FATAL: failed to get core properties from VHA device /dev/vha0: Invalid argument
[I 3/20 9:14:10.389 …driver/imagination_nna/imgdnn_manager.cc:31 error_callback] (nna_imgdnnCreateContext) Could not create NNA session, error
[F 3/20 9:14:10.389 …driver/imagination_nna/imgdnn_manager.cc:57 ImgdnnManager] Check failed: (static_cast(error_code) == static_cast(IMGDNN_SUCCESS)): 1!==0 Failed to call imgdnnCreateContext!
[F 3/20 9:14:10.389 …driver/imagination_nna/imgdnn_manager.cc:57 ImgdnnManager] Check failed: (static_cast(error_code) == static_cast(IMGDNN_SUCCESS)): 1!==0 Failed to call imgdnnCreateContext!
[F 3/20 9:14:10.392 …ter/nnadapter/src/runtime/compilation.cc:98 ~Program] Check failed: device_context: No device found.
[F 3/20 9:14:10.392 …ter/nnadapter/src/runtime/compilation.cc:98 ~Program] Check failed: device_context: No device found.
terminate called after throwing an instance of ‘nnadapter::logging::Exception’
what(): NNAdapter C++ Exception:
[F 3/20 9:14:10.392 …ter/nnadapter/src/runtime/compilation.cc:98 ~Program] Check failed: device_context: No device found.
./run.sh: line 177: 2405 Aborted (core dumped) ./$BUILD_DIR/demo ../assets/models/$MODEL_NAME ../assets/configs/$CONFIG_NAME ../assets/datasets/$DATASET_NAME $NNADAPTER_DEVICE_NAMES “$NNADAPTER_CONTEXT_PROPERTIES” $NNADAPTER_MODEL_CACHE_DIR $NNADAPTER_MODEL_CACHE_TOKEN $NNADAPTER_SUBGRAPH_PARTITION_CONFIG_PATH $NNADAPTER_MIXED_PRECISION_QUANTIZATION_CONFIG_PATH
I tried to run the NPU inference acceleration demo on the B100-A development board, but it cannot run. NNA deployment— Paddle-Lite documentation
root@paddlepi:/home/Paddle-Lite/PaddleLite-generic-demo/image_classification_demo/shell# dmesg | grep vha
[ 3647.373902] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_resume!
[ 3647.375040] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_suspend!
[ 3647.375079] misc vha0: vha_ioctl: code 0x80287100 unknown
[ 3647.390688] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_resume!
[ 3647.391858] ax21xx-nna 6fc00000.ipu: vha_session_pm_release pm_count is 1
[ 3647.392102] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_suspend!
[ 3894.907040] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_resume!
[ 3894.908396] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_suspend!
[ 3894.908445] misc vha0: vha_ioctl: code 0x80287100 unknown
[ 3894.924119] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_resume!
[ 3894.925288] ax21xx-nna 6fc00000.ipu: vha_session_pm_release pm_count is 1
[ 3894.925316] ax21xx-nna 6fc00000.ipu: runtime_pm: vha_plat_runtime_suspend!
I checked the logs and found that it seems the NNA is not functioning properly, and my VHA peripheral has no issues.
root@paddlepi:/home/Paddle-Lite/PaddleLite-generic-demo/image_classification_demo/shell# ls -l /dev/vha*
crw-rw-rw- 1 edgeboard users 10, 23 Mar 20 08:08 /dev/vha0
root@paddlepi:/home/Paddle-Lite/PaddleLite-generic-demo/image_classification_demo/shell# lsmod | grep -iE “vha|imgdnn”
vha 159744 0
img_mem 155648 1 vha
