drm/msm/gem: Evict active GEM objects when necessary

If we are under enough memory pressure, we should stall waiting for
active buffers to become idle in order to evict.

v2: Check for __GFP_ATOMIC before blocking

Signed-off-by: Rob Clark <robdclark@chromium.org>
Patchwork: https://patchwork.freedesktop.org/patch/496135/
Link: https://lore.kernel.org/r/20220802155152.1727594-14-robdclark@gmail.com

drivers/gpu/drm/msm/msm_gem_shrinker.c

@@ -24,6 +24,13 @@ static bool can_swap(void)
 	return enable_eviction && get_nr_swap_pages() > 0;
 }
 
+static bool can_block(struct shrink_control *sc)
+{
+	if (sc->gfp_mask & __GFP_ATOMIC)
+		return false;
+	return current_is_kswapd() || (sc->gfp_mask & __GFP_RECLAIM);
+}
+
 static unsigned long
 msm_gem_shrinker_count(struct shrinker *shrinker, struct shrink_control *sc)
 {
@@ -65,26 +72,65 @@ evict(struct drm_gem_object *obj)
 	return true;
 }
 
+static bool
+wait_for_idle(struct drm_gem_object *obj)
+{
+	enum dma_resv_usage usage = dma_resv_usage_rw(true);
+	return dma_resv_wait_timeout(obj->resv, usage, false, 1000) > 0;
+}
+
+static bool
+active_purge(struct drm_gem_object *obj)
+{
+	if (!wait_for_idle(obj))
+		return false;
+
+	return purge(obj);
+}
+
+static bool
+active_evict(struct drm_gem_object *obj)
+{
+	if (!wait_for_idle(obj))
+		return false;
+
+	return evict(obj);
+}
+
 static unsigned long
 msm_gem_shrinker_scan(struct shrinker *shrinker, struct shrink_control *sc)
 {
 	struct msm_drm_private *priv =
 		container_of(shrinker, struct msm_drm_private, shrinker);
+	struct {
+		struct drm_gem_lru *lru;
+		bool (*shrink)(struct drm_gem_object *obj);
+		bool cond;
+		unsigned long freed;
+	} stages[] = {
+		/* Stages of progressively more aggressive/expensive reclaim: */
+		{ &priv->lru.dontneed, purge,        true },
+		{ &priv->lru.willneed, evict,        can_swap() },
+		{ &priv->lru.dontneed, active_purge, can_block(sc) },
+		{ &priv->lru.willneed, active_evict, can_swap() && can_block(sc) },
+	};
 	long nr = sc->nr_to_scan;
-	unsigned long freed, purged, evicted = 0;
+	unsigned long freed = 0;
 
-	purged = drm_gem_lru_scan(&priv->lru.dontneed, nr, purge);
+	for (unsigned i = 0; (nr > 0) && (i < ARRAY_SIZE(stages)); i++) {
-	nr -= purged;
+		if (!stages[i].cond)
-
+			continue;
-	if (can_swap() && nr > 0) {
+		stages[i].freed =
-		evicted = drm_gem_lru_scan(&priv->lru.willneed, nr, evict);
+			drm_gem_lru_scan(stages[i].lru, nr, stages[i].shrink);
-		nr -= evicted;
+		nr -= stages[i].freed;
+		freed += stages[i].freed;
 	}
 
-	freed = purged + evicted;
+	if (freed) {
-
+		trace_msm_gem_shrink(sc->nr_to_scan, stages[0].freed,
-	if (freed)
+				     stages[1].freed, stages[2].freed,
-		trace_msm_gem_shrink(sc->nr_to_scan, purged, evicted);
+				     stages[3].freed);
+	}
 
 	return (freed > 0) ? freed : SHRINK_STOP;
 }

drivers/gpu/drm/msm/msm_gpu_trace.h

@@ -116,22 +116,26 @@ TRACE_EVENT(msm_gmu_freq_change,
 
 
 TRACE_EVENT(msm_gem_shrink,
-		TP_PROTO(u32 nr_to_scan, u32 purged, u32 evicted),
+		TP_PROTO(u32 nr_to_scan, u32 purged, u32 evicted,
-		TP_ARGS(nr_to_scan, purged, evicted),
+			 u32 active_purged, u32 active_evicted),
+		TP_ARGS(nr_to_scan, purged, evicted, active_purged, active_evicted),
 		TP_STRUCT__entry(
 			__field(u32, nr_to_scan)
 			__field(u32, purged)
 			__field(u32, evicted)
+			__field(u32, active_purged)
+			__field(u32, active_evicted)
 			),
 		TP_fast_assign(
 			__entry->nr_to_scan = nr_to_scan;
 			__entry->purged = purged;
 			__entry->evicted = evicted;
+			__entry->active_purged = active_purged;
+			__entry->active_evicted = active_evicted;
 			),
-		TP_printk("nr_to_scan=%u pages, purged=%u pages, evicted=%u pages",
+		TP_printk("nr_to_scan=%u pg, purged=%u pg, evicted=%u pg, active_purged=%u pg, active_evicted=%u pg",
-			  __entry->nr_to_scan,
+			  __entry->nr_to_scan, __entry->purged, __entry->evicted,
-			  __entry->purged,
+			  __entry->active_purged, __entry->active_evicted)
-			  __entry->evicted)
 );