Ubuntu 19.10 To Drop 32-bit x86 Packages

[F.Ultra]

But it still consumes some resources.

The only resource that Universe consumes for Canonical is the hosting of the files. That is all Canonical offers for Universe; storage and bandwith. And btw Gnome is also part of Universe and so is probably every single piece of software that you are upset that they "support" on s390x.

Yes that might be difficult to accept but packages such as Gnome is not officially supported by Canonical. This means that they can offer some support for say amd64 where people are using Gnome extensively while they can just ignore it for s390x but still have the package available on a "don't call us if it breaks" kind of support.

[the_scx]

The only resource that Universe consumes for Canonical is the hosting of the files. That is all Canonical offers for Universe; storage and bandwith.

It is simple not true. They are sharing the whole infrastructure, including build servers.

And btw Gnome is also part of Universe and so is probably every single piece of software that you are upset that they "support" on s390x.

No, it is not. All packages I mentioned earlier comes from the main repository.

http://ftp.tu-chemnitz.de/pub/linux/ubuntu-ports/dists/bionic/main/binary-s390x/Packages.xz

https://pastebin.com/raw/MdYYg7A9

Yes that might be difficult to accept but packages such as Gnome is not officially supported by Canonical.

This might be difficult to accept, but you are wrong.

[chithanh]

So you are saying that open source games on mainframes are more important than at least minimal multilib/multiarch support on x86-64 desktops?

What, where did you get that "more important" from? Certainly not me. What I'm saying why there is s390/ppc64le support is summarized best as:

the amount of extra work going from 0 to 1 i386 libraries is much more than for going from 1 to 100.

Their Ubuntu Server has to run on IBM mainframes.

there are big paying customers

So the brunt of the work has already been done for the paying customers. Flipping on ppc64le support for more packages isn't hard once the port exists.

If they don't have time to support Linux desktop properly (that includes support for multiarch/multilib) because they are currently focuses on servers, clouds and IoT solution, it is fine. They can say that they are not interested in Linux desktop anymore

I'd say that running 32-bit proprietary applications on a desktop really is a small minority use case. And there are ways to still make them work, just less convenient. I don't see how you can reach from "drop support for i386 from system libraries" to "not interested in Linux desktop anymore".

As I said, what now happens to multiarch/multilib, in a few years can also affect XWayland.

Maintaining XWayland, which is a single package, is not at all comparable work to maintaining an entire architecture port. It would need an update if Wayland protocol or related APIs change in an incompatible way (next opportunity would be the GBM/EGLStreams/UNIX device memory allocator thing). And even if Ubuntu developers lose interest, they can still move the package to universe and let the community do that work.

The reality is that WINE support for Windows 9x software and DirectX 1-8 is crap. You have a better chance to run patched game for Windows XP while using dgVoodoo 2 on a DVXK setup, than trying to run the original game in Windows 98 mode.

You have to distinguish here between general software, and the special breed of software that is games. Before Proton, even modern games were much more of a hit-and-miss with Wine. After Valve started pouring development resources into making those work the situation started improving considerably.

When it comes to general Win16 and early Win32 software, the support is actually pretty good. In case of Win16, infinitely better than Windows itself, which does not support those at all in x64.

[oiaohm]

They doesn't care either about driver quality and performance...
It doesn't support Vulkan at all (although they are trying to get it), and OpenGL is supported only up to 4.3. The overall driver quality is even worse than Nouveau.

https://web.archive.org/web/20190619191659/https://mesamatrix.net/

And performance is just terrible.

https://www.phoronix.com/scan.php?page=news_item&px=Virgil-3D-F24-Beta-Tests

To be honest, they have made significant progress since that time, but it is still not suitable from gamers.

So, could you just stop trolling and publishing nonsense? Thanks in advance.

As you said performance has improved a lot since then. https://www.phoronix.com/scan.php?pa...Transfer-Queue

Virgl is working on vulkan support and it should have lower overhead.

[the_scx]

So the brunt of the work has already been done for the paying customers. Flipping on ppc64le support for more packages isn't hard once the port exists.

It is not like they are creating Ubuntu on x86 for free. They have a lot money from paid support and ESM (Extended Security Maintenance).
What's more, they also have money from Snap Store.
This wouldn't be possible if they hadn't been so successful on Linux desktop. By successful I mean that Ubuntu is probably major Linux distribution for the average user.

Also, please keep in mind that the main Linux distribution used by IBM is not Ubuntu LTS but RHEL/CentOS. This also includes Linux on z Systems. Ubuntu is relatively new here. They didn't even exist on this market until 2015.
We have a similar situation with Linux on Power. Red Hat Enterprise Linux is 1st here, and SUSE Linux Enterprise Server is 2nd. Ubuntu is definitely not the most important here.

I'd say that running 32-bit proprietary applications on a desktop really is a small minority use case.

And I would say that most desktop users or at least a very significant group will come across such software.

And there are ways to still make them work, just less convenient.

You don't need to install Ubuntu on bare metal at all. You can run Linux software using WSL on Windows 10. Or you can install a Linux distribution on VirtualBox. It is "just less convenient" method, but no one should complain, right?

I don't see how you can reach from "drop support for i386 from system libraries" to "not interested in Linux desktop anymore".

Of course I used a little exaggeration here, but it doesn't change the fact that they really should rethink their decision if they don't want to piss off their users.

Maintaining XWayland, which is a single package, is not at all comparable work to maintaining an entire architecture port. It would need an update if Wayland protocol or related APIs change in an incompatible way (next opportunity would be the GBM/EGLStreams/UNIX device memory allocator thing). And even if Ubuntu developers lose interest, they can still move the package to universe and let the community do that work.

What if it will be the vendor developer who lost interest in supporting XWayland? Red Hat can do it on its own, but what about Canonical? Will they do the same or will they not care?
What is worse, no XWayland package in the main repo is much worst than you think. It could lead into dropping support for it from other software, i.e. Gtk+, Qt, SDL, etc. I can't even imagine how the community can handle such a situation. If you already have Gtk+ in the main repo, you wouldn't put an improved version in the contrib repo.

You have to distinguish here between general software, and the special breed of software that is games. Before Proton, even modern games were much more of a hit-and-miss with Wine. After Valve started pouring development resources into making those work the situation started improving considerably.

Like it or not, but games are driving the development of WINE. I am not saying that it is focused on games, but they have significant impact on it. Without games, WINE would never gain good support for DirectX. And today, DX is important not only in games, but also in many other software.
Anyway, compatibility with Windows 9x isn't too good, and probably won't be much better, because main development is focused on Windows 7+ now.

When it comes to general Win16 and early Win32 software, the support is actually pretty good. In case of Win16, infinitely better than Windows itself, which does not support those at all in x64.

It isn't entirely true. Win16 software is still well supported in Windows 10. You just need to run Windows 3.11 inside virtual machine. Right?
But to be honest, almost no one care about Win16 software anymore, so there is no reason to support it. And if you want to run some DOS games, you can use DOSBox. End of story.

[the_scx]

As you said performance has improved a lot since then. https://www.phoronix.com/scan.php?pa...Transfer-Queue

And it is still too low for games.

Virgl is working on vulkan support and it should have lower overhead.

End users don't care. They want working and polished solution, not something that may be usable in 5 or 10 years!

Virgl maybe good enough (just maybe!) for some desktop applications, but currently it is completely unsuitable for gaming. End of topic.

[oiaohm]

And it is still too low for games.

Its not too low for all games. Lot of the games that do not have a 64 bit port do not require a lot of GPU or CPU performance.

Games are not black and white performance requirements. You need to start sorting the game with 64 bit versions and games with only 32 bit versions. You will notice something. .

End users don't care. They want working and polished solution, not something that may be usable in 5 or 10 years!

Virgl maybe good enough (just maybe!) for some desktop applications, but currently it is completely unsuitable for gaming. End of topic.

19.10 is a development release. LTS release with problem is 20.04 if they decide to go that path long term. So we have at least 6 months to have a solution of some form for the LTS.

Reality Virgl even with its over head is good enough for 90% of the 32 bit only games. I will give you Virgl is not good enough for games that have 64 bit versions those are more modern..

[F.Ultra]

I'm not sure that I'm reading it well, but on this page:

https://packages.ubuntu.com/bionic/gnome/

..it appears that some GNOME packages are not from Universe, so perhaps from Main. Like "cheese" - "tool to take pictures and videos from your webcam". Pehaps that is an important piece of software?

Edit:
apt show cheese
Supported: 5y
Download-Size: 137 kB
APT-Sources: https://xxx/ubuntu/archive bionic/main amd64 Packages
Description: tool to take pictures and videos from your webcam
A webcam application that supports image and video capture. Makes
it easy to take photos and videos of you, your friends, pets or whatever
you want. Allows you to apply fancy visual effects, fine-control image
settings and has features such as Multi-Burst mode, Countdown timer
for photos.

Yes I didn't claim that every single piece of desktop software is in Universe. Of course you will find "strange for a mainframe" software in Main if you really look, such is the name of the game. Once a package is in Main they have to remain in Main for every arch so this whole argument is moot, Ubuntu didn't decide that Cheese was worth their resources for s390x while i386 was not, they decided that the whole of i386 was not worth it while they need s390x for their paying enterprise customers and Cheese just happens to be in Main and thus a package that they have to maintain for s390x even though no one will ever use it there.

[F.Ultra]

It is simple not true. They are sharing the whole infrastructure, including build servers.

Still does not take resources away from the Ubuntu maintainers which is the issue here, or rather the claimed issue.

No, it is not. All packages I mentioned earlier comes from the main repository.

http://ftp.tu-chemnitz.de/pub/linux/ubuntu-ports/dists/bionic/main/binary-s390x/Packages.xz

https://pastebin.com/raw/MdYYg7A9

This might be difficult to accept, but you are wrong.

The "gnome" package for the full Gnome desktop is in Universe, all the other packages that you have mentioned including tuxpaint is not in the packages file that you linked so which are the packages that you mentioned earlier that comes from main?

However all that is just one big straw man because Ubuntu have not said that they could maintain i386 if they dropped some packages for s390x and if they have to support s390x, which they have if they want to keep their paying enterprise customers, then they have to support Main for s390x regardless of which "stupid for s390x" packages it might contain.

You cannot seriously argue that they should drop s390x because you might find some offensive packages in Main, that have zero to do with the maintenance costs of i386.

[FireBurn]

It won't help, because you still need a 32-bit version of at least glibc and OpenGL/Vulkan libs. The Steam Runtime doesn't provide it and never will be, because it is out of the scope of this solution.

But they could add that in... they already have libstdc++ in there