UniqAI/nvidia-container-toolkit
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvidia-container-toolkit synced 2024-11-30 00:11:59 +00:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity
1,540 Commits 20 Branches 73 Tags 29 MiB
c90211e070
Commit Graph

1540 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Evan Lezar
7a533aeff3 Merge branch 'update-nvcdi-new-with-error' into 'main' 
Allow nvcdi.Option to return an error

See merge request nvidia/container-toolkit/container-toolkit!352
 2023-03-26 14:13:41 +00:00
Evan Lezar
226c54613e Also return an error from nvcdi.New 
This change allows nvcdi.New to return an error in addition to the
constructed library instead of panicing.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-26 16:13:12 +02:00
Evan Lezar
1ebbebf5de Merge branch 'CNT-3932/deduplicate-entries-in-cdi-spec' into 'main' 
Add transform to deduplicate entities in CDI spec

See merge request nvidia/container-toolkit/container-toolkit!345
 2023-03-24 19:04:43 +00:00
Evan Lezar
33f6fe0217 Generate a simplified CDI spec by default 
As simplified CDI spec has no duplicate entities in any single set of container edits.
Furthermore, contianer edits defined at a spec-level are not included in the container
edits for a device.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-24 11:01:46 +02:00
Evan Lezar
5ff206e1a9 Add transform to deduplicate entities in CDI spec 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-24 11:01:23 +02:00
Evan Lezar
df618d3cba Merge branch 'CNT-4052/fix-arm-management-containers' into 'main' 
Fix generation of management CDI spec in containers

See merge request nvidia/container-toolkit/container-toolkit!354
 2023-03-23 16:39:10 +00:00
Evan Lezar
9506bd9da0 Fix generation of management CDI spec in containers 
Since we relied on finding libcuda.so in the LDCache to determine both the CUDA
version and the expected directory for the driver libraries, the generation of the
management CDI specifications fails in containers where the LDCache has not been updated.

This change falls back to searching a set of predefined paths instead when the lookup of
libcuda.so in the cache fails.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-23 15:59:01 +02:00
Evan Lezar
5e0684e99d Merge branch 'update-libnvidia-container' into 'main' 
Update libnvidia-container

See merge request nvidia/container-toolkit/container-toolkit!353
 2023-03-23 08:50:18 +00:00
Evan Lezar
09a0cb24cc Remove fedora make targets 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-23 10:35:57 +02:00
Evan Lezar
ff92f1d799 Update libnvidia-container 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-23 10:33:26 +02:00
Christopher Desiniotis
b87703c503 Merge branch 'fix-nil-logger-in-library-locator' into 'main' 
Instantiate a logger when constructing a library Locator

See merge request nvidia/container-toolkit/container-toolkit!351
 2023-03-21 21:54:14 +00:00
Christopher Desiniotis
b2aaa21b0a Instantiate a logger when constructing a library Locator 
Signed-off-by: Christopher Desiniotis <cdesiniotis@nvidia.com>
 2023-03-21 13:38:36 -07:00
Evan Lezar
310c15b046 Merge branch 'CNT-4026/only-init-nvml-when-required' into 'main' 
Only init nvml as required when generating CDI specs

See merge request nvidia/container-toolkit/container-toolkit!344
 2023-03-20 13:26:07 +00:00
Evan Lezar
685802b1ce Only init nvml as required when generating CDI specs 
CDI generation modes such as management and wsl don't require
NVML. This change removes the top-level instantiation of nvmllib
and replaces it with an instanitation in the nvml CDI spec generation
code.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-20 14:24:08 +02:00
Evan Lezar
380eb8340a Merge branch 'blossom-ci' into 'main' 
Add blossom-ci github action

See merge request nvidia/container-toolkit/container-toolkit!349
 2023-03-20 09:56:23 +00:00
Evan Lezar
f98e1160f5 Update components with blossim-ci 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-20 11:06:44 +02:00
Evan Lezar
1962fd68df Merge branch 'locate-ipc-sockets-at-run' into 'main' 
Locate persistenced and fabricmanager sockets at /run instead of /var/run

See merge request nvidia/container-toolkit/container-toolkit!347
 2023-03-20 08:08:59 +00:00
Carlos Eduardo Arango Gutierrez
29813c1e14
Add blossom-ci github action 
Signed-off-by: Carlos Eduardo Arango Gutierrez <eduardoa@nvidia.com>
 2023-03-17 16:16:27 +01:00
Evan Lezar
df40fbe03e Locate persistenced and fabricmanager sockets at /run instead of /var/run 
This chagne prefers (non-symlink) sockets at /run over /var/run for
nvidia-persistenced and nvidia-fabricmanager sockets.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-17 09:23:48 +02:00
Carlos Eduardo Arango Gutierrez
7000c6074e Merge branch 'ci_rules' into 'main' 
Rework pipeline triggers for MRs

See merge request nvidia/container-toolkit/container-toolkit!346
 2023-03-15 13:15:23 +00:00
Evan Lezar
ef1fe3ab41 Rework pipeline triggers for MRs 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-15 14:15:20 +02:00
Evan Lezar
fdd198b0e8 Merge branch 'bump-v1.13.0-rc.3' into 'main' 
Bump version to v1.13.0-rc.3

See merge request nvidia/container-toolkit/container-toolkit!343
 2023-03-15 07:50:50 +00:00
Evan Lezar
e37f77e02d Update libnvidia-container 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-15 09:49:49 +02:00
Evan Lezar
3fcfee88be Bump version to v1.13.0-rc.3 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-15 09:26:19 +02:00
Evan Lezar
a082413d09 Merge branch 'trigger-ci-on-mrs-only' into 'main' 
Add workflow rule to only trigger on MRs

See merge request nvidia/container-toolkit/container-toolkit!342
 2023-03-15 07:10:30 +00:00
Evan Lezar
280f40508e Make pipeline manual on MRs 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-15 08:51:18 +02:00
Evan Lezar
e2be0e2ff0 Add workflow rule to only trigger on MRs 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-15 08:45:26 +02:00
Evan Lezar
dcff3118d9 Merge branch 'update-libnvidia-container' into 'main' 
Update libnvidia-container

See merge request nvidia/container-toolkit/container-toolkit!340
 2023-03-14 13:54:11 +00:00
Evan Lezar
731168ec8d Update changelog 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-14 15:05:36 +02:00
Evan Lezar
7b4435a0f8 Update libnvidia-container 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-14 15:05:29 +02:00
Evan Lezar
738af29724 Merge branch 'explicit-cdi-enabled-flag' into 'main' 
Add --cdi-enabled option to control generating CDI spec

See merge request nvidia/container-toolkit/container-toolkit!339
 2023-03-14 07:00:30 +00:00
Evan Lezar
08ef242afb Add --cdi-enabled option to control generating CDI spec 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-13 18:19:00 +02:00
Evan Lezar
92ea8be309 Merge branch 'fix-privileged-check-cdi-mode' into 'main' 
Return empty list of devices for unprivileged containers when...

See merge request nvidia/container-toolkit/container-toolkit!337
 2023-03-13 07:36:25 +00:00
Christopher Desiniotis
48414e97bb Return empty list of devices for unprivileged containers when 'accept-nvidia-visible-devices-envvar-unprivileged=false' 
Signed-off-by: Christopher Desiniotis <cdesiniotis@nvidia.com>
 2023-03-10 13:11:29 -08:00
Evan Lezar
77a2975524 Merge branch 'fix-kitmaker' into 'main' 
Use component name as folder name

See merge request nvidia/container-toolkit/container-toolkit!336
 2023-03-10 13:57:24 +00:00
Evan Lezar
ce9477966d Use component name as folder name 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-10 15:51:36 +02:00
Evan Lezar
fe02351c3a Merge branch 'bump-cuda-version' into 'main' 
Bump CUDA base image version to 12.1.0

See merge request nvidia/container-toolkit/container-toolkit!335
 2023-03-10 10:23:30 +00:00
Evan Lezar
9c2018a0dc Bump CUDA base image version to 12.1.0 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-10 11:31:23 +02:00
Evan Lezar
33e5b34fa1 Merge branch 'CNT-3999/legacy-cli-doesnt-work-in-cdi-mode' into 'main' 
Add nvidia-container-runtime-hook.skip-mode-detection option to config

See merge request nvidia/container-toolkit/container-toolkit!330
 2023-03-09 19:18:16 +00:00
Evan Lezar
ccf73f2505 Set skip-mode-detection in the toolkit-container by default 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 20:16:10 +02:00
Evan Lezar
3a11f6ee0a Add nvidia-container-runtime-hook.skip-mode-detection option to config 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 20:15:40 +02:00
Evan Lezar
8f694bbfb7 Merge branch 'set-nvidia-ctk-path' into 'main' 
Set nvidia-ctk.path config option based on installed path

See merge request nvidia/container-toolkit/container-toolkit!334
 2023-03-09 16:44:13 +00:00
Evan Lezar
4c2eff4865 Merge branch 'CNT-3998/cdi-accept-visible-devices-when-privileged' into 'main' 
Honor accept-nvidia-visible-devices-envvar-when-unprivileged setting in CDI mode

See merge request nvidia/container-toolkit/container-toolkit!331
 2023-03-09 15:59:08 +00:00
Evan Lezar
1fbdc17c40 Set nvidia-ctk.path config option based on installed path 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 17:53:08 +02:00
Evan Lezar
965d62f326 Merge branch 'fix-containerd-integration-tests' into 'main' 
Fix integration tests failing due to CDI spec generation

See merge request nvidia/container-toolkit/container-toolkit!333
 2023-03-09 14:41:52 +00:00
Evan Lezar
25ea7fa98e Remove whitespace in Makefile 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 15:32:07 +02:00
Evan Lezar
5ee040ba95 Disable CDI spec generation for integration tests 2023-03-09 15:32:07 +02:00
Evan Lezar
eb2aec9da8 Allow CDI options to be set by envvars 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 12:25:05 +02:00
Evan Lezar
973e7bda5e Check accept-nvidia-visible-devices-envvar-when-unprivileged option for CDI 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 11:15:53 +02:00
Evan Lezar
154cd4ecf3 Add to config struct 
Signed-off-by: Evan Lezar <elezar@nvidia.com>
 2023-03-09 11:15:53 +02:00