Add workaround for glTex(Sub)Image3D on Intel Macs

gpu/command_buffer/service/texture_manager.h

Index: gpu/command_buffer/service/texture_manager.h
diff --git a/gpu/command_buffer/service/texture_manager.h b/gpu/command_buffer/service/texture_manager.h
index 6781483049a2fcfa50576a4cfd6324be4b2ac353..11c8b6305050b6bff3fb6abacfa2c52b5bddf6e5 100644
--- a/gpu/command_buffer/service/texture_manager.h
+++ b/gpu/command_buffer/service/texture_manager.h
@@ -710,7 +710,9 @@ struct DecoderTextureState {
         unpack_alignment_workaround_with_unpack_buffer(
             workarounds.unpack_alignment_workaround_with_unpack_buffer),
         unpack_overlapping_rows_separately_unpack_buffer(
-            workarounds.unpack_overlapping_rows_separately_unpack_buffer) {}
+            workarounds.unpack_overlapping_rows_separately_unpack_buffer),
+        unpack_image_height_workaround_with_unpack_buffer(
+            workarounds.unpack_image_height_workaround_with_unpack_buffer) {}
 
   // This indicates all the following texSubImage*D calls that are part of the
   // failed texImage*D call should be ignored. The client calls have a lock
@@ -727,6 +729,7 @@ struct DecoderTextureState {
   bool force_cube_complete;
   bool unpack_alignment_workaround_with_unpack_buffer;
   bool unpack_overlapping_rows_separately_unpack_buffer;
+  bool unpack_image_height_workaround_with_unpack_buffer;
 };
 
 // This class keeps track of the textures and their sizes so we can do NPOT and
@@ -1172,6 +1175,12 @@ class GPU_EXPORT TextureManager : public base::trace_event::MemoryDumpProvider {
                                        const DoTexSubImageArguments& args,
                                        const PixelStoreParams& unpack_params);
 
+  void DoTexSubImageLayerByLayerWorkaround(
+      DecoderTextureState* texture_state,
+      ContextState* state,
+      const DoTexSubImageArguments& args,
+      const PixelStoreParams& unpack_params);
+
   void DoCubeMapWorkaround(
       DecoderTextureState* texture_state,
       ContextState* state,