DriveWorks camera_gmsl_ram&tensorRT&SSD&PX2 , result is error

Autonomous Machines DRIVE PX2 DriveWorks
#1 
nvidia@nvidia:~/samples$ ./sample_camera_gmsl_raw_custommodel_ssd 
nvrm_gpu: Bug 200215060 workaround enabled.
using 1th GPU device
ssd_init ...
/home/nvidia/TensorRT-3.0.2/data//ssd/test_plugin.prototxt
/home/nvidia/TensorRT-3.0.2/data/ssd/VGG_VOC0712_SSD.caffemodel
Bindings after deserializing:
Binding 0 (data): Input.
Binding 1 (detection_out): Output.
INPUT Tensor Shape is: C: 3  H: 300  W:  300
OUTPUT Tensor Shape is: C: 1  H: 200  W: 7
Allocate memory: input blob
Allocate memory: output blob
Program Arguments:
--camera-type=ar0231-rccb-ssc
--csi-port=ab
--fifo-size=3
--interpolationDemosaic=0
--serializer-type=raw
--slave=0
--write-file=

Initialize DriveWorks SDK v0.6.67
Release build with GNU 4.9.2 from v0.6.0-rc7-0-g51bd3aa against Vibrante PDK v5.0.5.0
TimeSource: monotonic epoch time offset is 1462513201851851
TimeSource: PTP ioctl returned error. Synchronized time will not be available.
TimeSource: Could not detect valid PTP time source at 'eth0'. Fallback to CLOCK_MONOTONIC.
SDK: No resources mounted, some modules will not function properly
SDK: Create NvMediaDevice
SDK: use EGL display as provided
SDK: number of GPU devices detected 2
SDK: currently selected GPU device discrete ID 0
SAL: identified board as DrivePX2-TegraA
SensorFactory::createSensor() -> camera.gmsl, camera-type=ar0231-rccb-ssc,csi-port=ab,fifo-size=3,interpolationDemosaic=0,serializer-type=raw,slave=0,write-file=,output-format=raw+data
CameraGMSL: required FPS = 30, resolution = 1920x1208
***** aurix state *****
 CAM_PWR_ON: 0x0
 TEGRA_GMSL: TEGRA B
 TEGRA_FPDL: TEGRA B
 TEGRA_A_HDMI: MXM
 TEGRA_B_HDMI: MXM
 TEGRA_PCI: TEGRA B
 FAN_STATE: UNKNOWN
***********************
***** aurix state *****
 CAM_PWR_ON: 0x1
 TEGRA_GMSL: UNKNOWN
 TEGRA_FPDL: UNKNOWN
 TEGRA_A_HDMI: UNKNOWN
 TEGRA_B_HDMI: UNKNOWN
 TEGRA_PCI: UNKNOWN
 FAN_STATE: UNKNOWN
***********************
Sensor AR0231 RCCB Rev7 detected!
Camera image with 1920x1208 at 30 FPS
gCameraHeight:1208 gCameraWidth:1920
create thread success
entry thread...
nvmEglStProducerCreateCommon: Entered
Exposure Time (s): 1801.83
data already get over,can ready next frame data
main : mutex unlock
thread : test detector start...
thread:resize image for fasterRCNN...
main:draw pre box coordinate!
Exposure Time (s): 1801.83
data already get over,can ready next frame data
main : mutex unlock
main:draw pre box coordinate!
Exposure Time (s): 1801.83
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Came into the image inference method here. 
Begin imageInference...
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
Exposure Time (s): 8164.99
main:draw pre box coordinate!
CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_small_nn_v1.
caskConvolutionLayer.cpp (410) - Cuda Error in execute: 33
caskConvolutionLayer.cpp (410) - Cuda Error in execute: 33

End imageInference...
Time: 1935.37 ms
ssd_detector :get result ...
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax 0.000000 ymax 0.000000 
ssd_detector :xmin 0.000000 ymin 0.000000 xmax -nan ymax 0.000000

as these log show, i run ssd caffemodel base on the camera_gmsl_raw(driveworks-0.6 samples) and tensorRT3.0.2,and get the result is all 0.but when i run it only on single image without camera ,the result is right.
so ,i want to know what does there mean ?and if there is an effect on result? thanks a lot
CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_small_nn_v1.
caskConvolutionLayer.cpp (410) - Cuda Error in execute: 33
caskConvolutionLayer.cpp (410) - Cuda Error in execute: 33

#2

Dear zhou-lw,
Can you confirm if you are resizing the input image according to network size.

#3

yes,i am sure the input image size is same with network size ,300*300

#4

Dear zhou-lw,
Can you please file a bug with sample code, model file to reproduce the error on our side. Please login to https://developer.nvidia.com/drive with your credentials. Please check MyAccount->MyBugs->Submit a new bug to file bug. Please share ID here for us to follow up.

#5

I confuse that sample_camera_gmsl_raw and tensorRT are both use GPU,are they use the same one ? is there collision?

#6

What happened with this issue?
Is it solved?

I do also have an error with tensorrt 4.1.3. (cuda 9, jetson tx2, Ubuntu 16.04):

Cuda error in file src/implicit_gemm.cu at line 1214: invalid resource handle
customWinogradConvActLayer.cpp (308) - Cuda Error in execute: 33
customWinogradConvActLayer.cpp (308) - Cuda Error in execute: 33

Also, when I run my application as sudo I am getting an:

CUDA cask failure at execution for trt_maxwell_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148m_nt_v1.
#7

Please help! I have the same problem, just copy everything from example Tensorrt 5.1.5.0

[TensorRT] ERROR: CUDA cask failure at execution for trt_maxwell_scudnn_128x64_relu_medium_nn_v1.
[TensorRT] ERROR: cuda/caskConvolutionLayer.cpp (355) - Cuda Error in execute: 33 (invalid resource handle)
[TensorRT] ERROR: cuda/caskConvolutionLayer.cpp (355) - Cuda Error in execute: 33 (invalid resource handle)