Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia releases open-source GPU kernel modules

developer.nvidia.com

Post reply on HN
Print view View original

271–280 of 415 posts

Re: Nvidia releases open-source GPU kernel modules

#271

Post by prash_ant »

greggsy wrote: post #266
mort96 wrote: post #238

Earlier quoted context omitted.

Sadly, it's also not the actual GPU driver, just the kernel part; you still need their massive closed blob in userspace. Almost had me excited there for a moment.

Has any rationale been provided around the blob being closed? Is it licensed, or simply too too complicated to pick apart? Graphics is a complicated field, but the techniques could still be patented - surely their market status isn’t dependent on trade secrets?

There have been reports earlier [1] that processing power of some GPUs is suppressed by software rather than the hardware capability itself. It is frequently easier to mass produce similar chips than to have different chips for different priced devices. I had come across comments in other online forums where the users alleged that some software flags restrict the capability. I hope someone else will link those webpages if they come across them

[1] https://www.tomshardware.com/news/nvidia-gpu-system-processo...

Re: Nvidia releases open-source GPU kernel modules

#272

Post by johnny22 »

moffkalast wrote: post #106
sylware wrote: post #18

The kernel modules are the first stage I guess, since a massive amount of hardware programing knowledge is in user space like with AMD/intel GPUs. I wonder how much LAPSUS$ hack has to do with it. I wonder if nvidia hardware programing interface is a mess like AMD one, just curious.

Likely very related, wasn't this one of their exact demands? Looks like Nvidia caved haha.

it's not at all. https://blogs.gnome.org/uraeus/2022/05/11/why-is-the-open-so...

Re: Nvidia releases open-source GPU kernel modules

#273

Post by johnny22 »

salawat wrote: post #143
kaladin-jasnah wrote: post #118

Earlier quoted context omitted.

This new driver targets GPUs with FALCON, and nouveau will be able to control it, as FALCON is GSP ( https://www.phoronix.com/scan.php?page=news_item&px=NVIDIA-R... ).

They're likely changing it because security researchers pwned FALCON. There is a way to leak the hash against which High Security mode FALCON compares the microcode for signature validation. So hopefully you'll forgive me if I maintain my skepticism over this being anything but Nvidia iterating to close a massive security defeat by redesigning their FALCON controller to mitigate the thorough pwning that's been achiev…

No post body was provided.

Re: Nvidia releases open-source GPU kernel modules

#274

Post by johnny22 »

evolve2k wrote: post #264
the_duke wrote: post #7

I wonder if this has been in the works for years, or if this is a reaction to the recent Lapsus hack.

Looking over the GitHub commit history of the two contributors on the repo, I’d say it’s a reaction. These folks look like they’ve barely touched much open source code before now. A planned run up would surely have these folks doing more GitHub based commits even if on very private repos.

c'mon y'all. read the other threads before commetning. It's not. https://blogs.gnome.org/uraeus/2022/05/11/why-is-the-open-so...

Re: Nvidia releases open-source GPU kernel modules

#275

Post by johnny22 »

black_puppydog wrote: post #27

Would love to hear from folks who worked on this! Was this an ongoing thing? Did it have to be pitched hard? What finally made the difference? I'm assuming here it's not because of LAPSUS as some speculate. This seems like it must have been in the making for quite a while.

No post body was provided.

Re: Nvidia releases open-source GPU kernel modules

#276

Post by johnny22 »

BeefySwain wrote: post #3

> Open kernel modules support all Ampere and Turing GPUs. Datacenter GPUs are supported for production, and support for GeForce and Workstation GPUs is alpha quality. Sounds like this is not going to be what I game on for at least a little bit?

Nope, but read this: https://blogs.gnome.org/uraeus/2022/05/11/why-is-the-open-so...

Re: Nvidia releases open-source GPU kernel modules

#277

Post by robocat »

smcl wrote: post #187
mjg59 wrote: post #53

1) This is unambiguously Good News 2) This is not upstreamable in its current form (nvidia admit this in their press release) 3) In an ideal world, nouveau (the open source driver for nvidia hardware) would be able to target this kernel code. Right now though there's no commitment for any sort of stable userland ABI, and that makes that difficult (moving to a new driver version may break interfaces that nouveau uses)…

> This is not upstreamable in its current form (nvidia admit this in their press release) True, I remember though that the semi-recent AMD Radeon drop was not immediately merge-able to mainline because they used a bespoke hardware abstraction layer. But (as alluded to by your point #1) it's a huge first step, and in AMD's case I think they eventually reworked it so that it could indeed be merged.

This is the key paragraph in the article relevant to your point:

  This open-source kernel code is currently split into OS-agnostic and kernel interface layer components. This stems from NVIDIA's proprietary driver on Linux largely being shared code across Windows / Linux / FreeBSD / Solaris. For it to be upstreamed in the Linux kernel it would likely need to be more re-factored to cater to Linux, just as AMD's DAL/DC originally had tough time upstreaming due to its numerous abstractions.

Re: Nvidia releases open-source GPU kernel modules

#278

Post by hyperionplays »

prash_ant wrote: post #271
greggsy wrote: post #266

Earlier quoted context omitted.

Has any rationale been provided around the blob being closed? Is it licensed, or simply too too complicated to pick apart? Graphics is a complicated field, but the techniques could still be patented - surely their market status isn’t dependent on trade secrets?

There have been reports earlier [1] that processing power of some GPUs is suppressed by software rather than the hardware capability itself. It is frequently easier to mass produce similar chips than to have different chips for different priced devices. I had come across comments in other online forums where the users alleged that some software flags restrict the capability. I hope someone else will link those webpag…

this was the case with RTX voice, a simple regedit made it work on non-RTX cards fine.

Re: Nvidia releases open-source GPU kernel modules

#279

Post by dmead »

marcodiego wrote: post #48

For those who didn't use Nvidia on linux in the old times: The driver was a proprietary binary. Since a kernel module requires interfacing with the kernel API, it could be considered a derivative work and a breach of the GPL license. So, Nvidia provided a small open source shim which interfaced between the kernel and the proprietary module. You had to compile that shim yourself with the right arcane command line inca…

the old times is what precipitated linus to give nvidia the finger.

Re: Nvidia releases open-source GPU kernel modules

#280

Post by lights0123 »

Humphrey wrote: post #265

I'm confused! Why would Nvidia want to keep any of this closed source. Surely they make money when people buy graphics cards, and having open source out-of-the-box graphics support in Linux would mean they would sell more graphics cards?

They only really care about Linux usage in embedded computers (Jetson) and in datacenters, and Free drivers would allow you to modify them to permit using consumer GPUs in virtual machines. Currently, you need to spend significantly more money for an enterprise GPU that has the same specs as a consumer GPU just so the driver will allow you to use GPU passthrough. They did recently allow consumers to pass a GPU to a single Windows VM guest so you could run a Windows-only game, but you can't split access to the single GPU among multiple VMs.
Post reply on HN