hi ,every one,我遇到了个问题是在ubuntu aarch64平台基于飞腾D2000平台，ubuntu版本是20.04版本，内核版本是5.10.226的版本；硬件使用华为atlas300iPro 测试samples-master/cplusplus/level2_simple_inference/2_object_detection/YOLOV3_coco_detection_video/scripts# bash sample_run.sh 模型时，报了如下错误，
[INFO] The sample starts to run
Current environment valid ip list:
127.0.0.1
192.168.1.2
Please choose one to show the presenter in browser:192.168.1.2
content_type config string:1
[INFO] Acl init success
[INFO] The program runs successfully, Please visit http://192.168.1.2:7007 for display server!
Enter any command to stop the application:[INFO] Open device 0 success
[INFO] load model ../model/yolov3.om success
[INFO] create model description success
[INFO] create model output success
[ERROR] execute model failed, modelId is 1
[ERROR] Execute model inference failed
[ERROR] Inference model inference output data failed
[INFO] unload model success, modelId is 1
跟踪内核dmesg打印错误如下
[ 1545.755573] [ascend] [drv_pcie] [ERROR] [devdrv_sync_non_trans_msg_chan_send 1528] <kworker/7:1:1243:1243> Device irq not resp. (dev_id=0; msg_type="common_msg_devmanager"; status=-1; retry_cnt=0)
[ 1553.247275] [ascend] [drv_pcie] [ERROR] [devdrv_sync_non_trans_msg_chan_send 1528] <kworker/u17:0:106:106> Device irq not resp. (dev_id=0; msg_type="common_msg_devmanager"; status=-1; retry_cnt=0)
[ 1560.603032] [ascend] [drv_pcie] [ERROR] [devdrv_sync_non_trans_msg_chan_send 1528] <kworker/7:1:1243:1243> Device irq not resp. (dev_id=0; msg_type="common_msg_devmanager"; status=-1; retry_cnt=0)
[ 1562.732159] [ascend] [drv_soft_fault] [ERROR] [heart_beat_judge 342] <kworker/7:1H:211,211> Heartbeat lost! (device=0; old_count=18446744073709551615; new_count=18446744073709551615; lost_count=4; total_lost_count=4; total_10s_count=0; total_15s_count 0).
[ 1562.818577] [ascend] [fault_manager] [ERROR] [dms_record_fault_event_log 1394] event_id=0x40f84e00; device_id=0; node_type=0x7c[BaseService]; node_id=0; sub_node_id=0; event_type=0x0[Heartbeat Lost.]; severity=3[Critical]; assertion=1[Event occur]; description=[device heartbeat lost]; os_id=0; event_serial_num=1; notify_serial_num=0; event_raised_time=1767947755083 ms.
[ 1566.922558] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <main:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1573.578586] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <main:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1578.698587] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <main:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1578.700550] [ascend] [devmm] [ERROR] [devmm_send_msg_to_agent 96] <main:1632,1632> Message send failed. (agent_id=0; ret=-19)
[ 1817.280398] [ascend] [drv_pcie] [ERROR] [devdrv_dma_chan_sync_wait_query 864] <MONITOR_0:1632:1649> Wait timeout. (dev_id=0; chan_id=7)
[ 1817.280421] [ascend] [drv_pcie] [ERROR] [devdrv_dma_chan_ptr_show 47] <MONITOR_0:1632:1649> dma_chan ptr show. (hardware_sq_tail=0xffffffff; cq_head=0xffffffff; cq_tail=0xffff; sq_head=0xffff; software_sq_tail=0xd; sq_head=0xc; cq_head=0xb)
[ 1817.280433] [ascend] [drv_pcie] [ERROR] [devdrv_dma_chan_sync_wait 896] <MONITOR_0:1632:1649> Dma copy failed. (dev_id=0; chan_id=7; valid=1; status=ffffffff)
[ 1817.280445] [ascend] [devmm] [ERROR] [devmm_dma_sync_link_copy 124] <MONITOR_0:1632,1649> Devdrv_dma_sync_link_copy fail. (dev_id=0; node_cnt=1; ret=-22)
[ 1817.280459] [ascend] [devmm] [ERROR] [devmm_sync_copy_process 2332] <MONITOR_0:1632,1649> Dma copy fail. (ret=-22; src=0x12c100000000; dst=0xffff48000b60; count=32)
[ 1817.280479] [ascend] [devmm] [ERROR] [devmm_ioctl_memcpy_process_frame 223] <MONITOR_0:1632,1649> Memcpy error. (ret=-22; src=0x12c100000000; dst=0xffff48000b60; count=32; direction=2)
[ 1817.280493] [ascend] [devmm] [ERROR] [devmm_print_pre_alloced_va 320] <MONITOR_0:1632,1649> Check alloced va. (hostpid=1632; va=0x12c100000000; start_va = 0x12c100000000; end_va = 0x12c100000fff)
[ 1817.280502] [ascend] [devmm] [ERROR] [devmm_print_pre_alloced_va 324] <MONITOR_0:1632,1649> Memory before check va is not alloced or not svm addr. (hostpid=1632; check_va=0xffff48000b60; is_svm_addr=0)
[ 1822.410573] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <MONITOR_0:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1827.530569] [ascend] [tsdrv] [ERROR] [devdrv_mailbox_kernel_sync_no_feedback 437] <main:1632> [dev_id=0 tsid=0]:devdrv_mailbox_message_send failed. ret=-62
请问内核这个错误的原因是什么？我该如何解决？
使用的驱动版本是：Ascend-hdk-310p-npu-driver_25.2.3_linux-aarch64.run，固件版本是：Ascend-hdk-310p-npu-firmware_7.7.0.10.220.run
+--------------------------------------------------------------------------------------------------------+
| npu-smi 25.2.3 Version: 25.2.3 |
+-------------------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===============================+=================+======================================================+
| 0 310P3 | OK | NA 50 0 / 0 |
| 0 0 | 0000:01:00.0 | 0 1855 / 21527 |
+===============================+=================+======================================================+
+-------------------------------+-----------------+------------------------------------------------------+
| NPU Chip | Process id | Process name | Process memory(MB) |
+===============================+=================+======================================================+
| No running processes found in NPU 0 |
+===============================+=================+======================================================+
运行过resnet50 demo模型是可以成功运行的。
错误的截图如下：
点赞
收藏
回复
分享
举报
已解决
hi ,every one,我遇到了个问题是在ubuntu aarch64平台基于飞腾D2000平台，ubuntu版本是20.04版本，内核版本是5.10.226的版本；硬件使用华为atlas300iPro 测试samples-master/cplusplus/level2_simple_inference/2_object_detection/YOLOV3_coco_detection_video/scripts# bash sample_run.sh 模型时，报了如下错误，
[INFO] The sample starts to run
Current environment valid ip list:
127.0.0.1
192.168.1.2
Please choose one to show the presenter in browser:192.168.1.2
content_type config string:1
[INFO] Acl init success
[INFO] The program runs successfully, Please visit http://192.168.1.2:7007 for display server!
Enter any command to stop the application:[INFO] Open device 0 success
[INFO] load model ../model/yolov3.om success
[INFO] create model description success
[INFO] create model output success
[ERROR] execute model failed, modelId is 1
[ERROR] Execute model inference failed
[ERROR] Inference model inference output data failed
[INFO] unload model success, modelId is 1
跟踪内核dmesg打印错误如下
[ 1545.755573] [ascend] [drv_pcie] [ERROR] [devdrv_sync_non_trans_msg_chan_send 1528] <kworker/7:1:1243:1243> Device irq not resp. (dev_id=0; msg_type="common_msg_devmanager"; status=-1; retry_cnt=0)
[ 1553.247275] [ascend] [drv_pcie] [ERROR] [devdrv_sync_non_trans_msg_chan_send 1528] <kworker/u17:0:106:106> Device irq not resp. (dev_id=0; msg_type="common_msg_devmanager"; status=-1; retry_cnt=0)
[ 1560.603032] [ascend] [drv_pcie] [ERROR] [devdrv_sync_non_trans_msg_chan_send 1528] <kworker/7:1:1243:1243> Device irq not resp. (dev_id=0; msg_type="common_msg_devmanager"; status=-1; retry_cnt=0)
[ 1562.732159] [ascend] [drv_soft_fault] [ERROR] [heart_beat_judge 342] <kworker/7:1H:211,211> Heartbeat lost! (device=0; old_count=18446744073709551615; new_count=18446744073709551615; lost_count=4; total_lost_count=4; total_10s_count=0; total_15s_count 0).
[ 1562.818577] [ascend] [fault_manager] [ERROR] [dms_record_fault_event_log 1394] event_id=0x40f84e00; device_id=0; node_type=0x7c[BaseService]; node_id=0; sub_node_id=0; event_type=0x0[Heartbeat Lost.]; severity=3[Critical]; assertion=1[Event occur]; description=[device heartbeat lost]; os_id=0; event_serial_num=1; notify_serial_num=0; event_raised_time=1767947755083 ms.
[ 1566.922558] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <main:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1573.578586] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <main:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1578.698587] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <main:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1578.700550] [ascend] [devmm] [ERROR] [devmm_send_msg_to_agent 96] <main:1632,1632> Message send failed. (agent_id=0; ret=-19)
[ 1817.280398] [ascend] [drv_pcie] [ERROR] [devdrv_dma_chan_sync_wait_query 864] <MONITOR_0:1632:1649> Wait timeout. (dev_id=0; chan_id=7)
[ 1817.280421] [ascend] [drv_pcie] [ERROR] [devdrv_dma_chan_ptr_show 47] <MONITOR_0:1632:1649> dma_chan ptr show. (hardware_sq_tail=0xffffffff; cq_head=0xffffffff; cq_tail=0xffff; sq_head=0xffff; software_sq_tail=0xd; sq_head=0xc; cq_head=0xb)
[ 1817.280433] [ascend] [drv_pcie] [ERROR] [devdrv_dma_chan_sync_wait 896] <MONITOR_0:1632:1649> Dma copy failed. (dev_id=0; chan_id=7; valid=1; status=ffffffff)
[ 1817.280445] [ascend] [devmm] [ERROR] [devmm_dma_sync_link_copy 124] <MONITOR_0:1632,1649> Devdrv_dma_sync_link_copy fail. (dev_id=0; node_cnt=1; ret=-22)
[ 1817.280459] [ascend] [devmm] [ERROR] [devmm_sync_copy_process 2332] <MONITOR_0:1632,1649> Dma copy fail. (ret=-22; src=0x12c100000000; dst=0xffff48000b60; count=32)
[ 1817.280479] [ascend] [devmm] [ERROR] [devmm_ioctl_memcpy_process_frame 223] <MONITOR_0:1632,1649> Memcpy error. (ret=-22; src=0x12c100000000; dst=0xffff48000b60; count=32; direction=2)
[ 1817.280493] [ascend] [devmm] [ERROR] [devmm_print_pre_alloced_va 320] <MONITOR_0:1632,1649> Check alloced va. (hostpid=1632; va=0x12c100000000; start_va = 0x12c100000000; end_va = 0x12c100000fff)
[ 1817.280502] [ascend] [devmm] [ERROR] [devmm_print_pre_alloced_va 324] <MONITOR_0:1632,1649> Memory before check va is not alloced or not svm addr. (hostpid=1632; check_va=0xffff48000b60; is_svm_addr=0)
[ 1822.410573] [ascend] [tsdrv] [ERROR] [logic_ioctl_cq_wait 1428] <MONITOR_0:1632> ts is down, devid(0) fid(0) tsid(0)
[ 1827.530569] [ascend] [tsdrv] [ERROR] [devdrv_mailbox_kernel_sync_no_feedback 437] <main:1632> [dev_id=0 tsid=0]:devdrv_mailbox_message_send failed. ret=-62
请问内核这个错误的原因是什么？我该如何解决？
使用的驱动版本是：Ascend-hdk-310p-npu-driver_25.2.3_linux-aarch64.run，固件版本是：Ascend-hdk-310p-npu-firmware_7.7.0.10.220.run
+--------------------------------------------------------------------------------------------------------+
| npu-smi 25.2.3 Version: 25.2.3 |
+-------------------------------+-----------------+------------------------------------------------------+
| NPU Name | Health | Power(W) Temp(C) Hugepages-Usage(page) |
| Chip Device | Bus-Id | AICore(%) Memory-Usage(MB) |
+===============================+=================+======================================================+
| 0 310P3 | OK | NA 50 0 / 0 |
| 0 0 | 0000:01:00.0 | 0 1855 / 21527 |
+===============================+=================+======================================================+
+-------------------------------+-----------------+------------------------------------------------------+
| NPU Chip | Process id | Process name | Process memory(MB) |
+===============================+=================+======================================================+
| No running processes found in NPU 0 |
+===============================+=================+======================================================+
运行过resnet50 demo模型是可以成功运行的。
错误的截图如下：