NV: Expose the nvdisp_disp0 device and a weak reference to the nvdrv:a service.

NVFlinger will call into the nvdisp_disp0 device to perform screen flips, bypassing the ioctl interface.

We now have the address of the framebuffer to draw, we just need to actually put it on the screen.

src/core/hle/service/vi/vi.cpp

@@ -6,6 +6,7 @@
 #include "common/scope_exit.h"
 #include "core/core_timing.h"
 #include "core/hle/ipc_helpers.h"
+#include "core/hle/service/nvdrv/nvdrv_a.h"
 #include "core/hle/service/vi/vi.h"
 #include "core/hle/service/vi/vi_m.h"
 
@@ -743,7 +744,19 @@ void NVFlinger::Compose() {
             continue;
         }
 
-        // TODO(Subv): Send the buffer to the GPU for drawing.
+        auto& igbp_buffer = buffer->igbp_buffer;
+
+        // Now send the buffer to the GPU for drawing.
+        auto nvdrv = NVDRV::nvdrv_a.lock();
+        ASSERT(nvdrv);
+
+        // TODO(Subv): Support more than just disp0. The display device selection is probably based
+        // on which display we're drawing (Default, Internal, External, etc)
+        auto nvdisp = nvdrv->GetDevice<NVDRV::nvdisp_disp0>("/dev/nvdisp_disp0");
+        ASSERT(nvdisp);
+
+        nvdisp->flip(igbp_buffer.gpu_buffer_id, igbp_buffer.offset, igbp_buffer.format,
+                     igbp_buffer.width, igbp_buffer.height, igbp_buffer.stride);
 
         buffer_queue->ReleaseBuffer(buffer->slot);
     }