hardware: DevelopKit/Custom board
software version: etPack5.1.5
core board:Orin Nano 4G
After running the deepstream app test program on Orin Nano, it was found that the CPU usage was 500% and the memory usage exceeded 1.7G. The following is the test process and configuration file.
config file:
root@tegra-ubuntu:/home/nvidia/code/rtsp_test# cat /vendor_app/bin/output/algo/fvs_ai_core/model/det/0/my_config_lt.txt
[application]
enable-perf-measurement=1
perf-measurement-interval-sec=5
#gie-kitti-output-dir=streamscl
[tiled-display]
enable=0
rows=1
columns=1
width=1280
height=720
gpu-id=0
nvbuf-memory-type=0
[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI 4=RTSP
# type=4
# uri=rtsp://foo.com/stream1.mp4
# 注意：uri 中的相对路径是相对于配置文件所在目录
# 当前配置文件在 model/det/0/，视频在 video/，所以需要 ../../../video/
# type=3
# uri=file:///home/nvidia/submit/fvs_ai_core/video/BlockedCarOnBridge.mp4
# uri=file:///home/nvidia/submit/fvs_ai_core/video/video-251217/DJI_20251217165019_0002_S.MP4
type=4
uri=rtsp://192.168.0.13:554/aibox_transfer_fpv
num-sources=1
gpu-id=0
latency=50
nvbuf-memory-type=0
[sink0]
enable=0
#Type - 1=FakeSink 2=EglSink 3=File
type=2
sync=0
source-id=0
gpu-id=0
nvbuf-memory-type=0
muxer-config=faststart=1
[sink1]
enable=0
type=6
msg-conv-config=/opt/nvidia/deepstream/deepstream/sources/apps/sample_apps/deepstream-test5/configs/dstest5_msgconv_sample_config.txt
msg-conv-payload-type=1
msg-conv-msg2p-new-api=1
msg-conv-frame-interval=1
msg-broker-proto-lib=/opt/nvidia/deepstream/deepstream/lib/libnvds_kafka_proto.so
msg-broker-conn-str=192.168.0.13;9092;dstest5
topic=dstest5
debug-payload-dir=/home/nvidia/code/fvs_ai_core/app/deepstream_yolo/result/payloads1
[sink2]
enable=0
type=3
#1=mp4 2=mkv
container=1
#1=h264 2=h265 3=mpeg4
## only SW mpeg4 is supported right now.
codec=1 # 3
#encoder type 0=Hardware 1=Software
enc-type=0 # 0 为硬件编码 1 为软件编码
sync=0
bitrate=20000000
output-file=/home/nvidia/submit/fvs_ai_core/result/20251217165019_0002_S_result.MP4
source-id=0
[sink3]
enable=1
type=4 # RTSP 推流
codec=1 # H264 编码 x264enc
enc-type=1 # 软件编码
sync=0
bitrate=4000000
profile=2
rtsp-port=8554
udp-port=5400
[osd]
enable=1
gpu-id=0
border-width=1
text-size=15
text-color=1;1;1;1;
text-bg-color=0.3;0.3;0.3;1
font=Arial
show-clock=0
clock-x-offset=800
clock-y-offset=820
clock-text-size=12
clock-color=1;0;0;0
nvbuf-memory-type=0
[streammux]
gpu-id=0
##Boolean property to inform muxer that sources are live
live-source=1
batch-size=1
##time out in usec, to wait after the first buffer is available
##to push the batch even if the complete batch is not formed
batched-push-timeout=10000
## Set muxer output width and height
width=1920
height=1080
##Enable to maintain aspect ratio wrt source, and allow black borders, works
##along with width, height properties
enable-padding=0
nvbuf-memory-type=0
## If set to TRUE, system timestamp will be attached as ntp timestamp
## If set to FALSE, ntp timestamp from rtspsrc, if available, will be attached
# attach-sys-ts-as-ntp=1
# config-file property is mandatory for any gie section.
# Other properties are optional and if set will override the properties set in
# the infer config file.
[primary-gie]
enable=1
gpu-id=0
#Required to display the PGIE labels, should be added even when using config-file
#property
batch-size=1 # 2
#Required by the app for OSD, not a plugin property
bbox-border-color0=1;0;0;1
bbox-border-color1=0;1;1;1
bbox-border-color2=0;0;1;1
bbox-border-color3=0;1;0;1
interval=0
#Required by the app for SGIE, when used along with config-file property
gie-unique-id=1
nvbuf-memory-type=0
# 模型配置（固定使用 algo_id=0 的模型）
model-engine-file=model_b1_gpu0_fp16.engine
labelfile-path=labels.txt
config-file=config_infer_primary_yolo11.txt
#infer-raw-output-dir=/home/nvidia/code/fvs_ai_core/primary_detector_raw_output/
[tracker]
enable=0
tracker-width=640
tracker-height=384
ll-lib-file=//opt/nvidia/deepstream/deepstream/lib/libByteTracker.so
gpu-id=0
display-tracking-id=1
[tests]
file-loop=0
jtop:
top：
After turning off sink 3 (turning off xh264enc), the CPU usage significantly decreased
Why does using x264enc in Deepstream consume so much CPU and memory? Is there an optimization method?
I need some help.
