add support for nvidia gpu access #5132

MondoGao · 2024-08-16T05:08:32Z

Add a new environment to enable Nvidia gpu access when creating containers. It's similar to /dev/dri support. See #1525.

Note: I didn't fully inspect and test this change since I don't have php development environment set up in my machine. Feel free to inherit this pr to enhance its maintainability.

szaimen · 2024-08-16T11:09:55Z

Hi @MondoGao thanks for your PR! Can you please fix DCO? Wee need that in order to be able to merge it later on. See https://github.com/nextcloud/all-in-one/pull/5132/checks?check_run_id=28844781358

MondoGao · 2024-08-16T19:48:10Z

Hi @MondoGao thanks for your PR! Can you please fix DCO? Wee need that in order to be able to merge it later on. See https://github.com/nextcloud/all-in-one/pull/5132/checks?check_run_id=28844781358

Sign added :)

szaimen · 2024-08-19T12:58:01Z

Hi, I had a fast look at this and I think it would add the capability to all containers that are controlled by AIO. Better would probably be to add this as a capability to containers-schema.json and add this only to certain containers via containers.json

Also this is missing some places, e.g. adding documentation on it in the readme. See #1659 as inspiration.

Additionally, do you know if this is also going to work with AMD and Intel GPUs? In best case we create only one setting that works for all of them.

MondoGao · 2024-08-19T19:04:42Z

Hi, I had a fast look at this and I think it would add the capability to all containers that are controlled by AIO. Better would probably be to add this as a capability to containers-schema.json and add this only to certain containers via containers.json

Also this is missing some places, e.g. adding documentation on it in the readme. See #1659 as inspiration.

Additionally, do you know if this is also going to work with AMD and Intel GPUs? In best case we create only one setting that works for all of them.

I believe docker runtime only supports Nvidia GPU passthrough. Intel & AMD CPU/GPU's hardware acceleration is exposed through /dev/dri.

Looks like I have to install php dev env to polish this pr, please expect a late response.

luzfcb · 2024-08-19T19:51:19Z

@MondoGao Thank you for starting this effort.

@MondoGao @szaimen

Besides supporting Nvidia GPU passthrough, there are cases where the container may require the same specific version of the NVIDIA driver installed on the host system. It might be a good idea to create an environment variable for the Nvidia driver version and let the container handle the Nvidia driver setup and configuration logic inside the container.

That is, Nextcloud AIO could either provide some way for the user to manually specify the version or, if it is not provided by the user, then have an automatic way to obtain this information and fill an NVIDIA_DRIVER_VERSION environment variable that will be available to all containers managed by Nextcloud AIO.

I have exactly this with docker-steam-headless, and it works great.

This is the driver download and installation script: https://github.com/Steam-Headless/docker-steam-headless/blob/860451da74b397385f1b1658545d2bb891aa8e46/overlay/etc/cont-init.d/60-configure_gpu_driver.sh

This script creates the X Server configuration files and other related configurations. It probably does not apply to Nextcloud and plugins since they do not use X Server, but I am including it here just for reference.

https://github.com/Steam-Headless/docker-steam-headless/blob/860451da74b397385f1b1658545d2bb891aa8e46/overlay/etc/cont-init.d/70-configure_xorg.sh

docjyJ · 2024-08-21T11:44:06Z

It would be nice to have this feature!

We can have Nextcloud Assistant servers with high performance easily!!!!!!

I agree with @szaimen the configuration should be done in the containers.json schema.

Otherwise it looks good to me.

AMD GPU : https://rocm.docs.amd.com/projects/install-on-linux/en/latest/how-to/docker.html

gbertolottiKS · 2024-12-12T14:29:01Z

Just to be sure to have understood this topic correctly: actually there's no way to use a dedicated GPU in Nextcloud AIO, since it is needed a development to make sure it can see (and then use) the graphic card. Is it correct?
This PR should resolve this : once finished, AIO can use "natively" the GPU, right?

Aren't some workaround to be able to use the GPU in the meantime? We are happy to help to test if it's necessary.

Thanks

szaimen · 2024-12-13T16:22:45Z

Just to be sure to have understood this topic correctly: actually there's no way to use a dedicated GPU in Nextcloud AIO, since it is needed a development to make sure it can see (and then use) the graphic card. Is it correct? This PR should resolve this : once finished, AIO can use "natively" the GPU, right?

Yes

Aren't some workaround to be able to use the GPU in the meantime?

Currently no as this PR is unfortunately not even close to being finished. You could try to find someone that takes over finishing this PR to speed up the development.

docjyJ

Any news?
What's blocking the merger?

php/src/Data/ConfigurationManager.php

szaimen · 2024-12-17T13:17:27Z

Any news? What's blocking the merger?

See #5132 (comment)

szaimen · 2024-12-17T13:17:44Z

Also, twig CI seems to fail...

docjyJ · 2024-12-17T15:12:50Z

Any news? What's blocking the merger?

See #5132 (comment)

I try to look this

docjyJ · 2024-12-17T15:43:14Z

Also, twig CI seems to fail...

Fix

docjyJ · 2024-12-17T15:54:19Z

Hi, I had a fast look at this and I think it would add the capability to all containers that are controlled by AIO. Better would probably be to add this as a capability to containers-schema.json and add this only to certain containers via containers.json

Done

Signed-off-by: Jean-Yves <[email protected]>

Signed-off-by: Simon L. <[email protected]>

… local-ai Signed-off-by: Simon L. <[email protected]>

Signed-off-by: Simon L. <[email protected]>

docjyJ

LGTM 🎉

docjyJ · 2024-12-20T10:08:13Z

Thank you @MondoGao

szaimen

LGTM. Thanks a lot @MondoGao and @docjyJ! 😊

szaimen · 2024-12-20T12:20:06Z

This is now released with v10.2.0 Beta. Testing and feedback is welcome! See https://github.com/nextcloud/all-in-one#how-to-switch-the-channel

luzfcb · 2024-12-20T20:10:28Z

@szaimen the local-ai community container is unable to start with Nextcloud AIO configured with the NEXTCLOUD_ENABLE_DRI_DEVICE=true, ENABLE_NVIDIA_GPU=true, AIO_COMMUNITY_CONTAINERS=local-ai environment variables.

This is the logs that I have in the AIO container:

Click me to see the logs

The error log is currently difficult to quickly understand what the cause is. Maybe there is some room for improvement somehow.

Initial startup of Nextcloud All-in-One complete!
You should be able to open the Nextcloud AIO Interface now on port 8080 of this server!
E.g. https://internal.ip.of.this.server:8080
⚠️ Important: do always use an ip-address if you access this port and not a domain as HSTS might block access to it later!

If your server has port 80 and 8443 open and you point a domain to your server, you can get a valid certificate automatically by opening the Nextcloud AIO Interface via:
https://your-domain-that-points-to-this-server.tld:8443
NOTICE: PHP message: Slim Application Error
Type: Exception
Code: 0
Message: Could not start container nextcloud-aio-local-ai: Server error: `POST http://127.0.0.1/v1.41/containers/nextcloud-aio-local-ai/start` resulted in a `500 Internal Server Error` response:
{"message":"failed to create task for container: failed to create shim task: OCI runtime create failed: runc create fail (truncated...)
File: /var/www/docker-aio/php/src/Docker/DockerActionManager.php
Line: 170
Trace: #0 /var/www/docker-aio/php/src/Controller/DockerController.php(59): AIO\Docker\DockerActionManager->StartContainer(Object(AIO\Container\Container))
#1 /var/www/docker-aio/php/src/Controller/DockerController.php(26): AIO\Controller\DockerController->PerformRecursiveContainerStart('nextcloud-aio-l...', true)
#2 /var/www/docker-aio/php/src/Controller/DockerController.php(209): AIO\Controller\DockerController->PerformRecursiveContainerStart('nextcloud-aio-a...', true)
#3 /var/www/docker-aio/php/src/Controller/DockerController.php(189): AIO\Controller\DockerController->startTopContainer(true)
#4 /var/www/docker-aio/php/vendor/slim/slim/Slim/Handlers/Strategies/RequestResponse.php(38): AIO\Controller\DockerController->StartContainer(Object(GuzzleHttp\Psr7\ServerRequest), Object(GuzzleHttp\Psr7\Response), Array)
#5 /var/www/docker-aio/php/vendor/slim/slim/Slim/Routing/Route.php(363): Slim\Handlers\Strategies\RequestResponse->__invoke(Array, Object(GuzzleHttp\Psr7\ServerRequest), Object(GuzzleHttp\Psr7\Response), Array)
#6 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(73): Slim\Routing\Route->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#7 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(73): Slim\MiddlewareDispatcher->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#8 /var/www/docker-aio/php/vendor/slim/slim/Slim/Routing/Route.php(321): Slim\MiddlewareDispatcher->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#9 /var/www/docker-aio/php/vendor/slim/slim/Slim/Routing/RouteRunner.php(74): Slim\Routing\Route->run(Object(GuzzleHttp\Psr7\ServerRequest))
#10 /var/www/docker-aio/php/vendor/slim/csrf/src/Guard.php(482): Slim\Routing\RouteRunner->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#11 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(177): Slim\Csrf\Guard->process(Object(GuzzleHttp\Psr7\ServerRequest), Object(Slim\Routing\RouteRunner))
#12 /var/www/docker-aio/php/vendor/slim/twig-view/src/TwigMiddleware.php(117): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#13 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(129): Slim\Views\TwigMiddleware->process(Object(GuzzleHttp\Psr7\ServerRequest), Object(Psr\Http\Server\RequestHandlerInterface@anonymous))
#14 /var/www/docker-aio/php/src/Middleware/AuthMiddleware.php(36): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#15 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(280): AIO\Middleware\AuthMiddleware->__invoke(Object(GuzzleHttp\Psr7\ServerRequest), Object(Psr\Http\Server\RequestHandlerInterface@anonymous))
#16 /var/www/docker-aio/php/vendor/slim/slim/Slim/Middleware/ErrorMiddleware.php(77): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#17 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(129): Slim\Middleware\ErrorMiddleware->process(Object(GuzzleHttp\Psr7\ServerRequest), Object(Psr\Http\Server\RequestHandlerInterface@anonymous))
#18 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(73): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#19 /var/www/docker-aio/php/vendor/slim/slim/Slim/App.php(209): Slim\MiddlewareDispatcher->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#20 /var/www/docker-aio/php/vendor/slim/slim/Slim/App.php(193): Slim\App->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#21 /var/www/docker-aio/php/public/index.php(189): Slim\App->run()
#22 {main}
Tips: To display error details in HTTP response set "displayErrorDetails" to true in the ErrorHandler constructor.
NOTICE: PHP message: Slim Application Error
Type: Exception
Code: 0
Message: Could not start container nextcloud-aio-local-ai: Server error: `POST http://127.0.0.1/v1.41/containers/nextcloud-aio-local-ai/start` resulted in a `500 Internal Server Error` response:
{"message":"failed to create task for container: failed to create shim task: OCI runtime create failed: runc create fail (truncated...)
File: /var/www/docker-aio/php/src/Docker/DockerActionManager.php
Line: 170
Trace: #0 /var/www/docker-aio/php/src/Controller/DockerController.php(59): AIO\Docker\DockerActionManager->StartContainer(Object(AIO\Container\Container))
#1 /var/www/docker-aio/php/src/Controller/DockerController.php(26): AIO\Controller\DockerController->PerformRecursiveContainerStart('nextcloud-aio-l...', true)
#2 /var/www/docker-aio/php/src/Controller/DockerController.php(209): AIO\Controller\DockerController->PerformRecursiveContainerStart('nextcloud-aio-a...', true)
#3 /var/www/docker-aio/php/src/Controller/DockerController.php(189): AIO\Controller\DockerController->startTopContainer(true)
#4 /var/www/docker-aio/php/vendor/slim/slim/Slim/Handlers/Strategies/RequestResponse.php(38): AIO\Controller\DockerController->StartContainer(Object(GuzzleHttp\Psr7\ServerRequest), Object(GuzzleHttp\Psr7\Response), Array)
#5 /var/www/docker-aio/php/vendor/slim/slim/Slim/Routing/Route.php(363): Slim\Handlers\Strategies\RequestResponse->__invoke(Array, Object(GuzzleHttp\Psr7\ServerRequest), Object(GuzzleHttp\Psr7\Response), Array)
#6 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(73): Slim\Routing\Route->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#7 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(73): Slim\MiddlewareDispatcher->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#8 /var/www/docker-aio/php/vendor/slim/slim/Slim/Routing/Route.php(321): Slim\MiddlewareDispatcher->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#9 /var/www/docker-aio/php/vendor/slim/slim/Slim/Routing/RouteRunner.php(74): Slim\Routing\Route->run(Object(GuzzleHttp\Psr7\ServerRequest))
#10 /var/www/docker-aio/php/vendor/slim/csrf/src/Guard.php(482): Slim\Routing\RouteRunner->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#11 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(177): Slim\Csrf\Guard->process(Object(GuzzleHttp\Psr7\ServerRequest), Object(Slim\Routing\RouteRunner))
#12 /var/www/docker-aio/php/vendor/slim/twig-view/src/TwigMiddleware.php(117): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#13 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(129): Slim\Views\TwigMiddleware->process(Object(GuzzleHttp\Psr7\ServerRequest), Object(Psr\Http\Server\RequestHandlerInterface@anonymous))
#14 /var/www/docker-aio/php/src/Middleware/AuthMiddleware.php(36): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#15 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(280): AIO\Middleware\AuthMiddleware->__invoke(Object(GuzzleHttp\Psr7\ServerRequest), Object(Psr\Http\Server\RequestHandlerInterface@anonymous))
#16 /var/www/docker-aio/php/vendor/slim/slim/Slim/Middleware/ErrorMiddleware.php(77): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#17 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(129): Slim\Middleware\ErrorMiddleware->process(Object(GuzzleHttp\Psr7\ServerRequest), Object(Psr\Http\Server\RequestHandlerInterface@anonymous))
#18 /var/www/docker-aio/php/vendor/slim/slim/Slim/MiddlewareDispatcher.php(73): Psr\Http\Server\RequestHandlerInterface@anonymous->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#19 /var/www/docker-aio/php/vendor/slim/slim/Slim/App.php(209): Slim\MiddlewareDispatcher->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#20 /var/www/docker-aio/php/vendor/slim/slim/Slim/App.php(193): Slim\App->handle(Object(GuzzleHttp\Psr7\ServerRequest))
#21 /var/www/docker-aio/php/public/index.php(189): Slim\App->run()
#22 {main}
Tips: To display error details in HTTP response set "displayErrorDetails" to true in the ErrorHandler constructor.

I think the reason is probably because the docker image used as a base for aio-local-ai is quay.io/go-skynet/local-ai:v2.24.2-aio-cpu , but when ENABLE_NVIDIA_GPU=true is used, it should be one of the images below:

quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-nvidia-cuda-12
quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-nvidia-cuda-11

Additionally, I saw that there are options for Intel GPU or something like this

quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-intel-f16
quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-intel-f32
quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-hipblas

Anyway, the main problem about the error is because it stop the all other containers to start.

szaimen · 2024-12-20T20:13:31Z

Hm... This specific error should not happen, it should still be able to start the container... Did you install the nvidia drivers correctly like mentioned im the readme?

luzfcb · 2024-12-20T20:26:41Z

Did you install the nvidia drivers correctly like mentioned im the readme?

Yes, since I can successfully run the Sample Workload with Docker

sudo docker run --rm --runtime=nvidia --gpus all ubuntu nvidia-smi

Note:

I'm using Unraid OS 6.12.3 and have the Nvidia Driver plugin by ich777 and the Nvidia proprietary driver v565.57.01 , and I am able to play games on Steam via docker container https://github.com/Steam-Headless/docker-steam-headless

Anyway, the main problem about the error on the local-ai community container is because it somehow block the all other nextcloud containers to start by Nextcloud AIO.

szaimen · 2024-12-20T20:57:46Z

So if you remove local-ai from the community containers, nextcloud starts correctly?

luzfcb · 2024-12-20T21:05:33Z

So if you remove local-ai from the community containers, nextcloud starts correctly?

Yes, and also if I keep local-ai on the community containers and set ENABLE_NVIDIA_GPU=false the nextcloud also starts correctly

szaimen · 2024-12-20T21:08:10Z

Okay this is weird. I guess we need to remove the setting for local-ai then. Doe the other containers that we added start correctly?

luzfcb · 2024-12-20T21:26:29Z

I guess we need to remove the setting for local-ai then

Until aio-local-ai provides a new image using quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-nvidia-cuda-12 as base instead of quay.io/go-skynet/local-ai:v2.24.2-aio-cpu , and community containers feature implements a way to change the image or tag defined on

all-in-one/community-containers/local-ai/local-ai.json

Lines 7 to 8 in 109b9dc

    
           "image": "szaimen/aio-local-ai", 
        
           "image_tag": "v2",

based on the existence of the ENABLE_NVIDIA_GPU=true, probably yes

Doe the other containers that we added start correctly?

facerecognition and memories start correctly.

Since I don't use plex and jellyfin, and I can't confirm if they work or not.

I'm also not sure if facerecognition and memories recognize the nvidia GPU immediately without any extra configuration.

I'll try to figure that out over the weekend.

luzfcb · 2024-12-20T21:33:54Z

I guess we need to remove the setting for local-ai then

Until aio-local-ai provides a new image using quay.io/go-skynet/local-ai:v2.24.2-aio-gpu-nvidia-cuda-12 as base instead of quay.io/go-skynet/local-ai:v2.24.2-aio-cpu , and community containers feature implements a way to change the image or tag defined on https://github.com/nextcloud/all-in-one/blob/109b9dc019ebb499a9571f8cf3129e6e26e1942a/community-containers/local-ai/local-ai.json#L7-L8 based on the existence of the ENABLE_NVIDIA_GPU=true, probably yes

A quick and simple solution is probably to create a new community container named local-ai-nvidia that uses a possible new aio-local-ai nvidia based image.

That is, if I want to use ENABLE_NVIDIA_GPU=true and local-ai, I should use local-ai-nvidia in the configuration instead of local-ai in the AIO_COMMUNITY_CONTAINERS Nextcloud AIO environment variable

szaimen added 2. developing Work in progress enhancement New feature or request labels Aug 16, 2024

szaimen added this to the next milestone Aug 16, 2024

szaimen mentioned this pull request Aug 16, 2024

publish new images after recent PRs are merged and tested #5086

Closed

28 tasks

MondoGao force-pushed the main branch from b774837 to 8fa7da0 Compare August 16, 2024 19:46

szaimen modified the milestones: v9.5.0, next, v9.5.1 Sep 4, 2024

szaimen modified the milestones: v9.6.0, next Sep 18, 2024

szaimen modified the milestones: v9.7.0, next Oct 10, 2024

szaimen removed this from the next milestone Oct 20, 2024

szaimen mentioned this pull request Oct 21, 2024

ver. 9.7.0: "nextcloud-aio-memories" doesn't have an access to Nvidia GPU and can't do HW transcoding #5446

Closed

szaimen marked this pull request as draft October 22, 2024 09:43

docjyJ reviewed Dec 17, 2024

View reviewed changes

php/src/Data/ConfigurationManager.php Outdated Show resolved Hide resolved

php/src/Data/ConfigurationManager.php Outdated Show resolved Hide resolved

docjyJ added 3 commits December 20, 2024 10:01

improuve doc

73ebbf6

Signed-off-by: Jean-Yves <[email protected]>

Rolback gpu mode

ccbe119

Signed-off-by: Jean-Yves <[email protected]>

Ready for review

4759ce2

Signed-off-by: Jean-Yves <[email protected]>

docjyJ force-pushed the main branch from a249069 to 4759ce2 Compare December 20, 2024 09:02

Fix typo

27787ab

Signed-off-by: Jean-Yves <[email protected]>

docjyJ requested a review from szaimen December 20, 2024 09:06

szaimen added 8 commits December 20, 2024 10:37

fix all links for transcoding->accelleration

0c9730c

Signed-off-by: Simon L. <[email protected]>

enable hardware accelleration for talk-recording, facerecognition and…

e066f56

… local-ai Signed-off-by: Simon L. <[email protected]>

remove enable_nvidia_gpu setting for update-yaml script

87c1733

Signed-off-by: Simon L. <[email protected]>

adjust some details

d7a9889

Signed-off-by: Simon L. <[email protected]>

adjust some details

f492ab2

Signed-off-by: Simon L. <[email protected]>

adjust some more details

9f99c0b

Signed-off-by: Simon L. <[email protected]>

more

396e1b8

Signed-off-by: Simon L. <[email protected]>

adjust more

6537f18

Signed-off-by: Simon L. <[email protected]>

docjyJ reviewed Dec 20, 2024

View reviewed changes

szaimen approved these changes Dec 20, 2024

View reviewed changes

szaimen added 3. to review Waiting for reviews and removed 2. developing Work in progress labels Dec 20, 2024

szaimen merged commit 53edc5d into nextcloud:main Dec 20, 2024
10 checks passed

szaimen mentioned this pull request Dec 20, 2024

Container.php: adjust casing of enable_nvidia_gpu to enableNvidiaGpu #5786

Draft

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

add support for nvidia gpu access #5132

add support for nvidia gpu access #5132

MondoGao commented Aug 16, 2024 •

edited by szaimen

Loading

szaimen commented Aug 16, 2024

MondoGao commented Aug 16, 2024

szaimen commented Aug 19, 2024 •

edited

Loading

MondoGao commented Aug 19, 2024

luzfcb commented Aug 19, 2024 •

edited

Loading

docjyJ commented Aug 21, 2024

gbertolottiKS commented Dec 12, 2024

szaimen commented Dec 13, 2024

docjyJ left a comment

szaimen commented Dec 17, 2024

szaimen commented Dec 17, 2024

docjyJ commented Dec 17, 2024

docjyJ commented Dec 17, 2024

docjyJ commented Dec 17, 2024

docjyJ left a comment

docjyJ commented Dec 20, 2024

szaimen left a comment •

edited

Loading

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024 •

edited

Loading

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024

luzfcb commented Dec 20, 2024 •

edited

Loading

add support for nvidia gpu access #5132

add support for nvidia gpu access #5132

Conversation

MondoGao commented Aug 16, 2024 • edited by szaimen Loading

szaimen commented Aug 16, 2024

MondoGao commented Aug 16, 2024

szaimen commented Aug 19, 2024 • edited Loading

MondoGao commented Aug 19, 2024

luzfcb commented Aug 19, 2024 • edited Loading

docjyJ commented Aug 21, 2024

gbertolottiKS commented Dec 12, 2024

szaimen commented Dec 13, 2024

docjyJ left a comment

Choose a reason for hiding this comment

szaimen commented Dec 17, 2024

szaimen commented Dec 17, 2024

docjyJ commented Dec 17, 2024

docjyJ commented Dec 17, 2024

docjyJ commented Dec 17, 2024

docjyJ left a comment

Choose a reason for hiding this comment

docjyJ commented Dec 20, 2024

szaimen left a comment • edited Loading

Choose a reason for hiding this comment

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024 • edited Loading

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024

szaimen commented Dec 20, 2024

luzfcb commented Dec 20, 2024

luzfcb commented Dec 20, 2024 • edited Loading

MondoGao commented Aug 16, 2024 •

edited by szaimen

Loading

szaimen commented Aug 19, 2024 •

edited

Loading

luzfcb commented Aug 19, 2024 •

edited

Loading

szaimen left a comment •

edited

Loading

luzfcb commented Dec 20, 2024 •

edited

Loading

luzfcb commented Dec 20, 2024 •

edited

Loading