public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 05/19] drm/i915: Use i915_vma_pin_iomap on the ringbuffer object
Date: Thu, 21 Apr 2016 10:19:05 +0300	[thread overview]
Message-ID: <1461223145.4381.19.camel@linux.intel.com> (raw)
In-Reply-To: <1461177750-20187-6-git-send-email-chris@chris-wilson.co.uk>

On ke, 2016-04-20 at 19:42 +0100, Chris Wilson wrote:
> Similarly to i915_gem_object_pin_map on LLC platforms, we can
> use the new VMA based io mapping on !LLC to amoritize the cost
> of ringbuffer pinning and unpinning.
> 
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

> ---
>  drivers/gpu/drm/i915/intel_ringbuffer.c | 16 +++++++++-------
>  1 file changed, 9 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
> index 245386e20c52..27ba15acae1e 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.c
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
> @@ -2084,20 +2084,23 @@ static int init_phys_status_page(struct intel_engine_cs *engine)
>  
>  void intel_unpin_ringbuffer_obj(struct intel_ringbuffer *ringbuf)
>  {
> +	GEM_BUG_ON(ringbuf->vma == NULL);
> +	GEM_BUG_ON(ringbuf->virtual_start == NULL);
> +
>  	if (HAS_LLC(ringbuf->obj->base.dev) && !ringbuf->obj->stolen)
>  		i915_gem_object_unpin_map(ringbuf->obj);
>  	else
> -		iounmap(ringbuf->virtual_start);
> +		i915_vma_unpin_iomap(ringbuf->vma);
>  	ringbuf->virtual_start = NULL;
> -	ringbuf->vma = NULL;
> +
>  	i915_gem_object_ggtt_unpin(ringbuf->obj);
> +	ringbuf->vma = NULL;
>  }
>  
>  int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
>  				     struct intel_ringbuffer *ringbuf)
>  {
>  	struct drm_i915_private *dev_priv = to_i915(dev);
> -	struct i915_ggtt *ggtt = &dev_priv->ggtt;
>  	struct drm_i915_gem_object *obj = ringbuf->obj;
>  	/* Ring wraparound at offset 0 sometimes hangs. No idea why. */
>  	unsigned flags = PIN_OFFSET_BIAS | 4096;
> @@ -2131,10 +2134,9 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
>  		/* Access through the GTT requires the device to be awake. */
>  		assert_rpm_wakelock_held(dev_priv);
>  
> -		addr = ioremap_wc(ggtt->mappable_base +
> -				  i915_gem_obj_ggtt_offset(obj), ringbuf->size);
> -		if (addr == NULL) {
> -			ret = -ENOMEM;
> +		addr = i915_vma_pin_iomap(i915_gem_obj_to_ggtt(obj));
> +		if (IS_ERR(addr)) {
> +			ret = PTR_ERR(addr);
>  			goto err_unpin;
>  		}
>  	}
-- 
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2016-04-21  7:17 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-20 18:42 Premature unpinning at last Chris Wilson
2016-04-20 18:42 ` [PATCH 01/19] drm/i915/overlay: Replace i915_gem_obj_ggtt_offset() with the known flip_addr Chris Wilson
2016-04-20 18:42 ` [PATCH 02/19] io-mapping: Specify mapping size for io_mapping_map_wc() Chris Wilson
2016-04-20 18:58   ` Luis R. Rodriguez
2016-04-20 19:14     ` Chris Wilson
2016-04-20 21:23       ` Luis R. Rodriguez
2016-04-20 18:42 ` [PATCH 03/19] drm/i915: Introduce i915_vm_to_ggtt() Chris Wilson
2016-04-20 18:42 ` [PATCH 04/19] drm/i915: Move ioremap_wc tracking onto VMA Chris Wilson
2016-04-20 18:42 ` [PATCH 05/19] drm/i915: Use i915_vma_pin_iomap on the ringbuffer object Chris Wilson
2016-04-21  7:19   ` Joonas Lahtinen [this message]
2016-04-20 18:42 ` [PATCH 06/19] drm/i915: Mark the current context as lost on suspend Chris Wilson
2016-04-20 18:42 ` [PATCH 07/19] drm/i915: L3 cache remapping is part of context switching Chris Wilson
2016-04-20 18:42 ` [PATCH 08/19] drm/i915: Consolidate L3 remapping LRI Chris Wilson
2016-04-20 18:42 ` [PATCH 09/19] drm/i915: Remove early l3-remap Chris Wilson
2016-04-20 18:42 ` [PATCH 10/19] drm/i915: Rearrange switch_context to load the aliasing ppgtt on first use Chris Wilson
2016-04-21  6:48   ` Joonas Lahtinen
2016-04-21  6:58     ` Chris Wilson
2016-04-21  7:01     ` Chris Wilson
2016-04-21  7:24       ` Chris Wilson
2016-04-21  7:32       ` Joonas Lahtinen
2016-04-20 18:42 ` [PATCH 11/19] drm/i915: Assign every HW context a unique ID Chris Wilson
2016-04-20 18:42 ` [PATCH 12/19] drm/i915: Replace the pinned context address with its " Chris Wilson
2016-04-20 18:42 ` [PATCH 13/19] drm/i915: Refactor execlists default context pinning Chris Wilson
2016-04-20 18:42 ` [PATCH 14/19] drm/i915: Move context initialisation to first-use Chris Wilson
2016-04-21  6:57   ` Joonas Lahtinen
2016-04-21  7:08     ` Chris Wilson
2016-04-21  7:47       ` Joonas Lahtinen
2016-04-21  7:56         ` Chris Wilson
2016-04-21  8:37           ` Joonas Lahtinen
2016-04-20 18:42 ` [PATCH 15/19] drm/i915: Move the magical deferred context allocation into the request Chris Wilson
2016-04-20 18:42 ` [PATCH 16/19] drm/i915: Move releasing of the GEM request from free to retire/cancel Chris Wilson
2016-04-20 18:42 ` [PATCH 17/19] drm/i915: Track the previous pinned context inside the request Chris Wilson
2016-04-21  7:07   ` Joonas Lahtinen
2016-04-21  7:22     ` Chris Wilson
2016-04-21  7:35       ` Joonas Lahtinen
2016-04-20 18:42 ` [PATCH 18/19] drm/i915: Store LRC hardware id in " Chris Wilson
2016-04-21  7:58   ` Tvrtko Ursulin
2016-04-21  9:02     ` Chris Wilson
2016-04-20 18:42 ` [PATCH 19/19] drm/i915: Stop tracking execlists retired requests Chris Wilson
2016-04-21 11:27 ` ✓ Fi.CI.BAT: success for series starting with [01/19] drm/i915/overlay: Replace i915_gem_obj_ggtt_offset() with the known flip_addr Patchwork
2016-04-21 12:05 ` ✗ Fi.CI.BAT: failure " Patchwork
2016-04-23 10:53 ` ✗ Fi.CI.BAT: warning " Patchwork
2016-04-25  6:51 ` Patchwork
  -- strict thread matches above, loose matches on Subject: below --
2016-04-21  8:58 [PATCH 01/19] " Chris Wilson
2016-04-21  8:58 ` [PATCH 05/19] drm/i915: Use i915_vma_pin_iomap on the ringbuffer object Chris Wilson
2016-04-21 14:57 Final CI pass for premature Chris Wilson
2016-04-21 14:57 ` [PATCH 05/19] drm/i915: Use i915_vma_pin_iomap on the ringbuffer object Chris Wilson

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=1461223145.4381.19.camel@linux.intel.com \
    --to=joonas.lahtinen@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox