That displaylink shit is basically dead. Look at the github repo, 3 or 4 people with 20 commits in the past year. There is obvsly no financial backing behind this whole mess. U should fully open source it or give up and let this stuff be. After 5 years you basically accomplished nothing besides vendors of docking stations using that stuff, but there is not even one release that one could call "stable".

Really, without an in-kernel Linux driver this display-link stuff does not make any sense at all if it breaks with every kernel update. (aside from the fact that I never managed to get the driver to work well)
I am not recommending DisplayLink until this driver gets upstream and fixed.

Sad. I've avoided getting a displaylink device for years because of the driver issue. My work environment changed and having to plug 5+ devices to my laptop every time I brought it back to the office got me to get a device that had display link. Sadly, the display link part of the device doesn't work. This really needs to be open sourced as more and more developers use laptops and Linux it is just going to make this product like AGP and other connection standards that are history.

Thank you for considering this request. With desktop Linux use now growing, it's a very good time to show your support. We remember! I'm hoping to buy 2 DisplayLink-based expanders, one for my Linux laptop from Dell, and one for my homebuilt workstation running Ubuntu 20.04 so that I can add additional monitors. Much appreciated!

I'm very disappointed that is currently not possible to install the "current" driver in Ubuntu 20.04 with the kernel 5.4.x and its flavours (e.g. MATE incl. the same kernel) . In addition i think its a shame because on Mac and Win it seems not to be problem to have a quick solution. Please fix this as soon as possible.

I work for a company that supports large Kubernetes deployments for organizations all over the world. In the Kubernetes and server side community, we live in Linux. After seeing the difficulty of running Zoom, Signal, Atom, Slack, and many other non graphic intensive apps or even just working with multiple monitors on a bad day, I have started heavily recommending to everyone that they not purchase anything dependent on DisplayLink drivers. This is a pity since I otherwise quite like ThinkPads.

Please please fix this driver issue or make it open, so we can. You're alienating a growing segment of the professional market that would otherwise be very happy to give you more money.

I bought a thinkpad docker usb c with 2 hdmi and I am very disapointend.
Its a same most of people buying this item usually they do not use windows.
Still doesn' t work well on Fedora 31 I have some many troubles to do it work. I have tried on ubuntu ass well and there just work on gonome but not on wayland it is imposible to rescale 125% or 150% . Only works well on windows 10. I am going to return to Amazon. If you do not offer a good support to linux I would not buy anymore to my bussines

I've written a patch for 5.2.14 dkms module for Linux 5.4.x compatibility: https://gist.github.com/alllexx88/1aebabd49da990089a489cb4f5c94698
You can apply it as follows:

wget -q -O- https://gist.githubusercontent.com/alllexx88/1aebabd49da990089a489cb4f5c94698/raw/c0856b2ab2d15f534686f8bbaa8ee48e8d93b4d6/DisplayLink-5.2.14-Linux-5.4.patch | sudo patch -p1 -d /usr/src/evdi-5.2.14

Then make sure your compiler (/usr/bin/cc) matches the version used to build your kernel (e.g., atm for kernel 5.4.x installed on Ubuntu using ukuu it has to be 9.2.1) and build+install the dkms module:
sudo dkms install evdi/5.2.14

Have a lot of issues with this.

Would be great if it was just integrated with Linux kernel hence not requiring DKMS crap which make kernel hard to boot in secure boot environment.

Still (18. Dec. '19) no support for Kernel 5.4.x !!!!!

ugh still poor linux support , good product crap compatibility because of lack of working drivers

It's a shame. Nothing less. In our days and age to have to manually download a
driver, broken moreover, and unable to follow modern kernels developement. This
is unacceptable.

It's not just about Linux.
There are other OpenSource UNIX like operating out there which also would like to implement an USB3 udl driver which gets you potential new customers.
OpenBSD e.g. has an USB2 udl driver, and we also would like to implement the USB3 one.

Why not doing it as suggested by Puchuu Punya and make a compression blob and release the driver itself as open source?

Would appreciate a feedback from you, private or public - Thanks.

I been using Linux desktop since 1999 and I cannot believe that today I still have to deal with such a product like this.
I started in a new company, installed Linux and plugged in the Type-C dock that I was provided just to learn that there is a company named DisplayLink that spreads this proprietary technology all over the place.
After doing some research on how to get this dock to light up on my Fedora Laptop, I actually figured I need something else, an alternative to this DisplayLink crap.

I use Fedora like many other developers and see that only Ubuntu is supported.
Can anyone recommend something that simply works on Linux? (Type-C x 2 Monitors and charge?)

Thanks

Has anyone figured out on how to flatpack or snap pack display link drivers along with dependencies? Using solus 4.0 here and a newbie.

+1

This would be a great idea! By making it open source, the overall experience will be better on Linux because you won't have to worry about getting the correct version anymore. Whatever distro you're running will have the version of Display Link you need ready to install a command away. This would also allow support for a newer distro like Ubuntu 19.04 to be available sooner.

It's great that DisplayLink shows support for GNU/Linux. Now, if they would release the source code as libre software that would be awesome. And make it easy or "automatic" to install. Awesome!

GPL v3 is a good license for driver software. (It's for hardware; that's exactly why Richard Stallman started GNU.)

> Agreed. If the compression technology is too proprietary to be released, then make 2 separate modules, one proprietary to load the compression algorithms, one open source so that kernel developers can easily port the module and update with each new kernel, while the compression algorithm remains closed source and proprietary. That way is best of both worlds.

It is 100% true. You can make a pure compression blob with simple input-output interface and nobody will touch your patents, you can keep it forever.

Do not forget that there are some bad countries like post USSR. Here we have no laws, we can make reverse engineering of your software. Unfortunately mobile monitors is not so popular and nobody wants to do reverse engineering today. But I have to warn you about it.