From d593d992a99c4945a183611354dded7578abab7b Mon Sep 17 00:00:00 2001
From: Ville Syrjälä <ville.syrjala@linux.intel.com>
Date: Fri, 13 Jun 2014 16:42:51 +0300
Subject: drm/i915: Fix __user sparse warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CHECK   linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c
linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c:1529:47: warning: incorrect type in initializer (different address spaces)
linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c:1529:47:    expected struct drm_i915_gem_exec_object2 *user_exec_list
linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c:1529:47:    got void [noderef] <asn:1>*
linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c:1533:61: warning: incorrect type in argument 1 (different address spaces)
linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c:1533:61:    expected void [noderef] <asn:1>*dst
linux/drivers/gpu/drm/i915/i915_gem_execbuffer.c:1533:61:    got unsigned long long *<noident>

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
---
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'drivers/gpu/drm/i915/i915_gem_execbuffer.c')

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 3a30133f93e8..93d7f7246588 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1525,7 +1525,7 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data,
 	ret = i915_gem_do_execbuffer(dev, data, file, args, exec2_list);
 	if (!ret) {
 		/* Copy the new buffer offsets back to the user's exec list. */
-		struct drm_i915_gem_exec_object2 *user_exec_list =
+		struct drm_i915_gem_exec_object2 __user *user_exec_list =
 				   to_user_ptr(args->buffers_ptr);
 		int i;
 
-- 
cgit v1.2.3


From f99d70690e0755696e4e68404c8993431efda469 Mon Sep 17 00:00:00 2001
From: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Thu, 19 Jun 2014 16:01:59 +0200
Subject: drm/i915: Track frontbuffer invalidation/flushing

So these are the guts of the new beast. This tracks when a frontbuffer
gets invalidated (due to frontbuffer rendering) and hence should be
constantly scaned out, and when it's flushed again and can be
compressed/one-shot-upload.

Rules for flushing are simple: The frontbuffer needs one more full
upload starting from the next vblank. Which means that the flushing
can _only_ be called once the frontbuffer update has been latched.

But this poses a problem for pageflips: We can't just delay the
flushing until the pageflip is latched, since that would pose the risk
that we override frontbuffer rendering that has been scheduled
in-between the pageflip ioctl and the actual latching.

To handle this track asynchronous invalidations (and also pageflip)
state per-ring and delay any in-between flushing until the rendering
has completed. And also cancel any delayed flushing if we get a new
invalidation request (whether delayed or not).

Also call intel_mark_fb_busy in both cases in all cases to make sure
that we keep the screen at the highest refresh rate both on flips,
synchronous plane updates and for frontbuffer rendering.

v2: Lots of improvements

Suggestions from Chris:
- Move invalidate/flush in flush_*_domain and set_to_*_domain.
- Drop the flush in busy_ioctl since it's redundant. Was a leftover
  from an earlier concept to track flips/delayed flushes.
- Don't forget about the initial modeset enable/final disable.
  Suggested by Chris.

Track flips accurately, too. Since flips complete independently of
rendering we need to track pending flips in a separate mask. Again if
an invalidate happens we need to cancel the evenutal flush to avoid
races.

v3:
Provide correct header declarations for flip functions. Currently not
needed outside of intel_display.c, but part of the proper interface.

v4: Add proper domain management to fbcon so that the fbcon buffer is
also tracked correctly.

v5: Fixup locking around the fbcon set_to_gtt_domain call.

v6: More comments from Chris:
- Split out fbcon changes.
- Drop superflous checks for potential scanout before calling intel_fb
  functions - we can micro-optimize this later.
- s/intel_fb_/intel_fb_obj_/ to make it clear that this deals in gem
  object. We already have precedence for fb_obj in the pin_and_fence
  functions.

v7: Clarify the semantics of the flip flush handling by renaming
things a bit:
- Don't go through a gem object but take the relevant frontbuffer bits
  directly. These functions center on the plane, the actual object is
  irrelevant - even a flip to the same object as already active should
  cause a flush.
- Add a new intel_frontbuffer_flip for synchronous plane updates. It
  currently just calls intel_frontbuffer_flush since the implemenation
  differs.

This way we achieve a clear split between one-shot update events on
one side and frontbuffer rendering with potentially a very long delay
between the invalidate and flush.

Chris and I also had some discussions about mark_busy and whether it
is appropriate to call from flush. But mark busy is a state which
should be derived from the 3 events (invalidate, flush, flip) we now
have by the users, like psr does by tracking relevant information in
psr.busy_frontbuffer_bits. DRRS (the only real use of mark_busy for
frontbuffer) needs to have similar logic. With that the overall
mark_busy in the core could be removed.

v8: Only when retiring gpu buffers only flush frontbuffer bits we
actually invalidated in a batch. Just for safety since before any
additional usage/invalidate we should always retire current rendering.
Suggested by Chris Wilson.

v9: Actually use intel_frontbuffer_flip in all appropriate places.
Spotted by Chris.

v10: Address more comments from Chris:
- Don't call _flip in set_base when the crtc is inactive, avoids redunancy
  in the modeset case with the initial enabling of all planes.
- Add comments explaining that the initial/final plane enable/disable
  still has work left to do before it's fully generic.

v11: Only invalidate for gtt/cpu access when writing. Spotted by Chris.

v12: s/_flush/_flip/ in intel_overlay.c per Chris' comment.

Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
---
 drivers/gpu/drm/i915/i915_drv.h            |  14 +++
 drivers/gpu/drm/i915/i915_gem.c            |  20 ++-
 drivers/gpu/drm/i915/i915_gem_execbuffer.c |   6 +-
 drivers/gpu/drm/i915/intel_display.c       | 191 +++++++++++++++++++++++++++--
 drivers/gpu/drm/i915/intel_drv.h           |  29 ++++-
 drivers/gpu/drm/i915/intel_overlay.c       |   3 +
 drivers/gpu/drm/i915/intel_sprite.c        |   4 +-
 7 files changed, 242 insertions(+), 25 deletions(-)

(limited to 'drivers/gpu/drm/i915/i915_gem_execbuffer.c')

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 17e90b371dbe..aa849f2a0d9f 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1330,6 +1330,17 @@ struct intel_pipe_crc {
 	wait_queue_head_t wq;
 };
 
+struct i915_frontbuffer_tracking {
+	struct mutex lock;
+
+	/*
+	 * Tracking bits for delayed frontbuffer flushing du to gpu activity or
+	 * scheduled flips.
+	 */
+	unsigned busy_bits;
+	unsigned flip_bits;
+};
+
 struct drm_i915_private {
 	struct drm_device *dev;
 	struct kmem_cache *slab;
@@ -1476,6 +1487,9 @@ struct drm_i915_private {
 	bool lvds_downclock_avail;
 	/* indicates the reduced downclock for LVDS*/
 	int lvds_downclock;
+
+	struct i915_frontbuffer_tracking fb_tracking;
+
 	u16 orig_clock;
 
 	bool mchbar_need_disable;
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index caed6621d71a..f6d123828926 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -1395,8 +1395,6 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data,
 		goto unlock;
 	}
 
-	intel_edp_psr_exit(dev);
-
 	/* Try to flush the object off the GPU without holding the lock.
 	 * We will repeat the flush holding the lock in the normal manner
 	 * to catch cases where we are gazumped.
@@ -1442,8 +1440,6 @@ i915_gem_sw_finish_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	intel_edp_psr_exit(dev);
-
 	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
@@ -2223,6 +2219,8 @@ i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj)
 			list_move_tail(&vma->mm_list, &vm->inactive_list);
 	}
 
+	intel_fb_obj_flush(obj, true);
+
 	list_del_init(&obj->ring_list);
 	obj->ring = NULL;
 
@@ -3552,6 +3550,8 @@ i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj)
 	old_write_domain = obj->base.write_domain;
 	obj->base.write_domain = 0;
 
+	intel_fb_obj_flush(obj, false);
+
 	trace_i915_gem_object_change_domain(obj,
 					    obj->base.read_domains,
 					    old_write_domain);
@@ -3573,6 +3573,8 @@ i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj,
 	old_write_domain = obj->base.write_domain;
 	obj->base.write_domain = 0;
 
+	intel_fb_obj_flush(obj, false);
+
 	trace_i915_gem_object_change_domain(obj,
 					    obj->base.read_domains,
 					    old_write_domain);
@@ -3626,6 +3628,9 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 		obj->dirty = 1;
 	}
 
+	if (write)
+		intel_fb_obj_invalidate(obj, NULL);
+
 	trace_i915_gem_object_change_domain(obj,
 					    old_read_domains,
 					    old_write_domain);
@@ -3962,6 +3967,9 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write)
 		obj->base.write_domain = I915_GEM_DOMAIN_CPU;
 	}
 
+	if (write)
+		intel_fb_obj_invalidate(obj, NULL);
+
 	trace_i915_gem_object_change_domain(obj,
 					    old_read_domains,
 					    old_write_domain);
@@ -4236,8 +4244,6 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	intel_edp_psr_exit(dev);
-
 	obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
 	if (&obj->base == NULL) {
 		ret = -ENOENT;
@@ -4937,6 +4943,8 @@ i915_gem_load(struct drm_device *dev)
 
 	dev_priv->mm.oom_notifier.notifier_call = i915_gem_shrinker_oom;
 	register_oom_notifier(&dev_priv->mm.oom_notifier);
+
+	mutex_init(&dev_priv->fb_tracking.lock);
 }
 
 void i915_gem_release(struct drm_device *dev, struct drm_file *file)
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 93d7f7246588..d815ef51a5ea 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -975,10 +975,8 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
 		if (obj->base.write_domain) {
 			obj->dirty = 1;
 			obj->last_write_seqno = intel_ring_get_seqno(ring);
-			/* check for potential scanout */
-			if (i915_gem_obj_ggtt_bound(obj) &&
-			    i915_gem_obj_to_ggtt(obj)->pin_count)
-				intel_mark_fb_busy(obj, ring);
+
+			intel_fb_obj_invalidate(obj, ring);
 
 			/* update for the implicit flush after a batch */
 			obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 9a3fa90660f4..ff6336793826 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -2756,6 +2756,9 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y,
 
 	dev_priv->display.update_primary_plane(crtc, fb, x, y);
 
+	if (intel_crtc->active)
+		intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_PRIMARY(pipe));
+
 	crtc->primary->fb = fb;
 	crtc->x = x;
 	crtc->y = y;
@@ -3950,6 +3953,13 @@ static void intel_crtc_enable_planes(struct drm_crtc *crtc)
 	mutex_lock(&dev->struct_mutex);
 	intel_update_fbc(dev);
 	mutex_unlock(&dev->struct_mutex);
+
+	/*
+	 * FIXME: Once we grow proper nuclear flip support out of this we need
+	 * to compute the mask of flip planes precisely. For the time being
+	 * consider this a flip from a NULL plane.
+	 */
+	intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_ALL_MASK(pipe));
 }
 
 static void intel_crtc_disable_planes(struct drm_crtc *crtc)
@@ -3972,6 +3982,13 @@ static void intel_crtc_disable_planes(struct drm_crtc *crtc)
 	intel_disable_planes(crtc);
 	intel_disable_primary_hw_plane(dev_priv, plane, pipe);
 
+	/*
+	 * FIXME: Once we grow proper nuclear flip support out of this we need
+	 * to compute the mask of flip planes precisely. For the time being
+	 * consider this a flip to a NULL plane.
+	 */
+	intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_ALL_MASK(pipe));
+
 	drm_vblank_off(dev, pipe);
 }
 
@@ -8212,6 +8229,8 @@ static int intel_crtc_cursor_set_obj(struct drm_crtc *crtc,
 		intel_crtc_update_cursor(crtc, intel_crtc->cursor_bo != NULL);
 	}
 
+	intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_CURSOR(pipe));
+
 	return 0;
 fail_unpin:
 	i915_gem_object_unpin_from_display_plane(obj);
@@ -8827,20 +8846,26 @@ out:
 }
 
 
-void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			struct intel_engine_cs *ring)
+/**
+ * intel_mark_fb_busy - mark given planes as busy
+ * @dev: DRM device
+ * @frontbuffer_bits: bits for the affected planes
+ * @ring: optional ring for asynchronous commands
+ *
+ * This function gets called every time the screen contents change. It can be
+ * used to keep e.g. the update rate at the nominal refresh rate with DRRS.
+ */
+static void intel_mark_fb_busy(struct drm_device *dev,
+			       unsigned frontbuffer_bits,
+			       struct intel_engine_cs *ring)
 {
-	struct drm_device *dev = obj->base.dev;
 	enum pipe pipe;
 
-	intel_edp_psr_exit(dev);
-
 	if (!i915.powersave)
 		return;
 
 	for_each_pipe(pipe) {
-		if (!(obj->frontbuffer_bits &
-		      INTEL_FRONTBUFFER_ALL_MASK(pipe)))
+		if (!(frontbuffer_bits & INTEL_FRONTBUFFER_ALL_MASK(pipe)))
 			continue;
 
 		intel_increase_pllclock(dev, pipe);
@@ -8849,6 +8874,150 @@ void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
 	}
 }
 
+/**
+ * intel_fb_obj_invalidate - invalidate frontbuffer object
+ * @obj: GEM object to invalidate
+ * @ring: set for asynchronous rendering
+ *
+ * This function gets called every time rendering on the given object starts and
+ * frontbuffer caching (fbc, low refresh rate for DRRS, panel self refresh) must
+ * be invalidated. If @ring is non-NULL any subsequent invalidation will be delayed
+ * until the rendering completes or a flip on this frontbuffer plane is
+ * scheduled.
+ */
+void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
+			     struct intel_engine_cs *ring)
+{
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	if (!obj->frontbuffer_bits)
+		return;
+
+	if (ring) {
+		mutex_lock(&dev_priv->fb_tracking.lock);
+		dev_priv->fb_tracking.busy_bits
+			|= obj->frontbuffer_bits;
+		dev_priv->fb_tracking.flip_bits
+			&= ~obj->frontbuffer_bits;
+		mutex_unlock(&dev_priv->fb_tracking.lock);
+	}
+
+	intel_mark_fb_busy(dev, obj->frontbuffer_bits, ring);
+
+	intel_edp_psr_exit(dev);
+}
+
+/**
+ * intel_frontbuffer_flush - flush frontbuffer
+ * @dev: DRM device
+ * @frontbuffer_bits: frontbuffer plane tracking bits
+ *
+ * This function gets called every time rendering on the given planes has
+ * completed and frontbuffer caching can be started again. Flushes will get
+ * delayed if they're blocked by some oustanding asynchronous rendering.
+ *
+ * Can be called without any locks held.
+ */
+void intel_frontbuffer_flush(struct drm_device *dev,
+			     unsigned frontbuffer_bits)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* Delay flushing when rings are still busy.*/
+	mutex_lock(&dev_priv->fb_tracking.lock);
+	frontbuffer_bits &= ~dev_priv->fb_tracking.busy_bits;
+	mutex_unlock(&dev_priv->fb_tracking.lock);
+
+	intel_mark_fb_busy(dev, frontbuffer_bits, NULL);
+
+	intel_edp_psr_exit(dev);
+}
+
+/**
+ * intel_fb_obj_flush - flush frontbuffer object
+ * @obj: GEM object to flush
+ * @retire: set when retiring asynchronous rendering
+ *
+ * This function gets called every time rendering on the given object has
+ * completed and frontbuffer caching can be started again. If @retire is true
+ * then any delayed flushes will be unblocked.
+ */
+void intel_fb_obj_flush(struct drm_i915_gem_object *obj,
+			bool retire)
+{
+	struct drm_device *dev = obj->base.dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	unsigned frontbuffer_bits;
+
+	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+	if (!obj->frontbuffer_bits)
+		return;
+
+	frontbuffer_bits = obj->frontbuffer_bits;
+
+	if (retire) {
+		mutex_lock(&dev_priv->fb_tracking.lock);
+		/* Filter out new bits since rendering started. */
+		frontbuffer_bits &= dev_priv->fb_tracking.busy_bits;
+
+		dev_priv->fb_tracking.busy_bits &= ~frontbuffer_bits;
+		mutex_unlock(&dev_priv->fb_tracking.lock);
+	}
+
+	intel_frontbuffer_flush(dev, frontbuffer_bits);
+}
+
+/**
+ * intel_frontbuffer_flip_prepare - prepare asnychronous frontbuffer flip
+ * @dev: DRM device
+ * @frontbuffer_bits: frontbuffer plane tracking bits
+ *
+ * This function gets called after scheduling a flip on @obj. The actual
+ * frontbuffer flushing will be delayed until completion is signalled with
+ * intel_frontbuffer_flip_complete. If an invalidate happens in between this
+ * flush will be cancelled.
+ *
+ * Can be called without any locks held.
+ */
+void intel_frontbuffer_flip_prepare(struct drm_device *dev,
+				    unsigned frontbuffer_bits)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	mutex_lock(&dev_priv->fb_tracking.lock);
+	dev_priv->fb_tracking.flip_bits
+		|= frontbuffer_bits;
+	mutex_unlock(&dev_priv->fb_tracking.lock);
+}
+
+/**
+ * intel_frontbuffer_flip_complete - complete asynchronous frontbuffer flush
+ * @dev: DRM device
+ * @frontbuffer_bits: frontbuffer plane tracking bits
+ *
+ * This function gets called after the flip has been latched and will complete
+ * on the next vblank. It will execute the fush if it hasn't been cancalled yet.
+ *
+ * Can be called without any locks held.
+ */
+void intel_frontbuffer_flip_complete(struct drm_device *dev,
+				     unsigned frontbuffer_bits)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	mutex_lock(&dev_priv->fb_tracking.lock);
+	/* Mask any cancelled flips. */
+	frontbuffer_bits &= dev_priv->fb_tracking.flip_bits;
+	dev_priv->fb_tracking.flip_bits &= ~frontbuffer_bits;
+	mutex_unlock(&dev_priv->fb_tracking.lock);
+
+	intel_frontbuffer_flush(dev, frontbuffer_bits);
+}
+
 static void intel_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -8876,6 +9045,7 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	struct intel_unpin_work *work =
 		container_of(__work, struct intel_unpin_work, work);
 	struct drm_device *dev = work->crtc->dev;
+	enum pipe pipe = to_intel_crtc(work->crtc)->pipe;
 
 	mutex_lock(&dev->struct_mutex);
 	intel_unpin_fb_obj(work->old_fb_obj);
@@ -8885,6 +9055,8 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	intel_update_fbc(dev);
 	mutex_unlock(&dev->struct_mutex);
 
+	intel_frontbuffer_flip_complete(dev, INTEL_FRONTBUFFER_PRIMARY(pipe));
+
 	BUG_ON(atomic_read(&to_intel_crtc(work->crtc)->unpin_work_count) == 0);
 	atomic_dec(&to_intel_crtc(work->crtc)->unpin_work_count);
 
@@ -9441,9 +9613,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	if (work == NULL)
 		return -ENOMEM;
 
-	/* Exit PSR early in page flip */
-	intel_edp_psr_exit(dev);
-
 	work->event = event;
 	work->crtc = crtc;
 	work->old_fb_obj = to_intel_framebuffer(old_fb)->obj;
@@ -9519,7 +9688,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 			  INTEL_FRONTBUFFER_PRIMARY(pipe));
 
 	intel_disable_fbc(dev);
-	intel_mark_fb_busy(obj, NULL);
+	intel_frontbuffer_flip_prepare(dev, INTEL_FRONTBUFFER_PRIMARY(pipe));
 	mutex_unlock(&dev->struct_mutex);
 
 	trace_i915_flip_request(intel_crtc->plane, obj);
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 5d20f719309a..bd0d10eeaf44 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -724,8 +724,33 @@ bool intel_has_pending_fb_unpin(struct drm_device *dev);
 int intel_pch_rawclk(struct drm_device *dev);
 int valleyview_cur_cdclk(struct drm_i915_private *dev_priv);
 void intel_mark_busy(struct drm_device *dev);
-void intel_mark_fb_busy(struct drm_i915_gem_object *obj,
-			struct intel_engine_cs *ring);
+void intel_fb_obj_invalidate(struct drm_i915_gem_object *obj,
+			     struct intel_engine_cs *ring);
+void intel_frontbuffer_flip_prepare(struct drm_device *dev,
+				    unsigned frontbuffer_bits);
+void intel_frontbuffer_flip_complete(struct drm_device *dev,
+				     unsigned frontbuffer_bits);
+void intel_frontbuffer_flush(struct drm_device *dev,
+			     unsigned frontbuffer_bits);
+/**
+ * intel_frontbuffer_flip - prepare frontbuffer flip
+ * @dev: DRM device
+ * @frontbuffer_bits: frontbuffer plane tracking bits
+ *
+ * This function gets called after scheduling a flip on @obj. This is for
+ * synchronous plane updates which will happen on the next vblank and which will
+ * not get delayed by pending gpu rendering.
+ *
+ * Can be called without any locks held.
+ */
+static inline
+void intel_frontbuffer_flip(struct drm_device *dev,
+			    unsigned frontbuffer_bits)
+{
+	intel_frontbuffer_flush(dev, frontbuffer_bits);
+}
+
+void intel_fb_obj_flush(struct drm_i915_gem_object *obj, bool retire);
 void intel_mark_idle(struct drm_device *dev);
 void intel_crtc_restore_mode(struct drm_crtc *crtc);
 void intel_crtc_update_dpms(struct drm_crtc *crtc);
diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c
index 99b6c142a095..307c2f1842b7 100644
--- a/drivers/gpu/drm/i915/intel_overlay.c
+++ b/drivers/gpu/drm/i915/intel_overlay.c
@@ -787,6 +787,9 @@ static int intel_overlay_do_put_image(struct intel_overlay *overlay,
 	overlay->old_vid_bo = overlay->vid_bo;
 	overlay->vid_bo = new_bo;
 
+	intel_frontbuffer_flip(dev,
+			       INTEL_FRONTBUFFER_OVERLAY(pipe));
+
 	return 0;
 
 out_unpin:
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 140bd8359f0e..0e3fd5c59e28 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -1034,6 +1034,8 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		else
 			intel_plane->disable_plane(plane, crtc);
 
+		intel_frontbuffer_flip(dev, INTEL_FRONTBUFFER_SPRITE(pipe));
+
 		if (!primary_was_enabled && primary_enabled)
 			intel_post_enable_primary(crtc);
 	}
@@ -1054,8 +1056,6 @@ intel_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 		mutex_unlock(&dev->struct_mutex);
 	}
 
-	intel_edp_psr_exit(dev);
-
 	return 0;
 }
 
-- 
cgit v1.2.3


From 821d66dd7ca3f59f68478d0d0ee1f28a1de74267 Mon Sep 17 00:00:00 2001
From: Oscar Mateo <oscar.mateo@intel.com>
Date: Thu, 3 Jul 2014 16:28:00 +0100
Subject: drm/i915: Emphasize that ctx->id is merely a user handle

This is an Execlists preparatory patch, since they make context ID become an
overloaded term:

- In the software, it was used to distinguish which context userspace was
  trying to use.
- In the BSpec, the term is used to describe the 20-bits long field the
  hardware uses to it to discriminate the contexts that are submitted to
  the ELSP and inform the driver about their current status (via Context
  Switch Interrupts and Context Status Buffers).

Initially, I tried to make the different meanings converge, but it proved
impossible:

- The software ctx->id is per-filp, while the hardware one needs to be
  globally unique.
- Also, we multiplex several backing states objects per intel_context,
  and all of them need unique HW IDs.
- I tried adding a per-filp ID and then composing the HW context ID as:
  ctx->id + file_priv->id + ring->id, but the fact that the hardware only
  uses 20-bits means we have to artificially limit the number of filps or
  contexts the userspace can create.

The ctx->user_handle renaming bits are done with this Cocci patch (plus
manual frobbing of the struct declaration):

    @@
    struct intel_context c;
    @@
    - (c).id
    + c.user_handle

    @@
    struct intel_context *c;
    @@
    - (c)->id
    + c->user_handle

Also, while we are at it, s/DEFAULT_CONTEXT_ID/DEFAULT_CONTEXT_HANDLE and
change the type to unsigned 32 bits.

v2: s/handle/user_handle and change the type to uint32_t as suggested by
Chris Wilson.

Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> (v1)
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
---
 drivers/gpu/drm/i915/i915_debugfs.c        |  2 +-
 drivers/gpu/drm/i915/i915_drv.h            |  6 +++---
 drivers/gpu/drm/i915/i915_gem_context.c    | 12 ++++++------
 drivers/gpu/drm/i915/i915_gem_execbuffer.c |  2 +-
 drivers/gpu/drm/i915/intel_uncore.c        |  2 +-
 5 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'drivers/gpu/drm/i915/i915_gem_execbuffer.c')

diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 31d5cdf30154..b3b56c46ef7e 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -1876,7 +1876,7 @@ static int per_file_ctx(int id, void *ptr, void *data)
 	if (i915_gem_context_is_default(ctx))
 		seq_puts(m, "  default context:\n");
 	else
-		seq_printf(m, "  context %d:\n", ctx->id);
+		seq_printf(m, "  context %d:\n", ctx->user_handle);
 	ppgtt->debug_dump(ppgtt, m);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index eeb3e4c27508..dcef38b1ea61 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -585,10 +585,10 @@ struct i915_ctx_hang_stats {
 };
 
 /* This must match up with the value previously used for execbuf2.rsvd1. */
-#define DEFAULT_CONTEXT_ID 0
+#define DEFAULT_CONTEXT_HANDLE 0
 struct intel_context {
 	struct kref ref;
-	int id;
+	int user_handle;
 	uint8_t remap_slice;
 	struct drm_i915_file_private *file_priv;
 	struct i915_ctx_hang_stats hang_stats;
@@ -2469,7 +2469,7 @@ static inline void i915_gem_context_unreference(struct intel_context *ctx)
 
 static inline bool i915_gem_context_is_default(const struct intel_context *c)
 {
-	return c->id == DEFAULT_CONTEXT_ID;
+	return c->user_handle == DEFAULT_CONTEXT_HANDLE;
 }
 
 int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 6b7648228757..de72a2859f32 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -276,14 +276,14 @@ __create_hw_context(struct drm_device *dev,
 	/* Default context will never have a file_priv */
 	if (file_priv != NULL) {
 		ret = idr_alloc(&file_priv->context_idr, ctx,
-				DEFAULT_CONTEXT_ID, 0, GFP_KERNEL);
+				DEFAULT_CONTEXT_HANDLE, 0, GFP_KERNEL);
 		if (ret < 0)
 			goto err_out;
 	} else
-		ret = DEFAULT_CONTEXT_ID;
+		ret = DEFAULT_CONTEXT_HANDLE;
 
 	ctx->file_priv = file_priv;
-	ctx->id = ret;
+	ctx->user_handle = ret;
 	/* NB: Mark all slices as needing a remap so that when the context first
 	 * loads it will restore whatever remap state already exists. If there
 	 * is no remap info, it will be a NOP. */
@@ -793,7 +793,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 	if (IS_ERR(ctx))
 		return PTR_ERR(ctx);
 
-	args->ctx_id = ctx->id;
+	args->ctx_id = ctx->user_handle;
 	DRM_DEBUG_DRIVER("HW context %d created\n", args->ctx_id);
 
 	return 0;
@@ -807,7 +807,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 	struct intel_context *ctx;
 	int ret;
 
-	if (args->ctx_id == DEFAULT_CONTEXT_ID)
+	if (args->ctx_id == DEFAULT_CONTEXT_HANDLE)
 		return -ENOENT;
 
 	ret = i915_mutex_lock_interruptible(dev);
@@ -820,7 +820,7 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 		return PTR_ERR(ctx);
 	}
 
-	idr_remove(&ctx->file_priv->context_idr, ctx->id);
+	idr_remove(&ctx->file_priv->context_idr, ctx->user_handle);
 	i915_gem_context_unreference(ctx);
 	mutex_unlock(&dev->struct_mutex);
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index d815ef51a5ea..c97178ebf2b5 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -938,7 +938,7 @@ i915_gem_validate_context(struct drm_device *dev, struct drm_file *file,
 	struct intel_context *ctx = NULL;
 	struct i915_ctx_hang_stats *hs;
 
-	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_ID)
+	if (ring->id != RCS && ctx_id != DEFAULT_CONTEXT_HANDLE)
 		return ERR_PTR(-EINVAL);
 
 	ctx = i915_gem_context_get(file->driver_priv, ctx_id);
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 29145df8ef64..e0f0843569a6 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -1010,7 +1010,7 @@ int i915_get_reset_stats_ioctl(struct drm_device *dev,
 	if (args->flags || args->pad)
 		return -EINVAL;
 
-	if (args->ctx_id == DEFAULT_CONTEXT_ID && !capable(CAP_SYS_ADMIN))
+	if (args->ctx_id == DEFAULT_CONTEXT_HANDLE && !capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
-- 
cgit v1.2.3


From 78382593e921c88371abd019aca8978db3248a8f Mon Sep 17 00:00:00 2001
From: Oscar Mateo <oscar.mateo@intel.com>
Date: Thu, 3 Jul 2014 16:28:05 +0100
Subject: drm/i915: Extract the actual workload submission mechanism from
 execbuffer

So that we isolate the legacy ringbuffer submission mechanism, which becomes
a good candidate to be abstracted away. This is prep-work for Execlists (which
will its own workload submission mechanism).

No functional changes.

Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
---
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 298 ++++++++++++++++-------------
 1 file changed, 162 insertions(+), 136 deletions(-)

(limited to 'drivers/gpu/drm/i915/i915_gem_execbuffer.c')

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index c97178ebf2b5..60998fc4e5b2 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1026,6 +1026,163 @@ i915_reset_gen7_sol_offsets(struct drm_device *dev,
 	return 0;
 }
 
+static int
+legacy_ringbuffer_submission(struct drm_device *dev, struct drm_file *file,
+			     struct intel_engine_cs *ring,
+			     struct intel_context *ctx,
+			     struct drm_i915_gem_execbuffer2 *args,
+			     struct list_head *vmas,
+			     struct drm_i915_gem_object *batch_obj,
+			     u64 exec_start, u32 flags)
+{
+	struct drm_clip_rect *cliprects = NULL;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	u64 exec_len;
+	int instp_mode;
+	u32 instp_mask;
+	int i, ret = 0;
+
+	if (args->num_cliprects != 0) {
+		if (ring != &dev_priv->ring[RCS]) {
+			DRM_DEBUG("clip rectangles are only valid with the render ring\n");
+			return -EINVAL;
+		}
+
+		if (INTEL_INFO(dev)->gen >= 5) {
+			DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
+			return -EINVAL;
+		}
+
+		if (args->num_cliprects > UINT_MAX / sizeof(*cliprects)) {
+			DRM_DEBUG("execbuf with %u cliprects\n",
+				  args->num_cliprects);
+			return -EINVAL;
+		}
+
+		cliprects = kcalloc(args->num_cliprects,
+				    sizeof(*cliprects),
+				    GFP_KERNEL);
+		if (cliprects == NULL) {
+			ret = -ENOMEM;
+			goto error;
+		}
+
+		if (copy_from_user(cliprects,
+				   to_user_ptr(args->cliprects_ptr),
+				   sizeof(*cliprects)*args->num_cliprects)) {
+			ret = -EFAULT;
+			goto error;
+		}
+	} else {
+		if (args->DR4 == 0xffffffff) {
+			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
+			args->DR4 = 0;
+		}
+
+		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
+			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
+			return -EINVAL;
+		}
+	}
+
+	ret = i915_gem_execbuffer_move_to_gpu(ring, vmas);
+	if (ret)
+		goto error;
+
+	ret = i915_switch_context(ring, ctx);
+	if (ret)
+		goto error;
+
+	instp_mode = args->flags & I915_EXEC_CONSTANTS_MASK;
+	instp_mask = I915_EXEC_CONSTANTS_MASK;
+	switch (instp_mode) {
+	case I915_EXEC_CONSTANTS_REL_GENERAL:
+	case I915_EXEC_CONSTANTS_ABSOLUTE:
+	case I915_EXEC_CONSTANTS_REL_SURFACE:
+		if (instp_mode != 0 && ring != &dev_priv->ring[RCS]) {
+			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
+			ret = -EINVAL;
+			goto error;
+		}
+
+		if (instp_mode != dev_priv->relative_constants_mode) {
+			if (INTEL_INFO(dev)->gen < 4) {
+				DRM_DEBUG("no rel constants on pre-gen4\n");
+				ret = -EINVAL;
+				goto error;
+			}
+
+			if (INTEL_INFO(dev)->gen > 5 &&
+			    instp_mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
+				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
+				ret = -EINVAL;
+				goto error;
+			}
+
+			/* The HW changed the meaning on this bit on gen6 */
+			if (INTEL_INFO(dev)->gen >= 6)
+				instp_mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
+		}
+		break;
+	default:
+		DRM_DEBUG("execbuf with unknown constants: %d\n", instp_mode);
+		ret = -EINVAL;
+		goto error;
+	}
+
+	if (ring == &dev_priv->ring[RCS] &&
+			instp_mode != dev_priv->relative_constants_mode) {
+		ret = intel_ring_begin(ring, 4);
+		if (ret)
+			goto error;
+
+		intel_ring_emit(ring, MI_NOOP);
+		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
+		intel_ring_emit(ring, INSTPM);
+		intel_ring_emit(ring, instp_mask << 16 | instp_mode);
+		intel_ring_advance(ring);
+
+		dev_priv->relative_constants_mode = instp_mode;
+	}
+
+	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
+		ret = i915_reset_gen7_sol_offsets(dev, ring);
+		if (ret)
+			goto error;
+	}
+
+	exec_len = args->batch_len;
+	if (cliprects) {
+		for (i = 0; i < args->num_cliprects; i++) {
+			ret = i915_emit_box(dev, &cliprects[i],
+					    args->DR1, args->DR4);
+			if (ret)
+				goto error;
+
+			ret = ring->dispatch_execbuffer(ring,
+							exec_start, exec_len,
+							flags);
+			if (ret)
+				goto error;
+		}
+	} else {
+		ret = ring->dispatch_execbuffer(ring,
+						exec_start, exec_len,
+						flags);
+		if (ret)
+			return ret;
+	}
+
+	trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
+
+	i915_gem_execbuffer_move_to_active(vmas, ring);
+	i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
+
+error:
+	kfree(cliprects);
+	return ret;
+}
+
 /**
  * Find one BSD ring to dispatch the corresponding BSD command.
  * The Ring ID is returned.
@@ -1085,14 +1242,13 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct eb_vmas *eb;
 	struct drm_i915_gem_object *batch_obj;
-	struct drm_clip_rect *cliprects = NULL;
 	struct intel_engine_cs *ring;
 	struct intel_context *ctx;
 	struct i915_address_space *vm;
 	const u32 ctx_id = i915_execbuffer2_get_context_id(*args);
-	u64 exec_start = args->batch_start_offset, exec_len;
-	u32 mask, flags;
-	int ret, mode, i;
+	u64 exec_start = args->batch_start_offset;
+	u32 flags;
+	int ret;
 	bool need_relocs;
 
 	if (!i915_gem_check_execbuffer(args))
@@ -1136,87 +1292,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	mode = args->flags & I915_EXEC_CONSTANTS_MASK;
-	mask = I915_EXEC_CONSTANTS_MASK;
-	switch (mode) {
-	case I915_EXEC_CONSTANTS_REL_GENERAL:
-	case I915_EXEC_CONSTANTS_ABSOLUTE:
-	case I915_EXEC_CONSTANTS_REL_SURFACE:
-		if (mode != 0 && ring != &dev_priv->ring[RCS]) {
-			DRM_DEBUG("non-0 rel constants mode on non-RCS\n");
-			return -EINVAL;
-		}
-
-		if (mode != dev_priv->relative_constants_mode) {
-			if (INTEL_INFO(dev)->gen < 4) {
-				DRM_DEBUG("no rel constants on pre-gen4\n");
-				return -EINVAL;
-			}
-
-			if (INTEL_INFO(dev)->gen > 5 &&
-			    mode == I915_EXEC_CONSTANTS_REL_SURFACE) {
-				DRM_DEBUG("rel surface constants mode invalid on gen5+\n");
-				return -EINVAL;
-			}
-
-			/* The HW changed the meaning on this bit on gen6 */
-			if (INTEL_INFO(dev)->gen >= 6)
-				mask &= ~I915_EXEC_CONSTANTS_REL_SURFACE;
-		}
-		break;
-	default:
-		DRM_DEBUG("execbuf with unknown constants: %d\n", mode);
-		return -EINVAL;
-	}
-
 	if (args->buffer_count < 1) {
 		DRM_DEBUG("execbuf with %d buffers\n", args->buffer_count);
 		return -EINVAL;
 	}
 
-	if (args->num_cliprects != 0) {
-		if (ring != &dev_priv->ring[RCS]) {
-			DRM_DEBUG("clip rectangles are only valid with the render ring\n");
-			return -EINVAL;
-		}
-
-		if (INTEL_INFO(dev)->gen >= 5) {
-			DRM_DEBUG("clip rectangles are only valid on pre-gen5\n");
-			return -EINVAL;
-		}
-
-		if (args->num_cliprects > UINT_MAX / sizeof(*cliprects)) {
-			DRM_DEBUG("execbuf with %u cliprects\n",
-				  args->num_cliprects);
-			return -EINVAL;
-		}
-
-		cliprects = kcalloc(args->num_cliprects,
-				    sizeof(*cliprects),
-				    GFP_KERNEL);
-		if (cliprects == NULL) {
-			ret = -ENOMEM;
-			goto pre_mutex_err;
-		}
-
-		if (copy_from_user(cliprects,
-				   to_user_ptr(args->cliprects_ptr),
-				   sizeof(*cliprects)*args->num_cliprects)) {
-			ret = -EFAULT;
-			goto pre_mutex_err;
-		}
-	} else {
-		if (args->DR4 == 0xffffffff) {
-			DRM_DEBUG("UXA submitting garbage DR4, fixing up\n");
-			args->DR4 = 0;
-		}
-
-		if (args->DR1 || args->DR4 || args->cliprects_ptr) {
-			DRM_DEBUG("0 cliprects but dirt in cliprects fields\n");
-			return -EINVAL;
-		}
-	}
-
 	intel_runtime_pm_get(dev_priv);
 
 	ret = i915_mutex_lock_interruptible(dev);
@@ -1320,63 +1400,11 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 	else
 		exec_start += i915_gem_obj_offset(batch_obj, vm);
 
-	ret = i915_gem_execbuffer_move_to_gpu(ring, &eb->vmas);
+	ret = legacy_ringbuffer_submission(dev, file, ring, ctx,
+			args, &eb->vmas, batch_obj, exec_start, flags);
 	if (ret)
 		goto err;
 
-	ret = i915_switch_context(ring, ctx);
-	if (ret)
-		goto err;
-
-	if (ring == &dev_priv->ring[RCS] &&
-	    mode != dev_priv->relative_constants_mode) {
-		ret = intel_ring_begin(ring, 4);
-		if (ret)
-				goto err;
-
-		intel_ring_emit(ring, MI_NOOP);
-		intel_ring_emit(ring, MI_LOAD_REGISTER_IMM(1));
-		intel_ring_emit(ring, INSTPM);
-		intel_ring_emit(ring, mask << 16 | mode);
-		intel_ring_advance(ring);
-
-		dev_priv->relative_constants_mode = mode;
-	}
-
-	if (args->flags & I915_EXEC_GEN7_SOL_RESET) {
-		ret = i915_reset_gen7_sol_offsets(dev, ring);
-		if (ret)
-			goto err;
-	}
-
-
-	exec_len = args->batch_len;
-	if (cliprects) {
-		for (i = 0; i < args->num_cliprects; i++) {
-			ret = i915_emit_box(dev, &cliprects[i],
-					    args->DR1, args->DR4);
-			if (ret)
-				goto err;
-
-			ret = ring->dispatch_execbuffer(ring,
-							exec_start, exec_len,
-							flags);
-			if (ret)
-				goto err;
-		}
-	} else {
-		ret = ring->dispatch_execbuffer(ring,
-						exec_start, exec_len,
-						flags);
-		if (ret)
-			goto err;
-	}
-
-	trace_i915_gem_ring_dispatch(ring, intel_ring_get_seqno(ring), flags);
-
-	i915_gem_execbuffer_move_to_active(&eb->vmas, ring);
-	i915_gem_execbuffer_retire_commands(dev, file, ring, batch_obj);
-
 err:
 	/* the request owns the ref now */
 	i915_gem_context_unreference(ctx);
@@ -1385,8 +1413,6 @@ err:
 	mutex_unlock(&dev->struct_mutex);
 
 pre_mutex_err:
-	kfree(cliprects);
-
 	/* intel_gpu_busy should also get a ref, so it will free when the device
 	 * is really idle. */
 	intel_runtime_pm_put(dev_priv);
-- 
cgit v1.2.3


From 48ba813701eb14b3008edefef4a0789b328e278c Mon Sep 17 00:00:00 2001
From: David Herrmann <dh.herrmann@gmail.com>
Date: Tue, 22 Jul 2014 18:46:09 +0200
Subject: drm: drop redundant drm_file->is_master

The drm_file->is_master field is redundant as it's equivalent to:
    drm_file->master && drm_file->master == drm_file->minor->master

1) "=>"
  Whenever we set drm_file->is_master, we also set:
      drm_file->minor->master = drm_file->master;

  Whenever we clear drm_file->is_master, we also call:
      drm_master_put(&drm_file->minor->master);
  which implicitly clears it to NULL.

2) "<="
  minor->master cannot be set if it is non-NULL. Therefore, it stays as
  is unless a file drops it.

  If minor->master is NULL, it is only set by places that also adjust
  drm_file->is_master.

Therefore, we can safely drop is_master and replace it by an inline helper
that matches:
    drm_file->master && drm_file->master == drm_file->minor->master

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
---
 drivers/gpu/drm/drm_crtc.c                 |  2 +-
 drivers/gpu/drm/drm_drv.c                  |  2 +-
 drivers/gpu/drm/drm_fops.c                 |  4 +---
 drivers/gpu/drm/drm_lock.c                 |  2 +-
 drivers/gpu/drm/drm_stub.c                 | 10 +++-------
 drivers/gpu/drm/i915/i915_gem_execbuffer.c |  4 ++--
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c        |  2 +-
 drivers/staging/imx-drm/imx-drm-core.c     |  2 +-
 include/drm/drmP.h                         | 19 ++++++++++++++++---
 9 files changed, 27 insertions(+), 20 deletions(-)

(limited to 'drivers/gpu/drm/i915/i915_gem_execbuffer.c')

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 3c4a62169f28..670b5eb13c87 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -3243,7 +3243,7 @@ int drm_mode_getfb(struct drm_device *dev,
 	r->bpp = fb->bits_per_pixel;
 	r->pitch = fb->pitches[0];
 	if (fb->funcs->create_handle) {
-		if (file_priv->is_master || capable(CAP_SYS_ADMIN) ||
+		if (drm_is_master(file_priv) || capable(CAP_SYS_ADMIN) ||
 		    drm_is_control_client(file_priv)) {
 			ret = fb->funcs->create_handle(fb, file_priv,
 						       &r->handle);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 0cc182745e31..7aa8121d1323 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -307,7 +307,7 @@ static int drm_ioctl_permit(u32 flags, struct drm_file *file_priv)
 		return -EACCES;
 
 	/* MASTER is only for master or control clients */
-	if (unlikely((flags & DRM_MASTER) && !file_priv->is_master &&
+	if (unlikely((flags & DRM_MASTER) && !drm_is_master(file_priv) &&
 		     !drm_is_control_client(file_priv)))
 		return -EACCES;
 
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 55143f7747f3..53435e07fa8d 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -233,7 +233,6 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 			goto out_close;
 		}
 
-		priv->is_master = 1;
 		/* take another reference for the copy in the local file priv */
 		priv->master = drm_master_get(priv->minor->master);
 		priv->authenticated = 1;
@@ -461,7 +460,7 @@ int drm_release(struct inode *inode, struct file *filp)
 
 	mutex_lock(&dev->master_mutex);
 
-	if (file_priv->is_master) {
+	if (drm_is_master(file_priv)) {
 		struct drm_master *master = file_priv->master;
 		struct drm_file *temp;
 
@@ -497,7 +496,6 @@ int drm_release(struct inode *inode, struct file *filp)
 	/* drop the master reference held by the file priv */
 	if (file_priv->master)
 		drm_master_put(&file_priv->master);
-	file_priv->is_master = 0;
 	mutex_unlock(&dev->master_mutex);
 
 	mutex_lock(&dev->struct_mutex);
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index f6452682141b..786401cd5f60 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -111,7 +111,7 @@ int drm_lock(struct drm_device *dev, void *data, struct drm_file *file_priv)
 	/* don't set the block all signals on the master process for now 
 	 * really probably not the correct answer but lets us debug xkb
  	 * xserver for now */
-	if (!file_priv->is_master) {
+	if (!drm_is_master(file_priv)) {
 		sigemptyset(&dev->sigmask);
 		sigaddset(&dev->sigmask, SIGSTOP);
 		sigaddset(&dev->sigmask, SIGTSTP);
diff --git a/drivers/gpu/drm/drm_stub.c b/drivers/gpu/drm/drm_stub.c
index 233ea208c9fe..18c9b3d8201e 100644
--- a/drivers/gpu/drm/drm_stub.c
+++ b/drivers/gpu/drm/drm_stub.c
@@ -177,7 +177,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
 	int ret = 0;
 
 	mutex_lock(&dev->master_mutex);
-	if (file_priv->is_master)
+	if (drm_is_master(file_priv))
 		goto out_unlock;
 
 	if (file_priv->minor->master) {
@@ -191,13 +191,10 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
 	}
 
 	file_priv->minor->master = drm_master_get(file_priv->master);
-	file_priv->is_master = 1;
 	if (dev->driver->master_set) {
 		ret = dev->driver->master_set(dev, file_priv, false);
-		if (unlikely(ret != 0)) {
-			file_priv->is_master = 0;
+		if (unlikely(ret != 0))
 			drm_master_put(&file_priv->minor->master);
-		}
 	}
 
 out_unlock:
@@ -211,7 +208,7 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	int ret = -EINVAL;
 
 	mutex_lock(&dev->master_mutex);
-	if (!file_priv->is_master)
+	if (!drm_is_master(file_priv))
 		goto out_unlock;
 
 	if (!file_priv->minor->master)
@@ -221,7 +218,6 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	if (dev->driver->master_drop)
 		dev->driver->master_drop(dev, file_priv, false);
 	drm_master_put(&file_priv->minor->master);
-	file_priv->is_master = 0;
 
 out_unlock:
 	mutex_unlock(&dev->master_mutex);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 60998fc4e5b2..2dd19da6b4b3 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1260,7 +1260,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	flags = 0;
 	if (args->flags & I915_EXEC_SECURE) {
-		if (!file->is_master || !capable(CAP_SYS_ADMIN))
+		if (!drm_is_master(file) || !capable(CAP_SYS_ADMIN))
 		    return -EPERM;
 
 		flags |= I915_DISPATCH_SECURE;
@@ -1369,7 +1369,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		ret = i915_parse_cmds(ring,
 				      batch_obj,
 				      args->batch_start_offset,
-				      file->is_master);
+				      drm_is_master(file));
 		if (ret)
 			goto err;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 18b54acacfbb..63c4d6f0281e 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -990,7 +990,7 @@ static struct vmw_master *vmw_master_check(struct drm_device *dev,
 	if (unlikely(ret != 0))
 		return ERR_PTR(-ERESTARTSYS);
 
-	if (file_priv->is_master) {
+	if (drm_is_master(file_priv)) {
 		mutex_unlock(&dev->master_mutex);
 		return NULL;
 	}
diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c
index 6f54ff4f9372..72913b299047 100644
--- a/drivers/staging/imx-drm/imx-drm-core.c
+++ b/drivers/staging/imx-drm/imx-drm-core.c
@@ -182,7 +182,7 @@ static void imx_drm_driver_preclose(struct drm_device *drm,
 {
 	int i;
 
-	if (!file->is_master)
+	if (!drm_is_master(file))
 		return;
 
 	for (i = 0; i < MAX_CRTC; i++)
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 80889982d196..6ede53712d7b 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -387,8 +387,6 @@ struct drm_prime_file_private {
 struct drm_file {
 	unsigned always_authenticated :1;
 	unsigned authenticated :1;
-	/* Whether we're master for a minor. Protected by master_mutex */
-	unsigned is_master :1;
 	/* true when the client has asked us to expose stereo 3D mode flags */
 	unsigned stereo_allowed :1;
 	/*
@@ -1034,7 +1032,7 @@ struct drm_device {
 	/** \name Locks */
 	/*@{ */
 	struct mutex struct_mutex;	/**< For others */
-	struct mutex master_mutex;      /**< For drm_minor::master and drm_file::is_master */
+	struct mutex master_mutex;      /**< For drm_minor::master */
 	/*@} */
 
 	/** \name Usage Counters */
@@ -1172,6 +1170,21 @@ static inline bool drm_is_primary_client(const struct drm_file *file_priv)
 	return file_priv->minor->type == DRM_MINOR_LEGACY;
 }
 
+/**
+ * drm_is_master() - Check whether a DRM open-file is DRM-Master
+ * @file: DRM open-file context
+ *
+ * This checks whether a DRM open-file context is owner of the master context
+ * attached to it. If a file owns a master context, it's called DRM-Master.
+ * Per DRM device, only one such file can be DRM-Master at a time.
+ *
+ * Returns: True if the file is DRM-Master, otherwise false.
+ */
+static inline bool drm_is_master(const struct drm_file *file)
+{
+	return file->master && file->master == file->minor->master;
+}
+
 /******************************************************************/
 /** \name Internal function definitions */
 /*@{*/
-- 
cgit v1.2.3


From 7963e9db1b1f842fdc53309baa8714d38e9f5681 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@gmail.com>
Date: Fri, 8 Aug 2014 07:30:53 +1000
Subject: Revert "drm: drop redundant drm_file->is_master"

This reverts commit 48ba813701eb14b3008edefef4a0789b328e278c.

Thanks to Chris:
"drm_file->is_master is not synomous with having drm_file->master ==
drm_file->minor->master. This is because drm_file->master is the same
for all drm_files of the same generation and so when there is a master,
every drm_file believes itself to be the master. Confusion ensues and
things go pear shaped when one file is closed and there is no master
anymore."

Conflicts:
	drivers/gpu/drm/drm_drv.c
	drivers/gpu/drm/drm_stub.c
---
 drivers/gpu/drm/drm_crtc.c                 |  2 +-
 drivers/gpu/drm/drm_drv.c                  | 10 +++++++---
 drivers/gpu/drm/drm_fops.c                 |  4 +++-
 drivers/gpu/drm/drm_ioctl.c                |  2 +-
 drivers/gpu/drm/drm_lock.c                 |  2 +-
 drivers/gpu/drm/i915/i915_gem_execbuffer.c |  4 ++--
 drivers/gpu/drm/vmwgfx/vmwgfx_drv.c        |  2 +-
 drivers/staging/imx-drm/imx-drm-core.c     |  2 +-
 include/drm/drmP.h                         | 19 +++----------------
 9 files changed, 20 insertions(+), 27 deletions(-)

(limited to 'drivers/gpu/drm/i915/i915_gem_execbuffer.c')

diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index ca8bb1bc92a0..fa2be249999c 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -3244,7 +3244,7 @@ int drm_mode_getfb(struct drm_device *dev,
 	r->bpp = fb->bits_per_pixel;
 	r->pitch = fb->pitches[0];
 	if (fb->funcs->create_handle) {
-		if (drm_is_master(file_priv) || capable(CAP_SYS_ADMIN) ||
+		if (file_priv->is_master || capable(CAP_SYS_ADMIN) ||
 		    drm_is_control_client(file_priv)) {
 			ret = fb->funcs->create_handle(fb, file_priv,
 						       &r->handle);
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 92bc6b1d9646..3242e208c0d0 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -179,7 +179,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
 	int ret = 0;
 
 	mutex_lock(&dev->master_mutex);
-	if (drm_is_master(file_priv))
+	if (file_priv->is_master)
 		goto out_unlock;
 
 	if (file_priv->minor->master) {
@@ -193,10 +193,13 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
 	}
 
 	file_priv->minor->master = drm_master_get(file_priv->master);
+	file_priv->is_master = 1;
 	if (dev->driver->master_set) {
 		ret = dev->driver->master_set(dev, file_priv, false);
-		if (unlikely(ret != 0))
+		if (unlikely(ret != 0)) {
+			file_priv->is_master = 0;
 			drm_master_put(&file_priv->minor->master);
+		}
 	}
 
 out_unlock:
@@ -210,7 +213,7 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	int ret = -EINVAL;
 
 	mutex_lock(&dev->master_mutex);
-	if (!drm_is_master(file_priv))
+	if (!file_priv->is_master)
 		goto out_unlock;
 
 	if (!file_priv->minor->master)
@@ -220,6 +223,7 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	if (dev->driver->master_drop)
 		dev->driver->master_drop(dev, file_priv, false);
 	drm_master_put(&file_priv->minor->master);
+	file_priv->is_master = 0;
 
 out_unlock:
 	mutex_unlock(&dev->master_mutex);
diff --git a/drivers/gpu/drm/drm_fops.c b/drivers/gpu/drm/drm_fops.c
index 4b060942cb3c..79d5221c6e41 100644
--- a/drivers/gpu/drm/drm_fops.c
+++ b/drivers/gpu/drm/drm_fops.c
@@ -194,6 +194,7 @@ static int drm_open_helper(struct file *filp, struct drm_minor *minor)
 			goto out_close;
 		}
 
+		priv->is_master = 1;
 		/* take another reference for the copy in the local file priv */
 		priv->master = drm_master_get(priv->minor->master);
 		priv->authenticated = 1;
@@ -425,7 +426,7 @@ int drm_release(struct inode *inode, struct file *filp)
 
 	mutex_lock(&dev->master_mutex);
 
-	if (drm_is_master(file_priv)) {
+	if (file_priv->is_master) {
 		struct drm_master *master = file_priv->master;
 
 		/**
@@ -453,6 +454,7 @@ int drm_release(struct inode *inode, struct file *filp)
 	/* drop the master reference held by the file priv */
 	if (file_priv->master)
 		drm_master_put(&file_priv->master);
+	file_priv->is_master = 0;
 	mutex_unlock(&dev->master_mutex);
 
 	if (dev->driver->postclose)
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index d3d1a8c72e98..40be746b7e68 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -607,7 +607,7 @@ static int drm_ioctl_permit(u32 flags, struct drm_file *file_priv)
 		return -EACCES;
 
 	/* MASTER is only for master or control clients */
-	if (unlikely((flags & DRM_MASTER) && !drm_is_master(file_priv) &&
+	if (unlikely((flags & DRM_MASTER) && !file_priv->is_master &&
 		     !drm_is_control_client(file_priv)))
 		return -EACCES;
 
diff --git a/drivers/gpu/drm/drm_lock.c b/drivers/gpu/drm/drm_lock.c
index ea1572596578..e26b59e385ff 100644
--- a/drivers/gpu/drm/drm_lock.c
+++ b/drivers/gpu/drm/drm_lock.c
@@ -112,7 +112,7 @@ int drm_lock(struct drm_device *dev, void *data, struct drm_file *file_priv)
 	/* don't set the block all signals on the master process for now 
 	 * really probably not the correct answer but lets us debug xkb
  	 * xserver for now */
-	if (!drm_is_master(file_priv)) {
+	if (!file_priv->is_master) {
 		sigemptyset(&dev->sigmask);
 		sigaddset(&dev->sigmask, SIGSTOP);
 		sigaddset(&dev->sigmask, SIGTSTP);
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 2dd19da6b4b3..60998fc4e5b2 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1260,7 +1260,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 
 	flags = 0;
 	if (args->flags & I915_EXEC_SECURE) {
-		if (!drm_is_master(file) || !capable(CAP_SYS_ADMIN))
+		if (!file->is_master || !capable(CAP_SYS_ADMIN))
 		    return -EPERM;
 
 		flags |= I915_DISPATCH_SECURE;
@@ -1369,7 +1369,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
 		ret = i915_parse_cmds(ring,
 				      batch_obj,
 				      args->batch_start_offset,
-				      drm_is_master(file));
+				      file->is_master);
 		if (ret)
 			goto err;
 
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 63c4d6f0281e..18b54acacfbb 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -990,7 +990,7 @@ static struct vmw_master *vmw_master_check(struct drm_device *dev,
 	if (unlikely(ret != 0))
 		return ERR_PTR(-ERESTARTSYS);
 
-	if (drm_is_master(file_priv)) {
+	if (file_priv->is_master) {
 		mutex_unlock(&dev->master_mutex);
 		return NULL;
 	}
diff --git a/drivers/staging/imx-drm/imx-drm-core.c b/drivers/staging/imx-drm/imx-drm-core.c
index 72913b299047..6f54ff4f9372 100644
--- a/drivers/staging/imx-drm/imx-drm-core.c
+++ b/drivers/staging/imx-drm/imx-drm-core.c
@@ -182,7 +182,7 @@ static void imx_drm_driver_preclose(struct drm_device *drm,
 {
 	int i;
 
-	if (!drm_is_master(file))
+	if (!file->is_master)
 		return;
 
 	for (i = 0; i < MAX_CRTC; i++)
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index a57646382086..a12fbbac373c 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -384,6 +384,8 @@ struct drm_prime_file_private {
 /** File private data */
 struct drm_file {
 	unsigned authenticated :1;
+	/* Whether we're master for a minor. Protected by master_mutex */
+	unsigned is_master :1;
 	/* true when the client has asked us to expose stereo 3D mode flags */
 	unsigned stereo_allowed :1;
 	/*
@@ -1020,7 +1022,7 @@ struct drm_device {
 	/** \name Locks */
 	/*@{ */
 	struct mutex struct_mutex;	/**< For others */
-	struct mutex master_mutex;      /**< For drm_minor::master */
+	struct mutex master_mutex;      /**< For drm_minor::master and drm_file::is_master */
 	/*@} */
 
 	/** \name Usage Counters */
@@ -1158,21 +1160,6 @@ static inline bool drm_is_primary_client(const struct drm_file *file_priv)
 	return file_priv->minor->type == DRM_MINOR_LEGACY;
 }
 
-/**
- * drm_is_master() - Check whether a DRM open-file is DRM-Master
- * @file: DRM open-file context
- *
- * This checks whether a DRM open-file context is owner of the master context
- * attached to it. If a file owns a master context, it's called DRM-Master.
- * Per DRM device, only one such file can be DRM-Master at a time.
- *
- * Returns: True if the file is DRM-Master, otherwise false.
- */
-static inline bool drm_is_master(const struct drm_file *file)
-{
-	return file->master && file->master == file->minor->master;
-}
-
 /******************************************************************/
 /** \name Internal function definitions */
 /*@{*/
-- 
cgit v1.2.3