drm/i915: Allocate a status page for each timeline

Allocate a page for use as a status page by a group of timelines, as we
only need a dword of storage for each (rounded up to the cacheline for
safety) we can pack multiple timelines into the same page. Each timeline
will then be able to track its own HW seqno.

v2: Reuse the common per-engine HWSP for the solitary ringbuffer
timeline, so that we do not have to emit (using per-gen specialised
vfuncs) the breadcrumb into the distinct timeline HWSP and instead can
keep on using the common MI_STORE_DWORD_INDEX. However, to maintain the
sleight-of-hand for the global/per-context seqno switchover, we will
store both temporarily (and so use a custom offset for the shared timeline
HWSP until the switch over).

v3: Keep things simple and allocate a page for each timeline, page
sharing comes next.

v4: I was caught repeating the same MI_STORE_DWORD_IMM over and over
again in selftests.

v5: And caught red handed copying create timeline + check.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190128181812.22804-3-chris@chris-wilson.co.uk

1 files changed, 112 insertions, 9 deletions

diff --git a/drivers/gpu/drm/i915/i915_timeline.c b/drivers/gpu/drm/i915/i915_timeline.c
index 84550f17d3df..8d5792311a8f 100644
--- a/drivers/gpu/drm/i915/i915_timeline.c
+++ b/drivers/gpu/drm/i915/i915_timeline.c
@@ -9,28 +9,78 @@
 #include "i915_timeline.h"
 #include "i915_syncmap.h"
 
-void i915_timeline_init(struct drm_i915_private *i915,
-			struct i915_timeline *timeline,
-			const char *name)
+static struct i915_vma *__hwsp_alloc(struct drm_i915_private *i915)
+{
+	struct drm_i915_gem_object *obj;
+	struct i915_vma *vma;
+
+	obj = i915_gem_object_create_internal(i915, PAGE_SIZE);
+	if (IS_ERR(obj))
+		return ERR_CAST(obj);
+
+	i915_gem_object_set_cache_coherency(obj, I915_CACHE_LLC);
+
+	vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL);
+	if (IS_ERR(vma))
+		i915_gem_object_put(obj);
+
+	return vma;
+}
+
+static int hwsp_alloc(struct i915_timeline *timeline)
+{
+	struct i915_vma *vma;
+
+	vma = __hwsp_alloc(timeline->i915);
+	if (IS_ERR(vma))
+		return PTR_ERR(vma);
+
+	timeline->hwsp_ggtt = vma;
+	timeline->hwsp_offset = 0;
+
+	return 0;
+}
+
+int i915_timeline_init(struct drm_i915_private *i915,
+		       struct i915_timeline *timeline,
+		       const char *name,
+		       struct i915_vma *global_hwsp)
 {
 	struct i915_gt_timelines *gt = &i915->gt.timelines;
+	void *vaddr;
+	int err;
 
 	/*
 	 * Ideally we want a set of engines on a single leaf as we expect
 	 * to mostly be tracking synchronisation between engines. It is not
 	 * a huge issue if this is not the case, but we may want to mitigate
 	 * any page crossing penalties if they become an issue.
+	 *
+	 * Called during early_init before we know how many engines there are.
 	 */
 	BUILD_BUG_ON(KSYNCMAP < I915_NUM_ENGINES);
 
 	timeline->i915 = i915;
 	timeline->name = name;
+	timeline->pin_count = 0;
+
+	if (global_hwsp) {
+		timeline->hwsp_ggtt = i915_vma_get(global_hwsp);
+		timeline->hwsp_offset = I915_GEM_HWS_SEQNO_ADDR;
+	} else {
+		err = hwsp_alloc(timeline);
+		if (err)
+			return err;
+	}
 
-	mutex_lock(&gt->mutex);
-	list_add(&timeline->link, &gt->list);
-	mutex_unlock(&gt->mutex);
+	vaddr = i915_gem_object_pin_map(timeline->hwsp_ggtt->obj, I915_MAP_WB);
+	if (IS_ERR(vaddr)) {
+		i915_vma_put(timeline->hwsp_ggtt);
+		return PTR_ERR(vaddr);
+	}
 
-	/* Called during early_init before we know how many engines there are */
+	timeline->hwsp_seqno =
+		memset(vaddr + timeline->hwsp_offset, 0, CACHELINE_BYTES);
 
 	timeline->fence_context = dma_fence_context_alloc(1);
 
@@ -40,6 +90,12 @@ void i915_timeline_init(struct drm_i915_private *i915,
 	INIT_LIST_HEAD(&timeline->requests);
 
 	i915_syncmap_init(&timeline->sync);
+
+	mutex_lock(&gt->mutex);
+	list_add(&timeline->link, &gt->list);
+	mutex_unlock(&gt->mutex);
+
+	return 0;
 }
 
 void i915_timelines_init(struct drm_i915_private *i915)
@@ -85,6 +141,7 @@ void i915_timeline_fini(struct i915_timeline *timeline)
 {
 	struct i915_gt_timelines *gt = &timeline->i915->gt.timelines;
 
+	GEM_BUG_ON(timeline->pin_count);
 	GEM_BUG_ON(!list_empty(&timeline->requests));
 
 	i915_syncmap_free(&timeline->sync);
@@ -92,23 +149,69 @@ void i915_timeline_fini(struct i915_timeline *timeline)
 	mutex_lock(&gt->mutex);
 	list_del(&timeline->link);
 	mutex_unlock(&gt->mutex);
+
+	i915_gem_object_unpin_map(timeline->hwsp_ggtt->obj);
+	i915_vma_put(timeline->hwsp_ggtt);
 }
 
 struct i915_timeline *
-i915_timeline_create(struct drm_i915_private *i915, const char *name)
+i915_timeline_create(struct drm_i915_private *i915,
+		     const char *name,
+		     struct i915_vma *global_hwsp)
 {
 	struct i915_timeline *timeline;
+	int err;
 
 	timeline = kzalloc(sizeof(*timeline), GFP_KERNEL);
 	if (!timeline)
 		return ERR_PTR(-ENOMEM);
 
-	i915_timeline_init(i915, timeline, name);
+	err = i915_timeline_init(i915, timeline, name, global_hwsp);
+	if (err) {
+		kfree(timeline);
+		return ERR_PTR(err);
+	}
+
 	kref_init(&timeline->kref);
 
 	return timeline;
 }
 
+int i915_timeline_pin(struct i915_timeline *tl)
+{
+	int err;
+
+	if (tl->pin_count++)
+		return 0;
+	GEM_BUG_ON(!tl->pin_count);
+
+	err = i915_vma_pin(tl->hwsp_ggtt, 0, 0, PIN_GLOBAL | PIN_HIGH);
+	if (err)
+		goto unpin;
+
+	return 0;
+
+unpin:
+	tl->pin_count = 0;
+	return err;
+}
+
+void i915_timeline_unpin(struct i915_timeline *tl)
+{
+	GEM_BUG_ON(!tl->pin_count);
+	if (--tl->pin_count)
+		return;
+
+	/*
+	 * Since this timeline is idle, all bariers upon which we were waiting
+	 * must also be complete and so we can discard the last used barriers
+	 * without loss of information.
+	 */
+	i915_syncmap_free(&tl->sync);
+
+	__i915_vma_unpin(tl->hwsp_ggtt);
+}
+
 void __i915_timeline_free(struct kref *kref)
 {
 	struct i915_timeline *timeline =