tu/kgsl: Revert "Remove zero CB queue submission fast path"

This reverts commit 0342d34bdb which
introduced a regression in the Turnip's KGSL backend, causing various
sync issues since KGSL doesn't advance the GPU timeline when a submit
without cmdbufs is made. A comment explaining the issue was added to the
code, and the fast path is reintroduced.

Signed-off-by: Mark Collins <mark@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34328>

src/freedreno/vulkan/tu_knl_kgsl.cc

@@ -1089,6 +1089,51 @@ kgsl_queue_submit(struct tu_queue *queue, void *_submit,
    uint64_t start_ts = tu_perfetto_begin_submit();
 #endif
 
+   if (submit->commands.size == 0) {
+      /* This handles the case where we have a wait and no commands to submit.
+       * It is necessary to handle this case separately as the kernel will not
+       * advance the GPU timeline if a submit with no commands is made, even
+       * though it will return an incremented fence timestamp (which will
+       * never be signaled).
+       */
+      const struct kgsl_syncobj *wait_semaphores[wait_count + 1];
+      for (uint32_t i = 0; i < wait_count; i++) {
+         wait_semaphores[i] = &container_of(waits[i].sync,
+                                            struct vk_kgsl_syncobj, vk)
+                                  ->syncobj;
+      }
+
+      struct kgsl_syncobj last_submit_sync;
+      if (queue->fence >= 0)
+         last_submit_sync = (struct kgsl_syncobj) {
+            .state = KGSL_SYNCOBJ_STATE_TS,
+            .queue = queue,
+            .timestamp = queue->fence,
+         };
+      else
+         last_submit_sync = (struct kgsl_syncobj) {
+            .state = KGSL_SYNCOBJ_STATE_SIGNALED,
+         };
+
+      wait_semaphores[wait_count] = &last_submit_sync;
+
+      struct kgsl_syncobj wait_sync =
+         kgsl_syncobj_merge(wait_semaphores, wait_count + 1);
+      assert(wait_sync.state !=
+             KGSL_SYNCOBJ_STATE_UNSIGNALED); // Would wait forever
+
+      for (uint32_t i = 0; i < signal_count; i++) {
+         struct kgsl_syncobj *signal_sync =
+            &container_of(signals[i].sync, struct vk_kgsl_syncobj, vk)
+                ->syncobj;
+
+         kgsl_syncobj_reset(signal_sync);
+         *signal_sync = wait_sync;
+      }
+
+      return VK_SUCCESS;
+   }
+
    VkResult result = VK_SUCCESS;
 
    if (u_trace_submission_data) {