video_core: skip empty fences; one lock per cache per fence; single-pass overlap subtract

Switch 2 titles signal a fence for each compute dispatch's release (5,000-7,000 a
second in title 1). With fast GPU fences the callback runs at once, so a fence with
no downloads, queries or queued operations does nothing: those (~20%) are no longer
made (one in 64 still is, for the operations fences carry). Each cache's lock is now
taken once per fence (the flush check took both together first). The overlap range
set's Subtract restarted its search after every erase (quadratic); the fencing thread
runs it for every downloaded range while holding the buffer cache. The GPU profiler
counts fences, skipped fences and buffer download bytes.

2 files changed, 13 insertions, 0 deletions

diff --git a/src/video_core/query_cache/query_cache.h b/src/video_core/query_cache/query_cache.h
index 245ab1996c..60ca50eb1f 100644
--- a/src/video_core/query_cache/query_cache.h
+++ b/src/video_core/query_cache/query_cache.h
@@ -520,6 +520,16 @@ bool QueryCacheBase<Traits>::HasUncommittedFlushes() const {
 }
 
 template <typename Traits>
+bool QueryCacheBase<Traits>::HasPendingFenceWork() const {
+    bool result = false;
+    impl->ForEachStreamer([&result](StreamerInterface* streamer) {
+        result |= streamer->HasUnsyncedQueries() || streamer->HasPendingSync();
+        return result;
+    });
+    return result;
+}
+
+template <typename Traits>
 bool QueryCacheBase<Traits>::ShouldWaitAsyncFlushes() {
     std::scoped_lock lk(impl->flush_guard);
     return !impl->flushes_pending.empty() && impl->flushes_pending.front() != 0ULL;
diff --git a/src/video_core/query_cache/query_cache_base.h b/src/video_core/query_cache/query_cache_base.h
index d73e9039f7..5a60e42986 100644
--- a/src/video_core/query_cache/query_cache_base.h
+++ b/src/video_core/query_cache/query_cache_base.h
@@ -109,6 +109,9 @@ public:
 
     bool HasUncommittedFlushes() const;
 
+    /// Whether a new fence has queries to flush or host writes to sync for this cache.
+    bool HasPendingFenceWork() const;
+
     bool ShouldWaitAsyncFlushes();
 
     void PopAsyncFlushes();