Linux Patch Pending To Fix Support For The Transmeta Crusoe CPU

[M.Bahr]

Hurling insults around does exactly zero to prove any points (whether material or ad hominems).

BTW, I tend to know what I'm talking about. In fact, part of my research work has influenced the AMD Ryzen 3000 series.

Please refrain from doing such claims. It does not shed a good light on a company when people learn what kind of flawed logic their engineers got.

[vladpetric]

I am not making stuff up but you are. It also becomes very clear, that you got logical fallacies in your reasoning. And you are ignoring important facts and conditions. Transmeta was a small company in comparison to intel. Their chips were quite expensive. In fact it was extremely difficult to compete against a big experienced, old and foremost rich corporation like intel. intel could generate much bigger margins with their chips.

Just because the Efficeon couldn't meet expectations doesn't mean the basic technology behind it was bad or would have already met it's limits genius. What kind of logic is this? That's like pointing to the world's first car and claiming :"Well look how slow this nonsense is. The horse is much faster. Let's forget about combustion engines. They got no future." This is the exact summary of your nonsensical statement.

Ok, so now you're kinda-but-not-quite backtracking on your claim that Intel somehow choked Transmeta in court with their x86 patents (clearly, they didn't; they merely countersued when Transmeta sued them, and that was way after their last processor release).

Also, since you're so keen on abusing logic as well: The one making a claim has the responsibility of backing such claim with evidence.

You take that Transmeta's designs were great as a given, when it's absolutely not so. In fact, there's no real-life evidence that their revolutionary approach to instruction execution (translating to VLIW) was great. All you quote is regurgitated hype, that's it. Guess what, there was a lot of hype in the past for many shitty designs, and some bought into it. Now hype in itself is not intrinsically bad, as most good designs were hyped as well, but it proves exactly nothing.

And your comparison is just that - a comparison. BTW, the main problem with horses is that in a big city they produced tons of horse shit (for real). Motorbikes and cars immediately addressed that, even if the original designs weren't fast at all (so yeah, the first motorbikes and cars slowly started addressing a significant pollution problem, as strange as that may sound today)

[vladpetric]

Please refrain from doing such claims. It does not shed a good light on a company when people learn what kind of flawed logic their engineers got.

Right, so because you hate me, my logic is bad?

You know who summarily dismisses stuff like you do? Good propagandists, like our prev president.

[M.Bahr]

Ok, so now you're kinda-but-not-quite backtracking on your claim that Intel somehow choked Transmeta in court with their x86 patents (clearly, they didn't; they merely countersued when Transmeta sued them, and that was way after their last processor release).

Also, since you're so keen on abusing logic as well: The one making a claim has the responsibility of backing such claim with evidence.

You take that Transmeta's designs were great as a given, when it's absolutely not so. In fact, there's no real-life evidence that their revolutionary approach to instruction execution (translating to VLIW) was great. All you quote is regurgitated hype, that's it. Guess what, there was a lot of hype in the past for many shitty designs, and some bought into it. Now hype in itself is not intrinsically bad, as most good designs were hyped as well, but it proves exactly nothing.

And your comparison is just that - a comparison. BTW, the main problem with horses is that in a big city they produced tons of horse shit (for real). Motorbikes and cars immediately addressed that, even if the original designs weren't fast at all (so yeah, the first motorbikes and cars slowly started addressing a significant pollution problem, as strange as that may sound today)

I am not backtracking anything but you are showing once again how confused you are about the events and what they actually mean. Transmeta suing intel was legit. Intel suing transmeta for infringing x86 patents though is controversial to say the least because transmeta hasn't stolen anything. They emulated x86 with their own revolutionary technology. Later court decision's like the google vs oracle one exposed those kind of patent infringements as baseless and nonsensical.

You want evidence? Learn how to use search engines. I am not going to play this childish game, There are really enough articles about this topic.

Just Hype? Look who is talking. X86-64 chips are all about hype. It is an horrible, antique, dangerous and completely overhyped ISA. It needs to be shooten to death, a mercy killing to end it's suffering. You as an allegedly ryzen 3000 engineer should know best how overcomplex x86-64 chips have become. But yet here you are defending x86 microcoded hardware in amd and intel chips as if they were la creme de la creme. Against your claims VLIW-architectures got a rich history and are still widely used in modern processors.

You somehow didn't get my comparison with horses and twisted it. I am the one voting for combustion engines while you are the one sticking to horses because you underestimate combustion engines. You made logical fallacies once again. Just because something doesn't exist in real life doesn't mean it could not be build or improved with further development. That's like pointing to the world's first air plane and claiming:"Well that's a good plane. But it can't get any better than that." If real engineers followed this principle, the most modern airplane would never have been created. Do you get it now? How much easier examples do you need to understand?

[vladpetric]

I am not backtracking anything but you are showing once again how confused you are about the events and what they actually mean. Transmeta suing intel was legit. Intel suing transmeta for infringing x86 patents though is controversial to say at the least because transmeta hasn't stolen anything. They emulated x86 with their own revolutionary technology. Later court decision's like the google vs oracle one exposed those kind of patent infringements as baseless and nonsensical.

You want evidence? Learn how to use search engines. I am not going to play this childish game, There are really enough articles about this topic.

Just Hype? Look who is talking. X86-64 chips are all about hype. It is an horrible, antique, dangerous and completely overhyped ISA. It needs to be shooten to death, a mercy killing to end it's suffering. You as an allegedly ryzen 3000 engineer should know best how overcomplex x86-64 chips have become. But yet here you are defending x86 microcoded hardware in amd and intel chips as if they were la creme de la creme. Against your claims VLIW-architectures got a rich history and are still widely used in modern processors. And ironically intel's intanium itself was partially based on a VLIW-architecture. By the way this is not the reason why the itantium failed. So don't even start this claim.

You somehow didn't get my comparison with horses and twisted it. I am the one voting for combustion engines while you are the one sticking to horses because you underestimate combustion engines. You made logical fallacies once again. Just because something doesn't exist in real life doesn't mean it could not be build or improved with further development. That's like pointing to the world's first air plane and claiming:"Well that's a good plane. But it can't get any better than that." If real engineers followed this principle, the most modern airplane would never have been created. Do you get it now? How much easier examples do you need to understand?

It really isn't that difficult to post links. Claiming that I should Google your arguments is absolutely laughable (again, burden of proof is on the person making the claim, twisting it around is both fallacious and, IMO, abusive). I don't think you actually have anything whatsoever on the choking in courts claim (as it never happened)

I am not a Ryzen 3000 engineer, I never actually said that (never worked for AMD). Maybe you should read again what I did say.

And I don't like x86 (I don't think anybody does), but the instruction set matters a lot less these days. What matters >10x (really) is the microarchitecture. When you have processors that can issue in the order of 10-15 instructions (or micro-ops) per cycle, keeping those execution units occupied is an insanely difficult task; most of that has very little to do with the instruction set actually. Also, there are actually things that x86-64 does well, namely a more compact instruction encoding than RISC instruction sets (for modern high performance RISC, all instructions take 32 bits, and on average the instruction density is actually lower vs x86-64).

Well, when you call for pitchforks, it's pretty clear where your hate lies.

Your knowledge of computer architecture is both spotty (as the saying goes, a little knowledge is a dangerous thing) and limited to the 90s btw. Things evolved significantly in the meanwhile. I would recommend a few academic books on the topic, but the likelihood of you trying to learn something is infinitesimal.

[rene]

To be fair the people who regularly use and test their precious Transmeta Crusoe collectable, likeI my world record smallest Oqo 01+ https://www.youtube.com/watch?v=s22Rk1xW9ME are just using a generic i586 kernel for everything, as building a kernel with Crusoe optimizations yield next to no performance difference, so running one build an all and your Sony Vaio P's Intel Atom is just way more comfortable and one build to test them all: https://www.youtube.com/watch?v=suCj4eulTJg

[M.Bahr]

It really isn't that difficult to post links. Claiming that I should Google your arguments is absolutely laughable (again, burden of proof is on the person making the claim, twisting it around is both fallacious and, IMO, abusive). I don't think you actually have anything whatsoever on the choking in courts claim (as it never happened)

I am not a Ryzen 3000 engineer, I never actually said that (never worked for AMD). Maybe you should read again what I did say.

And I don't like x86 (I don't think anybody does), but the instruction set matters a lot less these days. What matters >10x (really) is the microarchitecture. When you have processors that can issue in the order of 10-15 instructions (or micro-ops) per cycle, keeping those execution units occupied is an insanely difficult task; most of that has very little to do with the instruction set actually. Also, there are actually things that x86-64 does well, namely a more compact instruction encoding than RISC instruction sets (for modern high performance RISC, all instructions take 32 bits, and on average the instruction density is actually lower vs x86-64).

Well, when you call for pitchforks, it's pretty clear where your hate lies.

Your knowledge of computer architecture is both spotty (as the saying goes, a little knowledge is a dangerous thing) and limited to the 90s btw. Things evolved significantly in the meanwhile. I would recommend a few academic books on the topic, but the likelihood of you trying to learn something is infinitesimal.

Don't worry. I will never collect any links for you because your whole statements already exposed you as a dishonest person. Therefore even 1000 pieces of evidence will not convince you because you ego's own truth seems to be your only standard. Frankly said it is you who is constantly trying to attack me on a personal level. I don't feel anything like hate or whatever you are accusing me of. This is so low key seriously.

So suddenly you are backtracking from the impression you have invoked about yourself? You wrote the following before, i quote: " BTW, I tend to know what I'm talking about. In fact, part of my research work has influenced the AMD Ryzen 3000 series."

It was your statement that invoked assumptions about your qualification. And now that it becomes more and more obvious that you don't have any clue what you are talking about, you suddenly distance yourself from being knowledgeable. I got the impression from the very first moment that you are only guessing. And the latest when you talked against real and genuine experts who praise VLIW and it's potential it became very evident that you are just a show-off.

Unlike you a real expert would at least mention the pros and cons of VLIW-architecture and CISC cpus. But instead you are categorically against VLIW-architectures and this is again a proof that you are no real expert. Do you actually realize what you do? You are declaring all the VLIW engineers around the world as idiots but yourself as someone who knows better. How arrogant is that? You could at least mention the undeniable advantages of VLiW-architectures. This would be at least honest and serious, because every architecture has pros and cons. You could at least mention the role of VLIW in GPUS, how it is again more relevant today, because some disadvantages of VLIW are less effective today due to the increasing transistor count, how VLIW-architectures are used in todays chips like the Elbrus. At some point even intel themselves used VLIW in itanium together with HP. It was not a pure VLIW design but a mix. But nevertheless VLIW has undeniable advantages. You ignored all those facts again

I can't believe that you are still trying to defend the zombie x86-64. This is not a real ISA to me, sorry. Maybe it started as something sane and honest with the very first x86 design. But todays x86-64 chips carry a big and heavy load of legacy ballast around their necks. The only reason why x86-64 is being kept alive is to maintain the duopoly of amd and intel as patent trolls. If amd and intel wanted they could transition easily to RISC-V and emulate x86-64 for compatibility reasons in support of lagacy software. As a matter of fact intel plans to merley emulate the 32bit part of x86-64 in future. And amd even offers a RISC-V softcore in one of their server products namely their MicroBlaze V Processor.
In that regard it is quite amusing how you are trying to defend x86-64 vs RISC. Do you actually know as a so called expert and influencer of the Ryzen 3000 series that modern x86-64 are not native CISC cpus anymore? Intel uses a RISC core in their x86-64 cpus since the pentium pro. This is the inventor of x86 indirectly admitting how bad CISC is. And the newest stunt that amd and intel are trying to pull off are vector extentions. They know very well that x86-64 can absolutely not compete against the increasing threat of powerful GPUs which dominate the Super Computer segment for many years already. A real expert and x86-64 chipdesigner would have known that todays x86-64 chips are based on a RISC core.

But keep trying to defame me as somebody with little knowledge while presenting yourself as the big expert who allegedly influenced the ryzen 3000 development. This is so pathetic man. The more you write the more you are embarrassing yourself.

Edit: typo

[vladpetric]

Don't worry. I will never collect any links for you because your whole statements already exposed you as a dishonest person. Therefore even 1000 pieces of evidence will not convince you because you ego's own truth seems to be your only standard. Frankly said it is you who is constantly trying to attack me on a personal level. I don't feel anything like hate or whatever you are accusing me of. This is so low key seriously.

So suddenly you are backtracking from the impression you have invoked about yourself? You wrote the following before, i quote: " BTW, I tend to know what I'm talking about. In fact, part of my research work has influenced the AMD Ryzen 3000 series."

It was your statement that invoked assumptions about your qualification. And now that it becomes more and more obvious that you don't have any clue what you are talking about, you suddenly distance yourself from being knowledgeable. I got the impression from the very first moment that you are only guessing. And the latest when you talked against real and genuine experts who praise VLIW and it's potential it became very evident that you are just a show-off.

Unlike you a real expert would at least mention the pros and cons of VLIW-architecture and CISC cpus. But instead you are categorically against VLIW-architectures and this is again a proof that you are no real expert. Do you actually realize what you do? You are declaring all the VLIW engineers around the world as idiots but yourself as someone who knows better. How arrogant is that? You could at least mention the undeniable advantages of VLiW-architectures. This would be at least honest and serious, because every architecture has pros and cons. You could at least mention the role of VLIW in GPUS, how it is again more relevant today, because some disadvantages of VLIW are less effective today due to the increasing transistor count, how VLIW-architectures are used in todays chips like the Elbrus. At some point even intel themselves used VLIW in itanium together with HP. It was not a pure VLIW design but a mix. But nevertheless VLIW has undeniable advantages. You ignored all those facts again

I can't believe that you are still trying to defend the zombie x86-64. This is not a real ISA to me, sorry. Maybe it started as something sane and honest with the very first x86 design. But todays x86-64 chips carry a big and heavy load of legacy ballast around their necks. The only reason why x86-64 is being kept alive is to maintain the duopoly of amd and intel as patent trolls. If amd and intel wanted they could transition easily to RISC-V and emulate x86-64 for compatibility reasons in support of lagacy software. As a matter of fact intel plans to merley emulate the 32bit part of x86-64 in future. And amd even offers a RISC-V softcore in one of their server products namely their MicroBlaze V Processor.
In that regard it is quite amusing how you are trying to defend x86-64 vs RISC. Do you actually know as a so called expert and influencer of the Ryzen 3000 series that modern x86-64 are not native CISC cpus anymore? Intel uses a RISC core in their x86-64 cpus since the pentium pro. This is the inventor of x86 indirectly admitting how bad CISC is. And the newest stunt that amd and intel are trying to pull off are vector extentions. They know very well that x86-64 can absolutely not compete against the increasing threat of powerful GPUs which dominate the Super Computer segment for many years already. A real expert and x86-64 chipdesigner would have known that todays x86-64 chips are based on a RISC core.

But keep trying to defame me as somebody with little knowledge while presenting yourself as the big expert who allegedly influenced the ryzen 3000 development. This is so pathetic man. The more you write the more you are embarrassing yourself.

Edit: typo

You know what ... can you tell me your full name?

[muncrief]

Big big bullshit. You just don't want to admit, that you are wrong about the proper estimation of this technology. It was revolutionary at that time.

PS: i am tired and done with this nonsensical discussion. No matter what fact I refer to, some conceited person always seems to have something against it.

I understand M.Bahr.

My technology, which I believe surpassed even Transmeta, was also rejected because others simply could not understand it at the time. And it appears still cannot.

As I posted earlier, I had a much better idea called "Thrust Reconfigurable Processor", which used VLIW in a revolutionary way, allowing everything from custom FPGAs to full custom ICs to be reconfigured in an almost infinite number of ways, in one clock cycle. It also included a Thrust microprocessor that could be configured for any bit width, and executed every instruction in one clock cycle as well. It essentially worked by changing the address of RAM banks that drove registers to configure a versatile set of components. Its predecessor, Mimic, had some funding from Sun Microsystems, but later management couldn't understand it and ended their participation. But I kept working on Mimic and ended up with Thrust, which of course is quite complex. But it also failed because most people with the money to fund it couldn't understand it and I couldn't secure the additional funding I needed to complete its development.

As I said though, for those interested I did the best I could to convert the original FrameMaker 4 documentation to Word format awhile ago (though some graphics were blacked out and other parts omitted), and posted it on Mega. It's a long read, around 68 pages, and with the missing graphics is a bit more difficult to comprehend, but those knowledgeable in digital systems and microprocessor architecture will understand. Here's a link to the last architectural document I had for it, minus the header and table of contents ...

https://mega.nz/file/oThwHILa#4Wn55sxxnII_CckiSfV9SwL-x8Via7qRNZEmKeEzarc

[kiffmet]

M.Bahr and vladpetric : the both of you, just rent a room already and get it over with!

To put some oil into the fire: While the legacy ballast of X86_64 sucks, I do think that the general concept of CISC is still a good idea, because having an internal RISC core while the compiler emits CISC instructions, reduces compiler complexity, which is, as you can all remember, what ultimately killed VLIW - be it Itanum or AMD's Terrascale GPUs… Nvidia likely had an engineering reason aswell as to not use Transmeta's "X-to-VLIW" technology anymore after a certain point, despite "Project Denver" having been a huge success.

Anyhow, as soon as you're hitting a microcoded CISC instruction, you can be sure that the sub-instructions/µOPs or whatever this translates to, hits optimal performance paths on the specific chip it's executed on. Generic X86_64_v# instrs + microcode allows for compiler-independent "optimizations" happening during runtime and keeps code size down. Like why print a whole algorithm when you can just emit a keyword that resolves to a functionally equivalent, product specific macro?

You two available for a 3some btw?

muncrief Mega already deleted your virus…