All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mika Kuoppala <mika.kuoppala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH 2/3] drm/i915/gt: Set the PP_DIR registers upon enabling ring submission
Date: Thu, 06 Feb 2020 18:14:39 +0200	[thread overview]
Message-ID: <87ftfnae1c.fsf@gaia.fi.intel.com> (raw)
In-Reply-To: <20200206014439.2137800-2-chris@chris-wilson.co.uk>

Chris Wilson <chris@chris-wilson.co.uk> writes:

> Always prime the page table registers before starting the ring. Even
> though we will update these to the per-context page tables during
> dispatch, it is prudent to ensure that the registers always point to a
> valid PD.
>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
> ---
>  .../gpu/drm/i915/gt/intel_ring_submission.c   | 40 ++++++++++++-------
>  1 file changed, 26 insertions(+), 14 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
> index 42168d7cf5b5..f915a63e1110 100644
> --- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c
> +++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
> @@ -635,6 +635,27 @@ static bool stop_ring(struct intel_engine_cs *engine)
>  	return (ENGINE_READ(engine, RING_HEAD) & HEAD_ADDR) == 0;
>  }
>  
> +static struct i915_address_space *vm_alias(struct i915_address_space *vm)
> +{
> +	if (i915_is_ggtt(vm))
> +		vm = &i915_vm_to_ggtt(vm)->alias->vm;
> +
> +	return vm;
> +}
> +
> +static void set_pp_dir(struct intel_engine_cs *engine)
> +{
> +	struct i915_address_space *vm = vm_alias(engine->gt->vm);
> +
> +	if (vm) {
> +		struct i915_ppgtt *ppgtt = i915_vm_to_ppgtt(vm);
> +
> +		ENGINE_WRITE(engine, RING_PP_DIR_DCLV, PP_DIR_DCLV_2G);

I did think that for setup we set these zero first. But it seems
pointless. They should be zero after reset anywasy.

> +		ENGINE_WRITE(engine, RING_PP_DIR_BASE,
> +			     px_base(ppgtt->pd)->ggtt_offset << 10);

Shift for 16 and then index by cacheline so 16-6 it seems.

> +	}
> +}
> +
>  static int xcs_resume(struct intel_engine_cs *engine)
>  {
>  	struct drm_i915_private *dev_priv = engine->i915;
> @@ -693,6 +714,8 @@ static int xcs_resume(struct intel_engine_cs *engine)
>  	GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->tail));
>  	intel_ring_update_space(ring);
>  
> +	set_pp_dir(engine);
> +

Then rings are off and we start by setting up the pd.
Can't figure out a better spot.

Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

>  	/* First wake the ring up to an empty/idle ring */
>  	ENGINE_WRITE(engine, RING_HEAD, ring->head);
>  	ENGINE_WRITE(engine, RING_TAIL, ring->head);
> @@ -1169,23 +1192,12 @@ static void ring_context_destroy(struct kref *ref)
>  	intel_context_free(ce);
>  }
>  
> -static struct i915_address_space *vm_alias(struct intel_context *ce)
> -{
> -	struct i915_address_space *vm;
> -
> -	vm = ce->vm;
> -	if (i915_is_ggtt(vm))
> -		vm = &i915_vm_to_ggtt(vm)->alias->vm;
> -
> -	return vm;
> -}
> -
>  static int __context_pin_ppgtt(struct intel_context *ce)
>  {
>  	struct i915_address_space *vm;
>  	int err = 0;
>  
> -	vm = vm_alias(ce);
> +	vm = vm_alias(ce->vm);
>  	if (vm)
>  		err = gen6_ppgtt_pin(i915_vm_to_ppgtt((vm)));
>  
> @@ -1196,7 +1208,7 @@ static void __context_unpin_ppgtt(struct intel_context *ce)
>  {
>  	struct i915_address_space *vm;
>  
> -	vm = vm_alias(ce);
> +	vm = vm_alias(ce->vm);
>  	if (vm)
>  		gen6_ppgtt_unpin(i915_vm_to_ppgtt(vm));
>  }
> @@ -1553,7 +1565,7 @@ static int switch_context(struct i915_request *rq)
>  
>  	GEM_BUG_ON(HAS_EXECLISTS(rq->i915));
>  
> -	ret = switch_mm(rq, vm_alias(ce));
> +	ret = switch_mm(rq, vm_alias(ce->vm));
>  	if (ret)
>  		return ret;
>  
> -- 
> 2.25.0
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2020-02-06 16:15 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-06  1:44 [Intel-gfx] [PATCH 1/3] drm/i915/gt: Tweak gen7 xcs flushing Chris Wilson
2020-02-06  1:44 ` [Intel-gfx] [PATCH 2/3] drm/i915/gt: Set the PP_DIR registers upon enabling ring submission Chris Wilson
2020-02-06 16:14   ` Mika Kuoppala [this message]
2020-02-06  1:44 ` [Intel-gfx] [PATCH 3/3] drm/i915/gt: Stop invalidating the PD cachelines for gen7 Chris Wilson
2020-02-06 16:32   ` Mika Kuoppala
2020-02-06 19:26     ` Chris Wilson
2020-02-06  4:07 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/3] drm/i915/gt: Tweak gen7 xcs flushing Patchwork
2020-02-06  4:36 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2020-02-06 16:35 ` [Intel-gfx] [PATCH 1/3] " Mika Kuoppala
2020-02-06 19:27   ` Chris Wilson
2020-02-08 16:35 ` [Intel-gfx] ✓ Fi.CI.IGT: success for series starting with [1/3] " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87ftfnae1c.fsf@gaia.fi.intel.com \
    --to=mika.kuoppala@linux.intel.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.