Intel Itanium IA-64 Support To Be Deprecated By GCC 10, Planned Removal In GCC 11

[Guest]

Dead as a dodo. Intels best effort to sideline it's own x86 show, now defunct.
But I'm sure x86 will die any day now...

I don't know, x86 is good enough for airgapped gaming devices. Make it into something that people will only use in their PlayStation or something.

[bridgman]

A few good ideas? I can't even think of one. Except maybe some stuff like SMT, which other CPUs have too and which is basically contrary to the very concept of the Itanium. They implemented it as a band-aid trying to make up for a fundamentally terrible design.

I have a different take on this. Itanium was a good idea when it was conceived (before anyone realized just how far superscalar x86 designs could go) but had two major things going against it - one predictable, the other not so much:

- poor performance when running x86 code (this one was IMO predictable and hurt short term adoption, particularly after AMD64)
- superscalar x86 implementations became incredibly wide and capable (I suspect this surprised even the teams working on them)

There were also questions at the time IIRC about whether the compilers were really doing a sufficiently good job of extracting parallelism from the code. Not sure how that worked out but my impression was that there was simply not enough VLIW hardware on the market to build a critical mass of compiler technology to support it.

Modern x86 designs are mind-blowingly complex and clever. Not only are they able to pick 5-10 operations out of a single instruction stream to execute in parallel (at peak) but they pick through a 100+ instruction window in real time to accomplish that.

A lot of superscalar CPU technology builds on Robert Tomasulo's work at IBM in the 1960's on the 360/91... I never had a chance to meet him (big regret) but I imagine even he would have been surprised how far his ideas were able to be extended.

One of the more interesting things about x86 history is how out-of-order execution showed up more or less simultaneously in designs from Cyrix, Intel and AMD during late 1995 and early 1996... 30-ish years after it appeared in mainframes. Wikipedia says that Cyrix was first to market, which is pretty impressive.

[Bsdisbetter]

The Itanic will be remembered as one of the worst ideas in CPU design history, with a botched implementation to match.

That would be the same non-vulnerable intel chip series, would it?

[Bsdisbetter]

I have a different take on this. Itanium was a good idea when it was conceived (before anyone realized just how far superscalar x86 designs could go) but had two major things going against it - one predictable, the other not so much:

- poor performance when running x86 code (this one was IMO predictable and hurt short term adoption, particularly after AMD64)
- superscalar x86 implementations became incredibly wide and capable (I suspect this surprised even the teams working on them)

Modern x86 designs are mind-blowingly complex and clever. Not only are they able to pick 5-10 operations out of a single instruction stream to execute in parallel (at peak) but they pick through a 100+ instruction window in real time to accomplish that.

A lot of superscalar CPU technology builds on Robert Tomasulo's work at IBM in the 1960's on the 360/91... I never had a chance to meet him (big regret) but I imagine even he would have been surprised how far his ideas were able to be extended.

One of the more interesting things about x86 history is how out-of-order execution showed up more or less simultaneously in designs from Cyrix, Intel and AMD during late 1995 and early 1996... 30-ish years after it appeared in mainframes. Wikipedia says that Cyrix was first to market, which is pretty impressive.

The x86 emulator was a joke. True.

[Bsdisbetter]

Most customers are moving legacy apps on HP-UX IA-64 to containers. Then they can let the clock run out on the hardware as HP allows. This gives them time to develop the replacements on the platform of their choice. Eazy-peasy.

That might be true for hp/ux but not openvms. They have a long roadmap (traditionally) of support, so corporations have time. Similarly, vms software is migrating the code to x86. Some corporations are still running vax and/or alpha clusters - 20+ years after their prime.

[Bsdisbetter]

I smell HP landfill!

It's sad, but there's still a lot of IA64 out there, and now it's destined for the poop pile. One of the things I love about Linux is how it can keep really old equipment (especially equipment from a bad vendor) out of our landfills (yes, I supposed a good scrapper might be able to get something out of them).

You'd be surprised. More corporate pcs go to landfill than high end servers, that's just a matter of volume. On these systems with high costs the propensity is for corporations to keep them running for decades (and decades). If you pay HP enough money they'll provide support for dinosaur poo, if you need it.

[jpg44]

In ten years you'll still have those containers running on whatever hardware they moved it to, probably with only difference is whatever they've replaced failed hardware in that intervening period. It's also just as likely those old Itanium servers will still be in use because no one in management wants to either spend money on migration, or none of the IT support staff want to wake the sleeping giant of problems trying to replace legacy hardware can cause. Like the IRS is still using decades old IBM mainframes, and companies still using old punch card calculator systems. If it's not broke, don't run the risk of "fixing it" and having a technical catastrophe (failed/inept migration) and/or political nightmare on your head (raising taxes to pay for it and/or major budget increases to management-voters-shareholders).

https://www.nextgov.com/cio-briefing/2016/05/10-oldest-it-systems-federal-government/128599/

https://www.pcworld.com/article/249951/if-it-aint-broke-dont-fix-it-ancient-computers-in-use-today.html

If these aren't public facing systems, there's usually not any real risk to leaving them alone other than hardware failure.

IRS does not use decades old mainframe. IBM has a migration path, they migrated to more recent IBM hardware thats backward compatable

[jpg44]

"Considering the GCC compiler is used to compile the Linux kernel and IA-64 doesn't enjoy coverage from other compilers like Clang able to build the Linux kernel, it will effectively mean the end of the road for new Linux support moving forward."

Pretty sure when RHEL and SLES stopped supporting IA-64 a while back, that marked the end of the road for Linux on IA-64. This is enterprise hardware, nobody is running Gentoo on these things lol.

Linux is used in enterprises, but Itanium is sort of a dead platform, so no one is spending time porting new Linux distros to it, just keeping their legacy applications running and migrating off the platform.

[Bsdisbetter]

Linux is used in enterprises, but Itanium is sort of a dead platform, so no one is spending time porting new Linux distros to it, just keeping their legacy applications running and migrating off the platform.

It seems there's enterprises and there's enterprises. The sort of server hardware Itanium is in is totally locked down by the manufacturers.
The only time you would consider using them with *Linux derivatives is when they are sold off and go into private hands. And then you would use NetBSD for that.
Itanium based servers are a package of software/hardware; you cannot separate the two. You have a choice: black or black.

[torsionbar28]

It seems there's enterprises and there's enterprises. The sort of server hardware Itanium is in is totally locked down by the manufacturers.
The only time you would consider using them with *Linux derivatives is when they are sold off and go into private hands. And then you would use NetBSD for that.
Itanium based servers are a package of software/hardware; you cannot separate the two. You have a choice: black or black.

Complete nonsense, where do you get this bad info?? Plenty of Itanium hardware was sold with vendor installed and supported Linux distros. I worked for HP during this time, and I can assure you, RHEL was an officially supported OS on Itanium. Same goes for the Itanium servers from Dell, which were sold with your choice of Linux (RHEL or SLES) or Windows.