drm/i915: Use multiple VMs -- the point of no return

As with processes which run on the CPU, the goal of multiple VMs is to
provide process isolation. Specific to GEN, there is also the ability to
map more objects per process (2GB each instead of 2Gb-2k total).

For the most part, all the pipes have been laid, and all we need to do
is remove asserts and actually start changing address spaces with the
context switch. Since prior to this we've converted the setting of the
page tables to a streamed version, this is quite easy.

One important thing to point out (since it'd been hotly contested) is
that with this patch, every context created will have it's own address
space (provided the HW can do it).

v2: Disable BDW on rebase

NOTE: I tried to make this commit as small as possible. I needed one
place where I could "turn everything on" and that is here. It could be
split into finer commits, but I didn't really see much point.

Cc: Eric Anholt <eric@anholt.net>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>

1 files changed, 35 insertions, 25 deletions

diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 165a5c7d9424..ebe0f67eac08 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -288,17 +288,15 @@ i915_gem_create_context(struct drm_device *dev,
 				DRM_DEBUG_DRIVER("Couldn't pin %d\n", ret);
 				goto err_destroy;
 			}
+
+			ctx->vm = &dev_priv->mm.aliasing_ppgtt->base;
 		}
 	} else if (USES_ALIASING_PPGTT(dev)) {
 		/* For platforms which only have aliasing PPGTT, we fake the
 		 * address space and refcounting. */
-		kref_get(&dev_priv->mm.aliasing_ppgtt->ref);
-	}
-
-	/* TODO: Until full ppgtt... */
-	if (USES_ALIASING_PPGTT(dev))
 		ctx->vm = &dev_priv->mm.aliasing_ppgtt->base;
-	else
+		kref_get(&dev_priv->mm.aliasing_ppgtt->ref);
+	} else
 		ctx->vm = &dev_priv->gtt.base;
 
 	return ctx;
@@ -500,7 +498,7 @@ int i915_gem_context_open(struct drm_device *dev, struct drm_file *file)
 
 	mutex_lock(&dev->struct_mutex);
 	file_priv->private_default_ctx =
-		i915_gem_create_context(dev, file_priv, false);
+		i915_gem_create_context(dev, file_priv, USES_FULL_PPGTT(dev));
 	mutex_unlock(&dev->struct_mutex);
 
 	if (IS_ERR(file_priv->private_default_ctx)) {
@@ -587,6 +585,7 @@ static int do_switch(struct intel_ring_buffer *ring,
 {
 	struct drm_i915_private *dev_priv = ring->dev->dev_private;
 	struct i915_hw_context *from = ring->last_context;
+	struct i915_hw_ppgtt *ppgtt = ctx_to_ppgtt(to);
 	u32 hw_flags = 0;
 	int ret, i;
 
@@ -598,17 +597,15 @@ static int do_switch(struct intel_ring_buffer *ring,
 	if (from == to && from->last_ring == ring && !to->remap_slice)
 		return 0;
 
-	if (ring != &dev_priv->ring[RCS]) {
-		if (from)
-			i915_gem_context_unreference(from);
-		goto done;
+	/* Trying to pin first makes error handling easier. */
+	if (ring == &dev_priv->ring[RCS]) {
+		ret = i915_gem_obj_ggtt_pin(to->obj,
+					    get_context_alignment(ring->dev),
+					    false, false);
+		if (ret)
+			return ret;
 	}
 
-	ret = i915_gem_obj_ggtt_pin(to->obj, get_context_alignment(ring->dev),
-				    false, false);
-	if (ret)
-		return ret;
-
 	/*
 	 * Pin can switch back to the default context if we end up calling into
 	 * evict_everything - as a last ditch gtt defrag effort that also
@@ -616,6 +613,18 @@ static int do_switch(struct intel_ring_buffer *ring,
 	 */
 	from = ring->last_context;
 
+	if (USES_FULL_PPGTT(ring->dev)) {
+		ret = ppgtt->switch_mm(ppgtt, ring, false);
+		if (ret)
+			goto unpin_out;
+	}
+
+	if (ring != &dev_priv->ring[RCS]) {
+		if (from)
+			i915_gem_context_unreference(from);
+		goto done;
+	}
+
 	/*
 	 * Clear this page out of any CPU caches for coherent swap-in/out. Note
 	 * that thanks to write = false in this call and us not setting any gpu
@@ -625,10 +634,8 @@ static int do_switch(struct intel_ring_buffer *ring,
 	 * XXX: We need a real interface to do this instead of trickery.
 	 */
 	ret = i915_gem_object_set_to_gtt_domain(to->obj, false);
-	if (ret) {
-		i915_gem_object_ggtt_unpin(to->obj);
-		return ret;
-	}
+	if (ret)
+		goto unpin_out;
 
 	if (!to->obj->has_global_gtt_mapping) {
 		struct i915_vma *vma = i915_gem_obj_to_vma(to->obj,
@@ -640,10 +647,8 @@ static int do_switch(struct intel_ring_buffer *ring,
 		hw_flags |= MI_RESTORE_INHIBIT;
 
 	ret = mi_set_context(ring, to, hw_flags);
-	if (ret) {
-		i915_gem_object_ggtt_unpin(to->obj);
-		return ret;
-	}
+	if (ret)
+		goto unpin_out;
 
 	for (i = 0; i < MAX_L3_SLICES; i++) {
 		if (!(to->remap_slice & (1<<i)))
@@ -688,6 +693,11 @@ done:
 	to->last_ring = ring;
 
 	return 0;
+
+unpin_out:
+	if (ring->id == RCS)
+		i915_gem_object_ggtt_unpin(to->obj);
+	return ret;
 }
 
 /**
@@ -736,7 +746,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		return ret;
 
-	ctx = i915_gem_create_context(dev, file_priv, false);
+	ctx = i915_gem_create_context(dev, file_priv, USES_FULL_PPGTT(dev));
 	mutex_unlock(&dev->struct_mutex);
 	if (IS_ERR(ctx))
 		return PTR_ERR(ctx);