[PATCH 4/4] hohum

[Chris Wilson]

---
 .../gpu/drm/i915/gem/i915_gem_execbuffer.c    | 99 ++++++++++++++++---
 1 file changed, 86 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 414859fa2673..3cbd6bc5beb0 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -975,20 +975,90 @@ static inline struct i915_ggtt *cache_to_ggtt(struct reloc_cache *cache)
 	return &i915->ggtt;
 }
 
+static int reloc_gpu_chain(struct reloc_cache *cache)
+{
+	struct intel_gt_buffer_pool_node *pool;
+	struct i915_request *rq = cache->rq;
+	struct i915_vma *batch;
+	u32 *cmd;
+	int err;
+
+	pool = intel_gt_get_buffer_pool(rq->engine->gt, PAGE_SIZE);
+	if (IS_ERR(pool))
+		return PTR_ERR(pool);
+
+	batch = i915_vma_instance(pool->obj, rq->context->vm, NULL);
+	if (IS_ERR(batch)) {
+		err = PTR_ERR(batch);
+		goto out_pool;
+	}
+
+	err = i915_vma_pin(batch, 0, 0, PIN_USER | PIN_NONBLOCK);
+	if (err)
+		goto out_pool;
+
+	cmd = cache->rq_cmd + cache->rq_size;
+	if (cache->gen >= 8) {
+		*cmd++ = MI_BATCH_BUFFER_START_GEN8;
+		*cmd++ = lower_32_bits(batch->node.start);
+		*cmd++ = upper_32_bits(batch->node.start);
+	} else {
+		*cmd++ = MI_BATCH_BUFFER_START;
+		*cmd++ = lower_32_bits(batch->node.start);
+	}
+	i915_gem_object_flush_map(rq->batch->obj);
+	i915_gem_object_unpin_map(rq->batch->obj);
+	rq->batch = NULL;
+
+	err = intel_gt_buffer_pool_mark_active(pool, rq);
+	if (err == 0) {
+		i915_vma_lock(batch);
+		err = i915_request_await_object(rq, batch->obj, false);
+		if (err == 0)
+			err = i915_vma_move_to_active(batch, rq, 0);
+		i915_vma_unlock(batch);
+	}
+	i915_vma_unpin(batch);
+	if (err)
+		goto out_pool;
+
+	cmd = i915_gem_object_pin_map(pool->obj,
+				      cache->has_llc ?
+				      I915_MAP_FORCE_WB :
+				      I915_MAP_FORCE_WC);
+	if (IS_ERR(cmd)) {
+		err = PTR_ERR(cmd);
+		goto out_pool;
+	}
+
+	cache->rq_cmd = cmd;
+	cache->rq_size = 0;
+
+	/* Return with batch mapping (cmd) still pinned */
+	rq->batch = batch;
+	goto out_pool;
+
+out_pool:
+	intel_gt_buffer_pool_put(pool);
+	return err;
+}
+
 static void reloc_gpu_flush(struct reloc_cache *cache)
 {
-	struct drm_i915_gem_object *obj = cache->rq->batch->obj;
+	struct i915_request *rq = fetch_and_zero(&cache->rq);
 
-	GEM_BUG_ON(cache->rq_size >= obj->base.size / sizeof(u32));
-	cache->rq_cmd[cache->rq_size] = MI_BATCH_BUFFER_END;
+	if (rq->batch) {
+		struct drm_i915_gem_object *obj = rq->batch->obj;
 
-	__i915_gem_object_flush_map(obj, 0, sizeof(u32) * (cache->rq_size + 1));
-	i915_gem_object_unpin_map(obj);
+		GEM_BUG_ON(cache->rq_size >= obj->base.size / sizeof(u32));
+		cache->rq_cmd[cache->rq_size] = MI_BATCH_BUFFER_END;
 
-	intel_gt_chipset_flush(cache->rq->engine->gt);
+		__i915_gem_object_flush_map(obj, 0, sizeof(u32) * (cache->rq_size + 1));
+		i915_gem_object_unpin_map(obj);
+	}
 
-	i915_request_add(cache->rq);
-	cache->rq = NULL;
+	intel_gt_chipset_flush(rq->engine->gt);
+	i915_request_add(rq);
 }
 
 static void reloc_cache_reset(struct reloc_cache *cache)
@@ -1280,13 +1350,9 @@ static u32 *reloc_gpu(struct i915_execbuffer *eb,
 {
 	struct reloc_cache *cache = &eb->reloc_cache;
 	u32 *cmd;
-
-	if (cache->rq_size > PAGE_SIZE/sizeof(u32) - (len + 1))
-		reloc_gpu_flush(cache);
+	int err;
 
 	if (unlikely(!cache->rq)) {
-		int err;
-
 		if (!intel_engine_can_store_dword(eb->engine))
 			return ERR_PTR(-ENODEV);
 
@@ -1295,6 +1361,13 @@ static u32 *reloc_gpu(struct i915_execbuffer *eb,
 			return ERR_PTR(err);
 	}
 
+	if (cache->rq_size > PAGE_SIZE/sizeof(u32) - (len + 1) - 4) {
+		err = reloc_gpu_chain(cache);
+		if (unlikely(err))
+			return ERR_PTR(err);
+	}
+
+	GEM_BUG_ON(cache->rq_size + len > PAGE_SIZE/sizeof(u32));
 	cmd = cache->rq_cmd + cache->rq_size;
 	cache->rq_size += len;
 
-- 
2.20.1