Fedora 33's "Enterprise Linux Next" Effort Approved - Testbed For Raising CPU Requirements, Etc

[phoronix]

Phoronix: Fedora 33's "Enterprise Linux Next" Effort Approved - Testbed For Raising CPU Requirements, Etc

Fedora 33 later this year will see a new "Enterprise Linux Next" (ELN) buildroot and compose setup for testing new changes potentially destined for the next release of Red Hat Enterprise Linux. Most notable from the original ELN proposal is on potentially raising the x86_64 CPU requirements. ELN is now approved to take place...

http://www.phoronix.com/scan.php?page=news_item&px=Fedora-33-ELN-Approved

[skeevy420]

Dammit IBM Hat, -march=westmere is the farthest back you can go unless you intentionally want to lose customers who use cheap Intel workstations since new Celerons and cheap Pentiums are essentially -march=westmere on an updated platform 'cause Intel uses instruction sets as a way to fragment the market.

If y'all don't know that y'all don't deserve to be an industry leading Linux distribution.

Another similar example could be the usage of link-time-optimization.

Well, at least one of y'all over there ain't retarded.

[Setif]

Before you criticize RedHat/Fedora, This distribution is for Enterprises/Server.

[re:fi.64]

Dammit IBM Hat, -march=westmere is the farthest back you can go unless you intentionally want to lose customers who use cheap Intel workstations since new Celerons and cheap Pentiums are essentially -march=westmere on an updated platform 'cause Intel uses instruction sets as a way to fragment the market.

If y'all don't know that y'all don't deserve to be an industry leading Linux distribution.



Well, at least one of y'all over there ain't retarded.

This a is a test for raising the limits on servers. If you're running a production RHEL server on a Pentium, losing out on that would be the absolute least of your issues...

[skeevy420]

And enterprise tends to cheap out and buy cheapo Intels that are likely to not support AVX or AVX2.

LTO, OTOH, should ideally benefit everyone.

What Clear Linux does shouldn't be the standard.

Generic with LTO + What Clear Linux does tacked on with FatELF, OTHO....that's what industry leading Linux should be focusing on.

[zxy_thf]

Dammit IBM Hat, -march=westmere is the farthest back you can go unless you intentionally want to lose customers who use cheap Intel workstations since new Celerons and cheap Pentiums are essentially -march=westmere on an updated platform 'cause Intel uses instruction sets as a way to fragment the market.

If y'all don't know that y'all don't deserve to be an industry leading Linux distribution.

So every user is forced to use the slower binary targeting old ISA just because Intel is still selling crappy chips?
No thanks. Let the darn obsoleted chips die.

[pipe13]

I mean, it's not as if the current RHEL8 is going to just magically disappear...

[uxmkt]

Generic with LTO + What Clear Linux does tacked on with FatELF, OTHO....that's what industry leading Linux should be focusing on.

The industry focuses on what makes them money, not wishes from a random Phoronix user. If you think it should be done differently, you are free to make a product of your own and disprove all the fools that did not see it your way. Until then, you may remain silent.

[nils_]

And enterprise tends to cheap out and buy cheapo Intels that are likely to not support AVX or AVX2.

Depends on the enterprise and whether you look at servers or workstations. I would think those companies that buy cheapo Intels tend to use Windows.

[programmerjake]

I'm personally hoping that there will still be support for generic x86_64 since I would like to build an open-source x86_64 processor once the ISA patents expire (maybe a year or two), and there are other people who have similar intentions.