Default Content Mounted by Ascend Docker Runtime

Ascend Docker Runtime mounts the following directories and files to the container in read-only mode by default based on the actual environment.

Table 1 Default mounted directories and files (Atlas 200 AI accelerator module (RC))

Path

Description

/dev/davinciX

NPU device. X indicates the ID, for example, davinci0.

/dev/davinci_manager

Management device.

/usr/local/Ascend/driver/tools

Directory for storing the tool packages provided by the driver.

/usr/local/Ascend/driver/lib64

Directory for storing user-mode libraries provided by the driver.

/usr/local/sbin/npu-smi

npu-smi tool file.

/etc/hdcBasic.cfg

Basic HDC file.

/etc/sys_version.conf

File for storing the driver version information.

/dev/dvpp_cmdlist

Device files, which support inference services.

/var/queue_schedule

Management FlowGW scheduling framework.

NOTE:

To mount this directory, the following conditions must be met:

  • The MindCluster version must be 6.0.0 or later.
  • The HDK version must be 24.1.RC2 or later.
Table 2 Default mounted directories and files (Atlas 200I SoC A1 core board)

Path

Description

/dev/davinciX

NPU device. X indicates the ID, for example, davinci0.

/dev/davinci_manager

Device manager related to Da Vinci.

/usr/local/bin/npu-smi

npu-smi tool file.

/etc/hdcBasic.cfg

Basic HDC file.

/etc/sys_version.conf

File for storing the driver version information.

/dev/dvpp_cmdlist

Device files, which support inference services.

/var/queue_schedule

Management FlowGW scheduling framework.

NOTE:

To mount this directory, the following conditions must be met:

  • The MindCluster component version must be 6.0.0 or later.
  • The HDK version must be 24.1.RC2 or later.
Table 3 Default mounted directories and files (Atlas 200I A2 accelerator module and Atlas 200I DK A2 developer kit)

Path

Description

/dev/davinciX

NPU device. X indicates the ID, for example, davinci0.

/dev/davinci_manager

Device manager related to Da Vinci.

/dev/svm0

Device for memory management.

/dev/ts_aisle

aicpudrv driver device, which provides event-driven channel interfaces for job scheduling.

/dev/upgrade

Driver device.

/dev/sys

/dev/vdec

Device files, which support inference services.

/dev/vpc

/dev/pngd

/dev/venc

/dev/dvpp_cmdlist

/dev/log_drv

Log driver device.

/etc/sys_version.conf

File for storing the driver version information.

/etc/hdcBasic.cfg

Basic HDC file.

/usr/local/sbin/npu-smi

npu-smi tool file.

/usr/local/Ascend/driver/lib64

Directory for storing user-mode libraries provided by the driver.

/usr/lib64/aicpu_kernels/

/var/slogd

Log component file.

/var/dmp_daemon

DMP daemon process file.

/usr/lib64/libcrypto.so.1.1

Dynamic libraries required by the driver.

This path needs to be mounted to openEuler 22.03.

/usr/lib64/libyaml-0.so.2

/usr/lib/aarch64-linux-gnu/libcrypto.so.1.1

Dynamic libraries required by the driver.

This path needs to be mounted to Ubuntu 22.04.

/usr/lib/aarch64-linux-gnu/libyaml-0.so.2

/usr/lib64/libaicpu_processer.so

Dynamic libraries required by the driver.

/usr/lib64/libaicpu_prof.so

/usr/lib64/libaicpu_sharder.so

/usr/lib64/libadump.so

/usr/lib64/libtsd_eventclient.so

/usr/lib64/libaicpu_scheduler.so

/usr/lib64/libdcmi.so

/usr/lib64/libmpi_dvpp_adapter.so

/usr/lib64/libstackcore.so

/var/queue_schedule

Management FlowGW scheduling framework.

NOTE:

To mount this directory, the following conditions must be met:

  • The MindCluster component version must be 6.0.0 or later.
  • The HDK version must be 24.1.RC2 or later.
Table 4 Default mounted directories and files (Atlas 500 AI edge station (model 3000))

Path

Description

/dev/davinciX

NPU device. X indicates the ID, for example, davinci0.

/dev/davinci_manager

Management device.

/dev/hisi_hdc

Management device.

/dev/devmm_svm

Management device.

/home/data/miniD/driver/lib64

Directory for storing user-mode libraries provided by the driver.

/usr/local/dcmi

Directory for storing DCMI header files and libraries.

/usr/local/lib/libdcmi.so

File for storing the DCMI dynamic library.

/usr/local/bin/npu-smi

npu-smi tool file.

/dev/dvpp_cmdlist

Device files, which support inference services.

/var/queue_schedule

Management FlowGW scheduling framework.

NOTE:

To mount this directory, the following conditions must be met:

  • The MindCluster component version must be 6.0.0 or later.
  • The HDK version must be 24.1.RC2 or later.
Table 5 Default mounted directories and files (Atlas 500 A2 edge station)

Path

Description

/dev/davinciX

NPU device. X indicates the ID, for example, davinci0.

/dev/davinci_manager

Device manager related to Da Vinci.

/dev/svm0

Device for memory management.

/dev/ts_aisle

aicpudrv driver device, which provides event-driven channel interfaces for job scheduling.

/dev/upgrade

Driver device.

/dev/sys

/dev/vdec

Device files, which support inference services.

/dev/vpc

/dev/pngd

/dev/venc

/dev/dvpp_cmdlist

/dev/log_drv

Log driver device.

/usr/local/Ascend/driver/lib64

Directory for storing user-mode libraries provided by the driver.

/usr/lib64/aicpu_kernels

/usr/local/sbin/npu-smi

npu-smi tool file.

/etc/sys_version.conf

File for storing the driver version information.

/etc/ld.so.conf.d/mind_so.conf

File for configuring the dynamic library path of the driver.

/etc/hdcBasic.cfg

Basic HDC file.

/var/dmp_daemon

DMP daemon process file.

/var/slogd

Log component file.

/usr/lib64/libcrypto.so.1.1

Dynamic libraries required by the driver.

This path needs to be mounted to openEuler 22.03/EulerOS 2.11 and later.

/usr/lib64/libyaml-0.so.2

/usr/lib/aarch64-linux-gnu/libcrypto.so.1.1

Dynamic libraries required by the driver.

This path needs to be mounted to Ubuntu 22.04.

/usr/lib/aarch64-linux-gnu/libyaml-0.so.2

/usr/lib64/libsemanage.so.2

Dynamic libraries required by the driver.

/usr/lib64/libmmpa.so

/usr/lib64/libdrvdsmi.so

/usr/lib64/libdcmi.so

/usr/lib64/libstackcore.so

/usr/lib64/libmpi_dvpp_adapter.so

/usr/lib64/libaicpu_scheduler.so

/usr/lib64/libaicpu_processer.so

/usr/lib64/libaicpu_prof.so

/usr/lib64/libaicpu_sharder.so

/usr/lib64/libadump.so

/usr/lib64/libtsd_eventclient.so

/var/queue_schedule

Management FlowGW scheduling framework.

NOTE:

To mount this directory, the following conditions must be met:

  • The MindCluster component version must be 6.0.0 or later.
  • The HDK version must be 24.1.RC2 or later.
Table 6 Default mounting directories and files for other devices

Path

Description

/dev/davinciX

NPU device. X indicates the ID, for example, davinci0.

/dev/davinci_manager

Management device.

/dev/hisi_hdc

Management device.

/dev/devmm_svm

Management device.

/usr/local/Ascend/driver/lib64

Directory for storing user-mode libraries provided by the driver.

/usr/local/Ascend/driver/include

Directory for storing the header file provided by the driver.

/usr/local/dcmi

Directory for storing DCMI header files and libraries.

/usr/local/bin/npu-smi

npu-smi tool file.

/dev/dvpp_cmdlist

Device files, which support the digital vision preprocessing function.

/var/queue_schedule

Management FlowGW scheduling framework.

NOTE:

To mount this directory, the following conditions must be met:

  • The MindCluster component version must be 6.0.0 or later.
  • The HDK version must be 24.1.RC2 or later.