I run ./deepstream-app in sources/apps/sample_apps/deepstream-app
APP can start and load trt model, but osd can’t show on screen(all black)
And i end the app, i get error from sink_bin_queue
>>> Using TRT model serialized engine /home/ubuntu/DeepStreamInstall/DeepStreamSDK-Tesla-v3.0/DeepStreamSDK-Tesla-v3.0/DeepStream_Release/sources/apps/sample_apps/deepstream-app/sample_ssd_relu6.uff_b1_fp32.engine crypto flags(0)
cb_sourcesetup set 100 latency
Runtime commands:
h: Print this help
q: Quit
p: Pause
r: Resume
NOTE: To expand a source in the 2D tiled display and view object details, left-click on the source.
To go back to the tiled display, right-click anywhere on the window.
** INFO: <bus_callback:98>: Pipeline ready
** INFO: <bus_callback:84>: Pipeline running
cuGraphicsGLRegisterBuffer failed with error(304) gst_eglglessink_cuda_init texture = 1
____each start____stream 0 frame= 1_______
id:2, (249, 184, 36, 32 )
id:2, (381, 105, 42, 45 )
id:3, (392, 144, 46, 70 )
==================each end
____each start____stream 0 frame= 2_______
id:2, (249, 184, 36, 33 )
id:2, (381, 105, 42, 44 )
id:3, (392, 144, 44, 70 )
==================each end
____each start____stream 0 frame= 3_______
id:2, (251, 184, 38, 38 )
id:2, (381, 106, 40, 45 )
id:3, (234, 208, 78, 46 )
==================each end
** INFO: <bus_callback:84>: Pipeline running
** INFO: <bus_callback:84>: Pipeline running
**PERF: FPS 0 (Avg) FPS 1 (Avg)
**PERF: 138.77 (138.77) 30.66 (30.66)
**PERF: 136.41 (137.25) 30.11 (30.29)
**PERF: 135.16 (137.09) 30.18 (30.24)
**PERF: 0.00 (137.09) 30.07 (30.19)
**PERF: 0.00 (137.09) 30.02 (30.15)
**PERF: 0.00 (137.09) 30.02 (30.13)
Quitting
App run successful
ERROR from sink_bin_queue: Internal data flow error.
Debug info: gstqueue.c(992): gst_queue_handle_sink_event (): /GstPipeline:pipeline/GstBin:processing_bin_0/GstBin:sink_bin/GstQueue:sink_bin_queue:
streaming task paused, reason not-negotiated (-4)
0:00:00.027293019 20506 0x1760b20 WARN default nvinfer_property_parser.c:217:nvinfer_set_missing_default_class_attr:<primary_gie_classifier> warning: /home/ubuntu/DeepStreamInstall/DeepStreamSDK-Tesla-v3.0/DeepStreamSDK-Tesla-v3.0/DeepStream_Release/sources/apps/sample_apps/deepstream-app/config_infer_primary_ssd.txt: Setting missing ‘eps’ property for group [class-attrs-all] to 0.00
0:00:00.027345494 20506 0x1760b20 WARN default nvinfer_property_parser.c:240:nvinfer_set_missing_default_class_attr:<primary_gie_classifier> warning: /home/ubuntu/DeepStreamInstall/DeepStreamSDK-Tesla-v3.0/DeepStreamSDK-Tesla-v3.0/DeepStream_Release/sources/apps/sample_apps/deepstream-app/config_infer_primary_ssd.txt: Setting missing ‘group-threshold’ property for group [class-attrs-all] to 0
Using TRT model serialized engine /home/ubuntu/DeepStreamInstall/DeepStreamSDK-Tesla-v3.0/DeepStreamSDK-Tesla-v3.0/DeepStream_Release/sources/apps/sample_apps/deepstream-app/sample_ssd_relu6.uff_b1_fp32.engine crypto flags(0)
0:00:02.016218435 20506 0x1760b20 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_complete: pad not activated yet
0:00:02.016487031 20506 0x1760b20 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_complete: pad not activated yet
cb_sourcesetup set 100 latency
Runtime commands:
h: Print this help
q: Quit
p: Pause
r: Resume
NOTE: To expand a source in the 2D tiled display and view object details, left-click on the source.
To go back to the tiled display, right-click anywhere on the window.
** INFO: <bus_callback:98>: Pipeline ready
0:00:02.021400731 20506 0x7fd1980544f0 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex: failed to find fragment defaults for stream 1
0:00:02.021472914 20506 0x7fd1980544f0 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex: failed to find fragment defaults for stream 2
0:00:02.021562225 20506 0x7fd1980544f0 WARN basesrc gstbasesrc.c:2396:gst_base_src_update_length: processing at or past EOS
** INFO: <bus_callback:84>: Pipeline running
0:00:02.103324095 20506 0x7fd1cc003680 FIXME default gstutils.c:3766:gst_pad_create_stream_id_internal:fakesrc0:src Creating random stream-id, consider implementing a deterministic way of creating a stream-id
0:00:02.103716556 20506 0x7fd1cc0036d0 FIXME default gstutils.c:3766:gst_pad_create_stream_id_internal:fakesrc1:src Creating random stream-id, consider implementing a deterministic way of creating a stream-id
cuGraphicsGLRegisterBuffer failed with error(304) gst_eglglessink_cuda_init texture = 1
0:00:02.164114108 20506 0x175f0a0 ERROR nveglglessink ext/eglgles/gsteglglessink.c:2768:gst_eglglessink_configure_caps:<sink_sub_bin_sink1> Cuda Init failed
0:00:02.164125505 20506 0x175f0a0 ERROR nveglglessink ext/eglgles/gsteglglessink.c:2779:gst_eglglessink_configure_caps:<sink_sub_bin_sink1> Configuring caps failed
0:00:02.164189597 20506 0x1789320 ERROR nveglglessink ext/eglgles/gsteglglessink.c:2824:gst_eglglessink_setcaps:<sink_sub_bin_sink1> Failed to configure caps
0:00:02.164286194 20506 0x1789320 ERROR nveglglessink ext/eglgles/gsteglglessink.c:2824:gst_eglglessink_setcaps:<sink_sub_bin_sink1> Failed to configure caps
Can you run test1 ?
And then run: deepstream-app source30_720p_dec_infer-resnet_tiled_display_int8.txt
And then follow “sources/objectDetector_SSD/README” to run: deepstream-app deepstream_app_config_ssd.txt
I mean get dynamic link info for this plugin,
run this, ldd /deepstream plugin install path/libnvdsgst_eglglessink.so within your deepstream app running environments
seems your system path do not have nvidia gl library? it need nvidia gl library for nveglglessink to work.
when you install display driver, if you use deb file to install, it should extract into some system place, you need to create one soft link to it, here is my setup, for your reference.
tse@p4station:~$ ll /usr/lib/x86_64-linux-gnu/libEGL.so
lrwxrwxrwx 1 root root 29 5月 19 15:00 /usr/lib/x86_64-linux-gnu/libEGL.so → /usr/lib/nvidia-418/libEGL.so
lrwxrwxrwx 1 root root 16 4月 6 20:45 /usr/lib/nvidia-418/libEGL.so → libEGL.so.418.67
if you use run file to install, different version may behave differently, if promoted if need to setup nvidia x config, something like that, choose yes.
