[Intel-gfx] [PATCH v3] drm/i915: Initialize workarounds in logical ring mode too

[Mika Kuoppala]

Michel Thierry <michel.thierry at intel.com> writes:

> Following the legacy ring submission example, update the
> ring->init_context() hook to support the execlist submission mode.
>
> v2: update to use the new workaround macros and cleanup unused code.
> This takes care of both bdw and chv workarounds.
>
> v2.1: Add missing call to init_context() during deferred context creation.
>
> v3: Split init_context (emit) in legacy/lrc modes. For lrc, get the ringbuf
> from the context. (Mika/Daniel).
>
> Issue: VIZ-4092
> Issue: GMIN-3475
> Change-Id: Ie3d093b2542ab0e2a44b90460533e2f979788d6c
> Cc: Deepak S <deepak.s at intel.com>
> Cc: Mika Kuoppala <mika.kuoppala at linux.intel.com>
> Cc: Daniel Vetter <daniel at ffwll.ch>
> Signed-off-by: Michel Thierry <michel.thierry at intel.com>
> Signed-off-by: Arun Siluvery <arun.siluvery at linux.intel.com>
> ---
>  drivers/gpu/drm/i915/intel_lrc.c        | 47 ++++++++++++++++++++++++++++++++-
>  drivers/gpu/drm/i915/intel_ringbuffer.c |  2 +-
>  drivers/gpu/drm/i915/intel_ringbuffer.h |  4 +++
>  3 files changed, 51 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
> index 78c3dfc..69d672b 100644
> --- a/drivers/gpu/drm/i915/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/intel_lrc.c
> @@ -1072,6 +1072,44 @@ int intel_logical_ring_begin(struct intel_ringbuffer *ringbuf, int num_dwords)
>  	return 0;
>  }
>  
> +static int intel_logical_ring_workarounds_emit(struct intel_engine_cs *ring,
> +	       struct intel_context *ctx)
> +{
> +	int ret, i;
> +	struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
> +	struct drm_device *dev = ring->dev;
> +	struct drm_i915_private *dev_priv = dev->dev_private;
> +	struct i915_workarounds *w = &dev_priv->workarounds;
> +
> +	if (WARN_ON(w->count == 0))
> +		return 0;
> +
> +	ring->gpu_caches_dirty = true;
> +	ret = logical_ring_flush_all_caches(ringbuf);
> +	if (ret)
> +		return ret;
> +
> +	ret = intel_logical_ring_begin(ringbuf, w->count * 2 + 2);
> +	if (ret)
> +		return ret;
> +
> +	intel_logical_ring_emit(ringbuf, MI_LOAD_REGISTER_IMM(w->count));
> +	for (i = 0; i < w->count; i++) {
> +		intel_logical_ring_emit(ringbuf, w->reg[i].addr);
> +		intel_logical_ring_emit(ringbuf, w->reg[i].value);
> +	}
> +	intel_logical_ring_emit(ringbuf, MI_NOOP);
> +
> +	intel_logical_ring_advance(ringbuf);
> +
> +	ring->gpu_caches_dirty = true;
> +	ret = logical_ring_flush_all_caches(ringbuf);
> +	if (ret)
> +		return ret;
> +
> +	return 0;
> +}
> +
>  static int gen8_init_common_ring(struct intel_engine_cs *ring)
>  {
>  	struct drm_device *dev = ring->dev;
> @@ -1115,7 +1153,7 @@ static int gen8_init_render_ring(struct intel_engine_cs *ring)
>  
>  	I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
>  
> -	return ret;
> +	return init_workarounds_ring(ring);
>  }
>  
>  static int gen8_emit_bb_start(struct intel_ringbuffer *ringbuf,
> @@ -1366,6 +1404,7 @@ static int logical_render_ring_init(struct drm_device *dev)
>  		ring->irq_keep_mask |= GT_RENDER_L3_PARITY_ERROR_INTERRUPT;
>  
>  	ring->init = gen8_init_render_ring;
> +	ring->init_context_lrc = intel_logical_ring_workarounds_emit;
>  	ring->cleanup = intel_fini_pipe_control;
>  	ring->get_seqno = gen8_get_seqno;
>  	ring->set_seqno = gen8_set_seqno;
> @@ -1870,6 +1909,12 @@ int intel_lr_context_deferred_create(struct intel_context *ctx,
>  	}
>  
>  	if (ring->id == RCS && !ctx->rcs_initialized) {
> +		if (ring->init_context_lrc) {
> +			ret = ring->init_context_lrc(ring, ctx);
> +			if (ret)
> +				DRM_ERROR("ring init context: %d\n", ret);
> +		}
> +
>  		ret = intel_lr_context_render_state_init(ring, ctx);
>  		if (ret) {
>  			DRM_ERROR("Init render state failed: %d\n", ret);
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 98f2787..b7f32b9 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -818,7 +818,7 @@ static int chv_init_workarounds(struct intel_engine_cs *ring)
>  	return 0;
>  }
>  
> -static int init_workarounds_ring(struct intel_engine_cs *ring)
> +int init_workarounds_ring(struct intel_engine_cs *ring)
>  {
>  	struct drm_device *dev = ring->dev;
>  	struct drm_i915_private *dev_priv = dev->dev_private;
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
> index f0e7761..a6c4458 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.h
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
> @@ -149,6 +149,8 @@ struct  intel_engine_cs {
>  	int		(*init)(struct intel_engine_cs *ring);
>  
>  	int		(*init_context)(struct intel_engine_cs *ring);
> +	int		(*init_context_lrc)(struct intel_engine_cs *ring,
> +				      struct intel_context *ctx);

I was going to advertise earlier that using this abstraction will
make your patch smaller.

You don't need to introduce new interface here. Just
add struct intel_context *ctx as a parameter to (*init_context).

-Mika

>  	void		(*write_tail)(struct intel_engine_cs *ring,
>  				      u32 value);
> @@ -429,6 +431,8 @@ int intel_init_vebox_ring_buffer(struct drm_device *dev);
>  u64 intel_ring_get_active_head(struct intel_engine_cs *ring);
>  void intel_ring_setup_status_page(struct intel_engine_cs *ring);
>  
> +int init_workarounds_ring(struct intel_engine_cs *ring);
> +
>  static inline u32 intel_ring_get_tail(struct intel_ringbuffer *ringbuf)
>  {
>  	return ringbuf->tail;
> -- 
> 2.1.1