[GPU] Ownership-transfer-based RT cache, 3x3 resolution scaling

The ROV path is also disabled by default because of lower performance

src/xenia/gpu/shared_memory.h

@@ -74,7 +74,8 @@ class SharedMemory {
   // Checks if the range has been updated, uploads new data if needed and
   // ensures the host GPU memory backing the range are resident. Returns true if
   // the range has been fully updated and is usable.
-  bool RequestRange(uint32_t start, uint32_t length);
+  bool RequestRange(uint32_t start, uint32_t length,
+                    bool* any_data_resolved_out = nullptr);
 
   // Marks the range and, if not exact_range, potentially its surroundings
   // (to up to the first GPU-written page, as an access violation exception
@@ -90,7 +91,7 @@ class SharedMemory {
   // be called, to make sure, if the GPU writes don't overwrite *everything* in
   // the pages they touch, the CPU data is properly loaded to the unmodified
   // regions in those pages.
-  void RangeWrittenByGpu(uint32_t start, uint32_t length);
+  void RangeWrittenByGpu(uint32_t start, uint32_t length, bool is_resolve);
 
  protected:
   SharedMemory(Memory& memory);
@@ -119,7 +120,8 @@ class SharedMemory {
                                                 uint32_t length_allocations);
 
   // Mark the memory range as updated and protect it.
-  void MakeRangeValid(uint32_t start, uint32_t length, bool written_by_gpu);
+  void MakeRangeValid(uint32_t start, uint32_t length, bool written_by_gpu,
+                      bool written_by_gpu_resolve);
 
   // Uploads a range of host pages - only called if host GPU sparse memory
   // allocation succeeded if needed. While uploading, MarkRangeValid must be
@@ -188,6 +190,9 @@ class SharedMemory {
     // Subset of valid pages - whether each page in the GPU buffer contains data
     // that was written on the GPU, thus should not be invalidated spuriously.
     uint64_t valid_and_gpu_written;
+    // Subset of valid_and_gpu_written - whether each page in the GPU buffer
+    // contains data written specifically by resolving from EDRAM.
+    uint64_t valid_and_gpu_resolved;
   };
   // Flags for each 64 system pages, interleaved as blocks, so bit scan can be
   // used to quickly extract ranges.