JemmyChinese semiconductor outfit has Linux MPP repository on Github disabled after a DMCA takedown request & FFmpeg team accuses it of using libavcodec code without attribution | Tom's Hardware
www.tomshardware.com/software/chinese-semicondu…
14 Comments
Comments from other communities
rem26_artThe semiconductor manufacturer is Rockchip, who makes a ton of SoC’s for single board computers. Rockchip has apparently copied ffmpeg’s code without attribution and changed the license to a permissive one from LGPL
Terrible headline. Should have just been “Rockchip has…”.
Am I showing my age if I say that Tomshardware used to be decent?
Rusty ShacklefordI used to write a few things for HardOCP, and we always made fun of Tom’s for clickbait, chum, and straight up groveling to Nvidia. We straight up weren’t allowed to cite them.
That was like a decade ago.
…Maybe they were alright when Tom was still with them, but that was a long time ago. Too long for me to even know.
Sounds like it will be fixed soon
https://github.com/HermanChen/mpp/commit/fff87da91706d92913ba0254cee8c27eb093ae16
https://github.com/HermanChen/mpp/issues/73
Only after they were forced to. They’re criminal trash.
The verbiage of the first sentence is very telling.
We sincerely apologize for drawing so much attention…
…which to me reads as…
We’re sorry to our shareholders that we got caught with our pants down.
MorphitOh wow, so they actually stripped off the attribution to ffmpeg and slapped their own name and license in place of it. Now they’re forced to restored the license they’re working to rewrite it all themselves so they can delete the copyrighted code. They’re so sorry though.
I bought a Rockchip SBC (Orange Pi 5+), and when it worked it was awesome…but man, the software support (mainly kernel space) is just not there. Exercise in frustration to get everything working at the same time.
Currently running armbian. I don’t think HW acceleration is working, and I don’t think HDMI out is even working, but for my use case it’s a stable config…for now.
I think that might be the same chip as the Bananapi M7 which I did a write-up on last year. Trying a different kernel might help with your issue - it did for me.
https://blog.kumio.org/posts/2025/01/bananapim7-hvm.html#kernel-versions
I should probably update the table for trixie.
If I had published a popular library under LGPL, and then found out that a chip company stole my code by ignoring/removing the license (change to less restrictive in attribution) I would perhaps go as far as subtly block my code from ever properly functioning on the company’s chips, until the license is respected.
People might have forgot what happened in linux kernel with the “nvidia shim module”. Those were actually banned, non-gpl compatible kernel module cannot use gpl-only symbols from the kernel. What happened here is even worse, straight up violating the license from the authors.
GPL license should have a version that could cheaply be defended by the victim of the license violation, if a verbatim violating copy is found. Some €€/month bill could pileup while a violating copy is proven to be distributed.
edit: minor fixes.
Ffmpeg are kinda assholes and squelch innovation tbh.
N.E.P.T.RWydm? Rockchip copied their code, changed the license and didnt attribute FFmpeg. FFmpeg is a small team of enthusiasts who are responsible for plenty of important innovation and remain largely unpaid even with such substantial widespread use of their code in like SOOOO MANY big software projects. It isn’t their fault that people aren’t following the simple rules of the license to use their code.
In this case, sure, idk. I’ve had a project shut down because ffmpeg threatened law suits unless we open sourced our own custom libraries that work along side ffmpeg. And it’s not just the source code, they want full build environments. Our lawyers wouldn’t touch it, so we just shut it all down. Now I use gstreamer and avoid ffmpeg like the plague.
How would preventing companies from making their changes private squelch innovation? If people stop enforcing open source licenses, then companies would never contribute to the very same project they use. Ffmpeg is making innovation easier by forcing rockchip to publish their code.
If a company has to release their whole product as open source, they likely can’t make back their investment on development.
This particular case may be cut and dry, but I’ve had ffmpeg come after my company just for having an embedded Linux solution that uses off-the-shelf ffmpeg libraries. They claimed we had to release all of our custom libraries and app source and environment that were in the same product even though they didn’t have a line of ffmpeg code in them. As a small company, we couldn’t afford the litigation, so we just dropped the project.
Ah, so it’s personal, got it.
Well calling them assholes is kinda personal, no?
You’re complaining you couldn’t make back your investment on a project that leveraged other people’s code.
Code that was released under a GPL license.
What the fuck makes you think anyone gives a shit about your opinion of how unfair it is that you can’t profit from their code?
I contribute to open source plenty. My company hardly makes any profit. We put it all back into r&d because we like making stuff. Employing engineers then allows them to also contribute to open source. But we have to sell stuff to keep things going. And if we made improvement to the ffmpeg source, we’d definitely publish it. But we didn’t. We just used it in an embedded solution which had a bunch of custom apps gluing it all together. We did what many others have done, but ffmpeg decided to try to shake us down (they literally asked for a payout to make it go away). Corporate litigation is stupidly expensive and not what we want to spend our time on. I don’t really care. We had fun making the device and learned a bunch. We just also learned to avoid ffmpeg like the plague.
Dude, all Rockchip had to do to avoid this was a bit of license-related bookkeeping. They’re a corporation, so they’re used to dealing in bookkeeping and contracts. Someone at their end whose job it is to track this stuff either screwed up or let this through on purpose, assuming they wouldn’t be caught (more likely the latter, because China).
I would at least Google something before you state an opinion.
I have personal experience. I don’t need to Google it, thanks.
FAFO. and don’t expect sympathy.
I don’t need sympathy. I just warn others to steer clear of ffmpeg.
Couldn’t you rather provide the name of your company, so we know what software we should steer clear of?
Don’t worry, I’m sure you couldn’t afford our products. And I bet engineers at my company have contributed over 10x what you’ll ever contribute to useful open source libraries! But we can’t all live in mom and dad’s basement pretending we know what we’re talking about, can we?
Sounds like it will be fixed soon
https://github.com/HermanChen/mpp/commit/fff87da91706d92913ba0254cee8c27eb093ae16
https://github.com/HermanChen/mpp/issues/73
It’s kinda good but also reads a bit like
Such as?
Welcome to TomsHardware, where we do 0 research in our “journalism”
Random people saying shit like “why do open source, if you still sue?” or “Copyright needs to be abolished don’t use it for anything”
It’s really narrow minded to see it like that honestly, but they unfortunately do exist.
The first thing that came to my mind is that a DMCA takedown on GitHub doesn’t stop them from using it, but only from sharing their own additions with the world.
But sharing code is when a license like LGPL really has an effect in what you must do to comply..
I’m not sure I follow that sentence?
Everyone is free to edit, compile and use LGPL licensed code however you want on your PC. A DMCA can’t stop that, so it won’t make sense for someone to think that. At the time you share software (as a binary which used that LPGL licensed code, or the code itself) is when you are legally compelled to follow the license (hosting code on github).
Ah right. So I guess my point was: the DMCA takedown doesn’t necessarily force them to publish the code on GitHub, although luckily in this case they did end up doing that.
Usually they only comply by suing them, after nagging them for years..
Good morning TomsSlopware, maybe stop using Internet Explorer for your research.
This was already reported 2 weeks ago:
https://news.ycombinator.com/item?id=46394327
Sounds like RokChip really needed to put their hand on that hot stove top to understand they’d get burned. Hope they learn their lesson and improve their handling of FOSS code in the future.