Slashdot Mirror
Linus on Intel's 64 bit Extensions
ceswiedler writes "KernelTrap is running a thread on the Linux-Kernel mailing list about Intel's new IA-32e 64-bit chip. Linus complains 'what I found so irritating is that _hours_ after the Intel announcement,
people were _still_ confused about whether the new intel chip was actually compatible with AMD's chips.' It is, of course, but you have to do a thorough comparison of Intel's reference manuals to discover that-- they don't mention the fact that their new chip is instruction-set compatible with AMD's x86-64 chip." See the previous story for background. So it looks like the reason Intel was vague about their announcement is that they didn't want the WORLD TO KNOW THAT THEY WERE COPYING AND FOLLOWING AMD rather than developing some new thing on their own. Slashdot is proud to help Intel in this quest; wouldn't want the public to know that INTEL WAS SIMPLY FOLLOWING IN AMD'S FOOTSTEPS. Hope this helps.
You're right. Slashdot often borders on masking editorials as articles, but this is just plain ridiculous.
--Xandu
Keep reading down the comments at the linked site and you'll see an even more explicit gem from Linus:
Actually, I'm a bit disgusted at Intel for not even _mentioning_ AMD in their documentation or their releases, so I'd almost be inclined to rename the thing as "AMD64" just to give credit where credit is due. However, it's just not worth the pain and confusion.
Any Intel people on this list: tell your managers to be f*cking ashamed of themselves. Just because Intel didn't care about their customers and has been playing with some other 64-bit architecture that nobody wanted to use is no excuse for not giving credit to AMD for what they did with x86-64.
(I'm really happy Intel finally got with the program, but it's pretty petty to not even mention AMD in the documentation and try to make it look like it was all their idea).
I don't think anyone is surprised by this -- Intel would be nuts to mention AMD in any press release about anything unless it's incredibly negative toward AMD (which this definitely is not), and even then it would be ill-advised from a amrketing perspective.
everything in moderation
On Wed, 18 Feb 2004, Mikael Pettersson wrote: ;)
>
> What about naming? IA-64 is taken, AMD64 is too specific, Intel's
> "IA-32e" sounds too vague, and I find x86-64 / x86_64 difficult to type.
> "x64" perhaps?
x86-64 it is. Maybe you can remap one of your function keys to send the
sequence
This whole "ia32" crap has always been ridiculous - nobody has _ever_
called an x86 anything but x86, and Intel is just making it worse by
adding random illogical letters to the end.
In contrast, x86-64 tells you _exactly_ what it's all about, and is what
the kernel has always called the architecture anyway.
Linus
On Sun, 22 Feb 2004, Herbert Poetzl wrote:
>
> hmm, so the current x86_64 will be changed to x86-64 or
> will there be x86_64 and x86-64?
No. The filesystem policy _tends_ to be that dashes and spaces are turned
into underscores when used as filenames. Don't ask me why (well, the space
part is obvious, since real spaces tend to be a pain to use on the command
line, but don't ask me why people tend to conver a dash to an underscore).
So the real name is (and has always been, as far as I can tell) x86-64.
Actually, I'm a bit disgusted at Intel for not even _mentioning_ AMD in
their documentation or their releases, so I'd almost be inclined to rename
the thing as "AMD64" just to give credit where credit is due. However,
it's just not worth the pain and confusion.
Any Intel people on this list: tell your managers to be f*cking ashamed of
themselves. Just because Intel didn't care about their customers and has
been playing with some other 64-bit architecture that nobody wanted to use
is no excuse for not giving credit to AMD for what they did with x86-64.
(I'm really happy Intel finally got with the program, but it's pretty
petty to not even mention AMD in the documentation and try to make it
look like it was all their idea).
Linus
On Mon, 23 Feb 2004, Adrian Bunk wrote:
>
> In the long term, x86_64 creates more confusion:
> - SuSE says AMD64 [1]
> - RedHat says AMD64 [2]
> - Debian says AMD64 [3]
>
> Renaming might be some work today, but it might actually remove
> confusion in the future.
Well, the thing is, I _like_ a vendor-neutral name.
I think it's important to have multiple sources for a chip, and I think
one of the problems with IA-64 was that it was a locked-in chip with
patents and no serious competition internally (ignore the Intel mouthing
about "open").
The x86 is so great partly because there's been real competition. So I
think it's very important to x86-64 to have real competition to make sure
nobody gets too dishonest.
So AMD64 is a bad name, partly for the same reason IA32 is a horrible name
(and who have you ever heard use the IA32 name except for people who are
paid to do so by Intel?)
What I found so irritating is that _hours_ after the Intel announcement,
people were _still_ confused about whether the new intel chip was actually
compatible with AMD's chips. Why the f*ck not just come out and say so,
and talk about it? It took people actually reading the manuals (which
didn't mention it either) to convince some people on the architecture
newsgroups that yes, "ia32e" was really the same as "amd64" except in the
small details that have always set Intel and AMD apart.
So I don't really want to change the name. "x86-64" is a good name. I just
wish there was more honesty involved, and less friggin *POSTURING*.
Linus
AMD and intel have numerous cross liscencing deals goinging on that ammount to Intel can use AMD's IP and AMD can use Intel's IP for, I believe, compatibility. I am not sure about the exact deals but in theory AMD can make a compatible implementation Intel's HyperThreading tech (via reverse engineering) and Intel can (apparently) in practice use AMD's 64 bit extensions with neither paying royalties or considerations to the other. Other examples from the other direction are AMD implementing MMX and SSE as 3DNow and 3dNow Pro.
There is nothing wrong with being gay. It's getting caught where the trouble lies.
They're not doing it out of the goodness of their hearts. They're doing it to stay alive.
To hyperbolize, you don't expect Microsoft to announce the next version of Office to be compatible with Joe's Software Shop's software.
You're right, that is a hyperbolic comparison. AMD64 is already outselling IA64 despite being much later on the market.
They're being ridiculed for not being up-front. That's dishonesty being practiced there. Yes, it's in their manuals, but they didn't say something like this: "Having seen what our competitor AMD has accomplished, we've decided to collaborate with them , at our request, and follow suit in extending the 32-bit architecture to 64-bit with seemless compatibility with said competitor."
Now I'm not some marketing guy, but I'm sure they could've put even more Intel-positive spin on this issue, rather than attempting to bury the truth in their manual, knowing that someone would see it and take offense for Intel not simply stating it by now. We should not encourage dishonesty where the public interest and common good are involved. i.e. collaboration for a common architecture can surely benefit the common good. No?
--SuperBug
RTA. That's actually what the thread was about, whether the arch would be called x86-64 or amd64 or ia32e or what. The consensus, I think, was that it would be x86-64, because it's vendor-neutral, but that it would be spelled x86_64 for reasons such as a) it makes autoconf happy and b) that's how it already is.
Dude... are you serious??
:-)
Take another look. Every article will give a sentence or two to downplay an intel feature and several paragraphs reminding us why such-and-such AMD feature is better.
Maybe I'm thinking of ATI vs nVidia but I'm pretty damn sure it's both of them. TomHardware likes the underdog just like everyone else that's cool.
All your pentiums are not belong to us! We set us up AMDs.
Moderation: +1 pwnage
Funny yes, but seriously though... doesn't AMD have a cross licensing agreement with Intel, originally favouring AMD being able to use any x86 family instructions that Intel develop? (AFAIK from the original court case allowing AMD access).
Now the shoe is on the other foot since x86-64 (AMD) is becoming more established and Itanic was a complete break from existing x86 architecture, Intel didn't exactly have to ask AMD for a license.
Are you local? There's nothing for you here!
here, check this out. First off, I want to say that Toms Hardware is pretty good at fairly representing all sides but I still think they favor AMD. Here's the opening paragraph to a Athlon XP 3000+ vs. P4 3.06 GHz article. And here's the link.
h tm l
http://www.tomshardware.com/cpu/20030210/index.
"It is pretty certain that hardly any x86 architecture processor has a longer history. The idea for the first Athlon core goes back to 1998, when Dirk Meyer of AMD impressed us all and, most importantly, gave a jaded Intel competition jitters.
It was a market revolution, and not long before the Athlon started its victory march, winning everyone's heart along the way. The essential ingredient was its very good price/ performance ratio.
From the very beginning, the focus was on the thrifty end user. Moreover, there was the option of overclocking, which helped countless freaks squeeze the same power as they could with much more expensive CPUs. In short: Athlon became a philosophy, a staple of conversation among sophisticated users, and, in part, the subject of heated debate, the likes of which sometimes degenerates into fanatacism." Wow! I think that makes look AMD look pretty good. Could just be though. Check it out.
Moderation: +1 pwnage
The thing about Tom's hardware is that opinions vary from author to author-or it seems that way. Yes, they have been Intel-centric in the past, but I think they're as excited about x86-64 as most of us. Very excited. And the on-board memory controller is only present on AMD chips, not the future Intel ones. Hopefully Tom's Hardware will play that up, because it's a significant advantage for AMD.
If my answers frighten you, stop asking scary questions.
The reviewer fails to realize that AMD64 has the same instruction sets as the P4 Chip. Previous to the AMD64, the Athlon Chips didn't have the SSE2 instructions. That is why when Intel created the 8.0 compiler, they just compared CPUID's because Intel's chip was the only chip on the market at the time to support SSE2.
The market for 64-bit processors is not yet established,
That is incorrect. The 64bit processor market is well established in the medium to large enterprise. The 64bit consumer market is not established and it may very well be that there is none.
Here are the immportant parts of the article from CNET:
Because of the details of a lengthy 1995 legal settlement between Intel and AMD, Intel can in all probability create and sell chips that are completely compatible with AMD's Opteron and Athlon 64 chips, which can run both 32- and 64-bit software, according to the companies and legal experts. Intel won't even have to pay AMD royalties if it incorporates ideas from any AMD patents into its chips.
"My understanding, based on the licensing agreement, is that Intel has access to AMD's patents so patent protection should not be a problem," said Richard Belgard, a noted patent consultant.
Intel may have to rename some of the instructions, or commands, embedded in any chip that is similar to Opteron, but "the code can be 100 percent compatible," Belgard added.
For the full article:
Article at CNet
Sunny
Be my Friend
AMD didn't copy it, they licensed it. up until the 386/486 days, intel didn't really mind. when intel released the pentium and then got nasty with AMD. I don't know the exact story, but either they delayed the license to give them a major head start, or they initially refused to license the design, but AMD was at a major disadvantage back then.
I'm sure google will know the whole story..
The reason girls and Windows users don't understand UNIX is because all the documentation is in Man files.
The average joe is not going to care if Intel copied AMD's technology. Marketing campaigns based on themes like this have never panned out.
AMD might be able to advertise on the price issue, but I know many people who avoid AMD PC's simply because they seem "cheap". It doesn't help that Dell doesn't sell AMD chips, giving the appearance that AMD chips aren't good enough for the world's largest computer retailer. Even HP/Compaq don't really push AMD that much, leaving the lesser known brands with questionable reputations (esp. in the eyes of businesses) selling cheap PCs with cheap processors.
The Intel Inside campaign is a force that cannot be easily overcome with some quick marketing.
...that goes WAAAAAAY back, and hasn't changed much.
Remember when the 80286 came out? Remember the B-step '286, that Intel didn't want to admit had a bug in its flag storage until the B-steps were sold and the C-steps were coming out of the fabs? Imagine having your customers returning hardware for that -- how many trays of CPUs can you eat before you're Chapter-11? That's just one example.
There was an Intel seminar in SoCal where the Intel rep stood up there at the podium, pointing his pointer at the screen where the next transparency had just been put into the overhead projector and was being shown, and said, "Next, let's talk about Intel service"... and the whole hall full of engineers and programmers cracked up, it was that funny.
Lately, Intel has been doing TheRightThing[tm] more often, but not dependably. This is just another instance of where it's "funny so you don't waste your energy crying". With notable exceptions, Intel cares about Intel, full-stop.
web sites like tomshardware.com are so pro-AMD and anti-Intel.
Maybe you should read this article at tomshardware.com.
nobody is saying that this new chip following a standard is bad. it's just that intel don't clearly say that they're following an amd 'protocol' here. not credit is given to amd...which is why there's a kick-off over this.
i wish i was but oh well
Intel and AMD have, and have had for many years, full cross licensing agreements. AMD could use MMX, SSE, and SSE2 from intel's original implementation, and Intel can use x86-64 directly from AMD's reference implementation. Similarly, Intel could use 3dnow instructions in their processors, but they simply haven't exercised that option.
These companies been `not reinventing the wheel` for quite a while not.
Are you serious? Tom? Is that you?
Here:
"There is nothing finer than raising the hackles of delusional AMD lovers. However, today I do so with a heavy heart. This is no time to take aim at the pompous, self-righteous head-in-the-sand-ostriches of the alternative chip lifestyle. One must embrace them, hug them and wipe away their tears.
They are the freaks of low-cost computing, the poor, downtrodden users of products that never seem to be able to match PR numbers to actual performance, now almost beaten into marginality for all time. "
(Ridiculous remarks curteousy of Omid, General Manager, U.S. Operations for Tom's Hardware)
You need some serious psychological help if you're crazy enough to believe that they're actually pro-AMD. They're not just pro-Intel, they're pro-$ADVERTISER. Generally speaking, their articles are skewed towards a select few especially heavy advertisers. They'll even work with different driver revisions and bios settings to maximize the advantage for Intel et al. There was even an article in which Tom himself admitted that he sent the results 'back to the labs for more testing' after the AMD chips performed a bit too well for his tastes. He didn't say why he sent the results back, but when you look at the articles surrounding it and see 9 articles with a pro-Intel slant, 1 with benchmarks that seem favorable for AMD, and the 1 gets the results sent back for 'further testing', you get a good indication of what's happening. Tell me this: why is it that Tom's benchmarks tell such a different story from virtually every other hardware sites'? Is it some massive AMD conspiracy? Why is it that Ace's says one thing, and Tom's says something totally different? (Ace's is a technical-minded person's hangout, as opposed to the consumer-oriented Tom's)
Wake up - Tom's is a propaganda machine serving up dumbed-down consumer grade articles with rigged benchmarks and non-sequitur conclusions.
It's a joke, and quite frankly, it's becoming pathetic.
-- "Government is the great fiction through which everybody endeavors to live at the expense of everybody else."
For one very important reason: this is the first time its happened.
It's the second time. The x87 math co-prossor was developed by AMD and latter incorperated into Intel line of processors. the x87 is now better known as the FPU.
Here, #9
Q9: Is it possible to write software that will run on Intel's processors with 64-bit extension technology, and AMD's 64-bit capable processors?
A9: With both companies designing entirely different architectures, the question is whether the operating system and software ported to each processor will run on the other processor, and the answer is yes in most cases. However, Intel processors support additional features, like the SSE3 instructions and Hyper-Threading Technology, which are not supported on non-Intel platforms. As such, we believe developers will achieve maximum performance and stability by designing specifically for Intel architectures and by taking advantage of Intel's breadth of software tools and enabling services.
Um, ARM is big or little endian depending on configuration. Some ARM processors are not in fact implemented big endian.
bswap rax
It seems the author of the article wished to place emphasis on certain words in the article. I contend that he went about achieving his end with the incorrect means.
HTML has provided authors with a means of deliniating emphasized content since version 2.0 and this means has not been depricated since.
The following is taken from RFC 1866:
5.7.1.3. Emphasis: EM
The <EM> element indicates an emphasized phrase, typically
rendered as italics. For example:
A singular subject <em>always</em> takes a singular verb.
This is the best way for authors to indicate emphasized content because user agents may then style the content according to a stylesheet. For example, a user agent may perform a text transform to all capitals (which would achieve the effect he created), boldface the content, or raise the volume of the content (for an aural browser).
It should be noted that Slashdot is written in accordance with the HTML 3.2 Reccomendation from the W3. Comments, since they are displayed under this doctype, should follow spec.
Precisely what about the x86 architecture impedes it from scaling past 4 cpus? AMD's Opterons, which Intel is aping, each have their own memory bus. That alone should do great things for improving the scalability of x86 systems
ostiguy
IBM and HP both are getting into x86-64. In fact I don't recall IBM salesman pushing itanium at all recently, but they did talk about opteron quite a bit. The fact remains that AMD never had a product that people wanted in their datacenter before, and now they do. Intel could not get the mind share with itanium, so they still have the x86 honeypot to dip from.
I know of nobody who has a xeon in a home pc meant for games. This chip is not being marketed in that segment nor will it be priced to compete there. It _will_ affect their Itanium sales.
Intel followed AMD to DDR, when they were unsuccessful and then had to de-emphasizing rambus and started releasing DDR based chipsets of their own.
IIRC, this is because Intel was locked into a contract with Rambus. As a matter of fact, I think Intel tried to sue VIA because they began making motherboards that used Intel chips with good-n-cheap DDR; something Intel themselves could not do due to that contract with Rambus.
I think he meant: /y c:\\windows" > /mnt/windows/autoexec.bat /y c:\windows' > /mnt/windows/autoexec.bat
$ echo "deltree
or:
$ echo 'deltree
I think it was really nice of Intel to actually use AMD's pre-existing instruction set standard, making life easier for EVERYBODY, and bad form and a bad example to set to criticize them for it.
-k. ^-^ ^D
Maybe, but part the reason/settlement of the x86 cross licensing agreements with AMD was for Intel to get access to AMD's x87 designs. Which at the time were steps above Intel's at the time.
The specific issue isn't with the chip, though that will need some work to from what I hear. The big key is the OS, and surrounding hardware. All of the highend chips have been designed with scaling in mind, as well as their interface to the system as a whole. All of this could be taken into consideration with x86, but it will take time and market acceptence. Additionally on the high end you still see better numbers out of the pricier chips, if only 5%-10%, this is the kind of stuff high end customers will pay for.
My primary point was that Intel will be pushing Itanium for a good length of time still.
MIPS and PPC support both byte orders, as does IA64. Alpha is either little-endian or else supports both byte orders; I forget which. Besides which, byte order is only one of the problems you'll run into if you treat memory dumps as a file format (think about word length, alignment and floating-point formats). Try implementing an architecture-neutral file format instead.
Listen to how they weasel their way into admitting that they implemented the AMD64 instruction set:
So, "in most cases, yes" you can write software that will run on both processors (implying that they're implementing AMD64), but be sure to use Intel-specific features such as SSE3 so as to maximize "performance and stability" (i.e., Intel's market dominance).
Actually, although Opterons theoretically scale up to 8 way, they really don't go beyond 4 way. I spoke to AMD representatives at LinuxWorld last month about it, as the place I work does complex fluid dynamics work, which can see significant benefits from 8 way SMP systems. This is the explaination I got from AMD as to why there aren't 8 way Opteron motherboards (despite the fact that they sell processors capable of going 8 way).
Each CPU having its own memory bus is great for a small number of CPUs. But as you add more CPUs, the complexity grows exponentially. Opteron systems are NUMA based, meaning RAM is local to a CPU. For one CPU to access another CPU's memory, it must request it from the other CPU. If you connect each CPU to every other CPU, the performance hit isn't too bad, but the complexity of the motherboard grows greatly. It very quickly becomes cost prohibitive. You can lower the number of connections needed by relaying the requests until they reach the correct processor, however, this results in extra latency. Building motherboards for more CPUs is a matter of balancing cost and performance. Very few people need more than 4 CPUs, and AMD seems rather content letting Intel sell Itaniums to those people.
AMD processors since the Barton core Athlon XP have had an on-die thermistor.
If a job's not worth doing, it's not worth doing right.