mirror of https://github.com/keylase/nvidia-patch.git synced 2024-11-22 21:47:19 +00:00

This patch removes restriction on maximum number of simultaneous NVENC video encoding sessions imposed by Nvidia to consumer-grade GPUs.

Go to file

Snawoot ee9dd84e76 Merge pull request #132 from Snawoot/docker Docker support		2019-07-11 03:32:22 +03:00
.github/ISSUE_TEMPLATE	reorder issue templates	2019-02-14 01:03:50 +02:00
tools/nv-driver-locator	ndl: add request meta into response object	2019-06-09 22:41:49 +03:00
win	win10: new version 431.36	2019-07-09 17:00:28 +03:00
.gitmodules	win: pin latest stable patcher utility	2019-04-03 21:46:28 +03:00
docker-entrypoint.sh	docker-entrypoint: ensure output dir	2019-07-11 03:31:06 +03:00
Dockerfile	dockerfile: add some reasonable ENV defaults	2019-07-11 03:26:23 +03:00
FFMPEG-with-NVIDIA-Acceleration-on-Ubuntu_UG_v01.pdf	add doc	2016-11-16 13:37:39 +03:00
patch.sh	patch.sh: add custom output dir; get rid of `which` binary dependency	2019-07-11 02:44:17 +03:00
PULL_REQUEST_TEMPLATE.md	Add PR template	2018-12-26 18:43:19 +02:00
README.md	docs: docker support	2019-07-11 03:11:15 +03:00

README.md

nvidia-patch

This patch removes restriction on maximum number of simultaneous NVENC video encoding sessions imposed by Nvidia to consumer-grade GPUs.

Main target operating system is GNU/Linux, but for Windows support see win.

Requirements:

x86_64 system architecture
GNU/Linux operating system
nvenc-compatible gpu (https://developer.nvidia.com/video-encode-decode-gpu-support-matrix#Encoder)
Nvidia driver. Patch available for:
- 375.39
- 390.77
- 390.87
- 396.24
- 396.26
- 396.37
- 396.54
- 410.48
- 410.57
- 410.73
- 410.78
- 410.79
- 410.93
- 410.104
- 415.18
- 415.25
- 415.27
- 418.30
- 418.43
- 418.56
- 418.67
- 418.74
- 430.09
- 430.14
- 430.26
- 430.34

Tested on Ubuntu 18.04 LTS (GNU/Linux 4.15.0-23-generic x86_64)

Synopsis

# bash ./patch.sh -h

SYNOPSIS
       patch.sh [OPTION]...

DESCRIPTION
       The patch for Nvidia drivers to increase encoder sessions

       -s    Silent mode (No output)
       -r    Rollback to original (Restore lib from backup)
       -h    Print this help message

Step-by-Step guide

Examples are provided for driver version 430.34. All commands are runned as root.

Download driver

https://international.download.nvidia.com/XFree86/Linux-x86_64/430.34/NVIDIA-Linux-x86_64-430.34.run

Install driver

mkdir /opt/nvidia && cd /opt/nvidia
wget https://international.download.nvidia.com/XFree86/Linux-x86_64/430.34/NVIDIA-Linux-x86_64-430.34.run
chmod +x ./NVIDIA-Linux-x86_64-430.34.run
./NVIDIA-Linux-x86_64-430.34.run

Check driver

nvidia-smi

Output should show no errors and details about your driver and GPU.

Patch driver

This patch performs backup of original file prior to making changes.

bash ./patch.sh

You're all set!

Rollback

If something got broken you may restore patched driver from backup:

bash ./patch.sh -r

Docker support

It is possible to use this patch with nvidia-docker containers, even if host machine hasn't patched drivers. See Dockerfile for example.

Essentially all you need to do during build is:

COPY the patch.sh and docker-entrypoint.sh files into your container.
Make sure docker-entrypoint.sh is invoked on container start.

docker-entrypoint.sh script does on-the-fly patching by means of manipulating dynamic linker to workaround read-only mount of Nvidia runtime. Finally it passes original docker command to shell, like if entrypoint was not restricted by ENTRYPOINT directive. So docker run --runtime=nvidia -it mycontainer echo 123 will print 123. Also it can be just invoked from your entrypoint script, if you have any.