NVIDIA 495 Linux Beta Driver Released With GBM Support

[oiaohm]

I think you're on to something here. Whoever built Wayland, built it like they were in a vacuum, with little to no respect to user's needs.
I'm sure they had the best intentions, but I really have a problem with their attitude when people started to point out shortcomings in their creation and their response was mostly "that's not secure" or "not our job".

Some cases not Wayland role/job is the correct answer. Just because X11 protocol would take in everything and the kitchen sink.(I really wish that kitchen sink was just a joke kitchen sink did in fact make X11 protocol define at one point marked as a mandatory feature) does not mean wayland has to.

Yes that is not a secure path is also a valid option.

Imagine if IPv6 came along and said "yeah, we're not doing the whole subnet and broadcast address things because they're not secure/out of scope. But implementers are free to implement them on their own".

We don't need to imagine this dig out the first version of IPv6 it does not have subnet at all because that did call claim subnet was out of scope at first. Yes it was up to the implementers to define how IPv6 subnets worked in future versions of the standard. What wayland is doing with not having X feature until all parties who will be implementing come to agreement is lot more common than a lot would think. IPV6 did this for many bits of what we call IPV6 now.

First version of ipv6 was to deal with the running out of addresses. Heck it did not even include mappings between ipv4 and ipv6.

[birdie]

oiaohm

"This shows how out of the loop you are".

Sorry, pal, it's you who are terribly out of the loop.

Wayland was created as a replacement for X11/Xorg for the desktop first and foremost. Desktop users first and foremost need features which have been provided by Xorg/Windows/MacOS for over three decades now while Wayland developers continue to repeat, "You won't get this or that because it's insecure, you won't get this or that because we want to keep the protocol as simple as possible".

NO ONE F'ING CARES.
NO ONE F'ING CARES where Wayland could be used aside from the desktop.
NO ONE F'ING CARES that Xorg doesn't have a beautiful lock screen option: again, it's another "vulnerability" which has never been exploited in the wild. Yes, pop ups under Xorg/X11 can be visible on top of the lock screen - who the f cares? Show me ways to exploit the lock screen using pop ups, PLEASE. I can give you my desktop or laptop to try to get into my locked sessions - GOOD LUCK with that, sorry, you won't even try.

I've already told what Wayland fans love to talk about (Xorg's "insecurities") and you just keep doing absolutely the same.

Desktop is a very special use case where work has to be done, and security considerations, while important, take a backseat. I presume Wayland could be a brilliant option for kiosks, smartphones, embedded devices, etc. etc. etc. but people using Linux on the desktop are not the users of these types of devices.

And I'm sorry to break it to you, but we already have much better more secure OSes for the same type of devices: Android, ChromeOS and Fuchsia - all from an evil company. It's 10000 times easier to customize Android than to go through the hoops of dealing with the Linux software stack + Wayland + whatever you want to run on top of it. Oh, and writing and supporting an Android app is a lot easier and more straightforward than trying to do the same using ever changing Linux/GNU APIs.

And it's not as if Linux has a ton of options in terms of coding rich applications. Qt is currently the only full-featured toolkit for Linux but it comes with a very restrictive license or you need to pay for it. GTK alone is not enough for the majority of applications. Other toolkits are even more laughable.

[oiaohm]

oiaohm

"This shows how out of the loop you are".

Sorry, pal, it's you who are terribly out of the loop.

Wayland was created as a replacement for X11/Xorg for the desktop first and foremost.

https://people.freedesktop.org/~daniels/lca2013-wayland-x11.pdf

Try again. Neither of the lead developers of wayland were paid to work X11 used on the normal desktop. Instead they are working with Nokia and other parties that were used X11 in embedded.

http://events17.linuxfoundation.org/sites/events/files/slides/Issues%20and%20Solutions%20for%20applying%20Window%20System%20to%20Automotive%20Infotainment%20System_0.pdf

birdie like it arguements are not based on the facts. Wayland/Weston was was designed to replace X11 usage in embedded desktops like infotainment systems, phone ,cameras.....

Of course this is a lot less feature list. Also please look at the above PDF on the infotainment system there are a lot of things X11 cannot do that the embedded desktops want that the wayland solution does.

Desktop is a very special use case where work has to be done, and security considerations, while important, take a backseat. I presume Wayland could be a brilliant option for kiosks, smartphones, embedded devices, etc. etc. etc. but people using Linux on the desktop are not the users of these types of devices.

Yes Wayland is a brilliant option for kiosks, embedded devices, etc where X11 and direct framebuffer use to be used. The first form of desktop Wayland was designed to replace and has successfully done.

Security considerations with wayland cannot take a backseat because of it other use cases. Of course the lead wayland developers that X11 bare metal would be maintained next to wayland until Wayland caught up in features. This is also why Wayland developers did not care about being feature complete for the general desktop case and instead focused on the specialist case.

Has wayland been successful in the first stage removing X11 server from a lot of embedded usage yes. Has wayland been successful at killing off linux framebuffer usage in embedded yes in fact almost complete to the point the Linux kernel is getting to delete more and more direct framebuffer drivers and only have DRI/KMS drivers. This has got embedded and general desktop on Linux using very related graphics stack to each other.

birdie the reality here people using X11 on the desktop are not the ones who paid the developers of Wayland wages. Yes the people who use to use X11 in embeddded devices as well as directfb are the ones who paid the developers of Waylands wages.

birdie you have had totally the wrong expectation. Developers of wayland goal is a secure desktop for embedded usage and possible in future for general desktop and really don't care how long it takes for them to get to that results. That very different to your idea right birdie that wayland is designed for the general desktop right.

[birdie]

OK, I get it, the Linux desktop has been hijacked by people who don't care about Linux on the desktop. That kinda explains everything about the current state of graphics subsystems (Xorg is nearly dead and barely maintained/Wayland is barely functional outside of Gnome and it's not like it's bug-free and complete) or GTK, which has recently stopped looking like a desktop toolkit altogether. Thanks oiaohm!

Just stop peddling Wayland please because it sucks and I don't bloody care about theoretical Xorg issues it solves. Show me something practical and required which doesn't involve multi-monitor configurations.

For instance, I have a desktop with the only 144Hz Gsync compatible monitor using an NVIDIA GPU and I run Fedora 35 XFCE on it. I'm the only user of this PC and no one else uses it.

Why should I switch from perfectly working XFCE/Xorg to Wayland? Oh, wait, I cannot use XFCE any more.

Sorry and goodbye. I don't want to run Gnome or come close to it. KDE under Wayland continues to be a bugfest.

[Myownfriend]

https://people.freedesktop.org/~daniels/lca2013-wayland-x11.pdf

I posted the video of the guy delivering that presentation and birdie misunderstoof why I even posted it. Birdie's a fucking idiot lol

[Myownfriend]

Whoever built Wayland, built it like they were in a vacuum, with little to no respect to user's needs.

Again with the language that infers that Wayland is personally spiting you. Instead of saying "I think *insert feature* should be in Wayland and here's why" its, "Wayland has no respect for me and it's devs have an attitude problem."

[Myownfriend]

Just stop peddling Wayland please because it sucks and I don't bloody care about theoretical Xorg issues it solves.

Nobody's "peddling Wayland". You started yet another argument about this shit and we're correcting you. A while back I argued with a bunch (about 10) conspiracy theorists who were flat earthers, gravity deniers, moonlandering deniers, etc. for about three weeks. Every time I corrected them about something, explained something to them, broke something down for them, or provided a links and videos that disproved what they were saying, they would deny that I ever provided any information to begin with and/or start acting like I'm in on the conspiracy and I'm peddling the "ball-earth theory" to people. You give me sooooo much of that energy.

I provided you with some articles and directory of found X11 vulnerabilities and you sadi I haven't provided you with anything. Now you're claiming I'm peddling Wayland to you when you entered the conversation on your own free will because bitching about things that you don't know about makes you hard.

[oiaohm]

OK, I get it, the Linux desktop has been hijacked by people who don't care about Linux on the desktop. That kinda explains everything about the current state of graphics subsystems (Xorg is nearly dead and barely maintained/Wayland is barely functional outside of Gnome and it's not like it's bug-free and complete) or GTK, which has recently stopped looking like a desktop toolkit altogether. Thanks oiaohm!

Nothing like being totally backwards. People who cared about embedded work and secure desktops where having their time hijacked repairing up X.org X11 baremetal server. Yes redhat focus on Wayland going forwards for RHEL is because they need secure desktop to sell to the USA MIL.

When those embedded and secure desktop developers got a chance to step away from X.org X11 bare metal this left a vacuum because there was no one who was not either a embedded developer or a secure desktop developer who was working on X.org X11 server. This showed how poor the upstream involvement with x.org X11 server has been by those making desktop environments.

The reality is people who truly care about the Linux Desktop had not been in key projects for a very long time.

The reality here Linux Desktop developer and user have been hijacking development time from people who have no interest in the Linux Desktop. Of course this was going break at some point the fact this worked for almost 3 decades was lucky in the first place.

Please also note for over 25 years at every X11 conference there was a call for more involvement in the core X11 server by desktop environment developers. People took the point of view they did not need to step forwards because other parties were doing the work and totally missing those parties were getting no reward out of it.

The embedded and secure desktop developer stepping away from core X11 server as a possibility was on the wall for over 2 and half decades.

Why should I switch from perfectly working XFCE/Xorg to Wayland? Oh, wait, I cannot use XFCE any more.

https://wiki.xfce.org/releng/4.18/roadmap This state with XFCE may or may not change. Yes work on progressing applications of XFCE to support wayland are under way.

The reality here if you are remaining on X.org X11 server the lack of developer problem caused by the embedded and secure desktop developers moving over to wayland will have to be fixed. There is attempt at the moment.

Yes even remaining on x.org X11 server there need to be some serous questions. Like does x.org X11 server proceed down the plan of only have a modesetting driver for graphics so getting rid of the vendor unique drivers. Will x.org X11 server now work on adding means to restart the X11 server like that KDE developers are working on for Wayland.

birdie there is always a price to pay for a free lunch. Yes X11 desktop envornments for a long time have been getting a free lunch of x.org X11 server without having to put any developer time into the x.org X11 server and leaving the x.org X11 server to the embedded and secure desktop developers at best. birdie its time to pay the piper basically.

birdie the scary question here with the miss with x.org X11 bare metal example people were depending on developers who no direct interest in their use case that could walk away as soon better option appeared the question is how many other projects are in exactly the same boat.

Upstream involvement by effected parties is important. Lack of upstream involvement by effected parties is what leaded to the case of where x.org X11 bare metal developer went dead and now has to be restarted instead of a nice clean hand over.

Hopefully after the crunch if x.org X11 bare metal lives on the developers will be people interested in the general usage case. But it should not have required the embedded and secure desktop developers to step back to make the general desktop usage case developer to step up and take up the workload.

[RecursiveRose]

Time to derail the pointless bickering about Wayland (It's the only pony for the future, get on board or get left behind, X11 is consigned to knacker's yard)

Has anyone gotten SwayWM to work with the new Nvidia BETA yet? I heard the driver dev had gotten it working, but I'm still facing errors with this release.

[bug77]

Again with the language that infers that Wayland is personally spiting you. Instead of saying "I think *insert feature* should be in Wayland and here's why" its, "Wayland has no respect for me and it's devs have an attitude problem."

Fractional scaling, working clipboard and color management should have been part of Wayland since day 1, because they are features no modern desktop can do without.
Instead of that, devs took years to implement one of these and are still figuring out how to deal with the other two.

There, better?

Some cases not Wayland role/job is the correct answer. Just because X11 protocol would take in everything and the kitchen sink.(I really wish that kitchen sink was just a joke kitchen sink did in fact make X11 protocol define at one point marked as a mandatory feature) does not mean wayland has to.

Yes that is not a secure path is also a valid option.

Do any of the missing features listed above look like the kitchen sink to you?

We don't need to imagine this dig out the first version of IPv6 it does not have subnet at all because that did call claim subnet was out of scope at first. Yes it was up to the implementers to define how IPv6 subnets worked in future versions of the standard. What wayland is doing with not having X feature until all parties who will be implementing come to agreement is lot more common than a lot would think. IPV6 did this for many bits of what we call IPV6 now.

First version of ipv6 was to deal with the running out of addresses. Heck it did not even include mappings between ipv4 and ipv6.

I don't know what you mean by "first version" (first draft, first ratified RFC?), but you made me look. A. Tanenbaum explains IPv4 in IPv6 in 1998. RFC3177 talks about IPv6 subnets in 2001.
You'll have to agree this is in stark contrast to the above features missing in Wayland 13 years in.