mmio_bases                               -      12     +12
vcs1_mmio_bases                                -      12     +12
vcs3_mmio_bases                                -       8      +8
vcs2_mmio_bases                                -       8      +8
rcs_mmio_bases                                 -       8      +8
intel_engines_init_mmio                     1264    1272      +8
bcs_mmio_bases                                 -       4      +4
Total: Before=1479719, After=1479839, chg +0.01%

I have no idea what the compiler is doing to grow intel_engines, since 
by replacing and array of 3 u32s with a pointer I would expect a shrink 
of 4 bytes per-engine. Anyway, even without the compiler weirdness with 
this method we would grow the code size of at least 4 bytes per engine 
because we replace an array of 3 u32 (12B) with a pointer (8B) and an 
array of at 2 or more u32 (8B+). I guess we can reconsider if/when one 
engine reaches more than 3 mmio bases.

Thanks,
Daniele

> Regards,
> 
> Tvrtko
Tvrtko Ursulin March 9, 2018, 7:44 p.m. UTC | #6
On 09/03/2018 18:47, Daniele Ceraolo Spurio wrote:
> On 09/03/18 01:53, Tvrtko Ursulin wrote:
>>
>> On 08/03/2018 18:46, Daniele Ceraolo Spurio wrote:
>>> On 08/03/18 01:31, Tvrtko Ursulin wrote:
>>>>
>>>> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>>>>> The mmio bases we're currently storing in the intel_engines array are
>>>>> only valid for a subset of gens, so we need to ignore them and use
>>>>> different values in some cases. Instead of doing that, we can have a
>>>>> table of [starting gen, mmio base] pairs for each engine in
>>>>> intel_engines and select the correct one based on the gen we're 
>>>>> running
>>>>> on in a consistent way.
>>>>>
>>>>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>>> Signed-off-by: Daniele Ceraolo Spurio 
>>>>> <daniele.ceraolospurio@intel.com>
>>>>> ---
>>>>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>>>>> +++++++++++++++++++++------------
>>>>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>>>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>>>>
>>>>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>>>>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>> index 4ba139c27fba..1dd92cac8d67 100644
>>>>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>>>>> intel_engine_classes[] = {
>>>>>       },
>>>>>   };
>>>>> +#define MAX_MMIO_BASES 3
>>>>>   struct engine_info {
>>>>>       unsigned int hw_id;
>>>>>       unsigned int uabi_id;
>>>>>       u8 class;
>>>>>       u8 instance;
>>>>> -    u32 mmio_base;
>>>>> +    struct engine_mmio_base {
>>>>> +        u32 gen : 8;
>>>>> +        u32 base : 24;
>>>>> +    } mmio_bases[MAX_MMIO_BASES];
>>>>>       unsigned irq_shift;
>>>>>   };
>>>>
>>>> It's not bad, but I am just wondering if it is too complicated 
>>>> versus simply duplicating the tables.
>>>>
>>>> Duplicated tables would certainly be much less code and complexity, 
>>>> but what about the size of pure data?
>>>>
>>>> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
>>>> sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
>>>> NUM_ENGINES equals 240 bytes for intel_engines[] array with this 
>>>> scheme.
>>>>
>>>
>>> we remove a u32 (the old mmio base) so we only grow 8 per engine, but 
>>> the compiler rounds up to a multiple of u32 so 28 per engine, for a 
>>> total of 224.
>>>
>>>> Separate tables per gens would be:
>>>>
>>>> sizeof(struct engine_info) is 18 bytes.
>>>>
>>>> For < gen6 we'd need 3 engines * 18 = 54
>>>> < gen11 = 5 engines * 18 = 90
>>>>  >= gen11 = 8 engines * 18 = 144
>>>>
>>>> 54 + 90 + 144 = 288 bytes
>>>>
>>>> So a little bit bigger. Hm. Plus we would need to refactor so 
>>>> intel_engines[] is not indexed by engine->id but is contiguous array 
>>>> with engine->id in the elements. Code to lookup the compact array 
>>>> should be simpler than combined new code from this patch, especially 
>>>> if you add the test as Chris suggested. So separate engine info 
>>>> arrays would win I think overall - when considering size of text + 
>>>> size of data + size of source code.
>>>>
>>>
>>> Not sure. I would exclude the selftest, which is usually not compiled 
>>> for released kernels, which leads to:
>>
>> Yes, but we cannot exclude its source since selftests for separate 
>> tables wouldn't be needed.
>>
>>> add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
>>> Function                                     old     new   delta
>>> intel_engines                                160     224     +64
>>> __func__                                   13891   13910     +19
>>> intel_engines_init_mmio                     1247    1264     +17
>>> intel_init_bsd_ring_buffer                   142     135      -7
>>> Total: Before=1479626, After=1479719, chg +0.01%
>>>
>>> Total growth is 93, which is less then your estimation for the growth 
>>> introduced by replicating the table.
>>>
>>>> But on the other hand your solution might be more future proof. So I 
>>>> don't know. Use the crystal ball to decide? :)
>>>>
>>>
>>> I think we should expect that the total engine count could grow with 
>>> future gens. In this case to me adding a new entry to a unified table 
>>> seems much cleaner (and uses less data) than adding a completely new 
>>> table each time.
>>
>> Okay I was off in my estimates. But in general I was coming from the 
>> angle of thinking that every new mmio base difference in this scheme 
>> grows the size for all engines. So if just VCS grows one new base, 
>> impact is multiplied by NUM_ENGINES. But maybe separate tables are 
>> also not a solution. Perhaps pulling out mmio_base arrays to outside 
>> struct engine_info in another set of static tables, so they could be 
>> null terminated would be best of both worlds?
>>
>> struct engine_mmio_base {
>>      u32 gen : 8;
>>      u32 base : 24;
>> };
>>
>> static const struct engine_mmio_base vcs0_mmio_bases[] = {
>>      { .gen = 11, .base = GEN11_BSD_RING_BASE },
>>      { .gen = 6, .base = GEN6_BSD_RING_BASE },
>>      { .gen = 4, .base = BSD_RING_BASE },
>>      { },
>> };
>>
>> And then in intel_engines array, for BSD:
>>
>>     {
>>      ...
>>      .mmio_bases = &vcs0_mmio_bases;
>>      ...
>>     },
>>
>> This way we limit data growth only to engines which change their mmio 
>> bases frequently.
>>
>> Just an idea.
>>
> 
> I gave this a try and the code actually grows:
> 
> add/remove: 8/0 grow/shrink: 2/0 up/down: 120/0 (120)
> Function                                     old     new   delta
> intel_engines                                224     256     +32
> vcs0_mmio_bases                                -      16     +16
> vecs1_mmio_bases                               -      12     +12
> vecs0_mmio_bases                               -      12     +12
> vcs1_mmio_bases                                -      12     +12
> vcs3_mmio_bases                                -       8      +8
> vcs2_mmio_bases                                -       8      +8
> rcs_mmio_bases                                 -       8      +8
> intel_engines_init_mmio                     1264    1272      +8
> bcs_mmio_bases                                 -       4      +4
> Total: Before=1479719, After=1479839, chg +0.01%
> 
> I have no idea what the compiler is doing to grow intel_engines, since 
> by replacing and array of 3 u32s with a pointer I would expect a shrink 
> of 4 bytes per-engine. Anyway, even without the compiler weirdness with

It probably has to align the pointer to 8 bytes so that creates a hole. 
Moving mmio_base pointer higher up, either to the top or after two 
unsigned ints should work. Or packing the struct.

> this method we would grow the code size of at least 4 bytes per engine 
> because we replace an array of 3 u32 (12B) with a pointer (8B) and an 
> array of at 2 or more u32 (8B+). I guess we can reconsider if/when one 
> engine reaches more than 3 mmio bases.

Yeah it's fine. I was thinking that since you are almost there it makes 
sense to future proof it more, since no one will probably remember it 
later. But OK.

Regards,

Tvrtko
Tvrtko Ursulin March 12, 2018, 11:04 a.m. UTC | #7
On 09/03/2018 19:44, Tvrtko Ursulin wrote:
> 
> On 09/03/2018 18:47, Daniele Ceraolo Spurio wrote:
>> On 09/03/18 01:53, Tvrtko Ursulin wrote:
>>>
>>> On 08/03/2018 18:46, Daniele Ceraolo Spurio wrote:
>>>> On 08/03/18 01:31, Tvrtko Ursulin wrote:
>>>>>
>>>>> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>>>>>> The mmio bases we're currently storing in the intel_engines array are
>>>>>> only valid for a subset of gens, so we need to ignore them and use
>>>>>> different values in some cases. Instead of doing that, we can have a
>>>>>> table of [starting gen, mmio base] pairs for each engine in
>>>>>> intel_engines and select the correct one based on the gen we're 
>>>>>> running
>>>>>> on in a consistent way.
>>>>>>
>>>>>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>>>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>>>> Signed-off-by: Daniele Ceraolo Spurio 
>>>>>> <daniele.ceraolospurio@intel.com>
>>>>>> ---
>>>>>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>>>>>> +++++++++++++++++++++------------
>>>>>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>>>>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>>>>>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>> index 4ba139c27fba..1dd92cac8d67 100644
>>>>>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>>>>>> intel_engine_classes[] = {
>>>>>>       },
>>>>>>   };
>>>>>> +#define MAX_MMIO_BASES 3
>>>>>>   struct engine_info {
>>>>>>       unsigned int hw_id;
>>>>>>       unsigned int uabi_id;
>>>>>>       u8 class;
>>>>>>       u8 instance;
>>>>>> -    u32 mmio_base;
>>>>>> +    struct engine_mmio_base {
>>>>>> +        u32 gen : 8;
>>>>>> +        u32 base : 24;
>>>>>> +    } mmio_bases[MAX_MMIO_BASES];
>>>>>>       unsigned irq_shift;
>>>>>>   };
>>>>>
>>>>> It's not bad, but I am just wondering if it is too complicated 
>>>>> versus simply duplicating the tables.
>>>>>
>>>>> Duplicated tables would certainly be much less code and complexity, 
>>>>> but what about the size of pure data?
>>>>>
>>>>> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
>>>>> sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
>>>>> NUM_ENGINES equals 240 bytes for intel_engines[] array with this 
>>>>> scheme.
>>>>>
>>>>
>>>> we remove a u32 (the old mmio base) so we only grow 8 per engine, 
>>>> but the compiler rounds up to a multiple of u32 so 28 per engine, 
>>>> for a total of 224.
>>>>
>>>>> Separate tables per gens would be:
>>>>>
>>>>> sizeof(struct engine_info) is 18 bytes.
>>>>>
>>>>> For < gen6 we'd need 3 engines * 18 = 54
>>>>> < gen11 = 5 engines * 18 = 90
>>>>>  >= gen11 = 8 engines * 18 = 144
>>>>>
>>>>> 54 + 90 + 144 = 288 bytes
>>>>>
>>>>> So a little bit bigger. Hm. Plus we would need to refactor so 
>>>>> intel_engines[] is not indexed by engine->id but is contiguous 
>>>>> array with engine->id in the elements. Code to lookup the compact 
>>>>> array should be simpler than combined new code from this patch, 
>>>>> especially if you add the test as Chris suggested. So separate 
>>>>> engine info arrays would win I think overall - when considering 
>>>>> size of text + size of data + size of source code.
>>>>>
>>>>
>>>> Not sure. I would exclude the selftest, which is usually not 
>>>> compiled for released kernels, which leads to:
>>>
>>> Yes, but we cannot exclude its source since selftests for separate 
>>> tables wouldn't be needed.
>>>
>>>> add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
>>>> Function                                     old     new   delta
>>>> intel_engines                                160     224     +64
>>>> __func__                                   13891   13910     +19
>>>> intel_engines_init_mmio                     1247    1264     +17
>>>> intel_init_bsd_ring_buffer                   142     135      -7
>>>> Total: Before=1479626, After=1479719, chg +0.01%
>>>>
>>>> Total growth is 93, which is less then your estimation for the 
>>>> growth introduced by replicating the table.
>>>>
>>>>> But on the other hand your solution might be more future proof. So 
>>>>> I don't know. Use the crystal ball to decide? :)
>>>>>
>>>>
>>>> I think we should expect that the total engine count could grow with 
>>>> future gens. In this case to me adding a new entry to a unified 
>>>> table seems much cleaner (and uses less data) than adding a 
>>>> completely new table each time.
>>>
>>> Okay I was off in my estimates. But in general I was coming from the 
>>> angle of thinking that every new mmio base difference in this scheme 
>>> grows the size for all engines. So if just VCS grows one new base, 
>>> impact is multiplied by NUM_ENGINES. But maybe separate tables are 
>>> also not a solution. Perhaps pulling out mmio_base arrays to outside 
>>> struct engine_info in another set of static tables, so they could be 
>>> null terminated would be best of both worlds?
>>>
>>> struct engine_mmio_base {
>>>      u32 gen : 8;
>>>      u32 base : 24;
>>> };
>>>
>>> static const struct engine_mmio_base vcs0_mmio_bases[] = {
>>>      { .gen = 11, .base = GEN11_BSD_RING_BASE },
>>>      { .gen = 6, .base = GEN6_BSD_RING_BASE },
>>>      { .gen = 4, .base = BSD_RING_BASE },
>>>      { },
>>> };
>>>
>>> And then in intel_engines array, for BSD:
>>>
>>>     {
>>>      ...
>>>      .mmio_bases = &vcs0_mmio_bases;
>>>      ...
>>>     },
>>>
>>> This way we limit data growth only to engines which change their mmio 
>>> bases frequently.
>>>
>>> Just an idea.
>>>
>>
>> I gave this a try and the code actually grows:
>>
>> add/remove: 8/0 grow/shrink: 2/0 up/down: 120/0 (120)
>> Function                                     old     new   delta
>> intel_engines                                224     256     +32
>> vcs0_mmio_bases                                -      16     +16
>> vecs1_mmio_bases                               -      12     +12
>> vecs0_mmio_bases                               -      12     +12
>> vcs1_mmio_bases                                -      12     +12
>> vcs3_mmio_bases                                -       8      +8
>> vcs2_mmio_bases                                -       8      +8
>> rcs_mmio_bases                                 -       8      +8
>> intel_engines_init_mmio                     1264    1272      +8
>> bcs_mmio_bases                                 -       4      +4
>> Total: Before=1479719, After=1479839, chg +0.01%
>>
>> I have no idea what the compiler is doing to grow intel_engines, since 
>> by replacing and array of 3 u32s with a pointer I would expect a 
>> shrink of 4 bytes per-engine. Anyway, even without the compiler 
>> weirdness with
> 
> It probably has to align the pointer to 8 bytes so that creates a hole. 
> Moving mmio_base pointer higher up, either to the top or after two 
> unsigned ints should work. Or packing the struct.
> 
>> this method we would grow the code size of at least 4 bytes per engine 
>> because we replace an array of 3 u32 (12B) with a pointer (8B) and an 
>> array of at 2 or more u32 (8B+). I guess we can reconsider if/when one 
>> engine reaches more than 3 mmio bases.
> 
> Yeah it's fine. I was thinking that since you are almost there it makes 
> sense to future proof it more, since no one will probably remember it 
> later. But OK.

One idea to compact more, in addition to avoiding alignment holes, could 
be to store the engine_mmio_base directly in the engine_mmio_base 
pointer when it is a single entry, othewrise use a pointer to null 
terminated array. Actually we could store two without table indirection 
to be most optimal on 64-bit. Something like:

struct engine_mmio_base {
     u32 base : 24; /* Must be LSB, */
     u32 gen : 8;
};

union engine_mmio {
	struct engine_mmio_base mmio_base[2];
	struct engine_mmio_base *mmio_base_list;
};

#define MMIO_PTR_LIST BIT(0)

    {
     ... render engine ...
     .mmio = { (struct engine_mmio_base){.base = ..., ..gen = ...} },
     ...
    },
    {
     ...
     .mmio = { .mmio_base_list = &vcs0_mmio_bases | MMIO_PTR_LIST }
     ...
    },

This could be the best of both worlds, with up to two bases built-in, 
and engines with more point to an array.


Regards,

Tvrtko
Daniele Ceraolo Spurio March 12, 2018, 5:48 p.m. UTC | #8
On 12/03/18 04:04, Tvrtko Ursulin wrote:
> 
> On 09/03/2018 19:44, Tvrtko Ursulin wrote:
>>
>> On 09/03/2018 18:47, Daniele Ceraolo Spurio wrote:
>>> On 09/03/18 01:53, Tvrtko Ursulin wrote:
>>>>
>>>> On 08/03/2018 18:46, Daniele Ceraolo Spurio wrote:
>>>>> On 08/03/18 01:31, Tvrtko Ursulin wrote:
>>>>>>
>>>>>> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>>>>>>> The mmio bases we're currently storing in the intel_engines array 
>>>>>>> are
>>>>>>> only valid for a subset of gens, so we need to ignore them and use
>>>>>>> different values in some cases. Instead of doing that, we can have a
>>>>>>> table of [starting gen, mmio base] pairs for each engine in
>>>>>>> intel_engines and select the correct one based on the gen we're 
>>>>>>> running
>>>>>>> on in a consistent way.
>>>>>>>
>>>>>>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>>>>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>>>>> Signed-off-by: Daniele Ceraolo Spurio 
>>>>>>> <daniele.ceraolospurio@intel.com>
>>>>>>> ---
>>>>>>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>>>>>>> +++++++++++++++++++++------------
>>>>>>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>>>>>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>>>>>>
>>>>>>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>>>>>>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>>> index 4ba139c27fba..1dd92cac8d67 100644
>>>>>>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>>>>>>> intel_engine_classes[] = {
>>>>>>>       },
>>>>>>>   };
>>>>>>> +#define MAX_MMIO_BASES 3
>>>>>>>   struct engine_info {
>>>>>>>       unsigned int hw_id;
>>>>>>>       unsigned int uabi_id;
>>>>>>>       u8 class;
>>>>>>>       u8 instance;
>>>>>>> -    u32 mmio_base;
>>>>>>> +    struct engine_mmio_base {
>>>>>>> +        u32 gen : 8;
>>>>>>> +        u32 base : 24;
>>>>>>> +    } mmio_bases[MAX_MMIO_BASES];
>>>>>>>       unsigned irq_shift;
>>>>>>>   };
>>>>>>
>>>>>> It's not bad, but I am just wondering if it is too complicated 
>>>>>> versus simply duplicating the tables.
>>>>>>
>>>>>> Duplicated tables would certainly be much less code and 
>>>>>> complexity, but what about the size of pure data?
>>>>>>
>>>>>> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES * 
>>>>>> sizeof(u32), so 12, to the total of 30 if I am not mistaken. Times 
>>>>>> NUM_ENGINES equals 240 bytes for intel_engines[] array with this 
>>>>>> scheme.
>>>>>>
>>>>>
>>>>> we remove a u32 (the old mmio base) so we only grow 8 per engine, 
>>>>> but the compiler rounds up to a multiple of u32 so 28 per engine, 
>>>>> for a total of 224.
>>>>>
>>>>>> Separate tables per gens would be:
>>>>>>
>>>>>> sizeof(struct engine_info) is 18 bytes.
>>>>>>
>>>>>> For < gen6 we'd need 3 engines * 18 = 54
>>>>>> < gen11 = 5 engines * 18 = 90
>>>>>>  >= gen11 = 8 engines * 18 = 144
>>>>>>
>>>>>> 54 + 90 + 144 = 288 bytes
>>>>>>
>>>>>> So a little bit bigger. Hm. Plus we would need to refactor so 
>>>>>> intel_engines[] is not indexed by engine->id but is contiguous 
>>>>>> array with engine->id in the elements. Code to lookup the compact 
>>>>>> array should be simpler than combined new code from this patch, 
>>>>>> especially if you add the test as Chris suggested. So separate 
>>>>>> engine info arrays would win I think overall - when considering 
>>>>>> size of text + size of data + size of source code.
>>>>>>
>>>>>
>>>>> Not sure. I would exclude the selftest, which is usually not 
>>>>> compiled for released kernels, which leads to:
>>>>
>>>> Yes, but we cannot exclude its source since selftests for separate 
>>>> tables wouldn't be needed.
>>>>
>>>>> add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
>>>>> Function                                     old     new   delta
>>>>> intel_engines                                160     224     +64
>>>>> __func__                                   13891   13910     +19
>>>>> intel_engines_init_mmio                     1247    1264     +17
>>>>> intel_init_bsd_ring_buffer                   142     135      -7
>>>>> Total: Before=1479626, After=1479719, chg +0.01%
>>>>>
>>>>> Total growth is 93, which is less then your estimation for the 
>>>>> growth introduced by replicating the table.
>>>>>
>>>>>> But on the other hand your solution might be more future proof. So 
>>>>>> I don't know. Use the crystal ball to decide? :)
>>>>>>
>>>>>
>>>>> I think we should expect that the total engine count could grow 
>>>>> with future gens. In this case to me adding a new entry to a 
>>>>> unified table seems much cleaner (and uses less data) than adding a 
>>>>> completely new table each time.
>>>>
>>>> Okay I was off in my estimates. But in general I was coming from the 
>>>> angle of thinking that every new mmio base difference in this scheme 
>>>> grows the size for all engines. So if just VCS grows one new base, 
>>>> impact is multiplied by NUM_ENGINES. But maybe separate tables are 
>>>> also not a solution. Perhaps pulling out mmio_base arrays to outside 
>>>> struct engine_info in another set of static tables, so they could be 
>>>> null terminated would be best of both worlds?
>>>>
>>>> struct engine_mmio_base {
>>>>      u32 gen : 8;
>>>>      u32 base : 24;
>>>> };
>>>>
>>>> static const struct engine_mmio_base vcs0_mmio_bases[] = {
>>>>      { .gen = 11, .base = GEN11_BSD_RING_BASE },
>>>>      { .gen = 6, .base = GEN6_BSD_RING_BASE },
>>>>      { .gen = 4, .base = BSD_RING_BASE },
>>>>      { },
>>>> };
>>>>
>>>> And then in intel_engines array, for BSD:
>>>>
>>>>     {
>>>>      ...
>>>>      .mmio_bases = &vcs0_mmio_bases;
>>>>      ...
>>>>     },
>>>>
>>>> This way we limit data growth only to engines which change their 
>>>> mmio bases frequently.
>>>>
>>>> Just an idea.
>>>>
>>>
>>> I gave this a try and the code actually grows:
>>>
>>> add/remove: 8/0 grow/shrink: 2/0 up/down: 120/0 (120)
>>> Function                                     old     new   delta
>>> intel_engines                                224     256     +32
>>> vcs0_mmio_bases                                -      16     +16
>>> vecs1_mmio_bases                               -      12     +12
>>> vecs0_mmio_bases                               -      12     +12
>>> vcs1_mmio_bases                                -      12     +12
>>> vcs3_mmio_bases                                -       8      +8
>>> vcs2_mmio_bases                                -       8      +8
>>> rcs_mmio_bases                                 -       8      +8
>>> intel_engines_init_mmio                     1264    1272      +8
>>> bcs_mmio_bases                                 -       4      +4
>>> Total: Before=1479719, After=1479839, chg +0.01%
>>>
>>> I have no idea what the compiler is doing to grow intel_engines, 
>>> since by replacing and array of 3 u32s with a pointer I would expect 
>>> a shrink of 4 bytes per-engine. Anyway, even without the compiler 
>>> weirdness with
>>
>> It probably has to align the pointer to 8 bytes so that creates a 
>> hole. Moving mmio_base pointer higher up, either to the top or after 
>> two unsigned ints should work. Or packing the struct.
>>
>>> this method we would grow the code size of at least 4 bytes per 
>>> engine because we replace an array of 3 u32 (12B) with a pointer (8B) 
>>> and an array of at 2 or more u32 (8B+). I guess we can reconsider 
>>> if/when one engine reaches more than 3 mmio bases.
>>
>> Yeah it's fine. I was thinking that since you are almost there it 
>> makes sense to future proof it more, since no one will probably 
>> remember it later. But OK.
> 
> One idea to compact more, in addition to avoiding alignment holes, could 
> be to store the engine_mmio_base directly in the engine_mmio_base 
> pointer when it is a single entry, othewrise use a pointer to null 
> terminated array. Actually we could store two without table indirection 
> to be most optimal on 64-bit. Something like:
> 
> struct engine_mmio_base {
>      u32 base : 24; /* Must be LSB, */
>      u32 gen : 8;
> };
> 
> union engine_mmio {
>      struct engine_mmio_base mmio_base[2];
>      struct engine_mmio_base *mmio_base_list;
> };
> 
> #define MMIO_PTR_LIST BIT(0)
> 
>     {
>      ... render engine ...
>      .mmio = { (struct engine_mmio_base){.base = ..., ..gen = ...} },
>      ...
>     },
>     {
>      ...
>      .mmio = { .mmio_base_list = &vcs0_mmio_bases | MMIO_PTR_LIST }

This is giving a compiler error (even with the appropriate casting) for 
value not constant. I've solved by doing:

union engine_mmio {
	const struct engine_mmio_base bases[MAX_BUILTIN_MMIO_BASES];
	const struct engine_mmio_base *bases_list;
	union {
		const u64 is_list : 1;
		const u64 	  : 63;
	};
};

Code size is almost unchanged but still looks like a good compromise. 
Will update the selftest and send patches soon.

Daniele

>      ...
>     },
> 
> This could be the best of both worlds, with up to two bases built-in, 
> and engines with more point to an array.
> 
> 
> Regards,
> 
> Tvrtko
> 
> 
> 
>
Daniele Ceraolo Spurio March 12, 2018, 9:05 p.m. UTC | #9
On 12/03/18 10:48, Daniele Ceraolo Spurio wrote:
> 
> 
> On 12/03/18 04:04, Tvrtko Ursulin wrote:
>>
>> On 09/03/2018 19:44, Tvrtko Ursulin wrote:
>>>
>>> On 09/03/2018 18:47, Daniele Ceraolo Spurio wrote:
>>>> On 09/03/18 01:53, Tvrtko Ursulin wrote:
>>>>>
>>>>> On 08/03/2018 18:46, Daniele Ceraolo Spurio wrote:
>>>>>> On 08/03/18 01:31, Tvrtko Ursulin wrote:
>>>>>>>
>>>>>>> On 07/03/2018 19:45, Daniele Ceraolo Spurio wrote:
>>>>>>>> The mmio bases we're currently storing in the intel_engines 
>>>>>>>> array are
>>>>>>>> only valid for a subset of gens, so we need to ignore them and use
>>>>>>>> different values in some cases. Instead of doing that, we can 
>>>>>>>> have a
>>>>>>>> table of [starting gen, mmio base] pairs for each engine in
>>>>>>>> intel_engines and select the correct one based on the gen we're 
>>>>>>>> running
>>>>>>>> on in a consistent way.
>>>>>>>>
>>>>>>>> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
>>>>>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>>>>>> Signed-off-by: Daniele Ceraolo Spurio 
>>>>>>>> <daniele.ceraolospurio@intel.com>
>>>>>>>> ---
>>>>>>>>   drivers/gpu/drm/i915/intel_engine_cs.c  | 77 
>>>>>>>> +++++++++++++++++++++------------
>>>>>>>>   drivers/gpu/drm/i915/intel_ringbuffer.c |  1 -
>>>>>>>>   2 files changed, 49 insertions(+), 29 deletions(-)
>>>>>>>>
>>>>>>>> diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c 
>>>>>>>> b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>>>> index 4ba139c27fba..1dd92cac8d67 100644
>>>>>>>> --- a/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>>>> +++ b/drivers/gpu/drm/i915/intel_engine_cs.c
>>>>>>>> @@ -81,12 +81,16 @@ static const struct engine_class_info 
>>>>>>>> intel_engine_classes[] = {
>>>>>>>>       },
>>>>>>>>   };
>>>>>>>> +#define MAX_MMIO_BASES 3
>>>>>>>>   struct engine_info {
>>>>>>>>       unsigned int hw_id;
>>>>>>>>       unsigned int uabi_id;
>>>>>>>>       u8 class;
>>>>>>>>       u8 instance;
>>>>>>>> -    u32 mmio_base;
>>>>>>>> +    struct engine_mmio_base {
>>>>>>>> +        u32 gen : 8;
>>>>>>>> +        u32 base : 24;
>>>>>>>> +    } mmio_bases[MAX_MMIO_BASES];
>>>>>>>>       unsigned irq_shift;
>>>>>>>>   };
>>>>>>>
>>>>>>> It's not bad, but I am just wondering if it is too complicated 
>>>>>>> versus simply duplicating the tables.
>>>>>>>
>>>>>>> Duplicated tables would certainly be much less code and 
>>>>>>> complexity, but what about the size of pure data?
>>>>>>>
>>>>>>> In this patch sizeof(struct engine_info) grows by MAX_MMIO_BASES 
>>>>>>> * sizeof(u32), so 12, to the total of 30 if I am not mistaken. 
>>>>>>> Times NUM_ENGINES equals 240 bytes for intel_engines[] array with 
>>>>>>> this scheme.
>>>>>>>
>>>>>>
>>>>>> we remove a u32 (the old mmio base) so we only grow 8 per engine, 
>>>>>> but the compiler rounds up to a multiple of u32 so 28 per engine, 
>>>>>> for a total of 224.
>>>>>>
>>>>>>> Separate tables per gens would be:
>>>>>>>
>>>>>>> sizeof(struct engine_info) is 18 bytes.
>>>>>>>
>>>>>>> For < gen6 we'd need 3 engines * 18 = 54
>>>>>>> < gen11 = 5 engines * 18 = 90
>>>>>>>  >= gen11 = 8 engines * 18 = 144
>>>>>>>
>>>>>>> 54 + 90 + 144 = 288 bytes
>>>>>>>
>>>>>>> So a little bit bigger. Hm. Plus we would need to refactor so 
>>>>>>> intel_engines[] is not indexed by engine->id but is contiguous 
>>>>>>> array with engine->id in the elements. Code to lookup the compact 
>>>>>>> array should be simpler than combined new code from this patch, 
>>>>>>> especially if you add the test as Chris suggested. So separate 
>>>>>>> engine info arrays would win I think overall - when considering 
>>>>>>> size of text + size of data + size of source code.
>>>>>>>
>>>>>>
>>>>>> Not sure. I would exclude the selftest, which is usually not 
>>>>>> compiled for released kernels, which leads to:
>>>>>
>>>>> Yes, but we cannot exclude its source since selftests for separate 
>>>>> tables wouldn't be needed.
>>>>>
>>>>>> add/remove: 0/0 grow/shrink: 3/1 up/down: 100/-7 (93)
>>>>>> Function                                     old     new   delta
>>>>>> intel_engines                                160     224     +64
>>>>>> __func__                                   13891   13910     +19
>>>>>> intel_engines_init_mmio                     1247    1264     +17
>>>>>> intel_init_bsd_ring_buffer                   142     135      -7
>>>>>> Total: Before=1479626, After=1479719, chg +0.01%
>>>>>>
>>>>>> Total growth is 93, which is less then your estimation for the 
>>>>>> growth introduced by replicating the table.
>>>>>>
>>>>>>> But on the other hand your solution might be more future proof. 
>>>>>>> So I don't know. Use the crystal ball to decide? :)
>>>>>>>
>>>>>>
>>>>>> I think we should expect that the total engine count could grow 
>>>>>> with future gens. In this case to me adding a new entry to a 
>>>>>> unified table seems much cleaner (and uses less data) than adding 
>>>>>> a completely new table each time.
>>>>>
>>>>> Okay I was off in my estimates. But in general I was coming from 
>>>>> the angle of thinking that every new mmio base difference in this 
>>>>> scheme grows the size for all engines. So if just VCS grows one new 
>>>>> base, impact is multiplied by NUM_ENGINES. But maybe separate 
>>>>> tables are also not a solution. Perhaps pulling out mmio_base 
>>>>> arrays to outside struct engine_info in another set of static 
>>>>> tables, so they could be null terminated would be best of both worlds?
>>>>>
>>>>> struct engine_mmio_base {
>>>>>      u32 gen : 8;
>>>>>      u32 base : 24;
>>>>> };
>>>>>
>>>>> static const struct engine_mmio_base vcs0_mmio_bases[] = {
>>>>>      { .gen = 11, .base = GEN11_BSD_RING_BASE },
>>>>>      { .gen = 6, .base = GEN6_BSD_RING_BASE },
>>>>>      { .gen = 4, .base = BSD_RING_BASE },
>>>>>      { },
>>>>> };
>>>>>
>>>>> And then in intel_engines array, for BSD:
>>>>>
>>>>>     {
>>>>>      ...
>>>>>      .mmio_bases = &vcs0_mmio_bases;
>>>>>      ...
>>>>>     },
>>>>>
>>>>> This way we limit data growth only to engines which change their 
>>>>> mmio bases frequently.
>>>>>
>>>>> Just an idea.
>>>>>
>>>>
>>>> I gave this a try and the code actually grows:
>>>>
>>>> add/remove: 8/0 grow/shrink: 2/0 up/down: 120/0 (120)
>>>> Function                                     old     new   delta
>>>> intel_engines                                224     256     +32
>>>> vcs0_mmio_bases                                -      16     +16
>>>> vecs1_mmio_bases                               -      12     +12
>>>> vecs0_mmio_bases                               -      12     +12
>>>> vcs1_mmio_bases                                -      12     +12
>>>> vcs3_mmio_bases                                -       8      +8
>>>> vcs2_mmio_bases                                -       8      +8
>>>> rcs_mmio_bases                                 -       8      +8
>>>> intel_engines_init_mmio                     1264    1272      +8
>>>> bcs_mmio_bases                                 -       4      +4
>>>> Total: Before=1479719, After=1479839, chg +0.01%
>>>>
>>>> I have no idea what the compiler is doing to grow intel_engines, 
>>>> since by replacing and array of 3 u32s with a pointer I would expect 
>>>> a shrink of 4 bytes per-engine. Anyway, even without the compiler 
>>>> weirdness with
>>>
>>> It probably has to align the pointer to 8 bytes so that creates a 
>>> hole. Moving mmio_base pointer higher up, either to the top or after 
>>> two unsigned ints should work. Or packing the struct.
>>>
>>>> this method we would grow the code size of at least 4 bytes per 
>>>> engine because we replace an array of 3 u32 (12B) with a pointer 
>>>> (8B) and an array of at 2 or more u32 (8B+). I guess we can 
>>>> reconsider if/when one engine reaches more than 3 mmio bases.
>>>
>>> Yeah it's fine. I was thinking that since you are almost there it 
>>> makes sense to future proof it more, since no one will probably 
>>> remember it later. But OK.
>>
>> One idea to compact more, in addition to avoiding alignment holes, 
>> could be to store the engine_mmio_base directly in the 
>> engine_mmio_base pointer when it is a single entry, othewrise use a 
>> pointer to null terminated array. Actually we could store two without 
>> table indirection to be most optimal on 64-bit. Something like:
>>
>> struct engine_mmio_base {
>>      u32 base : 24; /* Must be LSB, */
>>      u32 gen : 8;
>> };
>>
>> union engine_mmio {
>>      struct engine_mmio_base mmio_base[2];
>>      struct engine_mmio_base *mmio_base_list;
>> };
>>
>> #define MMIO_PTR_LIST BIT(0)
>>
>>     {
>>      ... render engine ...
>>      .mmio = { (struct engine_mmio_base){.base = ..., ..gen = ...} },
>>      ...
>>     },
>>     {
>>      ...
>>      .mmio = { .mmio_base_list = &vcs0_mmio_bases | MMIO_PTR_LIST }
> 
> This is giving a compiler error (even with the appropriate casting) for 
> value not constant. I've solved by doing:
> 
> union engine_mmio {
>      const struct engine_mmio_base bases[MAX_BUILTIN_MMIO_BASES];
>      const struct engine_mmio_base *bases_list;
>      union {
>          const u64 is_list : 1;
>          const u64       : 63;
>      };
> };
>

I retract this statement. The trick with the union compiles fine but 
only one of the 2 assignments (between bases_list and is_list) actually 
ends up in the structure.
I've experimented with other possible solutions with shifts and/or ORs, 
but they all lead to a compiler error for value not constant. At this 
point I'll stick with the original implementation as I want to avoid 
over-engineering this.

Daniele

> Code size is almost unchanged but still looks like a good compromise. 
> Will update the selftest and send patches soon.
> 
> Daniele
> 
>>      ...
>>     },
>>
>> This could be the best of both worlds, with up to two bases built-in, 
>> and engines with more point to an array.
>>
>>
>> Regards,
>>
>> Tvrtko
>>
>>
>>
>>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Patch
diff mbox

diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index 4ba139c27fba..1dd92cac8d67 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -81,12 +81,16 @@  static const struct engine_class_info intel_engine_classes[] = {
 	},
 };
 
+#define MAX_MMIO_BASES 3
 struct engine_info {
 	unsigned int hw_id;
 	unsigned int uabi_id;
 	u8 class;
 	u8 instance;
-	u32 mmio_base;
+	struct engine_mmio_base {
+		u32 gen : 8;
+		u32 base : 24;
+	} mmio_bases[MAX_MMIO_BASES];
 	unsigned irq_shift;
 };
 
@@ -96,7 +100,9 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_RENDER,
 		.class = RENDER_CLASS,
 		.instance = 0,
-		.mmio_base = RENDER_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 1, .base = RENDER_RING_BASE }
+		},
 		.irq_shift = GEN8_RCS_IRQ_SHIFT,
 	},
 	[BCS] = {
@@ -104,7 +110,9 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_BLT,
 		.class = COPY_ENGINE_CLASS,
 		.instance = 0,
-		.mmio_base = BLT_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 6, .base = BLT_RING_BASE }
+		},
 		.irq_shift = GEN8_BCS_IRQ_SHIFT,
 	},
 	[VCS] = {
@@ -112,7 +120,11 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_BSD,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 0,
-		.mmio_base = GEN6_BSD_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 11, .base = GEN11_BSD_RING_BASE },
+			{ .gen = 6, .base = GEN6_BSD_RING_BASE },
+			{ .gen = 4, .base = BSD_RING_BASE }
+		},
 		.irq_shift = GEN8_VCS1_IRQ_SHIFT,
 	},
 	[VCS2] = {
@@ -120,7 +132,10 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_BSD,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 1,
-		.mmio_base = GEN8_BSD2_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 11, .base = GEN11_BSD2_RING_BASE },
+			{ .gen = 8, .base = GEN8_BSD2_RING_BASE }
+		},
 		.irq_shift = GEN8_VCS2_IRQ_SHIFT,
 	},
 	[VCS3] = {
@@ -128,7 +143,9 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_BSD,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 2,
-		.mmio_base = GEN11_BSD3_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 11, .base = GEN11_BSD3_RING_BASE }
+		},
 		.irq_shift = 0, /* not used */
 	},
 	[VCS4] = {
@@ -136,7 +153,9 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_BSD,
 		.class = VIDEO_DECODE_CLASS,
 		.instance = 3,
-		.mmio_base = GEN11_BSD4_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 11, .base = GEN11_BSD4_RING_BASE }
+		},
 		.irq_shift = 0, /* not used */
 	},
 	[VECS] = {
@@ -144,7 +163,10 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_VEBOX,
 		.class = VIDEO_ENHANCEMENT_CLASS,
 		.instance = 0,
-		.mmio_base = VEBOX_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 11, .base = GEN11_VEBOX_RING_BASE },
+			{ .gen = 7, .base = VEBOX_RING_BASE }
+		},
 		.irq_shift = GEN8_VECS_IRQ_SHIFT,
 	},
 	[VECS2] = {
@@ -152,7 +174,9 @@  static const struct engine_info intel_engines[] = {
 		.uabi_id = I915_EXEC_VEBOX,
 		.class = VIDEO_ENHANCEMENT_CLASS,
 		.instance = 1,
-		.mmio_base = GEN11_VEBOX2_RING_BASE,
+		.mmio_bases = {
+			{ .gen = 11, .base = GEN11_VEBOX2_RING_BASE }
+		},
 		.irq_shift = 0, /* not used */
 	},
 };
@@ -223,6 +247,21 @@  __intel_engine_context_size(struct drm_i915_private *dev_priv, u8 class)
 	}
 }
 
+static u32 __engine_mmio_base(struct drm_i915_private *i915,
+			      const struct engine_mmio_base* bases)
+{
+	int i;
+
+	for (i = 0; i < MAX_MMIO_BASES; i++)
+		if (INTEL_GEN(i915) >= bases[i].gen)
+			break;
+
+	GEM_BUG_ON(i == MAX_MMIO_BASES);
+	GEM_BUG_ON(!bases[i].base);
+
+	return bases[i].base;
+}
+
 static int
 intel_engine_setup(struct drm_i915_private *dev_priv,
 		   enum intel_engine_id id)
@@ -257,25 +296,7 @@  intel_engine_setup(struct drm_i915_private *dev_priv,
 			 class_info->name, info->instance) >=
 		sizeof(engine->name));
 	engine->hw_id = engine->guc_id = info->hw_id;
-	if (INTEL_GEN(dev_priv) >= 11) {
-		switch (engine->id) {
-		case VCS:
-			engine->mmio_base = GEN11_BSD_RING_BASE;
-			break;
-		case VCS2:
-			engine->mmio_base = GEN11_BSD2_RING_BASE;
-			break;
-		case VECS:
-			engine->mmio_base = GEN11_VEBOX_RING_BASE;
-			break;
-		default:
-			/* take the original value for all other engines  */
-			engine->mmio_base = info->mmio_base;
-			break;
-		}
-	} else {
-		engine->mmio_base = info->mmio_base;
-	}
+	engine->mmio_base = __engine_mmio_base(dev_priv, info->mmio_bases);
 	engine->irq_shift = info->irq_shift;
 	engine->class = info->class;
 	engine->instance = info->instance;
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 1d599524a759..2e4408477ab5 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -2079,7 +2079,6 @@  int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine)
 		engine->emit_flush = gen6_bsd_ring_flush;
 		engine->irq_enable_mask = GT_BSD_USER_INTERRUPT;
 	} else {
-		engine->mmio_base = BSD_RING_BASE;
 		engine->emit_flush = bsd_ring_flush;
 		if (IS_GEN5(dev_priv))
 			engine->irq_enable_mask = ILK_BSD_USER_INTERRUPT;