Slashdot Mirror
Linux Kernel Patch Hints At At 32-Core Support For AMD Zen Chips
New submitter Iamthecheese points to an article which says that a patch published on the Linux Kernel Mailing List indicates that AMD's forthcoming Zen processors will have as many as 32 cores per socket, but notes that while the article's headline says "Confirms," "the article text doesn't bear that out." Still, he writes,
There are hints of such from last year. A leaked patch for the 14 nanometer AMD Zeppelin (Family 17h, Model 00h) reveals support for up to 32 cores. Another blog says pretty much the same thing. We recently discussed an announced 4+8 core AMD chip, but nothing like this.
https://xkcd.com/619/
AMD best hope this CPU has some actual guts to it for performance / power efficiency. They haven't had a great CPU since the Duron / Thunderbird days when they were (arguably) the leader on the desktop.
Their CPU's have gotten progressively worse compared to intel, to a point where it's pretty much complete folly to go AMD at this point, which is a big shame.
Let's hope they close the gap significantly, very significantly. They've almost always been behind, even if it's only slightly (yet had to hugely undercut prices)
At their current rate, I do not see them lasting a hell of a lot longer. So this one better be the one for a couple of years.
I hear that rm -rf has some success with that. I'm not sure though as I don't care much for the argument.
This. The number of cores the system can use is even configurable.
Zen is (according to AMD, so I guess it is optimistic) is supposed to bring 40% improvement in instructions per clock. That would put it around Sandy Bridge level. They would have to pull off 100% improvement to be competitive at high levels once again.
I predicted that I will see a kilo-core machine in my lifetime.
Do GPU shader cores count?
Linux already supports 32 cores very well, (Or 1024...) Still, driver support is needed for the details of adressing (up to) 32 cores on that particular architecture. Hence the patch. AMD may or may not turn out a 32-core chip in the near future - but at least their architecture supports that number of cores. Which is a bit interesting. If it turns out too hard to make, expect chips with 12-20 cores and gradually more as production quality ramps up.
CMOS technology, is static meaning that there is no current flow through a gate when it is on or off. Current only flows while the transistor is transitioning states. P = I x V and as I (current) increases, so does power. All 'digital' circuits are actually analog and you can show that I is proportional to frequency squared. Instead of having a power (^) increase in energy use, you have a linear relation.
It appears to be dead, but a very interesting technology.
Look, MikeeUSA:
I don't like systemd. I even go to some lengths to have a Debian running without. But every time I see your dirty drivel, I feel the urge of embracing systemd [1] [2] [3], just to avoid being associated with you.
One could get the impression that you are a pro-systemd false-flag operation except... I think you're just an idiot. One way or the other, I won't give up my stance just because of you.
[1] https://twitter.com/zacchiro/s...
[2] http://etbe.coker.com.au/2015/...
[3] https://identi.ca/cwebber/note...
If the previous max cores per socket was 16, and the value in the kernel needs to be a power of 2, then at most this tells us that they have a 17 (more likely 18 or 20) core CPU on the way.
https://www.phoronix.com/scan.php?page=news_item&px=Intel-KNL-Perf-In-Linux-4.5/
AMD long ago gave up competing on raw CPU performance with Intel. They compete on price and integration. They have better on-board GPUs than Intel, and they cost less. The XBOne and PS4 both use AMD CPUs and GPUs.
The question is if these markets are enough.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
AMD's dual-core, partially shared, but partially independent has been a confusing thing. Better than hypethreading, but worse than real cores, claiming performance of real cores.
Note for all those desktop enthusiasts out there, don't get too excited. To look at Intel as an example, they go up to 4 cores per desktop socket, but go to 18 cores per socket in servers (at 150W per socket) as of this moment (can't talk about unreleased product). AMD does 8 'core' desktop processors (4 modules) and 16 'core' opteron (really 8 modules), so it's not just an Intel thing.
XML is like violence. If it doesn't solve the problem, use more.
The question is if these markets are enough.
Yes.
For people who simply don't care about specs the AMD will come in at the right price point. For people who do care about specs but are honest about what they're using their systems for it'll also likely be a no brainer.
Aside from a few real power users the only people who care about maximum power are lunkheads who want to make teh 1337 b0xen for no other reason than bragging rights. No one else really pays for the high end processors.
While that's true, AMD hasn't had a strong product launch since 2012. You can't just let the market go for 4 years and expect to be in business. They sold high end chips to complete with mid range intel and soon their chips only competed with core i3 CPUs. Without a major catch up effort, they're out of the x86 business.
R&D matters.
From what I've read about AMD's Zen architecture, they've dispensed with the "two single threaded cores per module" architecture and now have SMT allowing two threads in each core according to this, much like "hyper threading" on Intel chips.
If that's the case, and we can expect a 32 core chip to execute 64 threads, then that's an awful lot of threads to keep supplied with data and instructions. In comparison, the biggest Intel Xeon I know about, the E5-2699 v3 has 18 cores, 36 threads, 45MB of last level cache, and 4 memory channels (68GB/sec to RAM). Intel sticks pretty close to that 1.25MB cache per core in their big Xeons. So if you adhered to Intel's apparent rules, a 32 core 64 thread chip would need 80MB of LLC and maybe 6 memory channels. Anandtech estimates 5.7 billion transistors for the big Xeon. Scaling the Intel design from 18 to 32 cores would require over 10 billion transistors! That number leads me to believe that an SMT 32 core 64 thread chip built with 2016 technology would not be practical.
What might be practical is a chip with some "heavy" cores optimized for balls-to-the-wall floating point execution, and other "lighter" cores for lower power integer tasks. This has been done in "octocore" mobile phone chips and called a big LITTLE architecture. The idea is that the OS and various decoding and checksumming tasks can stay resident on the low power light cores, while the heavy cores do things like game physics and photo noise reduction. Because the multiprocessing is not symmetric, the OS kernel needs special rules to assign tasks to cores. Which leads me to wonder if AMD has something like big LITTLE up its sleeve for 32 core Zens.
--- Often in error; never in doubt!
Perhaps cores-schmores is one way to approach this? Lots of small cores with relatively slow clocks, as higher clocks tend to worsen power efficiency.
Which is also the road that Intel themselves pursue with Xeon Phi (the currently used descendant of their failed GPU).
I'm not discounting Intel's success with single-core performance per se, but I sometimes feel it's aimed at speeding up legacy applications
Yup, the drawback is that not a lot of current application are able to run on tons of separate threads.
Not only "legacy" but even applications recently produced or currently being produced.
But the architecture can have some success on servers, and some scientific workloads.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Yes. Especially those. You can get a decent CPU for everyday use, _and_ graphics that leaves Intels integrated offerings and quite a few discrete chips in the dust on a pretty damned decent power budget.
Not saying you can't get better performance with a discrete cpu / gpu combo, but not in the same monetary/power budget space.
This discussion *really* took a weird turn. How the fuck do you get from config.sys to pedos?
That and there is a good chance with that many cores there is a NUMA architecture... at some level.
Those were *my* config.sys and autoexec.bat files, you insensitive clod!
This is like looking at a hardware register in a generic register layout that leaves 8 bits for "core index" and deducing that the manufacturer must be intending on delivering a 256-core CPU.
Then you find the documentation for the specific family and find out that bits 7-3 are "reserved and will be read as zero".
But the driver patch they submitted doesn't make that assumption "just in case".
Because it's easier to plan ahead in the driver than it is to actually deliver a 256-core CPU.
"Still, driver support is needed for the details of adressing (up to) 32 cores on that particular architecture."
The details of x86-64 have been well known for more than a decade. The architecture hasn't changed.
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
"While that's true, AMD hasn't had a strong product launch since 2012."
AMD hasn't NEEDED a strong product launch, considering the lack-luster performance of the mobile i5 versus an old as dirt Athlon II.
Still waiting on Serviscope_minor to wake up to fucking reality and realize that Jessica Price isn't going to fuck him.
The architecture hasn't changed.
Unless AMD pulls a fast one by presenting a 32-core processor with eight general processors and 20 graphic processors.
Why does the article call it "leaked patch"? That seems like a normal public patch to Linux Kernel Mailing List.
Also when I read the source code, I do not see anything suggesting 32 cores, and instead the patch adds support for an "instructions retired" register which is introduced in the Zeppelin architecture.
So is the article rubbish or am I rubbish? Once again I get the feeling that by even just slightly scraping the surface, things turn out to be completely different than what is described. :D
That's 12 general processors and 20 graphic processors. My skinny vanilla latte haven't kicked in today for me to do math this early in the morning.
No, the prevailing view is that the translation of the Hebrew should be closer to "If a man find a lady who is a virgin, who is not pledged to be married, and lay hold on her, and lies with her, and they be found" rather than "and rapes her" (It's two words in Hebrew, and different words, like those for "seized" or "forced" would more likely be used to describe rape).
Context also suggests that the intent was for a case where the woman was willing or seduced.
Also, Exodus 22:16-17 says almost the same thing, including the same punishment of requiring the paying of a bride-price and forcing the man to marry the woman, but uses a word translated as "entices" or "seduces" rather than "lays hold of".
Which is why I almost always buy AMD - Even if current Intel technology can beat it, I see it as an investment in the future. Keeps Intel on their toes having competition. If AMD dies, then Intel innovation slows WAY down.
That would be awesome if Intel comes out with a 64 core chip just before AMD releases Zen, competition in action. If I truly needed 64 cores, then I would buy Intel, if not, then I will buy a cheaper 32 core AMD. Works perfectly for me.
You've always got to think long term, life as it is now is guaranteed to change - do what you can to make sure it changes for the better.
A mobile i5 from 2012, with half as many cores, running at half the clock speed, probably at half the voltage of the Athlon.
You're benching Intel's 17 watt netbook CPU from 2012 versus AMD's 95 watt underdesk spaceheater from 2010, and think that's a good thing for AMD when they nearly tie?
0 1 - just my two bits
True. It seems that AMD learned their lesson when they held the performance crown - while they might be able to outcompete Intel on raw performance and affordability, they'll never be able to afford to compete against Intel's dirty tricks and anticompetive behavior. So instead they target the mainstream and console market, and let those who are actually interested in the price:performance ratio to come to them.
--- Most topics have many sides worth arguing, allow me to take one opposite you.
I hope it's enough. As is, it's almost impossible to find a decent AMD laptop. Your choices are HP, HP, HP, HP, sometimes Lenovo or Asus. Yet on Lenovo's own damned website they don't even offer|list the mid-to-higher-end machines you can buy with AMD. Lenovo lists 5 (3 models) and they are all low-end junk.
well, amd is almost dead and intel's innovation has slowed down to a crawl. apart from new instructions, there is almost no reason to upgrade a processor bought in the last 4-5 years. you simply don't gain much.
for me, the magic trigger words are "hardware accelerated hevc/vp9 encoding". if a cpu that allows for real time 4k hevc/vp9 encoding comes out, i'll switch immediately. otherwise, i'll stay with my current cpu for a good few more years.
Is Intel planning any ARM chips? I doubt they've done that since they sold off XScale to Marvel
x4 640 is a mobile ultra low power 17W cpu?
Who logs in to gdm? Not I, said the duck.
> So you want to go back to shell scripts? A system in the style of your father's CONFIG.SYS and AUTOEXEC.BAT is what you want?
That presupposes that a DOS batch file is anything like a Unix shell script. All you've really done is demonstrate how utterly clueless you are about either of the things you're whining about.
People who have no clue, should be in no position to force anyone else to "abandon the past". They simply aren't qualified to judge. This is the fundemental problem with the SystemD crowd. They are idiots distracted by shiny objects.
A Pirate and a Puritan look the same on a balance sheet.
is power in your country free? fx-83xx needs roughly twice as much of it as comparably priced (and 15-20% less powerful) i5-46xx.
Welcome to Slashdot.
CMOS technology, is static meaning that there is no current flow through a gate when it is on or off. Current only flows while the transistor is transitioning states.
That's the idea but it has never quite worked that way. There is always a small current flow from vdd to ground even when the gate is "off". At smaller geometries, this leakage becomes not just significant but can be the majority of the power drain. Thus, having lots of cores ready but not active does not help. They still suck power. Finfets help a great deal but only for a while and at 14nm and below the problem is coming back. The work around is to actually turn off the power to inactive regions. This works but shutting down and restarting units is complicated and time consuming, making it more difficult to respond to transient demands.
Intel's upcoming chips top out at 44 threads:
SKU Name Cores/Threads Base Clock Boost Clock L3 Cache (LLC) TDP
Intel Xeon E5-2699 V4 22/44 2.2 GHz ~3.6 GHz 55 MB 145 W
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
You're right. Notebook, not netbook.
Still it's a long way from benchmarking AMD's enthusiast desktop chip versus Intel's enthusiast desktop chip.
0 1 - just my two bits
That particular patch only tells us, that AMD might support up to 32C/64T per socket for the Zen family. 2P systems could theoretically support 64C/128T then. Of course this depends on the core size and power consumption.
At the cache level, usually. I haven't checked, but I would hope that the Linux scheduler is smart enough to know that it's cheaper to migrate a thread between cores on the same die than between dies because the cache is likely to be warmer.
sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
The problem is that the hardware encoders all produce shitty fucking results compared to the software encoders. Yes, they're much faster, but so what? The only purpose for fast encoding is for real time streams. 99% of the people interested in that are gamers, and both AMD and nVidia offer live encoding shit with EZ-PZ integration for the popular streaming services.
As a gamer (who doesn't stream), I'd use my real GPU. If I was worried about losing a small fraction of gaming performance, I'd split my outputs to a separate box entirely.
As someone who rips and encodes his blurays, I simply throw it at x264 with the ultra slow preset and my desired bitrate and let it run. It finishes when it finishes.
We might see an actual 32 core chip. It will be an Opteron that is optimized for running large numbers of servers in a single box. The clock rate of the cores will be relatively low for thermal reasons and the chip will have no graphics processors.
I try to buy AMD for that reason too - I dread to see a market without them. My guest desktop at home is running an AMD FX 8120 (with a GTX 580, SSD, and 16GB RAM) and it's good enough for when people use it. Other times it just runs BOINC.
I'm starting to think GNU is the problem with "GNU/Linux" these days.
They haven't done this kind of chicanery with the APUs. It'd get them so much bad press that it's simply not worth it.