UniqAI/nvidia-container-toolkit
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvidia-container-toolkit synced 2024-11-24 13:05:17 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
1,965 Commits 21 Branches 73 Tags 29 MiB
b12bdfc52a
Commit Graph

1965 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Evan Lezar
0938576618 Remove NVIDIA experimental runtime from toolkit container 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-10 11:44:55 +02:00
Evan Lezar
4ca8d4173a Merge branch 'revert-d5cbe48d' into 'main' 
Revert "Merge branch 'bump-golang-1.20.5' into 'main'"

See merge request nvidia/container-toolkit/container-toolkit!437
 2023-07-05 15:12:01 +00:00
Evan Lezar
978549dc58 Revert "Merge branch 'bump-golang-1.20.5' into 'main'" 
This reverts merge request !436
 2023-07-05 15:11:41 +00:00
Evan Lezar
d5cbe48d59 Merge branch 'bump-golang-1.20.5' into 'main' 
Bump golang version to 1.20.5

See merge request nvidia/container-toolkit/container-toolkit!436
 2023-07-05 14:07:58 +00:00
Evan Lezar
e8ec795883 Bump golang version to 1.20.5 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-05 16:07:41 +02:00
Evan Lezar
62bc6b211f Merge branch 'bump-cuda-12.2.0' into 'main' 
Bump cuda base image to 12.2.0

See merge request nvidia/container-toolkit/container-toolkit!435
 2023-07-05 10:11:27 +00:00
Evan Lezar
6fac6c237b Bump cuda base image to 12.2.0 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-05 11:28:32 +02:00
Evan Lezar
20ff4e2fb9 Merge branch 'generate-default-config-post-install' into 'main' 
Ensure that default config is created on the file system as a post-install step

See merge request nvidia/container-toolkit/container-toolkit!431
 2023-07-05 09:27:29 +00:00
Evan Lezar
f78d3a858f Rework default config generation to not use toml 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-05 11:26:55 +02:00
Evan Lezar
bc6ca7ff88 Generate default config post-install 
The debian and rpm packages are updated to trigger the generation of
of a default config if no config exists at the expected location.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-05 11:26:53 +02:00
Evan Lezar
65ae6f1dab Fix generation of default config 
This change ensures that the nvidia-ctk config default command
generates a config file that is compatible with the official documentation
to, for example, disable cgroups in the NVIDIA Container CLI.

This requires that whitespace around comments is stripped before outputing the
contets.

This also adds an option to load a config and modify it in-place instead. This can
be triggered as a post-install step, for example.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-05 11:26:04 +02:00
Evan Lezar
ba24338122 Add quiet mode to nvidia-ctk cli 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-05 11:26:04 +02:00
Evan Lezar
2299c9588d Merge branch 'create-config-folders' into 'main' 
Ensure that folders exist when creating config files

See merge request nvidia/container-toolkit/container-toolkit!433
 2023-07-05 09:25:28 +00:00
Evan Lezar
ba80d0318f Merge branch 'rpm-fix-missing-coreutils-during-install' into 'main' 
RPM spec: Avoid scriptlet failure during initial system installation

See merge request nvidia/container-toolkit/container-toolkit!432
 2023-07-05 08:43:26 +00:00
Evan Lezar
6342dae0e9 Ensure that parent directories exist for config files 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-03 15:30:31 +02:00
Evan Lezar
baf94181aa Add engine.Config to encapsulate writing 
This change adds an engine.Config type to encapsulate the writing
of config files for container engines.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-07-03 15:26:47 +02:00
Evan Lezar
bbe9742c46 Merge branch 'switch-to-latest-dind' into 'main' 
Switch to latest dind image for tests

See merge request nvidia/container-toolkit/container-toolkit!430
 2023-06-30 09:46:49 +00:00
Evan Lezar
1447ef3818 Switch to latest dind image for tests 
The stable-dind image is out of date and has not been updated for 3 years.
This change updates to the latest dind image.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-30 11:03:07 +02:00
Claudius Volz
5598dbf9d7 RPM spec: Only run fixup code if the package is being upgraded, to avoid a scenario where the coreutils (mkdir, cp) are not available yet during a fresh system installation. 
Signed-off-by: Claudius Volz <c.volz@gmx.de>
 2023-06-29 00:23:24 +02:00
Evan Lezar
8967e851c4 Merge branch 'fix-multiple-driver-roots-wsl' into 'main' 
Fix bug with multiple driver store paths

See merge request nvidia/container-toolkit/container-toolkit!425
 2023-06-27 14:15:38 +00:00
Evan Lezar
15378f6ced Merge branch 'fix-ordering-of-envvars' into 'main' 
Ensure common envvars have higher precedence

See merge request nvidia/container-toolkit/container-toolkit!426
 2023-06-27 13:26:34 +00:00
Evan Lezar
4d2e8d1913 Ensure common envvars have higher precedence 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-27 14:45:15 +02:00
Evan Lezar
4feaee0fe6 Merge branch 'bump-version' into 'main' 
Bump version to v1.14.0-rc.2

See merge request nvidia/container-toolkit/container-toolkit!427
 2023-06-27 12:38:05 +00:00
Evan Lezar
51984d49cf Update libnvidia-container 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-27 14:37:26 +02:00
Evan Lezar
a6a8bb940c Bump version to v1.14.0-rc.2 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-27 14:02:59 +02:00
Evan Lezar
6265e34afb Fix bug with multiple driver store paths 
This change uses the actual discovered path of nvidia-smi when
creating a symlink to the binary on WSL2 platforms.

This ensures that cases where multiple driver store paths are
detected are supported.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-26 21:37:14 +02:00
Evan Lezar
d08a2394b3 Merge branch 'fix-package-archive-script' into 'main' 
Fix package archive script

See merge request nvidia/container-toolkit/container-toolkit!424
 2023-06-26 11:46:26 +00:00
Evan Lezar
c0f1263d78 Fix package archive script 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-26 13:46:08 +02:00
Evan Lezar
a25b1c1048 Merge branch 'fix-load-kernel-modules' into 'main' 
Split internal system package

See merge request nvidia/container-toolkit/container-toolkit!420
 2023-06-26 08:30:51 +00:00
Evan Lezar
99859e461d Merge branch 'import-wrapper-and-runtime' into 'main' 
Import NVIDIA Docker and NVIDIA Container Runtime to in-tree folders

See merge request nvidia/container-toolkit/container-toolkit!418
 2023-06-21 12:33:05 +00:00
Evan Lezar
d52dbeaa7a Split internal system package 
This changes splits the functionality in the internal system package
into two packages: one for dealing with devices and one for dealing
with kernel modules. This removes ambiguity around the meaning of
driver / device roots in each case.

In each case, a root can be specified where device nodes are created
or kernel modules loaded.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-15 09:01:13 +02:00
Evan Lezar
c11c7695cb Merge branch 'update-go-nvlib' into 'main' 
Update go-nvlib with new constructor API

See merge request nvidia/container-toolkit/container-toolkit!422
 2023-06-14 22:50:12 +00:00
Evan Lezar
c4d3b13ae2 Update go-nvlib with new constructor 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-14 17:55:33 +02:00
Evan Lezar
bcf3a70174 Update vendoring 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-14 17:55:18 +02:00
Evan Lezar
743d290577 Merge branch 'CNT-4301/resolve-auto-to-cdi' into 'main' 
Resolve auto mode to cdi if all devices are cdi devices

See merge request nvidia/container-toolkit/container-toolkit!421
 2023-06-13 14:48:32 +00:00
Evan Lezar
82347eb9bc Resolve auto mode as cdi for fully-qualified names 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-13 16:05:37 +02:00
Evan Lezar
84c7bf8b18 Minor refactor of mode resolver 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-13 16:04:03 +02:00
Evan Lezar
d92300506c Construct CUDA image object once 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-13 10:36:02 +02:00
Evan Lezar
2da32970b9 Merge branch 'refactor-logger' into 'main' 
Refactor Logging

See merge request nvidia/container-toolkit/container-toolkit!416
 2023-06-12 08:59:41 +00:00
Evan Lezar
1d0a733487 Replace logger.Warn(f) with logger.Warning(f) 
This aligns better with klog used in other projects.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-12 10:48:04 +02:00
Evan Lezar
9464953924 Use logger.Interface when resolving auto mode 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-12 10:46:11 +02:00
Evan Lezar
c9b05d8fed Use logger Interface in runtime configuration 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-12 10:46:11 +02:00
Evan Lezar
a02bc27c3e Define a basic logger interface 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-12 10:46:10 +02:00
Carlos Eduardo Arango Gutierrez
6a04e97bca Merge branch 'use-same-envvars-for-runtime-config' into 'main' 
Allow same envars for all runtime configs

See merge request nvidia/container-toolkit/container-toolkit!357
 2023-06-12 08:32:34 +00:00
Evan Lezar
0780621024 Ensure runtime dir is set for crio cleanup 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-11 12:46:08 +02:00
Evan Lezar
2bc0f45a52 Remove unused constants and variables 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-11 11:38:22 +02:00
Evan Lezar
178eb5c5a8 Rework restart logic 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-10 12:41:53 +02:00
Evan Lezar
761fc29567 Add version info to config CLIs 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-09 18:49:17 +02:00
Evan Lezar
9f5c82420a Refactor toolking to setup and cleanup configs 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-09 18:49:15 +02:00
Evan Lezar
23041be511 Add runtimeDir as argument 
Thsi change adds the --nvidia-runtime-dir as a command line
argument when configuring container runtimes in the toolkit container.
This removes the need to set it via the command line.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-06-09 18:48:34 +02:00