I pulled the DeepStream 3.0-18.11, and launched it on a server with Tegra P4.
docker pull nvcr.io/nvidia/deepstream:3.0-18.11
It runs, and X11 is set up (I can launch firefox to my desktop via SSH X11 tunneling).
when I try to start the sample, it errors out:
if I leave the default DISPLAY env variable (which is set automatically by the ssh tunneling, and firefox works with this setting), I got:
root@c1c2b60a9b63:~/DeepStream_Release/samples/configs/deepstream-app# deepstream-app --gst-debug-level=3 -c so_display_int8.txt
0:00:02.826477321 11940 0x1a3f890 ERROR egladaption ext/eglgles/gstegladaptation_egl.c:144:gst_1> Could not get EGL display connection
0:00:02.826540467 11940 0x1a3f890 ERROR egladaption ext/eglgles/gstegladaptation_egl.c:185:gst_1> Couldn't setup window/surface from handle
0:00:02.826552569 11940 0x1a3f890 ERROR nveglglessink ext/eglgles/gsteglglessink.c:532:egl_init:<
0:00:02.826571446 11940 0x1a3f890 ERROR nveglglessink ext/eglgles/gsteglglessink.c:558:egl_init:<
** ERROR: <main:564>: Failed to set pipeline to PAUSED
Quitting
App run failed
According to this https://devtalk.nvidia.com/default/topic/1046131/jetson-tx2/nvivafilter-could-not-get-egl-display-connection/
I tried to remove DISPLAY
unset DISPLAY
then I got this:
>>> Warning. Could not open model engine file /root/DeepStream_Release/samples/configs/deepstream-app/../../mod_b16_int8.engine
>>> Generating new TRT model engine
Using INT8 data type.
Warning: Flatten layer ignored. TensorRT implicitly flattens input to FullyConnected layers, but in other circur.
***** Storing serialized engine file as /root/DeepStream_Release/samples/configs/deepstream-app/../../models/Sint8.engine batchsize = 16 *****
>>> Warning. Could not open model engine file /root/DeepStream_Release/samples/configs/deepstream-app/../../modt8.engine
>>> Generating new TRT model engine
Using INT8 data type.
***** Storing serialized engine file as /root/DeepStream_Release/samples/configs/deepstream-app/../../models/Pgine batchsize = 4 *****
0:00:32.929765563 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.930098535 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.930559687 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.931045826 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.931517207 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.931975514 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.932731767 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
0:00:32.932874680 11941 0xa73d440 WARN basesrc gstbasesrc.c:3489:gst_base_src_start_comple
** ERROR: <main:583>: Could not open X Display
Quitting
0:00:32.938881276 11941 0x7fedd0004a80 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.938911297 11941 0x7fedd0004a80 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.938930618 11941 0x7fedd0004a80 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux21
0:00:32.939015703 11941 0x7fedd0004a80 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux22
0:00:32.939105302 11941 0x7fedd0004a80 WARN basesrc gstbasesrc.c:2396:gst_base_src_update_lengt
0:00:32.939358473 11941 0x1f82ad0 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.939431925 11941 0x1f82ad0 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.939483933 11941 0x1f82ad0 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux11
0:00:32.939704106 11941 0x1c76b70 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.939907679 11941 0x1c76b70 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.940002792 11941 0x1c76b70 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux01
0:00:32.940214718 11941 0x1f78000 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.940534442 11941 0x1f78000 WARN qtdemux qtdemux_types.c:222:qtdemux_type_get: unkno
0:00:32.940626028 11941 0x1f78000 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux31
0:00:32.940641800 11941 0x1c76b70 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux02
0:00:32.940756154 11941 0x1f82ad0 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux12
0:00:32.940978623 11941 0x1c76b70 WARN basesrc gstbasesrc.c:2396:gst_base_src_update_lengt
0:00:32.941043127 11941 0x1f78000 WARN qtdemux qtdemux.c:2651:qtdemux_parse_trex:<qtdemux32
0:00:32.941166322 11941 0x1f78000 WARN basesrc gstbasesrc.c:2396:gst_base_src_update_lengt
0:00:32.941395854 11941 0x1f82ad0 WARN basesrc gstbasesrc.c:2396:gst_base_src_update_lengt
0:00:33.042499404 11941 0x2177850 WARN GST_PADS gstpad.c:4092:gst_pad_peer_query:<osd_queue
App run failed
my env:
$nvidia-smi
Wed Apr 10 16:53:20 2019
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.39 Driver Version: 418.39 CUDA Version: 10.1 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla P4 Off | 00000000:00:04.0 Off | 0 |
| N/A 36C P0 22W / 75W | 0MiB / 7611MiB | 4% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| No running processes found |
+-----------------------------------------------------------------------------+
Anyone knows what can be the cause?
A docker container supposed to make it “build once, run everywhere”, is that only me who can’t see it run?