[FFmpeg-devel] [PATCH v2 2/3] avfilter/dnn_backend_torch: implement async execution logic

[Raja Rathour via ffmpeg-devel <ffmpeg-devel@ffmpeg.org>]

Update execute_model_th to support asynchronous execution by pushing tasks to the pending queue and notifying the worker thread.

Signed-off-by: Raja Rathour <imraja729@gmail.com>
---
 libavfilter/dnn/dnn_backend_torch.cpp | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/libavfilter/dnn/dnn_backend_torch.cpp b/libavfilter/dnn/dnn_backend_torch.cpp
index af92cd13da..b5ff5b44d1 100644
--- a/libavfilter/dnn/dnn_backend_torch.cpp
+++ b/libavfilter/dnn/dnn_backend_torch.cpp
@@ -375,14 +375,12 @@ static int execute_model_th(THRequestItem *request, Queue *lltask_queue)
         goto err;
     }
     if (task->async) {
-        avpriv_report_missing_feature(th_model->ctx, "LibTorch async");
-    } else {
-        ret = th_start_inference((void *)(request));
-        if (ret != 0) {
-            goto err;
+        std::lock_guard<std::mutex> lock(*th_model->mutex);
+        if (ff_safe_queue_push_back(th_model->pending_queue, request) < 0) {
+            return AVERROR(ENOMEM);
         }
-        infer_completion_callback(request);
-        return (task->inference_done == task->inference_todo) ? 0 : DNN_GENERIC_ERROR;
+        th_model->cond->notify_one();
+        return 0;
     }
 
 err:
-- 
2.48.1

_______________________________________________
ffmpeg-devel mailing list -- ffmpeg-devel@ffmpeg.org
To unsubscribe send an email to ffmpeg-devel-leave@ffmpeg.org