diff --git a/app/src/main/cpp/skyline/gpu/interconnect/command_executor.cpp b/app/src/main/cpp/skyline/gpu/interconnect/command_executor.cpp
index c259c634..01b3f7ec 100644
--- a/app/src/main/cpp/skyline/gpu/interconnect/command_executor.cpp
+++ b/app/src/main/cpp/skyline/gpu/interconnect/command_executor.cpp
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: MPL-2.0
 // Copyright © 2021 Skyline Team and Contributors (https://github.com/skyline-emu/)
 
+#include <condition_variable>
+#include <mutex>
 #include <range/v3/view.hpp>
 #include <adrenotools/driver.h>
 #include <common/settings.h>
@@ -555,7 +557,7 @@ namespace skyline::gpu::interconnect {
         }
     }
 
-    void CommandExecutor::Submit(std::function<void()> &&callback) {
+    void CommandExecutor::Submit(std::function<void()> &&callback, bool wait) {
         for (const auto &flushCallback : flushCallbacks)
             flushCallback();
 
@@ -581,6 +583,21 @@ namespace skyline::gpu::interconnect {
             callback();
 
         ResetInternal();
+
+        if (wait) {
+            std::condition_variable cv;
+            std::mutex mutex;
+            bool gpuDone{};
+
+            waiterThread.Queue(nullptr, [&cv, &mutex, &gpuDone] {
+                std::scoped_lock lock{mutex};
+                gpuDone = true;
+                cv.notify_one();
+            });
+
+            std::unique_lock lock{mutex};
+            cv.wait(lock, [&gpuDone] { return gpuDone; });
+        }
     }
 
     void CommandExecutor::LockPreserve() {
diff --git a/app/src/main/cpp/skyline/gpu/interconnect/command_executor.h b/app/src/main/cpp/skyline/gpu/interconnect/command_executor.h
index 4fd4d0bd..d1225b76 100644
--- a/app/src/main/cpp/skyline/gpu/interconnect/command_executor.h
+++ b/app/src/main/cpp/skyline/gpu/interconnect/command_executor.h
@@ -301,8 +301,9 @@ namespace skyline::gpu::interconnect {
         /**
          * @brief Execute all the nodes and submit the resulting command buffer to the GPU
          * @param callback A function to call upon GPU completion of the submission
+         * @param wait Whether to wait synchronously for GPU completion of the submit
          */
-        void Submit(std::function<void()> &&callback = {});
+        void Submit(std::function<void()> &&callback = {}, bool wait = false);
 
         /**
          * @brief Locks all preserve attached buffers/textures
diff --git a/app/src/main/cpp/skyline/soc/gm20b/gpfifo.cpp b/app/src/main/cpp/skyline/soc/gm20b/gpfifo.cpp
index dfa48b73..49fe7b94 100644
--- a/app/src/main/cpp/skyline/soc/gm20b/gpfifo.cpp
+++ b/app/src/main/cpp/skyline/soc/gm20b/gpfifo.cpp
@@ -1,7 +1,9 @@
 // SPDX-License-Identifier: MPL-2.0
 // Copyright © 2020 Skyline Team and Contributors (https://github.com/skyline-emu/)
 
+#include <gpu.h>
 #include <common/signal.h>
+#include <common/settings.h>
 #include <loader/loader.h>
 #include <kernel/types/KProcess.h>
 #include <soc.h>
@@ -156,7 +158,7 @@ namespace skyline::soc::gm20b {
     void ChannelGpfifo::Process(GpEntry gpEntry) {
         // Submit if required by the GpEntry, this is needed as some games dynamically generate pushbuffer contents
         if (gpEntry.sync == GpEntry::Sync::Wait)
-            channelCtx.executor.Submit();
+            channelCtx.executor.Submit({}, *state.settings->useDirectMemoryImport);
 
         if (!gpEntry.size) {
             // This is a GPFIFO control entry, all control entries have a zero length and contain no pushbuffers