Delegated License Service Release Notes

NVIDIA Delegated License Service Release Notes

Release information for all users of NVIDIA Delegated License Service.

This document summarizes current status, information on supported platforms, and known issues with NVIDIA® License System release 2.1.0.

1.1. Updates in this Release

New Features in this Release

  • Automated conversion from the NVIDIA Licensing Portal of legacy NVIDIA vGPU software license servers to NLS license servers
  • Node-locked licensing
  • Email notifications for licensing events
  • Support for non-persistent desktop pools
  • Suspension of logins to a DLS appliance after four failed login attempts
  • Changes to the user interface of the NVIDIA Licensing Portal for accessing the following features:
    • DLS settings
    • Software downloads
  • Miscellaneous bug fixes as listed in Resolved Issues

Software Support Introduced in this Release

  • Support for the DLS on the following hypervisor software releases:
    • Red Hat Enterprise Linux Kernel-based Virtual Machine (KVM) 9.0 and 9.1
    • Ubuntu Hypervisor 22.04
    • VMware vSphere Hypervisor (ESXi) 7.0.1

Feature Support Withdrawn in this Release

  • The DLS is no longer supported on VMware vSphere Hypervisor (ESXi) 6.7 and 7.0.
  • TLS 1.0 and 1.1 are disabled for the DLS.

1.2. Supported Platforms

1.2.1. Supported Hypervisors

For deployment in a virtual machine, the Delegated License Server (DLS) component of the NVIDIA License System is supplied as a virtual appliance. The virtual appliance must be installed on a supported hypervisor software release.

The following hypervisor software releases are supported:

  • Citrix Hypervisor 8.2
  • Linux Kernel-based Virtual Machine (KVM) hypervisors with QEMU 2.12.0 (qemu-kvm-2.12.0-64.el8.2.27782638)
  • Microsoft Windows Server with Hyper-V 2019 Datacenter edition
  • Red Hat Enterprise Linux Kernel-based Virtual Machine (KVM) 9.0 and 9.1
  • Red Hat Virtualization 4.3
  • Ubuntu Hypervisor 22.04
  • VMware vSphere Hypervisor (ESXi) 7.0.1, 7.0.2, and 7.0.3

1.2.2. Supported Container Orchestration Platforms

For deployment on a bare-metal OS, the Delegated License Server (DLS) component of the NVIDIA License System is supplied as a containerized software image. The containerized software image must be deployed on a supported container orchestration platform.

The following container orchestration platform releases are supported:

  • Docker 20.10.17 with Docker Compose 2.6.0
  • Kubernetes 1.23.8
  • Red Hat OpenShift Container Platform 4.10.18 with Kubernetes 1.23.5
  • VMware Tanzu Application Platform 1.1 with Kubernetes 1.23.6

1.2.3. Licensed Client Support

NVIDIA License System supports specific releases of several NVIDIA software products as licensed clients.

Software Product Supported Releases
NVIDIA® vGPU™ software graphics drivers NVIDIA vGPU software starting with release 13.0

1.2.4. Web Browser Requirements

NVIDIA License System and NVIDIA Licensing Portal were tested with Google Chrome version 86.0.4240.111 (Official Build) (64-bit).

A container orchestration platform cannot control or restrict access to the OS on which the platform is running. Therefore, containerized DLS software images cannot support the features of VM-based DLS virtual appliances that rely on the ability of the appliance to control the underlying OS.

Containerized DLS software images do not support the following features, for which equivalent functionality is available through standard OS interfaces:

  • Log archive settings
  • NTP configuration
  • Static IP address configuration
  • DLS diagnostics user configuration
  • Disk expansion

Because a container orchestration platform cannot control the underlying OS, the following limitations also apply to containerized DLS software images:

  • Online migration from a VM-based DLS virtual appliance to a containerized DLS software image is not supported because the destination containerized DLS software image retains its IP address even after data migration.

    Instead, you must use offline migration when migrating from a VM-based DLS virtual appliance to a containerized DLS software image.

  • When the secondary node is removed from an HA cluster, the containerized DLS software image that hosts the node is not shut down.

    Instead, you must shut down the DLS software container manually.

Only resolved issues that have been previously noted as known issues or had a noticeable user impact are listed. The summary and description for each resolved issue indicate the effect of the issue on NVIDIA License System before the issue was resolved.

Bug ID Summary
3508713

After DLS migration, renewal of licenses checked out before the migration fails

After a DLS instance is migrated to a new DLS appliance, the migrated DLS instance fails to renew licenses checked out before the migration. However, the checked out licenses remain active even after the failure to renew them. This issue does not prevent licensed clients from checking out licenses.

The complete resolution of this issue requires the following licensed clients:

  • NVIDIA vGPU software 13 starting with release 13.6
  • NVIDIA vGPU software 15 starting with release 15.0
3797900

Licensed client on IPv4 network fails to acquire a license

A licensed client on an IPv4 network fails to acquire a license from a DLS instance. This issue occurs because the IP address in the client configuration token through which the DLS instance is accessed is an IPv6 address.

3804294

Online DLS migration fails with HTTP error 500 for IPv4 address configuration on mixed network

On a mixed IPv4 and IPv6 network infrastructure, online migration of a DLS instance hosted on a VM with an IPv4 address fails. After the data has been migrated from the existing DLS appliance, the migration cannot be confirmed on the new DLS appliance because the ACKNOWLEDGE MIGRATION button does not appear in the management interface on the new DLS appliance. Furthermore, the IPv4 address of the VM that hosts the new DLS appliance is not displayed in the management interface on the new DLS appliance. Any attempt to set IP address of the VM from the management interface fails with HTTP error 500. This issue occurs because, during initialization, the DLS application ignores IPv4 addresses on an network in which IPv6 is also configured.


4.1. The DLS appliance has an expired SSL certificate and weak code cipher suites

Description

A security scan of the DLS appliance reveals that the DLS appliance has an expired SSL certificate and weak message authentication code cipher suites.

By default, a DLS virtual appliance is configured with a self-signed X.509 SSL server certificate that is included in the DLS virtual appliance image from which the DLS virtual appliance is created. This certificate expired on November 14, 2023.

The DLS appliance supports Transport Layer Security (TLS) version 1.2 but the weak message authentication code cipher suites have not been removed. The TLS/SSL server in the DLS appliance also supports the use of static key ciphers.

Workaround

To address the issue of the expired SSL certificate, install a new SSL certificate. You can use a self-signed SSL certificate or a certificate that is signed by a third party, such as a certificate authority (CA), for this purpose.

To address the issue of weak code cipher suites, customize the ciphers that the DLS appliance should use.

Ensure that the sudo DLS user account rsu_admin has been created.

  1. Use the hypervisor management console of the appliance to log in as the user rsu_admin to the VM that hosts the DLS virtual appliance.
  2. As root, open the file /var/lib/docker/volumes/configurations/_data/ui/dls-web-ui.conf.template for editing in the nano plain-text editor.
    Copy
    Copied!
                

    $ sudo nano /var/lib/docker/volumes/configurations/_data/ui/dls-web-ui.conf.template

  3. In the file /var/lib/docker/volumes/configurations/_data/ui/dls-web-ui.conf.template, remove the weak message authentication code cipher suites and add the directive ssl_ciphers followed by the ciphers that you want to use. For example, to use the TLS_RSA_WITH_AES_128_CCM_8 cipher, add the following directive:
    Copy
    Copied!
                

    ssl_ciphers "TLS_RSA_WITH_AES_128_CCM_8";

  4. Save your changes to the /var/lib/docker/volumes/configurations/_data/ui/dls-web-ui.conf.template file and quit the editor.
  5. Restart the DLS appliance.

Status

Open

Ref. #

4016523

4.2. With Windows DNS, internal DLS services cannot be restarted from the DLS management interface

Description

If a Windows DNS server is used for name resolution, and the DLS instance is accessed through the fully qualified domain name of its VM, the internal DLS services of the instance cannot be restarted from the management interface of the NVIDIA Licensing application of the DLS instance. On the Service Instance page, the RESTART button remains deactivated even if the status of the services is displayed as inactive.

You can restart these services from the management interface only if you are logged in to the specific instance whose services must be restarted. The DLS determines the instance by matching the fully qualified domain name of the instance with browser's URL origin property.

This issue occurs when the case of the fully qualified domain name in the DNS entry that maps the default host name to the fully qualified domain name is different from the case of the browser's URL origin property. This issue occurs with Windows DNS servers because the fully qualified domain name is in uppercase but the browser's URL origin property is typically in lowercase.

Workaround

Use any one of the following workarounds for this issue.

  • When logging in to the DLS appliance, specify the IP address of the VM on which the DLS virtual appliance is installed instead of its fully qualified domain name or CNAME.
  • Modify the DNS entry that maps the default host name to the fully qualified domain name to use a lowercase fully qualified domain name.

Status

Open

Ref. #

4351000

4.3. The DLS cannot be administered when its root file system is full

Description

When the root file system of the DLS virtual appliance is full, administrative operations on the DLS virtual appliance, such as creating the DLS sudo user account rsu_admin or expanding the disk space on the DLS virtual appliance, fail. This issue can occur if client registration data for licensed clients is being created faster than the DLS appliance can delete data that is no longer required.

Workaround

Determine if the root file system of the DLS virtual appliance is full.

  1. Use the hypervisor management console of the appliance to log in as the user dls_system to the VM that hosts the DLS virtual appliance.

    You don't need to provide a password to log in as the dls_system user.

  2. Determine the amount of space available on the /dev/mapper/vgnls–si–0--root file system.
    Copy
    Copied!
                

    $ df -h /dev/mapper/vgnls–si–0--root

If the root file system of the DLS virtual appliance is full, increase its size.

  1. Use the hypervisor management console of the appliance to log in as the user rsu_admin to the VM that hosts the DLS virtual appliance.

    If the rsu_admin user has not been registered, you can still log in to the VM as the rsu_admin user with the default password rsu_admin.

  2. Stop the applicationStartup service.
    Copy
    Copied!
                

    $ systemctl stop applicationStartup

  3. As root, delete the files in which client registration data for licensed clients is logged.
    Copy
    Copied!
                

    $ sudo rm -rf /var/lib/docker/volumes/logs/_data/*

  4. Expand the disk space on the DLS virtual appliance as explained in Expanding the Disk Space on a DLS Virtual Appliance in NVIDIA Delegated License Service User Guide.

After increasing the size of the root file system of the DLS virtual appliance, repeat any administrative operations on the DLS virtual appliance that failed.

Status

Resolved in NVIDIA License System 3.1.

Ref. #

4230517

4345127

4.4. Login to the DLS appliance fails with an unexpected error

Description

Login to the DLS appliance fails with an unexpected error because a connection to the appliance cannot be created. This issue occurs if the root file system of the DLS virtual appliance is full, typically because client registration data for licensed clients is accumulating over time. When the root file system of the DLS virtual appliance is full, the database container begins terminating connections, preventing users from logging in to the DLS appliance.

Workaround

  1. Use the hypervisor management console of the appliance to log in as the user rsu_admin to the VM that hosts the DLS virtual appliance.

    If the rsu_admin user has not been registered, you can still log in to the VM as the rsu_admin user with the default password rsu_admin.

  2. From the patch-bin.tar.gz archive, extract the configuration volume /var/lib/docker/volumes/configurations/_data/.
  3. Create a cron job to run the following command periodically, for example, once each day, on this release of the DLS virtual appliance (2.1.0) with a day limit of 0.
    Copy
    Copied!
                

    sudo docker exec -it config-nls-si-0-1 /etc/dls/network/patch-bin/run.sh 2.1.0 0

Status

Resolved in NVIDIA License System 3.1.

Ref. #

4103600

4.5. Migration of a DLS instance fails

Description

Migration of a DLS instance can fail if a large quantity of data is to be migrated. This issue affects both online and offline migration of a DLS instance. When this issue occurs, the NVIDIA Licensing application on the new DLS virtual appliance is affected in one of the following ways:

  • The NVIDIA Licensing Dashboard does not show license server details.
  • The ACKNOWLEDGE MIGRATION button is absent from Maintenance page.

Workaround

Contact NVIDIA Enterprise Support to obtain a workaround for this issue.

Status

Open

Ref. #

3961380

4.6. HA cluster creation fails after migration of a DLS instance

Description

HA cluster creation after migration of a DLS instance can fail if a large quantity of data is to be migrated.

Workaround

Contact NVIDIA Enterprise Support to obtain a workaround for this issue.

Status

Open

Ref. #

3931610

4.7. BadRequestError error is displayed on the Events page of a DLS instance

Description

When a licensed client requests a license from a DLS instance, the following error is displayed on the Events page of the DLS instance:

Copy
Copied!
            

BadRequestError(origin reference reference already in use by different fingerprint)

Workaround

Ignore this error because it is a transient error. After the licensed client repeats the request, this issue is resolved.

Status

Not a bug

Ref. #

3966221

4.8. Node-locked licenses are missing from the DLS instance Leases page

Description

Node-locked licenses are missing from the Leases page for the DLS instance that is bound to the license server from which the node-locked licenses were generated. However, node-locked licenses are correctly listed on the Leases tab of the NVIDIA Licensing Dashboard for the DLS instance. Node-locked licenses can be generated only from license servers for which the Disconnected leasing mode? option is set.

Workaround

Use the Leases tab of the NVIDIA Licensing Dashboard for the DLS instance to list node-locked licenses.

Status

Open

Ref. #

3936853

4.9. VM-based DLS appliance has security vulnerabilities

Description

The VM-based DLS appliance for each supported hypervisor has security vulnerabilities related to options set for file-system partitions and access permissions for some files. The vulnerabilities are as follows:

  • The nodev option is not set on the /boot/efi partition.
  • Every time the VM that hosts the DLS appliance is started, Docker creates the following temporary partitions without the nodev option:
    • /var/lib/docker/overlay2/21e110eda56bbc76b7342e24497f357a7409e4effb924177e04cba0cb28e9f2f/merged
    • /run/docker/netns/a8b1b321de4b
    • /var/lib/docker/overlay2/9c4fd6f22f7998ffaead43c21e5cda184f28fb1c2445e2ed00bbe8dc8be08339/merged
    • /run/docker/netns/88c61763a83f
  • The mode of the following files is -rwxrwxrwx, which allows write access by other users (world):
    • /etc/resolv.conf
    • /home/rsu_admin/Rapid7/agent_installer.sh
    • /home/rsu_admin/Rapid7/cafile.pem
    • /home/rsu_admin/Rapid7/client.crt
    • /home/rsu_admin/Rapid7/client.key
    • /home/rsu_admin/Rapid7/config.json
    • /run/NetworkManager/no-stub-resolv.conf
  • Every time the VM that hosts the DLS appliance is started, Docker creates the following files with the mode -rwxrwxrwx, which allows write access by other users (world):
    • /home/dls_system/device
    • /home/dls_system/dns
    • /home/dls_system/gateway
    • /home/dls_system/ip_address
    • /home/dls_system/static-ip-ova-logs

Workaround

You can mitigate these vulnerabilities by setting the nodev option on the affected file-system partitions and restricting write access to the affected files.

Change the affected files and partitions that are not created after every reboot of the VM only once.

  1. Use the hypervisor management console of the appliance to log in as the user dls_system to the VM that hosts the DLS appliance.

    You don't need to provide a password to log in as the dls_system user.

  2. Add the nodev mount option to the entry in /etc/fstab for the /boot/efi partition.

  3. Set the mode of the affected files that are not created after every reboot of the VM to allow access only by owner and root.

    1. Change to the /etc directory.

      Copy
      Copied!
                  

      $ cd /etc

    2. Change the mode of the resolv.conf file to -rwxr-xr-x.

      Copy
      Copied!
                  

      $ sudo chmod 755 resolv.conf

    3. Change to the /home/rsu_admin/Rapid7 directory.

      Copy
      Copied!
                  

      $ cd /home/rsu_admin/Rapid7

    4. Change the mode of the affected files in this directory to -rwxr-xr-x.

      Copy
      Copied!
                  

      $ sudo chmod 755 \ agent_installer.sh cafile.pem client.crt client.key config.json

    5. Change to the /run/NetworkManager directory.

      Copy
      Copied!
                  

      $ cd /run/NetworkManager

    6. Change the mode of the no-stub-resolv.conf file to -rwxr-xr-x.

      Copy
      Copied!
                  

      $ sudo chmod 755 no-stub-resolv.conf

  4. Restart the VM that hosts the DLS appliance.

Change the affected files and partitions that are recreated after every reboot of the VM every time the VM is rebooted.

  1. Use the hypervisor management console of the appliance to log in as the user dls_system to the VM that hosts the DLS appliance.

    You don't need to provide a password to log in as the dls_system user.

  2. Add the nodev mount option to the entry in /etc/fstab for each of the following partitions:

    • /var/lib/docker/overlay2/21e110eda56bbc76b7342e24497f357a7409e4effb924177e04cba0cb28e9f2f/merged
    • /run/docker/netns/a8b1b321de4b
    • /var/lib/docker/overlay2/9c4fd6f22f7998ffaead43c21e5cda184f28fb1c2445e2ed00bbe8dc8be08339/merged
    • /run/docker/netns/88c61763a83f
  3. Remount the partitions for which you added the nodev mount option in the previous step.

    For each partition, run the following mount command on the partition.

    Copy
    Copied!
                

    $ sudo mount -o remount,nodev partition

    partition
    The partition that you are remounting.
  4. Set the mode of the affected files that are recreated after every reboot of the VM to allow access only by owner and root.

    1. Change to the /home/dls_system directory.

      Copy
      Copied!
                  

      $ cd /home/dls_system

    2. Change the mode of the affected files in this directory to -rwxr-xr-x.

      Copy
      Copied!
                  

      $ sudo chmod 755 \ device dns gateway ip_address static-ip-ova-logs

Status

Not a bug

Ref. #

3923943

4.10. Events cannot be exported from a DLS instance on a Kubernetes platform

Description

Events cannot be exported from a DLS instance hosted by a container-based DLS appliance running on Kubernetes, Red Hat OpenShift Container Platform, or VMware Tanzu Application Platform. When a user tries to export events, the attempt fails and the error message Export file generation failed. Please try again. is displayed.

This issue occurs because the user that is exporting events does not have the permissions required to create the directories to which the events are exported.

Workaround

Create the directories to which the events are exported and grant write access to all users to these directories.

  1. Change to the directory where the configurations volume is mounted.

  2. Create the enc and unenc directories.

    1. Create the enc directory.

      Copy
      Copied!
                  

      $ mkdir enc

    2. Create the unenc directory.

      Copy
      Copied!
                  

      $ mkdir unenc

  3. Grant write access to all users to the directories that you created in the previous step.

    1. Grant write access to all users to the enc directory.

      Copy
      Copied!
                  

      $ chmod -R 707 enc

    2. Grant write access to all users to the unenc directory.

      Copy
      Copied!
                  

      $ chmod -R 707 unenc

Status

Open

Ref. #

3917695

4.11. Validation of the client configuration token fails

Description

When Network Time Protocol (NTP) servers are configured for a VM-based DLS instance, the system times on the DLS instance and the licensed client might still be different. In this situation, validation of the client configuration token fails. When the licensed client requests a license, the request fails and the following error message is displayed:

Copy
Copied!
            

Client Configuration Token Validation Failed. Either the token is not activated or system clock settings are tampered

Workaround

When configuring NTP servers on a DLS virtual appliance, specify the IP addresses, not the fully qualified domain names, of the NTP servers.

Status

Open

Ref. #

3718863

4.12. VM hosting a DLS appliance cannot be reached

Description

After a VM-based DLS appliance has been installed, the VM that is hosting the DLS appliance cannot be reached after it has been started. This issue occurs when a static IP address has been assigned to the VM that is hosting the DLS appliance and the subnet mask of the VM's network was specified in an incorrect format. The subnet mask of the VM's network must be specified in classless inter-domain routing (CIDR) format without the leading slash character (/).

Workaround

Reinstall the VM-based DLS appliance, specifying the subnet mask in the correct format, namely, CIDR format without the leading slash character (/).

To get a subnet mask in CIDR format from its decimal equivalent, refer to the table on page 2 of IETF RFC 1878: Variable Length Subnet Table For IPv4. For example, the subnet mask in CIDR format of the decimal equivalent 255.255.255.0 is 24.

Status

Not a bug

Ref. #

3741535

4.13. Client fails to acquire offline license when rebooted

Description

When a licensed client that is configured with an offline license is rebooted, the client might fail to acquire a license. When this issue occurs, the following message is written to the licensing event log file on the client:

Copy
Copied!
            

Client fingerprint mismatch - No valid lease found in local trusted store


This issue occurs when the MAC addresses of the network adapters for a client change when the client is rebooted. When the MAC addresses change, the NVIDIA vGPU software graphics driver treats the client as a new client and the offline license in the client's trusted storage database is discarded.

Typically, the MAC addresses change because the network configuration of the client has been explicitly changed by an administrator. However, the MAC address of a client can unexpectedly change when the client is rebooted for several reasons, for example:

  • The client requests a license before the client's network interfaces are initialized.
  • Docker or the NVIDIA Container Runtime for Docker is installed on the client and the ifconfig command lists it as a network interface.

Status

Open

Ref. #

200665895

If necessary, you can create a predefined sudo user account for a DLS virtual appliance. This user account has the elevated privileges required to update and upgrade the Ubuntu GPL/LGPL v3 licensed Open Source Software (OSS) libraries within the DLS virtual appliance. This account provides no other access to a DLS virtual appliance.

This account is provided to comply with the terms of the GPL/LGPL v3 license under which some libraries in the Ubuntu operating system in the DLS virtual appliance are released.

CAUTION:

Any changes to the Ubuntu GPL/LGPL v3 licensed OSS libraries within the DLS virtual appliance might impair the performance of the DLS virtual appliance or prevent it from functioning as required. If you make any changes to these libraries, the affected DLS instance is no longer eligible for support from NVIDIA. It is your responsibility to ensure that the DLS instance continues to perform and function as required.

Perform this task for each DLS virtual appliance for which you want to create the DLS sudo user account. If the DLS virtual appliance is hosting a node in an HA cluster, the creation of the user is not automatically propagated to the other node in the cluster.

  1. From the hypervisor, log in as the user dls_system to the VM that hosts the DLS virtual appliance. You don't need to provide a password to log in as the dls_system user.
  2. Run the /etc/adminscripts/enable_sudo.sh script.
    Copy
    Copied!
                

    $ /etc/adminscripts/enable_sudo.sh

  3. When prompted, provide a password for this user.

The sudo user with elevated privileges rsu_admin is created.

The file /var/dls/sudouser is created to indicate that the Ubuntu GPL/LGPL v3 licensed OSS libraries within the DLS virtual appliance have been updated or upgraded. If the DLS virtual appliance is hosting a node in an HA cluster, this file is automatically copied to the other node in the cluster.

In addition, to update the OS or install packages on Ubuntu:

  1. Test to make sure the Ubuntu repositories are reachable with the existing network settings.
  2. If the repository URLs are not getting resolved, you should delete the symlink /etc/resolv.conf, and copy the default resolv.conf located at /run/NetworkManager to /etc/resolv.conf.
    Copy
    Copied!
                

    sudo rm -f /etc/resolv.conf

    Copy
    Copied!
                

    sudo cp /run/NetworkManager/no-stub-resolv.conf /etc/resolv.conf

    Note:

    Before you copy /run/NetworkManager/resolv.conf, make sure you make a DNS server entry that will resolve the Ubuntu (or any custom) package repositories.

  3. Once the OS update/upgrade is done, please restore the network configurations to their original state by executing the following commands:
    Copy
    Copied!
                

    sudo rm -f /etc/resolv.conf

    Copy
    Copied!
                

    sudo ln -s /run/NetworkManager/no-stub-resolv.conf /etc/resolv.conf

Notice

This document is provided for information purposes only and shall not be regarded as a warranty of a certain functionality, condition, or quality of a product. NVIDIA Corporation (“NVIDIA”) makes no representations or warranties, expressed or implied, as to the accuracy or completeness of the information contained in this document and assumes no responsibility for any errors contained herein. NVIDIA shall have no liability for the consequences or use of such information or for any infringement of patents or other rights of third parties that may result from its use. This document is not a commitment to develop, release, or deliver any Material (defined below), code, or functionality.

NVIDIA reserves the right to make corrections, modifications, enhancements, improvements, and any other changes to this document, at any time without notice.

Customer should obtain the latest relevant information before placing orders and should verify that such information is current and complete.

NVIDIA products are sold subject to the NVIDIA standard terms and conditions of sale supplied at the time of order acknowledgement, unless otherwise agreed in an individual sales agreement signed by authorized representatives of NVIDIA and customer (“Terms of Sale”). NVIDIA hereby expressly objects to applying any customer general terms and conditions with regards to the purchase of the NVIDIA product referenced in this document. No contractual obligations are formed either directly or indirectly by this document.

NVIDIA products are not designed, authorized, or warranted to be suitable for use in medical, military, aircraft, space, or life support equipment, nor in applications where failure or malfunction of the NVIDIA product can reasonably be expected to result in personal injury, death, or property or environmental damage. NVIDIA accepts no liability for inclusion and/or use of NVIDIA products in such equipment or applications and therefore such inclusion and/or use is at customer’s own risk.

NVIDIA makes no representation or warranty that products based on this document will be suitable for any specified use. Testing of all parameters of each product is not necessarily performed by NVIDIA. It is customer’s sole responsibility to evaluate and determine the applicability of any information contained in this document, ensure the product is suitable and fit for the application planned by customer, and perform the necessary testing for the application in order to avoid a default of the application or the product. Weaknesses in customer’s product designs may affect the quality and reliability of the NVIDIA product and may result in additional or different conditions and/or requirements beyond those contained in this document. NVIDIA accepts no liability related to any default, damage, costs, or problem which may be based on or attributable to: (i) the use of the NVIDIA product in any manner that is contrary to this document or (ii) customer product designs.

No license, either expressed or implied, is granted under any NVIDIA patent right, copyright, or other NVIDIA intellectual property right under this document. Information published by NVIDIA regarding third-party products or services does not constitute a license from NVIDIA to use such products or services or a warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property rights of the third party, or a license from NVIDIA under the patents or other intellectual property rights of NVIDIA.

Reproduction of information in this document is permissible only if approved in advance by NVIDIA in writing, reproduced without alteration and in full compliance with all applicable export laws and regulations, and accompanied by all associated conditions, limitations, and notices.

THIS DOCUMENT AND ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS, DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY, “MATERIALS”) ARE BEING PROVIDED “AS IS.” NVIDIA MAKES NO WARRANTIES, EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THE MATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OF NONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULAR PURPOSE. TO THE EXTENT NOT PROHIBITED BY LAW, IN NO EVENT WILL NVIDIA BE LIABLE FOR ANY DAMAGES, INCLUDING WITHOUT LIMITATION ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, PUNITIVE, OR CONSEQUENTIAL DAMAGES, HOWEVER CAUSED AND REGARDLESS OF THE THEORY OF LIABILITY, ARISING OUT OF ANY USE OF THIS DOCUMENT, EVEN IF NVIDIA HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Notwithstanding any damages that customer might incur for any reason whatsoever, NVIDIA’s aggregate and cumulative liability towards customer for the products described herein shall be limited in accordance with the Terms of Sale for the product.

VESA DisplayPort

DisplayPort and DisplayPort Compliance Logo, DisplayPort Compliance Logo for Dual-mode Sources, and DisplayPort Compliance Logo for Active Cables are trademarks owned by the Video Electronics Standards Association in the United States and other countries.

HDMI

HDMI, the HDMI logo, and High-Definition Multimedia Interface are trademarks or registered trademarks of HDMI Licensing LLC.

OpenCL

OpenCL is a trademark of Apple Inc. used under license to the Khronos Group Inc.

Trademarks

NVIDIA, the NVIDIA logo, NVIDIA Maxwell, NVIDIA Pascal, NVIDIA Turing, NVIDIA Volta, Quadro, and Tesla are trademarks or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which they are associated.

© 2021-2024 NVIDIA Corporation. All rights reserved. Last updated on Jan 19, 2024.