Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
V100 32G
• DeepStream Version
5.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
7.2.3
• NVIDIA GPU Driver Version (valid for GPU only)
460
• Issue Type( questions, new requirements, bugs)
wonder if there were bugs
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
I first used the code below to get nvstreammux sink pad prepared.
Blockquote
for (guint i = 0; i < MAX_NUM_STREAMS; i++)
{
gchar mux_sink_pad_name[32];
g_snprintf(mux_sink_pad_name, 31, “sink_%u”, i);
std::string str(mux_sink_pad_name);
GstPad *mux_sink_pad = gst_element_get_request_pad(this->app_context_.main_bin.nvstreammux, mux_sink_pad_name);
gst_object_unref(mux_sink_pad);
}
Blockquote
for (guint i = 0; i < MAX_NUM_STREAMS; i++)
{
gchar demux_src_pad_name[32];
g_snprintf(demux_src_pad_name, 31, “src_%u”, i);
std::string str(demux_src_pad_name);
GstPad *demux_src_pad = gst_element_get_request_pad(this->app_context_.main_bin.nvstreamdemux, demux_src_pad_name);
gst_object_unref(demux_src_pad);
}
when a new stream came, dynamic link source bin src pad to nvstreammux sink pad, using gst_element_get_static_pad
I tryied three ways，
Blockquote
when MAX_NUM_STREAMS = 1, after pads linked, the memory cost very much and fast, about 300M / s.
when MAX_NUM_STREAMS = 4, if all 4 streams added and pads linked, it was normal, no memory leak or cost too much.
when MAX_NUM_STREAM = 4, not all 4 streams added and pads linked, the memory also cost very much and fast, same as the first one.
while adding the stream, sometimes there would be a warning
Blockquote
GStreamer-WARNING **: 20:19:11.825: gstpad.c:5203:store_sticky_event:nvstreamdemux0:src_3 Sticky event misordering, got ‘segment’ before ‘caps’
I also tried valgrind to check the location of momery problem.
the log was below
Blockquote
==78925== 20,209,104 bytes in 421,023 blocks are indirectly lost in loss record 7,825 of 7,836
==78925== at 0x4C33B25: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==78925== by 0x4E8FC30: g_malloc0 (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x5B0DDF2: nvds_create_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0DCB3: nvds_create_user_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0B610: nvds_create_batch_meta (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x11DD9021: gst_nvstreammux_src_push_loop(void*) (gstnvstreammux.cpp:966)
==78925== by 0x5200278: ??? (in /usr/lib/x86_64-linux-gnu/libgstreamer-1.0.so.0.1405.0)
==78925== by 0x4EB2C6F: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EB22A4: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x56EB6DA: start_thread (pthread_create.c:463)
==78925==
==78925== 20,211,480 bytes in 842,145 blocks are indirectly lost in loss record 7,826 of 7,836
==78925== at 0x4C31B0F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==78925== by 0x4E8FBD8: g_malloc (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EA7A85: g_slice_alloc (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4E85E25: g_list_prepend (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x5B0DE3A: nvds_create_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0DB13: nvds_create_obj_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0B58C: nvds_create_batch_meta (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x11DD9021: gst_nvstreammux_src_push_loop(void*) (gstnvstreammux.cpp:966)
==78925== by 0x5200278: ??? (in /usr/lib/x86_64-linux-gnu/libgstreamer-1.0.so.0.1405.0)
==78925== by 0x4EB2C6F: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EB22A4: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x56EB6DA: start_thread (pthread_create.c:463)
==78925==
==78925== 20,696,832 bytes in 53,895 blocks are indirectly lost in loss record 7,827 of 7,836
==78925== at 0x4C31B0F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==78925== by 0x11DDB7D0: GstBatchBufferWrapper::GstBatchBufferWrapper(_GstNvStreamMux*, unsigned int, bool) (gstnvstreammux_impl.h:97)
==78925== by 0x11DD9003: gst_nvstreammux_src_push_loop(void*) (gstnvstreammux.cpp:965)
==78925== by 0x5200278: ??? (in /usr/lib/x86_64-linux-gnu/libgstreamer-1.0.so.0.1405.0)
==78925== by 0x4EB2C6F: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EB22A4: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x56EB6DA: start_thread (pthread_create.c:463)
==78925==
==78925== 23,578,352 bytes in 421,042 blocks are indirectly lost in loss record 7,828 of 7,836
==78925== at 0x4C33B25: calloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==78925== by 0x4E8FC30: g_malloc0 (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x5B0DDF2: nvds_create_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0DB9F: nvds_create_classifier_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0B5B8: nvds_create_batch_meta (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x11DD9021: gst_nvstreammux_src_push_loop(void*) (gstnvstreammux.cpp:966)
==78925== by 0x5200278: ??? (in /usr/lib/x86_64-linux-gnu/libgstreamer-1.0.so.0.1405.0)
==78925== by 0x4EB2C6F: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EB22A4: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x56EB6DA: start_thread (pthread_create.c:463)
==78925==
==78925== 30,436,056 (10,584 direct, 30,425,472 indirect) bytes in 441 blocks are definitely lost in loss record 7,829 of 7,836
==78925== at 0x4C31B0F: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==78925== by 0x4E8FBD8: g_malloc (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EA7A85: g_slice_alloc (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4E85E25: g_list_prepend (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x5B0DE3A: nvds_create_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0D9FB: nvds_create_frame_meta_pool (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x5B0B560: nvds_create_batch_meta (in /opt/nvidia/deepstream/deepstream-5.1/lib/libnvds_meta.so)
==78925== by 0x11DD9021: gst_nvstreammux_src_push_loop(void*) (gstnvstreammux.cpp:966)
==78925== by 0x5200278: ??? (in /usr/lib/x86_64-linux-gnu/libgstreamer-1.0.so.0.1405.0)
==78925== by 0x4EB2C6F: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x4EB22A4: ??? (in /usr/lib/x86_64-linux-gnu/libglib-2.0.so.0.5600.4)
==78925== by 0x56EB6DA: start_thread (pthread_create.c:463)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)