GlobalFoundries to Acquire MIPS
mips.com
mips.com
This could be interesting to see how much they try to loss-lead to get market share in the low-end
In terms of process, it's ancient.
5G wasn't really that necessary, although I know it's critical as a broadband replacement for many people.
I would miss TikTok, of course.
Must be the most moronic decision ever.
and it's not like 20/20 hindsight either, because every hardware enthusiast knew at the time Intel was having troubles and was worried TSMC (and Samsung at the time) were going to be the only fabs producing leading edge lithographies.
These nm values are really bullshit anyway, but the tech node that was supposed to be Intel’s 7nm, which ended up being called “Intel 4” (because they branded some 10nm tech as Intel 7), only came out in like 2023. Given they Global Foundries was always behind Intel, suddenly leapfrogging them by 2-3 years would be quite a feat.
> These nm values are really bullshit anyway, but the tech node that was supposed to be Intel’s 7nm, which ended up being called “Intel 4” (because they branded some 10nm tech as Intel 7), only came out in like 2023. Given they Global Foundries was always behind Intel, suddenly leapfrogging them by 2-3 years would be quite a feat.
This is a very weak argument. Intel was ahead of everyone, now everyone is ahead of Intel. Remember TSMC's blunder processes like 20nm? How they turned around after that? Or how GloFo has had always mediocre processes but they finally hit the nail in the head with their 14/12nm? Fab business has always had companies leapfrogging each other, it turns out the worst sin is not trying. GloFo's greedy investors chose to bury the business in the ground for their short term profits.
First, nobody knew if even TSMC was going to succeed at bringing a 7nm process to market. 02018 was maybe the height of the "Moore's Law is over" belief. There was a lot of debate about whether planar semiconductor scaling had finally reached the limit of practical feasibility, although clearly it was still two orders of magnitude from the single-atom physical limit, which had been reached by Xie's lab in 02002. Like Intel, SMIC didn't reach 7nm until 02023 (with the HiSilicon processor for Huawei's Mate60 cellphone) despite having the full backing of the world's most technically productive country, and when they did, it was a shocking surprise in international relations with the US.
Second, even if GF had brought 7nm to market, there was no guarantee it would be profitable. The most profitable companies in a market are not always the most technically advanced; often the pioneers die with arrows in their backs. If you can make 7nm chips in volume, but the price for them is so high that almost everyone sticks with 12nm processes (maybe from your competitors), you can still lose money on the R&D. Moore's Law as originally stated in "Cramming" was about how the minimum price per transistor kept moving to smaller and smaller transistors, and historically that has been an immensely strong impetus to move to smaller processes, but it's clearly weakened in recent years, with many successful semiconductor products like high-end FPGAs still shipping on very old process nodes. (Leaving aside analog, which is a huge market that doesn't benefit from smaller feature size.)
Third, we don't know what the situation inside GF was, and maybe GF's CEO did. Maybe they'd just lost all their most important talent to TSMC or Samsung, so their 7nm project was doomed. Maybe their management politics were internally dysfunctional in a way that blocked progress on 7nm, even if it hadn't been canceled. There's no guarantee that GF would have been successful at mass production of 7nm chips even in a technical sense, no matter how much money they spent on it.
In the end it seems like GF lost the bet pretty badly. But that doesn't necessarily imply that it was the wrong bet. Just, probably.
They had previously signed a contract with IBM to produce silicon at these more advanced nodes that they could not honor, and there was legal action between them.
https://www.anandtech.com/show/13277/globalfoundries-stops-a...
https://newsroom.ibm.com/2025-01-02-GlobalFoundries-and-IBM-...
In any case, at the time and still I think GF was probably correct in that they would not be able to compete at the leading edge and make money at it. Remember, AMD and IBM separated fabs out for a reason and not having the scale necessary to compete was probably a big part of that. AMD has succeeded on TSMC and IBM seems to be doing ok on Samsung. Most chips are not at the leading edge and don't need to be, and so most fabs don't need to be leading edge to serve customers. There are all kinds of applications where a more mature and better characterized process is better, whether for harsh environments, mixed signal applications, or just low volume parts where $20M of tooling cost is not worth it.
Do you have any evidence, besides GF's own PR/IR department, that the process ever actually worked in volume? Because from my point of view, how they ended things looks exactly how I would spin away a multibillion-dollar investment into a failed process.
Since it didn't happen, the only thing we know is what they said and they said it was because of "strategic shift"
> Tom Caulfield also mentioned GF needed $3 billion dollars of additional capital to get to 12,000 wpm and they could only fund half of it through cash flow, they would have to borrow the other half and the projected return wasn’t good.
> When Tom took over as CEO he went out on the road and visited GF’s customers. What he found was a lack of commitment to GF’s 7 nm process in the customer base. Many customers were never going to go to 7 nm and of the customers who were, GF wouldn’t have enough capacity to meet their demands. There was also concern in the customer base that 7 nm would take up all the R&D and capital budgets and starve the other processes they wanted to use of investment.
(https://semiwiki.com/wikis/company-wikis/globalfoundries-wik...)
If 7LP worked, given this market and its hunger for capacity, it'd be in production at at least small scale. Equipment costs are down and knowledge has disseminated, making it a lot cheaper to launch, especially as "7nm" isn't the leading edge any more.
I don't think it works.
Duh. Of course it doesn't work, because they cancelled it in 2018!
> making it a lot cheaper to launch, especially as "7nm" isn't the leading edge any more.
Same logic cuts both ways. If they didn't think it was financially viable in 2018 when it'd be a leading edge process and their customers would be willing to pay top dollar for it, why would they think it'd be feasible now when it isn't the leading edge lithography and nobody would be paying top dollar for it?
On top of that I doubt even your claim that it'd be cheaper to do the investment now would hold given how everything got more expensive since 2018. I'm also doubtful that machines got cheaper since ASML is still the only ones building them and they've probably got their hands full with their existing customers. They'd probably laugh at GloFo if they'd come with a request like that "Sorry GloFo, we're already booked until 2030 building machines for TSMC, Intel and Samsung maybe try at 2032" :P
GloFo got off the train and there's no going back.
The reason one would expect 7LP to be cheaper to launch now is that their competitors have got equivalent processes into production that can be learned from, or even "learned from" (ripped off). Equipment suppliers have debugged their offerings and pruned their lines to what's useful. In short, someone else has derisked it and found what works. That is a major advantage. In other industries, one company doing the derisking can launch an entire industry (see, Apple, iPhone) if moats are low. Moats are very, very, very high in the foundry space, so there are not many companies that could copy TSMC 7FF or Intel 7 even if they wanted to. GlobalFoundries could do it. But they choose not to. If they were on the cusp of a node introduction, they'd love to see their competition swoop in and solve the last problems for them. Sure, it makes them late to market, but at a vastly lower total spend to enter the market (one with tremendous moats and limited competition!). They could probably profit off that.
But they don't want to. So, either, leading-edge process nodes are uneconomical (in which case, good riddance, leave the market), or they don't actually have significant R&D effort completed and are still billions of dollars in R&D opex away from having anything viable. In which case... nothing of real value was lost.
So, yeah, it sucks that we lost a competitor. But I don't think we lost GF on the leading edge because they didn't like the color of paint on the new ion-implanter's frame. I think we lost them because they didn't have a product and they knew it. In which case there is nothing to mourn.
Name company making chips with EUV that is not TSMC, Samsung, or Intel?
I don't fault them for failing to predict the chip shortage and huge opportunity to acquire customers that would result. The fact remains: they will eventually fade away.
My guess is that the guys in Abu Dhabi did not want to do the investments needed to bring 7nm into production. They lost a huge opportunity because of that. At the time, it probably looked like the right financial decision to them, even though practically everyone affected downstream thought it was myopic.
Pursuing 7nm would have likely bankrupted GloFo.
https://www.eetimes.com/samsung-globalfoundries-prep-14nm-pr...
"Samsung expects to be in production late this year with a 14 nm FinFET process it has developed. GlobalFoundries has licensed the process and will have it in production early next year."
GlobalFoundries licensed 14nm from Samsung. How do you know GlobalFoundries is capable of 7nm?
kragen thinks making most of his readers glitch for a second every time they read one of his dates is worth it on order to advertise for the Long Now. Really unfortunate choice, since he often has decent information to share.
However OCTAL (leading zero) prefixing of a text mode number fails on a number of points:
* It's still a fixed register size (5 characters), which will overflow on the year 100000 AD.
* It's confusing, everyone else.
* It's not technically correct. (human behavior)
Truncating to two year digits was confusing because ambiguity. There is no ambiguity if a number encoded in decimal uses precisely the number of characters it needs. That's how normal humans normally write numbers.
The burden of proof is on you to support your claim that they could have executed a 7nm process profitably, as opposed to them looking at the data and coming to a rational conclusion that they couldn't.
Why the fuck I'd have to prove that given that GloFo themselves claimed that they pulled out of it because it'd be unprofitable? Some people in this subthread are very eager to put words into my mouth.
What?
The chip shortage was a shortage of cheap but inferior 28nm, 40nm, 65nm and 80nm chips that GlobalFoundries was (and still is) well positioned to profit from.
Keep in mind that your iphone only has very few chips in <10nm technology. The rest is using much larger groundrules, even the memory.
The automobile industry showed us that there is demand for older nodes.
Glo-flo is leading edge for anyone without EUV.
https://en.wikipedia.org/wiki/Delay_slot
I'm surprised by how many other architectures use it.
Stanford MIPS was extremely influential, which was undoubtedly a major factor in many RISC architectures copying the delay-slot feature, including SPARC, the PA-RISC, and the i860. But the delay slot really only simplifies a particular narrow range of microarchitectures, those with almost exactly the same pipeline structure as the original. If you want to lengthen the pipeline, either you have to add the interlocks back in, or you have to add extra delay slots, breaking binary compatibility. So delay slots fell out of favor fairly quickly in the 80s. Maybe they were never a good tradeoff.
One of the main things pushing people to RISC in the 80s was virtual memory, specifically, the necessity of being able to restart a faulted instruction after a page fault. (See Mashey's masterful explanation of why this doomed the VAX in https://yarchive.net/comp/vax.html.) RISC architectures generally didn't have multiple memory accesses or multiple writes per instruction (ARM being a notable exception), so all the information you needed to restart the failed instruction successfully was in the saved program counter.
But delay slots pose a problem here! Suppose the faulting instruction is the delay-slot instruction following a branch. The next instruction to execute after resuming that one could either be the instruction that was branched to, or the instruction at the address after the delay-slot instruction, depending on whether the branch was taken or not. That means you need to either take the fault before the branch, or the fault handler needs to save at least the branch-taken bit. I've never programmed a page-fault handler for MIPS, the SPARC, PA-RISC, or the i860, so I don't know how they handle this, but it seems like it implies extra implementation complexity of precisely the kind Hennessy was trying to weasel out of.
The WP page also mentions that MIPS had load delay slots, where the datum you loaded wasn't available in the very next instruction. I'm reminded that the Tera MTA actually had a variable number of load delay slots, specified in a field in the load instruction, to allow the compiler to allow as many instructions as it could for the memory reference to come back from RAM over the packet-switching network. (The CPU would then stall your thread if the load took longer than the allotted number of instructions, but the idea was that a compiler that prefetched enough stuff into your thread's huge register set could make such stalls very rare.)
One more thing about branch delay slots: It seems original SuperH went for very minimal solution. It prevents interrupts being taken between branch and delay slot, and not much else. PC-relative accesses are relative to the branch target, and faults are also reported with branch target address. As far I can see this makes faults in branch delay slots unrecoverable. In SH-3 they patched that by reporting faults in delay slots for taken branches with branch address itself, so things can be fixed up in the fault handler.
As for SH2, ouch! So SH2 got pretty badly screwed by delay slots, eh?
As for SuperH I don't think they cared too much. Primary use of handling faults is memory paging, and MMU was added only in SH-3, so that's probably the reason they also fixed delay slot fault recovery. Before that faults were either illegal opcodes or alignment violations, probably the answer for that was "don't do that".
I didn't remember that the SH2 didn't support virtual memory (perhaps because I've never used SuperH). That makes sense, then.
I think that, for the ways people most commonly use CPUs, it's acceptable if the value you read from a register in a load delay slot is nondeterministic, for example depending on whether you resumed from a page fault or not, or whether you had a cache miss or not. It could really impede debugging if it happened in practice, and it could impede reverse-engineering of malware, but I believe that such things are actually relatively common. (IIRC you could detect the difference between an 8086 and an 8088 by modifying the next instruction in the program, which would have been already loaded by the 8086 but not the 8088. But I'm guessing that under a single-stepping debugger the 8086 would act like an 8088 in this case.) The solution would probably be "Stop lifting your arm like that if it hurts;" it's easy enough to not emit the offending instruction sequences from your compiler in this case.
The case where people really worry about nondeterminism is where it exposes information in a security-violating way, as in Spectre, which isn't even nondeterminism at the register-contents level, just the timing level.
Myself, I have a strong preference for strongly deterministic CPU semantics, and I've been working on a portable strongly deterministic (but not for timing) virtual machine for archival purposes. But clearly strong determinism isn't required for a usable CPU.
Apparently so. Maybe the logic is that it is available one instruction later if it's a hit, but when it's a miss it's stalls entire pipeline anyway, and resumes only when result is available.
One source of non-determinism that stayed for long time in various architectures were LL/SC linked atomics. It mostly didn't matter but eg. rr recording debugger on AArch64 doesn't work on applications using these instead of newer CAS extension atomics.
WRT LL/SC, I don't think it's dead yet—isn't RISC-V's A extension using a sort of LL/SC? rr is indeed exactly the kind of collateral damage that I deplore. rr is amazing.
Switch(mipsarch): Case 1: Nop.
Case 2: Noop.
Case 10: Noooooooooop.
https://www.jwhitham.org/2016/02/risc-instruction-sets-i-hav...
Itanium was the really odd one — it not only used register windows but could offload some of the prior windows onto the heap. Most people would probably never notice… unless you’re trying to get a conservative scanning GC working and are stumped why values in some registers seem to not be traced…
'Anyway this chip architect guy is standing up in front of this group promising the moon and stars. And I finally put my hand up and said I just could not see how you're proposing to get to those kind of performance levels. And he said well we've got a simulation, and I thought Ah, ok. That shut me up for a little bit, but then something occurred to me and I interrupted him again. I said, wait I am sorry to derail this meeting. But how would you use a simulator if you don't have a compiler? He said, well that's true we don't have a compiler yet, so I hand assembled my simulations. I asked "How did you do thousands of line of code that way?" He said “No, I did 30 lines of code”. Flabbergasted, I said, "You're predicting the entire future of this architecture on 30 lines of hand generated code?" [chuckle], I said it just like that, I did not mean to be insulting but I was just thunderstruck. Andy Grove piped up and said "we are not here right now to reconsider the future of this effort, so let’s move on".'
And a lesser known thing that I couldn't find much information on is that Sun also worked on VLIW chip during the 90s. Apparently Bill Joy was convinced that VLIW was the future so they did a VLIW chip, and the project was lead by David Ditzel. As far as I am aware this was never released. If any Sun veterans have any idea about this, I would love to know.
https://www.oracle.com/servers/technologies/opensparc-t1-pag...
The T2 is also open, and places an FPU in each core.
https://www.oracle.com/servers/technologies/opensparc-t2-pag...
When there is such complaint about closed firmware in the Raspberry Pi, and the risk of the Intel ME and other closed CPU features, I wonder why these open designs are ignored. Yes, the performance and power consumption would be poor by modern standards.
By now there are much better more modern design out-there and for RISC-V.
Like if you tried to debug a software crash on Itanium. The customer provided core dump was useless as you could not see what was going on. Intel added a debug mode to their compilers which disabled all that EPIC so hopefully you could reproduce the crash there, or on other CPU architectures. Otherwise you were basically screwed.
That HP-Intel arrangement was weird. One time, an Intel-badged employee came out to change a tape drive on a (Compaq->HP->HPE) Compaq SSL2020 tape robot. Okay, I guess they shared employees. ¯\_(ツ)_/¯
At least one toolchain would just pad the slots with nops
Is this the very beginning of a market consolidation?
For most places that kind of high-cost work doesn't make much sense when their product isn't "a CPU", and they also typically have to buy other IP anyway like memory controllers or I/O blocks -- so buying a CPU core isn't that strange in the grand scheme.
MIPS has also hitched it's horse to RISC-V now, and I am seeing a critical mass of talent and capital forming in that space.
AFAIK MIPS still hasn't shipped a high-end processor competitive with the XuanTie 910 that article is about. And I think the billions of RISC-V microcontroller cores that have shipped already (10 billion as of 02022 according to https://wccftech.com/x86-arm-rival-risc-v-architecture-ships...) are also mostly not from MIPS.
At least the Long Now Foundation stuff comes with that context built-in.
http://textfiles.com/occult/MOOISM/
I have some retro-reading to do.
https://longnow.org/ideas/long-now-years-five-digit-dates-an...
I don't believe I'm actually doing those people any injury, so while they're obviously free to continue requesting different formatting of my posts, I'm free to ignore them.
I think it's important for people to be able to complain about things that bother them, for the reasons described in https://news.ycombinator.com/item?id=44501817. In that thread, we were discussing a different commenter requesting that an author please not use AI for editing his own books, although the request was made in a particularly obnoxious fashion. Consider "Please don't play your music so loud at night", "Please don't look at my sister", or "Please don't throw your trash out your car window". But "please format your dates differently" doesn't seem like a very important request, even if it were phrased politely, to the point that it makes me (and, as I've seen, others) think less of the people who are making it.
If my date formatting really bothers them, they're free to stop reading the site. After having looked at their comment histories, I wish some of them would, because the only thing they ever post are similarly vacuous complaints. If people had to choose between reading a site where I posted and they didn't, and a site where they posted and I didn't, 100% of people would choose the former. (Others do occasionally post something worthwhile, but nothing that inspires me to wonder how I could earn their admiration.)
So, these days, I can easily ignore it.
Especially the day after this comment of mine got voted up to +151: https://news.ycombinator.com/item?id=44491713
This was stronger before you edited in the longer paragraphs telling us that you look through their comment histories.
Look, you do you, but I'd rather hear your passionate promotion of the leading zero approach since you apparently know it irritates people.
Edit: PS Isn't it interesting that the comment that you are justly proud of doesn't have any dates in it? :)
Generally the comments I'm proud of tend to have more dates rather than less dates, because often they either provide verifiable information about things that have happened in the past at specific times, or cite specific sources, including the publication date. I think comments like that are more valuable than comments that simply take a position or state a complaint.
Consider, for example, this overview of the historical development of delay slots in RISC architectures https://news.ycombinator.com/item?id=44503609, or this discussion of what GlobalFoundries' prospects for developing a 7nm node really were when they abandoned it https://news.ycombinator.com/item?id=44503245, or this summary of the causes for the antibacterial properties of honey and the evidence for its astounding shelf-stability https://news.ycombinator.com/item?id=44495658, or this brief history of the ideology of pre-Satoshi cryptocurrencies https://news.ycombinator.com/item?id=44471683.
These are all within the last week. I'm pleased with how each of those comments came out, and, although the last one went relatively unnoticed (probably because it was deep in a contentious thread), all of the others were very well received.
However, presumably not by the sort of people who are more interested in how other people format their comments than they are in biochemistry, archaeology, or computer architecture. Which is fine with me. Different people have different interests, and some people evidently have a very strong interest in how other people format their dates, and don't have much to contribute on topics like computer architecture. Possibly this isn't the right site for them, and if so, they'll figure that out sooner or later, and go criticize other people's comment formatting elsewhere.
Sooner, I hope, because I'm really not interested in their opinions.
Frankly, something about that leading 0 makes me grit my teeth and stop reading. I can't explain why it affects me like that. Perhaps I'm the only one who does, although threads like this seem to pop up whenever they post so I don't think so. If HN had a mute button, I'd probably use it just because it annoys me to that level.
Edit: And now that we're talking about it, they seem to have the need to mention a specific year way more than most, as though deliberately looking for opportunities to draw attention to themselves. Oof. That just made it about 10x more grating to me.
I think the C910 looks better on paper than it performs in practice. I hope that isn't the case for MIPS.
Do you have any examples? Apple Silicon cores took pieces of the pwerficient cores, and everything else I know of either tweaked an official ARM design or started more or less from scratch.
I don't know the details of how much rework they had to do.
As I understand it cavium still sells processors but only directly to hyperscalers.
Ironically Cavium may have gone through the same process with their previous design, but given that they then acquired this one from broadcom, perhaps it didn't go very well! I have no concrete information though.
That is a frustrating pattern in the RISC-V world. Many companies that boast having x wide cores with y SPECint numbers but nothing that has been independently verified.
The last high end MIPS was in the SGI times, 30 years ago.
https://www.theregister.com/2025/05/06/loongson_inspur_cloud...
Lots of companies had their own mips implementation, but still might use an implementation from mips-the-company because even if you have your own team, you probably don't want to implement every core size that you might need. But then for some reason lots of them switched to using ARM, within a few years (in some cases getting an architecture licence and keeping their CPU team).
It seems like RV has a more stable structure, as the foundation doesn't licence cores, so even if one or two of the implementors die it won't necessarily reflect on the viability of the ecosystem
You want to burn your initial capital on lawyers? This is MIPS we're talking about.