SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
If anyone is interested, I have a build script for CUDA 10.
I recently got a new GPU (the very cool NVIDIA TITAN Xp Star Wars Collector's Edition; sorry I'm a geek ) to work with at the university and I had to upgrade to CUDA 10 in order to access all the hardware features available.
So, had to write my own CUDA build script since the SlackBuilds.org CUDA is at 8.0.61 and much have changed since.
You can check out the CUDA build script here: https://github.com/aragorn2101/Slack...er/cudatoolkit
I used the 430.34 driver as the SlackBuilds.org driver was a little bit behind when I was working on this, but I think any 430.xx driver will work.
The CUDA version is 10.1.168, but feel free to modify the script and adapt to your needs. For example, there is a new 10.1.243 toolkit which might need a test.
The CUDA runfiles have names like "cuda_10.1.243_418.87.00_linux.run". It means that they include the driver 418.87 within and it does not mean that it will not work with drivers 430.xx or higher.
The CUDA runfiles have names like "cuda_10.1.243_418.87.00_linux.run". It means that they include the driver 418.87 within and it does not mean that it will not work with drivers 430.xx or higher.
Cheers
Does cuda_10.1.243 work with gcc 9.2 (while the officially supported gcc versions are 8.x and below)?
Sorry but I just happened to have all my computers with NVIDIA GPUs in production use or I'll have tried gcc-9.2 by myself.
I can only tell that gcc 8.3 and 7.4 work well with it.
Does cuda_10.1.243 work with gcc 9.2 (while the officially supported gcc versions are 8.x and below)?
Sorry but I just happened to have all my computers with NVIDIA GPUs in production use or I'll have tried gcc-9.2 by myself.
I can only tell that gcc 8.3 and 7.4 work well with it.
As far as I can tell, it does not work with versions 9 of gcc, for now at least. At first, I had to upgrade the kernel on Slackware 14.2 since the 4.4.189 kernel does not have support for my IntelGraphics. I took the kernel from Current, and also upgrade gcc to the 9.x.x that Current had. Turned out that nvcc wouldn't work. I saw this in the past with CUDA 8 as well and it is usually because the in-built CUDA compiler is mismatched with your machine's gcc and glibc. If Nvidia released the complete source of their compiler we could have recompiled everything using the new gcc, but that's not the case.
As far as I can tell, it does not work with versions 9 of gcc, for now at least. At first, I had to upgrade the kernel on Slackware 14.2 since the 4.4.189 kernel does not have support for my IntelGraphics. I took the kernel from Current, and also upgrade gcc to the 9.x.x that Current had. Turned out that nvcc wouldn't work. I saw this in the past with CUDA 8 as well and it is usually because the in-built CUDA compiler is mismatched with your machine's gcc and glibc. If Nvidia released the complete source of their compiler we could have recompiled everything using the new gcc, but that's not the case.
Probably more complicated than recompiling the source. Working as a preprocessor, nvcc has to support new syntax introduced with the new version of gcc and libstdc++, if any.
The safe solution seems now to be installing an older version along with gcc-9, and specify -ccbin on the nvcc command line.
They released a vulkan beta driver rebased around the 435 series, too. I've been using it for about an hour on current and no issues. (Works with No Man's Sky! The older vulkan beta driver didn't.)
They released a vulkan beta driver rebased around the 435 series, too. I've been using it for about an hour on current and no issues. (Works with No Man's Sky! The older vulkan beta driver didn't.)
do you by chance know how long it usually is before the official vulkan driver comes out?
I have playback issues (video pausing when starting video or jumping to a different position in the video with odd playback behavior for a few seconds when it starts back) with UltraHD video files I grabbed off YouTube with the 435.19.02 and 435.21 drivers.
Here is the video information regarding one of the videos in question:
I have playback issues (video pausing when starting video or jumping to a different position in the video with odd playback behavior for a few seconds when it starts back) with UltraHD video files I grabbed off YouTube with the 435.19.02 and 435.21 drivers.
Here is the video information regarding one of the videos in question:
I'm just letting people know that there is a bug in the new drivers because the 418 and 430 drivers work fine. If I made another thread it would be in the NVIDIA forum complaining about this issue.
Distribution: Slackware64-current with "True Multilib" and KDE4Town.
Posts: 9,098
Original Poster
Rep:
Quote:
Originally Posted by RadicalDreamer
I have playback issues (video pausing when starting video or jumping to a different position in the video with odd playback behavior for a few seconds when it starts back) with UltraHD video files I grabbed off YouTube with the 435.19.02 and 435.21 drivers.
Here is the video information regarding one of the videos in question:
I've posted the SBo updates to nvidia-430.50, the long lived branch. Please note that this update will not yet work properly on Slackware-current. Also:
Mats reminded me of a Nvidia post from 2018 for Nvidia driver
end-of-life schedule:
nvidia-legacy304 304.137 EOL 2017 no further updates
nvidia-legacy340 340.107 EOL 2019 in pending
nvidia-legacy390 390.129 EOL 2022 released
nvidia-legacy304 is UNMAINTAINED. I will treat bug reports as
"Deprecated; won't fix".
nvidia-legacy340 is also UNMAINTAINED, since Nvidia has not released a
fix for recent kernel compile issues. Mats sent me a patch (also on 2 other threads here on LQ). I will attempt updates as time
permits with SBo community assistance until December 31, 2019.
Fixed display color range handling on pre-Turing GPUs, such that when limited color range is selected through the display controls page in nvidia-settings, output pixel values will be correctly clamped to Consumer Technology Association (CTA) range.
Distribution: Slackware64-current with "True Multilib" and KDE4Town.
Posts: 9,098
Original Poster
Rep:
440.26
Last Thursday, 17 October 2019, Nvidia released a new beta series with several new features and bug fixes.
I've been running it for the last four days with no coughs or hiccups.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.