Open-Source Processors 125
clay pigeon writes "This EE Times article covers the development of open-source processors. No doubt exciting news for hardware hackers and those with a need to know about every last detail of their systems, but how will this effect the hardware industry? Can open-source hardware duplicate the success of the open-source software movement?" I'm not holding my breath. Fabrication facilities are a lot more expensive then a CD-ROM presses (or more accurately, internet connections). But I still hope it happens. It would be an interesting market if everyone worked together on the designs, but built their own chips.
Need Open-source EDA Tools for FPGAs first (Score:1)
If open-source hardware is ever to become as widespread as open-source software it is likely going to be via the use of FPGAs.
But just as gcc helped spur open-source software, we need open-source tools for hardware design (tools for Electronic Design Automation (EDA)).
This includes schematic capture tools, hardware description language (HDL) parsers, HDL simulators, and circuit synthesis tools.
http://www.opencollector.org [opencollector.org] has a list of some open-source EDA software under development.
Once there is an infrastructure of open-source EDA tools, this will spur on the development of open-source hardware building blocks or cores (see http://www.opencores.org [opencores.org]).
Looking forward to the day open-source hardware will have as vibrant a community as open-source software...
Re:Micro Fab Facillities! (Score:1)
Re:I'm disappointed (Score:1)
My current project is an example of that. I'm working on a cockpit for a model airplane controller. The model airplane controls are usually just two sticks and some buttons on a standard radio. I'm interfacing full rudder pedals, and throttles to the controller via microcontroller. There is a video feed from the model going into the display unit (looking for a cheap fresnel for it). Stage 2 is to feed telemetry back over the audio channel, and use a small FPGA to do video overlay and drive the instruments. If I have room, I may ditch the microcontroller and move it all to the FPGA after it's all working.
The point is... (Score:1)
The point is to learn and have fun.
Hardware designs for FPGAs can be shared with other FPGA designers and enthusiasts. There is no requirement for mass production.
Following your reasoning, there would be no point in Linus Torvalds writing Linux. If he had anything to contribute to OS design, he should have gotten a job at Microsoft or Sun or Apple or wherever. (Well, arguably he didn't contribute anything new to OS design... but why would you discourage him from trying to write Linux in the first place??)
WinBoards? (Score:1)
Only partly true. I'm guessing you haven't seen the "Winboard". (I think that's what they were called...) Much like the famous "winmodem", these are motherboards which have been stripped down to make the hardware cheaper, with the guts of the controllers moved into software. Due to the fact that the drivers are (of course) proprietary, these computers will run only under Windows!
To be exact, I think the CPU's are general purpose, but the chipsets, and perhaps even the memory, do not have standard, open interfaces, and therefore cannot be used under any other "alternative" OS.
Neat stuff, eh?
--
Re:Yeah but, chip making isn't as easy as writing (Score:1)
>Designed chips is very much like programing.
Yeah, if you want gigantic, slow, hot, wasteful chips.
Of course, when it comes to performance and efficiency, designing chips is not very much like programming (I assume, I'm not a chip designer). But that also applies to high performance software like e.g. kernel scheduler or memory management. Then programming isn't very much like programming either.
For the most time you will get away with standard knowledge and not looking beyond the horizon of your high level language. When it comes to performance then you have to know stuff about pipelining and cache lines and help the ignorant compiler by modifying your source to fit the wanted assembler output or directly include a few lines of assembler in the critical parts. So, in that sense, designing chips is like programming.
Fab, my ass! (Reconfigurable Hardware) (Score:1)
There are lots of people here talking how open source won't work for hardware designs since it's so damn a) cost expensive and b) time expensive (does not allow experimenting) to get a chip design through a fab. So what? Reconfigurable hardware is out there and has been for quite some time.
You get a standard programmable logic device from some company like Altera (there are more, this is just the one I remember right now) and then you can program it with any chip design you want (within the complexity boundary set by the chip in question). These can do from 25MHz up to 100MHz, so you won't be able to replace your 1GHz Athlon with it, tough. Still it's usable for a lot of things. Remember, PCI runs at 33MHz or 66MHz. An MP3 decoder can be done with a lot less than 25MHz.
The software (Windows, since the control interface of the PLD is usually proprietary) compiles your Verilog or VHDL code into the required form which can be uploaded into the chip. So hardware can be designed much like software with the same round trip times. And no, you won't have to write an adder using NAND gates, you just say in VHDL "a And once you have a nice working design you can use the same source to get it through a fab.
But often you don't want a fab at all. Fabs are expensive and production has a long latency (something like 4 to 6 weeks). You can just give the PLD a finished ROM and use that as a production system. One MP3 player was only two months on market before being replaced by a successor, so it was shipped with PLDs. Only if you have really large numbers to produce, the product stays long enough on the market or the speed of a PLD isn't enough is a fab really cost efficient. I guess we'll see a lot more PLDs in shipped products in the future.
There are quite a number of CPUs for PLDs available already. Some PLD companies license CPU designs for use with their devices, e.g. I have an Altera APEX device here in front of me (not mine, a bit expensive) which comes with a NIOS softcore. Which is configurable: choose 16 or 32 bit, the number of address bits, how many registers, number of positions the shifter can shift in one cycle, ... depending on how much room you need on the device for your own design (I'm not a hw designer myself (yet), I have to program the NIOS).
Other designs are freely available under open source / free licenses (lots of stuff, not restricted to CPUs), including a SPARC CPU with peripherals by the European Space Agency under LGPL. If your PLD is large enough, you can put a whole computer on that chip (CPU, ROM, RAM, serial port etc.).
And for starters, you can get a complete development board with a PLD and software for about $150. A small CPU (no 32 bit thing with lots of registers) would fit, and it's enough for some hacking to get up to speed.
Some links for open source hardware: OpenCores [opencores.org], Free IP [free-ip.com], Google Web Directory - Computers > Hardware > Open Source [google.com], LEON-1 [estec.esa.nl] (the ESA SPARC core).
Re:Does not compute (Score:1)
The startup costs for me to start making open source software are very low (assuming I have a computer)
Startup costs to start making open source hardware: use a simulator or/and get a board with a programmable logic device (less than two hundred dollars). Low cost, assuming you have a computer (with a parallel or serial port for downloading to hardware).
the cost of failure/screwup is low (an hour or so of my time to compile).
Cost of failure/screwup: time to fix and recompile. Then simulate or download to PLD. Low cost.
What are you going to start with? A chip that beeps out 'hello world' in morse over a pizeo?
Apparantly making CPUs is rather easy so this what most textbooks on hardware design in, say, VHDL give as the first complete example that goes beyond some simple ALU. Go look at the source of an 8 bit processor that is available for download. It's amazingly simple in a high level language like VHDL.
Who will pay for the testing, manufacturing, etc? (Score:1)
Whereas with software it's just electronic bits and bytes being worked with, the expense of working with tangible materials seems to me to be cost prohibitive.
Re:I'm disappointed (Score:1)
So now you've done that. You got a working FPGA. What are you going to do with it? Riiiight...we should all go about building our own audio hardware accelerators.
Well, it might work if you actually make CPU's, where people can just plug it into their motherboards. Even then, it's really kind of pointless. If I have the "code" for a SPARC, and I give it to you, that's still almost worthless.
1) You have to Fabricate this damn chip. Do you have $250,000US kicking around? That's about how much you'll need for a 0.25um mask set (yup, not even cutting edge). Minimum 6-8 weeks to get this done.
2) You want to make changes to a working design? HAHA! Another quarter million to Fab that change. And another 6-8 weeks.
3) Ooops! The new design didn't quite work. Why not? Do you know how to debug? Let's say yes. Do you have access to a high-speed IC tester to check it out? $5million a pop. Or maybe you can rent one. That's about $300/hour.
By this time, you should be thinking "Why don't I just walk over the the local computer shop 10minutes away and plunk down $200 for a new K7?"
I mean, if I think I had something to contribute to CPU design, I might (wild idea), GO GET A JOB at AMD, or Intel or wherever...
There's not too much value in open-sourcing your CPU design. The Verilog or VHDL code may be verified to be 100% correct, but that code doesn't mean that your silicon is 100% correct.
open standard DO matter (Score:1)
You start of with Intellectual Property (ie, the design, documentation etc) of parts of a system. That means buses, CPU's (ie ARM, MIPS...), memory controllers, peripheral IO. You mungle it together, and voila. Because you have to pay for the IP, the chip costs more. ARM Co., Ltd is waging war against copycats of their instruction sets. An open standard would help, in other words.
Btw, licensing costs of blocks of IP for large numbers are about $1 each (order of).
Re:The problem with OS cores .... (Score:1)
--
You don't need factories to design chips (Score:1)
I also don't agree with the complaint about chips being too complicated for a self educated "hacker" to design. You can break chips down into functional units that are easy to understand, and create them using a high-level design language. The interesting part is that some self-educated kid might actually find a better way of doing something than the industry "standard".
jc
Embedded systems. (Score:1)
Re:Last thing I want to do... (Score:1)
I never said anything about changing the CPU...I was replying to a comment that "...the last thing I want to do is recompile my CPU each time there is an upgrade". Recompiling is software, not hardware. And my reply was clearly talking about how Transmeta recompiles their source
I never said it was Open Source hardware. It's not.
Re:It is becoming very possible!!! (Score:1)
Opensource quality standard cell library (Score:1)
Re:Yeah but, chip making isn't as easy as writing (Score:1)
The logic portion is exactly like writing
code, you're just writing in verilog.
And there's plenty of room for self-expression
and style.
List Of open source projects (Score:1)
http://www.ics.uci.edu/~sumitg/CadPages.html#Be
Open source projects are big boon for research groups. The biggest problem that research groups have is that they do not have access to industry standard processors and do no have the resources to design one from scratch. So a whole set of these in public domain will mean a lot to the research community.
Also, to all the software people here, who do not know much about hardware design: open source hardware designs are just as exciting to the hardware community as open source software is to the software community.
To clarify/correct what some posters have said - we do NOT fiddle around with gates and low level stuff like that. That is only done to optimize design (its like writing assembly line code for core kernels). A lot of processor blocks are designed in high level hardware description languages, such as VHDL and Verilog.
Increasingly, people are beginning to use C or variants of C (see http://www.cecs.uci.edu/~specc)
for hardware design and simulation. We are working on a high level synthesis tool that generates hardware starting from C - check it out
at
http://www.cecs.uci.edu/~spark
Sumit
Re:This would help me out! (Score:1)
doh.. how the hell did I hit a 'g' instead of a 'd'? Thats what I get for rushing.. oh well.. no Aggie jokes plz
JOhn
Re:The problem is that Processors are only the sta (Score:1)
Think about it. Instead of using a crate ICs consisting of AND gates, NOR gates, etc.. to implement digital logic just put them in a nice FPGA and you're done. In addition, I have even seen some FPGAs which mount on breadboards.
JOhn
Assumptions on development cycle for CPU's (Score:1)
Having the fab in the debugging and rough optimization cycle is crazy. That's what software is for.
The fab is further down the road than some people think. It's for nailing down the 'real world' results.
I hope this help clarify some things, and perhaps those who know more can elaborate.
--Phil
SPARC is semi-open (Score:1)
It always amuses me when I read some pundit going on about about the Sun "proprietary" architecture. Yeah, right. And Intel is an OpenSource company.
Anyway, there's a good repository of documents [sparc.org] on the SPARC architecture available for download.
Desktop Manufacturing (Score:1)
Re:If you think about it... (Score:1)
--
Re:Last thing I want to do... (Score:1)
Agreed, it would be cool, but it's not open-source hardware.
----------
There's no VB for chip design (Score:1)
In the same sense, anyone can get lock a couple of sheep in a barn and wait for them to do what the nature intended them to do, but to do it more efficiently you have to pretty up the sheep to make it attractive to the ram.
Wait.. what was my point again?
Oh, yes. I think your point is very moot. You don't need much skill to get simple, yet effective, software to work (VB), while you do need a lot of skill to get a chip to work at all. Even if you get past the HDL writing, the synthesis and physical compiler tools are not nearly as easy to use as VB. There is no "Build" button. It takes skill to produce something that can be sent to the fab.
Now, FPGAs usually come with VC-like IDEs, and it's possible to make very simple designs with only basic HDL/digital logic knowledge. The software equivalent in complexity would be a CGI counter, though. Hardly useful for a future open-source hardware community.
----------
Re:Saddam can sell back the PX2s. (Score:1)
US is trying to keep the manufactured CPUs away, though. Because that's the real difference between a high-tech country, and a low-tech one -- the capability to produce these chips. The bulk of the expense is the manufacturing cost.
Open-source is not going to help a poor country much. They could use it to educated people, maybe, but they still wouldn't be able to produce any designs they come up with.
----------
Re:an outsider's perspective (Score:1)
The simulation tools are a little on the pricey side. Students can get cheap, but stripped down, tools from FPGA companies, like Altera, though. Enough to play around, implement and simulate simple designs.
----------
Re:I agree... don't hold your breath. (Score:1)
You can't just "compile" your scematics and test them.
Actually, yes, you can, and that's in fact how it works :). You run a compiler, and then you run a test in a simulator to see if it works.
You have a point regarding the rest, though. The cost of actually producing a chip is prohibitive to enthusiasts.
I imagine very few people on the street could tell you how to create an XOR operation with a simple set of logic gates.
Not to sound nit-picky, but XOR is a simple logic gate :).
----------
Print your CPU. (Score:1)
Re:I'm disappointed (Score:1)
Writing a verilog model of a piece of hardware in an open source style with a community of designers is completely possible. In fact, it would be very easy, but there are issues that have prevented this from actually taking off in any type of widespread fashion.
Getting the model and code together is not the problem. Organizing that code into a CVS structure for the open source community is also not a problem. It works just like software.
ISSUE NUMBER 1
I have worked with FPGAs from both Xilinx and Altera, and I don't think you are considering the drawbacks of these systems.
The main one of these is memory. Most FPGA design kits do not have any type of off-chip storage. (ie, no SRAM on the FPGA board.) This means that you are forced to store any variables and data either
Anyone who has worked with FPGAs knows that you begin to lose a lot of your available resources as soon as you try to implement any kind of RAM in the gate array. This makes it not an attractive option.
The only other option then is to store the data used by the custom hardware in the system memory/disks. This will work, but it will means a hell of a lot of system CPUFPGA transactions. Slow and cumbersome, but workable.
Lets back off of the whole FPGA thing for now, because we don't really need any FPGA to come up with the design for some hardware.
It is possible to build the model only in Verilog (or VHDL). This can be simulated and debugged on a computer with the right tools. (There are free tools available out there for Linux for all of this.)
But... even though we could do this... what then? The project would have to be useful anough and draw enough interest for someone to foot the manufacturing bill. Either that, or it gets sold as a non-efficient FPGA-based design that the user has to compile and upload to the FPGA board.
The performance of this hardware would suck. It would need to built as an ASIC to work optimally.
ISSUE NUMBER 2
Even if someone does foot the bill for manufacturing, the performance would be horrible. The major reason for this is that the project is going to built on some type of silicon. What type? That all depends on who does the manufacturing.
Anyone who works in hardware design knows that many pieces of circuitry have to be tweaked depending on the electrical characteristics of the silicon. This would be a hinderance to an open source hardware project for which the contributers don't know anything about the silicon that will eventually house this device.
ISSUE NUMBER 3
Hardware clocking? Obviously, this thing will need some sort of clock in it to push it along and make it operate. What kind of clock? How fast?
This (again) is something that will depend on:
- silicon characteristics
- system architecture
If you don't know how much time you have to accomplish (for example) a binary comparison of two values, then you don't know how to build this comparator. Do you have enough time to do it all in one clock cycle? Does it need to accomplish the task across four clock cycles? I don't know. Neither will anybody else.SUMMARY
Opensource designs are possible, but not easy, and not likely. There are so many aspects of the design that are directly influenced by the final materials, packaging, and system board, that it becomes too problematic with trying to develop a useful, stable, robust device.
A piece of software is not affected by the type of disk that it is stored on. A piece of hardware is influenced dramatically by the type of silicon it will be stored on.
Opensource hardware will never be.
SirPoopsalot
To send me an email, remove the SPAM's and replace the -at- with @.
The Problem: Variable Cost (Score:1)
How much does it cost to recompile a piece of code after modifying the source? Basically nothing. How much does it cost to fab a new chip after modifying the design? Millions of dollars.
I'm not saying that there aren't advantages to open source hardware (and there are precedents [vhdl.org]). But the advantages aren't as big as for software because not as many people can realistically participate. Let's look at some of the classic stated advantages of open source software:
Of course if you're talking about reconfigurable computing [io.com] then everything changes.
[Just in case you're wondering, I work in the Alpha mciroprocessor group at Compaq [compaq-alpha.com]. Read into that whatever you like.]
Re:Won't work (Score:1)
Re:Yeah but, chip making isn't as easy as writing (Score:1)
module FA(s,cout,a,b,cin);
output s,cout;
input a,b,cin;
NAND2 n1(na1,a,b), n2(na2,xr1, cin), n3(cout,na1,na2);
XOR2 x1(xr1,a,b), x2(s,cin,xr1);
endmodule
This is so far from actual hardware. You would never even use nand and xor gates for adders, you'd design it at transistor level. A mirror adder, for example, uses something like 28 CMOS transistors.
But that is still so far from actual hardware, you have to do layout. Sizing, routing, it's a lot of work. And verilog code has next to no correlation with it. (Quick google search turns up this guys [berkeley.edu] assignment, which is a good example.)
God does not play dice with the universe. Albert Einstein
Maybe.. 3d printing (Score:1)
http://slashdot.org/articles/00/09/28/175207.shtm
This is where I read about it.
Alternative idea.. (Score:1)
wtf? (Score:1)
Re:Micro Fab Facillities! (Score:1)
http://xputers.informatik.uni-kl.de/ [uni-kl.de]
http://xputers.informatik.uni-kl.de/xputer/index_
full custom chips (Score:1)
In hardware design you can't use 'hacking' in the regular sense: you have to do testing (simulation) too and this is where another problem arises: that software uses a lot of experience (both human and in electronics, like test runs, which are even more expensive). Only then, if you are completely sure that your design (in VHDL, Verilog, etc.) is correct, it goes out for production. If the chip doesn't work, you loose a lot of money and time (about 3-4 months). I doubt that anybody would invest that much money (about 250000$ for one ASIC) and then release the result to the Community!
As for Open Source: a year ago I searched the web and found next to nothing (please correct me if I am wrong). As I said above, those software tools have to be very clean and bugfree (betas are next to useless!). So you'd have to use non-free software (which is very non-free, like very expensive) to create a free design. I doubt that the Open Source gurus would welcome that
Oh yes, if you are talking about a CPU: in those projects are about 300 people and to be honest: I don't think that the bazar aproach works here, because: some parts are done in hand layout, testing is very important (and who wants to be testing? I am not talking about compiling, trying some new window manager, etc. I mean testing of waveforms, interpretation of those results, etc. that testing is just a pain in the
I don't know too much about programmable chips, but I assume that the situation will be much better here...
Oh, one more thing: I hope you don't think I tried to flame the Open Source Community or something like that --- I tried to state my experience...
Forgive me for being obvious but... (Score:1)
So maybe if every linux user on the planet contributed $10 we could get enough money and... oh wait... never mind.
Re:Won't work (Score:1)
Bit late, but.... (Score:1)
open source and polymers (Score:1)
Re:Micro Fab Facillities! (Score:1)
The only problem I see with reprogramming is the heat output. What happens when dnetc and the GIMP get into reconfiguration wars?
Re:Give it time. (Score:1)
Show me a coder who really codes for free, and I'll show you a derelict bum without a house, decent clothing, healthcare, and food, let alone a computer.
Re:Open Source Hardware will not change a lot. (Score:1)
The problem with the chip market is not the size, it's trying to break your way in when you have such high initial infrastructure costs. Secondly, people seem to be a bit leary of buying silicon from an untrusted producer.
Therefore, holding copyright on the design of a chip is largely irrelevant, especially when the manufacturers essentially operate as a cartel. Have you seen Motorolla designing an x86 chip recently? Didn't think so. With a few public exceptions the processor manufacturers largely stay away from each other.
Exactly what defines an "x86 processor"? I would contend that the instruction set is mainly responsible for setting one processor apart from another. If you want to do something different and offer programmers something different from x86 (e.g. RISC), you use a different ISA more suited to your objectives, and the result is by definition not an x86. If, however, you're like AMD and want to break into the consumer market, you emulate the market leader and give customers what they're used to, i.e. the x86 ISA. That doesn't mean the core of your part has to look anything like a vanilla x86. In fact, you'll want something better and therefore something different, otherwise why would people bother switching to you?
However, they will be receptive to open source as it would allow them to downgrade their expensive design engineering teams, outsource and up the shares dividends accordingly.
Hmmm. Most people involved in OSS seem to do so because they can see something going back into the community, although I don't mean to generalize across the entire group of OSS developers. I don't see how you could get that with processors, unless the manufacturers start giving away chips for free. Either way, the differences between hardware and software production are too great to make any predictions at this stage.
I think Open Source designed processors will be higher quality, but it won't make any difference to price or competition in the field.
Well, we'll see :). The fact that you will have a lot of ideas about processors just floating around should see a big reduction in the R&D budgets of any startup chip manufacturers, which should ameliorate the problem outlined at the beginning of this post. I have a hard time predicting the viability of open source hardware when the verdict is still out on open source software.
The problem is not processor design (Score:1)
You can implement a processor in a FPGA, of
course, but such an implementation wouldn't
run at speeds above the 33 MHz range.
(I'm actually just doing this, so I know that's a
realistic number.)
In order to achieve higher speed you need custom
chips - for which the plant will charge you
in the $10k range (that's for getting *one*
prototype made). Even that is still the ASIC
range, for full-custom chips you need to pay more.
Something which is in the Intel/AMD league
requires access to specialized plants - such as
Intel and AMD have.
This is were the true challenge in processor
design is, too - not the architecture (cache,
ALU, floating point units...) but the manufacturing
technology. To make chips which can transport
data from a registers through some logic and
into another register in 2 ns!
So if you consider how close you are in making
your own RAM chip as an open source project, that's
about the same challenge as making an open source
processor.
It's a different story if you're looking for
embedded processors which go on a FPGA - but
what's the benefit to the user? Presumably
you want to get something out of it, not just
saving some chip design company a bit of money...
FPGA not a large barrier to entry... (Score:1)
You mention Altera. If you are a student you can get the evaluation board (for 5k to 70k gate FPGAs--maybe larger now because the technology is so rapidly advancing) for very cheap--like $200--and it is still a good deal less even if you are not a student.
Furthermore, you can freely download (as in beer, not in speech--the source is closed) the basic MAX+PLUS and Synopsis software [altera.com] for use with the evaluation board. This is perfect for the hobbyist. In fact, these boards are ideal for implementing simple microprocessor cores plus added custom logic, and certainly beats wire-wrapping a bunch of 74hct-series gates and registers and stuff.
The good thing is that FPGA technology is moving so incredibly fast that eventually many more hobbyists could get involved. Plus, VHDL is supposed to be modular, so even if you can't synthesize an Athlon on your FLEX70K you could test and synthesize modules that perform some of its functions at a slower speed. Then if a bunch of hobbyists combined their modules they could come up with one or more open-source VHDL-based microprocessor designs.
If companies with the resources or money to access fabs like the design, with open source they would have ZERO design costs. This isn't so far fetched--I'd say that the Pentium III and PowerPCs could be the Windows 2000 and Mac OS X of the hardware world, and it could be possible to do a Linux or BSD of the hardware world as well. Sure, fab costs are very high, but for a project of that scale, engineering/design costs also play a gigantic factor. I think it could result in lower cost, more interoperable hardware designs. The only resistance is the closed attitude of chip makers (even their support software is closed source despite being downloadable on the net--wouldn't Altera gain a huge amount of support and boost sales of their FPGAs if they allowed the open-source community to develop software tools for their products?)
Open Source ISA's for Crusoe? (Score:1)
If this idea of opensource Processor designs takes off, wouldn't it be possible to test (atleast the Instruction Set Architecture) on Transmeta's Crusoe (or something similar)? It would seem to me that having an environment where people could design and implement their designs in software for testing would allow people alot more flexibility.
This is asking for incompatibility issues (Score:1)
Re:Micro Fab Facillities! (Score:1)
Effectivly, this would make a CPU-Burner possible, although I think it would be a while before we can make high-end processors this way.
Re:Micro Fab Facillities! (Score:1)
But than again, I only saw an hour long talk on the subject, so I don't have any idea how practicl e these ideas are.
Re:SPARC is semi-open (Score:1)
Re:Verilog (Cish), VHDL (adaish) (Score:1)
I think that you might consider the whole HDL scene similar (though not at all the same) to code "generation" tools. I've seen several tools which generate code, but you could never use what it generates directly. It was either very suboptimal or sub-correct :) or both, and that means that software expertise is necessary to get meaningful results from the process.
Perhaps as the state of the art progresses, these tools will become smart enough to be "trusted" like a compiler, so engineers can focus on other things. Although this is a danger, too. We might just have to focus on good designs, or (*gasp*) sane process methods.
Desktop MEMS (Score:1)
I hate to trot out this cliche again, but it's possible that in 20 years a desktop MEMS construction set could perform low-production-rate personal manufacturing of custom chips from standard blanks.
Re:FP? (Score:1)
Re:The problem with OS cores .... (Score:1)
Re:WinBoards? (Score:2)
Down that path lies madness. On the other hand, the road to hell is paved with melting snowballs.
I build processors in FPGAs, perhaps you can too (Score:2)
Excellent basic synthesis and FPGA implementation tools are now $0-$55, several proto boards are $100-$200, the devices themselves are <$20, and with them you can build 16- and 32-bit processors that run at 33-67 MHz and beyond. Perfect for many embedded systems projects.
Visit our web site, FPGA CPU News [www.fpgacpu.org] for further information. And/or join the mailing list [www.yahoogroups.com/group/fpga-cpu].
See my Circuit Cellar magazine article series, "Building a RISC CPU and a System-on-a-Chip in an FPGA" [www.fpgacpu.org/xsoc/cc.html]. The corresponding free kit (which requires a $100 FPGA proto board from another company) is at [www.fpgacpu.org/xsoc/index.html].
Or see my recent DesignCon'2001 paper on "Building a Simple FPGA-Optimized RISC Processor and System-on-a-Chip" [www.fpgacpu.org/soc-gr0040-001201.pdf]. This latter paper includes the annotated synthesizable Verilog source code of a simple CPU -- less than 200 lines of code.
Learning to design digital systems with a hardware description language, and then progressing to design your own peripherals, and perhaps processors, is not a trivial undertaking, but neither is it anymore an exclusive and unobtainable art, practiced only by high priests in well funded semiconductor companies with in house fabs.
Jan Gray, Gray Research LLC
Re:Success (Score:2)
You are wrong, it will succeed to the same extent as open source software for all the same reasons. Remember, this hardware is just a different kind of software, it's software for silicon. It's cost is asymtoticly approaching zero, just like sofware. The design sofware is now free, and fabbing costs a tiny fraction of what it used to, due to competition. You can load it into fpga's if you want to, and especially, you can simulate it.
--
Possible, but not via inkjet. (Score:2)
It is a very interesting read and even speaks to the possibility of this allowing open source cpu's in the future. Imagine, downloading and printing the latest version of your cpu before upgrading to the n.n kernel it is targeted at
The problem with this technology is that even if you can shrink an inkjet printer head to the point where you're laying down 0.18-micron dots, you'll have one heck of a time keeping the environment clean enough for your fabbed chip to work. This is one of the main reasons why semiconductor fabs are so expensive (though other factors definitely contribute).
If you're printing at anything other than deep submicron linewidth... your chips will be useless. Maximum operating frequency, to a rough approximation, goes up as the inverse square of the linewidth. Make the linewidth 100 times wider, and your chip will run about 10000 times slower. You're not going to be playing quake on an inkjet-printed chip any time soon.
This technology is extremely useful for making things like active-matrix displays and cheap "smart" devices, but that's about it.
Now, the silver lining. You _can_ fab moderately complex chips using Field Programmable Gate Arrays. These have been around for a long time, and are frequently used for prototyping. You'll get something a tenth the complexity and a fifth the speed of a real chip, but that's still enough for many applications (program yourself a sound card with nifty new features, or emulate an older gaming system or computer in hardware). This will also give you a training tool for gearing up for the real thing (true IC design and fabrication).
Fabbing ICs is impractical for individuals, but possible for organizations. It'll cost you about $500k or so once you have a finished, debugged design to produce a few hundred thousand chips (a couple of wafer runs; the first one or two runs will be buggy alpha silicon which you test and revise).
Current FPGA technology. (Score:2)
Sure, easily. Current top-of-the-line FPGAs let you synthesize designs with on the order of a million gates or so.
People have been writing papers about chips with reconfigurable parts for many years. The idea continues to show promise, but the practical problems with actually _using_ something like this are great, and the benefits aren't amazingly spectacular.
You can buy modules containing integrated processor cores and FPGA logic already from Altera and the other big FPGA companies.
Re:I'm disappointed (Score:2)
FPGAs are an alternative, but still not that cheap and they too have their drawbacks. Ever tried routing one so that all the registers get their data on time? Not very fun. Ever tried pushing one past 50MHz without having to pay $1000 a chip? (If my knowledge on the current state of FPGAs is incorrect, somebody please correct me.)
If you have a small design, yes, you can use an FPGA. As a matter of fact, it's a neat idea. Too bad simulators cost so much (that $5000 thing, though Altera and Xylinx tools might be cheaper). If you want high-volume - another poster correctly stated - $250k NRE (non-recurring engineering) charge just to set up, plus a per-unit charge for each ASIC (usually a couple of hundred dollars).
Don't think that it's all that easy. In some ways Open Source could probably help - on a module level. As in, hey, I just wrote a really cool, high speed SDRAM controller. But on a full-chip level I think you'll find that it's a bit more difficult.
Re:Last thing I want to do... (Score:2)
--
Re:no need (Score:2)
The *real* hardware industry (pick a few uProcs, DSPs, comm chips, etc) generally does a really good job with design... The fact that Winmodems are crippled hardware isn't much more than limited requirements, with low cost being the driving factor. Processors, in most cases, can run any number of operating systems, given the right motivation on the part of the programing teams. Most O/Ss need the same basic hooks, so it usually isn't a big stretch (except for those recodes of the asm).
The most important "openness" that hardware companies can provide is full interface specs on their products... register sets, timings, etc... with that information, a person or group could do almost whatever they want with a given chip (even those winmodems).
As for bz2 on modems... software compression is nice, and easily added (of course both phone-link endpoints need to support it). The hardware compression offered by (for example) USR/3Com has been pretty impressive, relatively speaking, since you have a limited block size that you will work with. The adaptive compression used in bz2 may or may not be suited for that task, but that's another debate 8^)
--
Re:Open Source - it's not a panacea (Score:2)
What are you talking about? Free Software is often of quality as good as or better than proprietary software. Some companies (such as Red Hat) have put out poor products such as RH7.0 (but, remember, the file handle leak was caused by a piece of new Red Hat software which had not been released and subjected to the standard Free Software "many eyes"). This ought not to reflect poorly on the movement as a whole - the software they chose (such as the notorious beta compiler) was simply not the best of free software.
FH continues: "Anyway, companies like Sun have tried this with sparc international, (no doubt someone will whinge about the license) but this is essentially an open source CPU."
I have no idea what the license was, but it's almost certainly not anything like open source. Open source does *NOT* mean "you can look at the source code". It has a *SPECIFIC* meaning, which you can find at www.opensource.org. This *is* important. That you make this mistake indicates a deep lack of understanding of Free Software and Open Source software.
Here are the html links for this one ! (Score:2)
(maybe its the dilbert zone time)
list of open source projects and benchmarks is at
http://www.ics.uci.edu/~sumitg/CadPages.html#Benc
people are beginning to use C or variants of C (see http://www.cecs.uci.edu/~specc [uci.edu])
for hardware design and simulation.
synthesis tool that generates hardware starting from C at
http://www.cecs.uci.edu/~spark [uci.edu]
This would help me out! (Score:2)
Altera has many designs which you can download as MegaFunctions, but you can't see how they work. You just incorporate them into your design by feeding in inputs into the black box and reading outputs.
I can't see why people would spend lots of time and money into developing open cores though unless they wanted quicker adoption by downstream developers. Perhaps a reference core which other people could tweek. It would be nice to see GPL of some sort make sure the cores stay open... cuz if it doesn't stay open then whats the point? You could just be doing the fun grunt work for some other putz who will probably come along and market your work without giving you credit.
Anyway, you guys can also check out opencores.org [opencores.org] for more info.
JOhn
The problem with OS cores .... (Score:2)
Software you can do at home - hardware takes infrastructure that most people don't have access to (fabs, NRE, synthesis and layout tools etc)
After all to build a real high-end CPU you need a the work of maybe 100 people for a couple of years ...
For these reasons I think that things like OpenCores make sense for companies but not so much for individuals (however much I wish it were otherwise :-( )
Micro Fab Facillities! (Score:2)
no need (Score:2)
Re:Does not compute (Score:2)
Misunderstanding (Score:2)
The basic idea here is that high-volume embedded devices today usually consist of a CPU and various application-specific elements on a single chip. When designing an embedded system, therefore, you don't buy CPU chips, you buy the right to replicate a CPU design on your own chip. Since in many cases all you need is a rather vanilla CPU, having an open-source description of some reasonable CPU reduces product cost.
This isn't for desktop systems. It's for appliances, handhelds, and such.
The need is not for opensource hardware... (Score:2)
MIPS anyone? (Score:2)
Lemme see, a open standards based processor that can be individually implemented by the individual companies, many of whom are willing to release the very details specifications of how their processor works?
Because isn't that what we want? I'm no electrical engineer, I don't understand how the implementation goes, but as long as it supports a common set, and I can get precise documentation on how to bring the board up, thats pretty much all I care about.
What befuddles me is that the MIPS processor, (which is so well documented, easy to use, and widely available from companies that don't have a monopoly) has fallen so far behind the others, both in Linux kernel implementation and its wide range of usage. Any opinions as to why?
Re:Yeah but, chip making isn't as easy as writing (Score:2)
Motorolas new coldfire is 100% Synthesized, including the on chip cache. Of course they have probably helped the router a bit while placing the cache blocks and clock distribution. But the core byself is a sea of gates.
Taking it too far (Score:2)
No, I haven't read the story, but who's going to dedicate factory time to building a processor that is open source versus factory time to build a proprietary processor designed with YEARS of experience.
Also, don't forget that the lifecycle of chips and software are drastically different. Upgrades to software tend to come every 6 months to a year - with patches in between. Upgrades to processors come more like every 3 to 5 years. Who's going to back the open source chip when they find a bug that calls for a major replacement?
Re:no need (Score:2)
All humor aside, I agree, sort of. You dont see processors which are *designed* at a hardware level to only work for a particular operating system. This is demonstrated best by the fact that dreamcasts run linux, amongst other computing devices, which were only ever designed to run the SEGA proprietry operating system. Admittedly some hardware peripheral companies are being less cooperative with releasing drivers for other operating systems... perhaps *there* we might be better off open sourcing hardware. Opensource modems, for instance, might get better compression with bz2, and we could get up to 64k on a standard phone line (I dont need someone to tell me I'm talking out my ass, its a pie in the sky example). But processors? I think the companies there are already doing a pretty good job.
---
Success (Score:2)
Re:Yeah but, chip making isn't as easy as writing (Score:2)
Yeah, if you want gigantic, slow, hot, wasteful chips. If you want high-performace chips, like we're talking about here, you would at least have to design the datapath by hand. And have you ever synthesized memory, by chance? Probably not. I bet you would say "just get a memory compiler". What about clock generation and distribution? What about high speed digital signaling?
On the other hand, attitudes like yours are good for people like me who design chips for a living. You see, when people that think "designing chips is like programming" start trying to design chips, that is true job security for people who actually know what they're doing.
I agree... don't hold your breath. (Score:2)
Because of these factors, who would be willing to open up their processor designs? Also, if processor designs were open, how many people would realistically be able to make meaningful contributions? I imagine very few people on the street could tell you how to create an XOR operation with a simple set of logic gates. :-) (And I'm not saying I could either.) However, I'd say a large majority of people could pick up C++ in a few weeks (and they do) versus the years up years of studying that hardware engineers suffer though.
Who knows. It could work. It'd just take some completely new thinking.
RMS on Free Hardware (Score:2)
He thinks it's a good idea, but he points out several key issues that differentiate free hardware from free software.
Saddam can sell back the PX2s. (Score:2)
Seriously, though, this model, if it works, will end the current conception of consumer processors as a secret technology as worthy of hiding as much as an Enigma machine would be.
Furthermore, it gives less developed countries a chance to advance with innovative designs. Hardware making is expensive, but with a relatively small investment, a poor country could put out processors on open-source models, perhaps specializing rather than trying to compete with the big boys Intel and AMD.
an outsider's perspective (Score:2)
That is beside the point. The ability to design the chip openly is valuable in and of itself. Perhaps it would never see the light of day. Who cares - if the idea behind the design is sound enough, it could very well be adapted into a product from a "viable" chipmaker. Maybe I'm not being realistic, but I don't see this as one of those "doing something because you can" deals... this could have merit, even if only from a pure science point of view. There's nothing wrong with cold, hard research - something can always be learned.
Here's a question, though: Is there a way that if someone has a chip design, that design could be emulated in software to test the concept? Again, not being my field, I have no idea if doing such an emulation would even pan out anything useful... just an idea.
Open Source Hardware will not change a lot. (Score:2)
Therefore, holding copyright on the design of a chip is largely irrelevant, especially when the manufacturers essentially operate as a cartel. Have you seen Motorolla designing an x86 chip recently? Didn't think so. With a few public exceptions the processor manufacturers largely stay away from each other.
However, they will be receptive to open source as it would allow them to downgrade their expensive design engineering teams, outsource and up the shares dividends accordingly.
I think Open Source designed processors will be higher quality, but it won't make any difference to price or competition in the field.
They fuck you up, your mum and dad.
FPGAs can't be tested for races (Score:2)
Having said that, rapid prototyping using FPGAs, such as Xilinx's contribution to artificial intelligence research [nanodot.org] can be neat.
Re:The Problem: Variable Cost (Score:2)
Multi million gate, very fast FPGAs are already available, and I personally don't think it will be very many years before they are suitable for general purpose microprocessors, rather than just special purpose reconfigurable computers. And when that happens then:
Anybody can fix it when it breaks. I have seen in the last couple of years a dramatic increase in the number of VHDL simulators available, and a corresponding decrease in the cost of them. When someone identifies the problem, they will just write a patch, and then you will download into your FPGA/microprocessor/whatever hardware. This idea is already available with Xilinx Internet Reconfigurable Logic.
You can tweak it to your needs. Just as with software, your typical hacker is not going to do extensive verification. They will do a quick simulation to test the changes they made, load the design into the FPGA, and try it out. And if it doesn't work right, just do another iteration.
How many people are going to learn Verilog/VHDL? How many people have learned JAVA, Python, Perl... (this list is almost endless). And my HP16500 mostly sets in a corner gathering dust. I am now seeing the practice of imbedding a logic analyzer into the FPGA, and having a GUI frontend on your computer.
Re:Alternative idea.. (Score:2)
The problem is that Processors are only the start (Score:3)
More people might be using PPC, MIPS, or StrongARM if there were inexpensive motherboards; the fact that there aren't should give the observer cause to say "Hmmm..."
I'm not sure just how far this issue extends...
The point is that it's not good enough to have a slick new CPU design; you need a system around it to take advantage of it, or, quite frankly, just to allow it to function.
Those that were prepared to "roll their own MIPS variant" to build some highly customized embedded system may find this a quite acceptable scenario. But anybody thinking that this leads to having VA Linux Systems selling F-CPU [f-cpu.org]-based systems any time soon is severely delusional...
programmable logic (Score:3)
There are already companies working on this. At the design automation conference, I saw a company that was working with an addon card that would accelerate different algorithms. And it ran under Linux!!
RTC magazine just had an article on the 5 big technologies. One of these was Linux, and the predicted 6th technology is CPUs with embedded FPGA.
Everything is impossible until you do it. (Score:3)
Sure open source hardware may take a while to get going as the curve to get into it is higher in both dollars and education but there have already been stabs at such things for years and it's succeeded in some lesser hardware hacking areas pretty well. Just remember if it wasn't for the home hackers we might not have PC's at all. A lot of people never thought the PC could take off for similar reasons but it seems to be doing okay. They didn't think Linux could take off but again it seems to have made a place for itself.
Re:Last thing I want to do... (Score:3)
Except for having to recompile the code, Transmeta is doing exactly that. You can upgrade your CPU. If they open their code-morphing software, they you could recompile your CPU (code). That would be cool.
Yeah but, chip making isn't as easy as writing cod (Score:3)
This doesn't scale well to actual chip making though. You really need a thorough grounding in digital logic before you start throwing ands, nands, xors and ors around. There are no higher language equivalents like Perl or VB for chip making, it's just tedious gate and run after gate and run.
Plus, there isn't as much room for self expression in chip making either. Taking the Perl example again, using hte language and reading the Camel book you get a good idea of Larry Wall's mindset. Can you get this from a cip? No?
But, if you're a geek with delusions of grandeur and have a few thousand dollars to throw away in a fab, don't let me stop you.
Open Source - it's not a panacea (Score:3)
Hardware design, and chip design is an esoteric and hard to understand skill, debugging a race condition in a silicon chip can take a skilled technician many many hours of painstaking labor.
Contrast this to software development which is more often than not simply drawing a pretty screen, and filling in a few callbacks.
What will happen if the notoriously poor quality control standards of 'open source' reach the hardware level ? I mean, if your kernel doesn't compile, you simply back out the changes, but if your CPU wont boot, what do you do ?
Anyway, companies like Sun have tried this with sparc international, (no doubt someone will whinge about the license) but this is essentially an open source CPU. It hasn't really caught on.
I think this is just an attempt to garner publicity.
Open Source should confine itself to the realm of software, where it makes eminent good sense.
Verilog (Cish), VHDL (adaish) (Score:4)
You can learn either language relatively quickly if you know some other programming langauages - but using them for logic design does require a grounding in hardware that's outside of normal programming experience (wires, low level concurrency etc etc) - and converting a working simulation into a physical device requires a lot more infrastructure than most people have on hand ....
Re:Yeah but, chip making isn't as easy as writing (Score:4)
Um....no.
There are many high level design languages for chip design, VHDL [uni-hamburg.de] and simulators for testing designs prior to fab.
Designed chips is very much like programing.
I'm disappointed (Score:5)
Nobody seems to know about the equivalent for hardware: Designs written in a Hardware Description language such as Verilog or VHDL can be worked on as a group. When you want to test it, you download it to an FPGA. Complete development kits including software and a protoboard can be had from Altera or Xylinx for a few hundred dollars (less if you are a student). If you make a mistake, fix your VHDL and recompile.
Also, people fail to consider that the designs for this type of thing rarely are on the level of AMD or Intel. We don't make 22 million-transistor designs, but if you want a custom hardware accelerator (say, an Ogg Vorbis accelerator, or hardware accelerated encryption where you KNOW EXACTLY what it is doing. No NSA backdoors. No worry about getting specs from OEMs.) this is the way to go about it. If your project gets popular, you can get them made in quantity as ASICs from any number of companies.
Please, people, look into these things before you start flaming a project like this.
Last thing I want to do... (Score:5)
It is becoming very possible!!! (Score:5)
It is a very interesting read and even speaks to the possibility of this allowing open source cpu's in the future. Imagine, downloading and printing the latest version of your cpu before upgrading to the n.n kernel it is targeted at
The speed is not yet that of a Pentium, but the researcher believes it will be someday. Wish