Atlas 300V Pro 安装失败:dcmi module initialize failed. ret is -8005
收藏回复举报
Atlas 300V Pro 安装失败:dcmi module initialize failed. ret is -8005
t('forum.solved') 已解决
新人帖
发表于2025-05-26 11:28:58
0 查看

服务器:KunTai R722  + Atlas 300V Pro * 7

操作系统:

[root@localhost ~]# hostnamectl 

   Static hostname: localhost.localdomain 

         Icon name: computer-server 

           Chassis: server 

        Machine ID: ba9549085f2945889ffbbd61603dc1d3 

           Boot ID: 36c4a3d090b94db78e6aefec9e3ae234 

  Operating System: Kylin Linux Advanced Server V10 (Tercel) 

            Kernel: Linux 4.19.90-17.ky10.aarch64 

      Architecture: arm64 

安装驱动提示:

[root@localhost ~]# rpm -ivh Atlas-300v-pro-npu-driver-21.0.3.3-1.aarch64.rpm 

Verifying...                          ################################# [100%] 

准备中...                          ################################# [100%] 

正在升级/安装... 

   1:Atlas-300v-pro-npu-driver-21.0.3.################################# [100%] 

[Driver] [2025-05-26 10:23:37] [INFO]Start time: 2025-05-26 10:23:37 

[Driver] [2025-05-26 10:23:37] [INFO]LogFile: /var/log/ascend_seclog/ascend_install.log 

[Driver] [2025-05-26 10:23:37] [INFO]OperationLogFile: /var/log/ascend_seclog/operation.log 

[Driver] [2025-05-26 10:23:37] [WARNING]Do not power off or restart the system during the installation/upgrade 

[Driver] [2025-05-26 10:23:37] [INFO]set username and usergroup, HwHiAiUser:HwHiAiUser 

[Driver] [2025-05-26 10:23:38] [INFO]driver install type: Rebuild 

[Driver] [2025-05-26 10:23:38] [INFO]Rebuilding by kernel path(/lib/modules/4.19.90-17.ky10.aarch64/build), detail in /var/log/ascend_               seclog/ascend_rebuild.log 

[Driver] [2025-05-26 10:24:50] [INFO]upgradePercentage:10% 

[Driver] [2025-05-26 10:24:52] [INFO]upgradePercentage:30% 

[Driver] [2025-05-26 10:24:52] [INFO]upgradePercentage:40% 

[Driver] [2025-05-26 10:24:53] [INFO]upgradePercentage:90% 

[Driver] [2025-05-26 10:24:53] [INFO]Waiting for device startup... 

[Driver] [2025-05-26 10:24:56] [INFO]Device startup success 

***********all scan davinci num 0 

[Driver] [2025-05-26 10:29:57] [INFO]upgradePercentage:100% 

[Driver] [2025-05-26 10:32:33] [WARNING]There are some chips not ready, please check after reboot 

[Driver] [2025-05-26 10:32:33] [INFO]Driver package install success! Reboot needed for installation/upgrade to take effect! 

[Driver] [2025-05-26 10:32:33] [INFO]End time: 2025-05-26 10:32:33 

警告:%post(Atlas-300v-pro-npu-driver-21.0.3.3-1.aarch64) 脚本执行失败,退出状态码为 2 

[root@localhost ~]# rpm -ivh Atlas-300v-pro-npu-firmware-1.79.22.7.220-1.noarch.rpm 

Verifying...                          ################################# [100%] 

准备中...                          ################################# [100%] 

正在升级/安装... 

   1:Atlas-300v-pro-npu-firmware-1:1.7################################# [100%] 

[Firmware] [2025-05-26 10:33:07] [INFO]Start time: 2025-05-26 10:33:07 

[Firmware] [2025-05-26 10:33:07] [INFO]LogFile: /var/log/ascend_seclog/ascend_install.log 

[Firmware] [2025-05-26 10:33:07] [INFO]OperationLogFile: /var/log/ascend_seclog/operation.log 

[Firmware] [2025-05-26 10:33:07] [WARNING]Do not power off or restart the system during the installation/upgrade 

[Firmware] [2025-05-26 10:35:48] [ERROR]None of the chips are up. 

[Firmware] [2025-05-26 10:35:48] [INFO]End time: 2025-05-26 10:35:48 

运行命令报错:

[root@localhost ~]# npu-smi info 

dcmi module initialize failed. ret is -8005 

dmesg

[  280.803164] [ascend] [devdrv] [devdrv_manager_first_get_devnum 274] <npu-smi:11846> pcie report dev_num = 7 

[  280.813304] [ascend] [devdrv] [devdrv_manager_container_get_devnum 1324] <npu-smi:11846> weird device number, dev_num = 0 

[  281.850304] [ascend] [devdrv] [devdrv_manager_container_get_devnum 1324] <npu-smi:11846> weird device number, dev_num = 0 

[  282.890266] [ascend] [devdrv] [devdrv_manager_container_get_devnum 1324] <npu-smi:11846> weird device number, dev_num = 0 

[  283.930226] [ascend] [devdrv] [devdrv_manager_container_get_devnum 1324] <npu-smi:11846> weird device number, dev_num = 0 

[  284.970185] [ascend] [devdrv] [devdrv_manager_container_get_devnum 1324] <npu-smi:11846> weird device number, dev_num = 0 

[ 2424.556102] [ascend] [ERROR] [drv_pcie] [devdrv_get_rd_flag 466] <kworker/48:4:3607:3607> Load timeout. (dev_id=12; flag_r=6666666666666666) 

[ 2424.569084] [ascend] [drv_pcie] [devdrv_notify_blackbox_err 741] <kworker/48:4:3607:3607> Get blaclbox code. (dev_id=12; blaclbox_code=1744965632) 

[ 2424.582790] [ascend] [drv_pcie] [devdrv_load_file_copy 541] <kworker/48:0:254:254> Get file size. (dev_id=10; file_name="/usr/local/Ascend/driver/device/ascend710_dt.img"; size=949565) 

我要发帖子