Re: [f-cpu] F-CPU vs ALPHA and licence

[Yann Guidon <whygee@f-cpu.org>]

hello,

nicO wrote:
> Yann Guidon a écrit :
> > hello,
> > nicO wrote:
> > > Yann Guidon a écrit :
etc.


> > let's sum it up : you don't want extra pins and i don't want to go through
> > memory. We can still try to "multiplex" the semaphores on the front side bus.
> > Note that in a mutli-CPU system with tens or hundreds of chips, because
> > you have to go through the "normal" path, the latency of the semaphores
> > will be this of the memory.
> > Now imagine we had around ten or twelve pins anyway. A "cheap" FPGA can have
> > around 160 user pins, so we can connect more than ten CPU DIRECTLY to each
> > others. With a good PCB layout, a good topology, some VHDL and a few more
> > pins per CPU, we can increase the "cut" of the system and drastically reduce
> > the semaphore latency.
> 
> So you add another chip it cost a lot. Why using a new port ? What does
> it add ? Nothing ! Just few more problem ! Canonical CPU are always put
> on one bus (or switch, it's the same).
and that's their problem.
note that your "always" shows that you don't know alternatives,
there are other computers that have different architectures than
the SMP/MPP you learn at school.

> > Concerning SoCs : would you call the P4, the G4, the US3, the R18K... SoC ?
> > maybe, maybe not. i stick to the definitions that were defined before i came.
> > i recently browsed the net and found that on f-cpu.tux.org :
> We could _never_  fight against G4, US3, ... because our design flow are
> semi-custom. So we lose 30% of the performance. We fight against R10000,
> SH4, SH5, ARM9-10, even a kind of PPC, all this are made for SOC only.

i do not "fight against", i "fight for" a free, ununcumbered platform.
i do not work as SoC expert. i'm just a wannabe-computer-architect.

> If you want create a complete new computer never forget what happen to
> Next. they were the best computer during there life, but now is dead :
> too new !

they didn't obviously found their market.
and Next is not the only "fallen" company around.
you should read the book i showed you :-)

> > Our goal is not to make a Nth Sparc clone or a LEON.
> > it is not either our goal to use Meiko's FPU.
> > Btw, have a look at Opencores.org : they have already a FPU
> > for add/mul.
> I know it's very slow (60 Mhz compare to 450 Mhz for michael's mul unit)
so either you use a slow foreign IP, or we redesign design it from scratch
for speed, freedom and all the bells and whistles.

> > > In your case, to make a Chip (or a SOC, it the same !), all vhdl code
> > > must be in GPL. All the code ! Which compagny would make a fcpu if it
> > > couldn't add something to create more value than the other ?
> >
> > Our goal is not to let "companies" make money "for free" with our work.
> > If they want to play the faire game, they use the GPL. If they want to
> > make more value than others, they still have to respect the GPL.
> > My understanding of what the LGPL would do is : companies will not
> > respect the architecture and add non-standard extensions. It may be
> > "worth" for an individual company but it's not a benefit for everybody.
> 
> _NO_ ! LGPL just said you could _link_ with proprietary code. If you
> touch the fcpu it-self, it's like touching GPL code ! You absolutely
> can't touch the code as you want ! (that's the goal of LGPL)

remove your pink glases. Of course, i know that the f-cpu itseld must not
modified. But you forget one detail : as you said, "the chip is the system"
and what's the use of a system where the CPU is surrounded with "cores"
that you can't access as a programmer ? LGPL does not tell anybody to
release the API/ABI/interface of the other cores.

So imagine i am Intel or whatever "big bad company".
i come, d/l the code, add my "proprietary" stuff to the f-cpu core,
such as the SDRAM controller, the interrupt controller and a secret
backdoor or two. now, who will bug me about that ???

i try to have a long sighted vision, and i am not impressed by short-
or middle-term issues about revenue or whatever. I have seen that the lack
of long-term vision does to my own life and other's.
I would like to find a good solution but IMHO it is not a good solution
to help companies stay in their short-term revenue goals.

> > > That said
> > > that we will never use all proprietary interface (IEEE1394, ethernet,
> > > USB, PCI, hyperband,...) so nobody could be connected to a fcpu at low
> > > cost. Never forget that board=chip, todays.
> > again : look at opencores : they have an increasing number of free cores.
> Yes, but what about IP right ?...What about the choice ?
what what ?
you want everything ready for you ?
you have access to the source, you have direct connexion to the authors,
so instead of paying an inhouse engineer for customizing the external
cores, you can pay (often "less" than the inhouse engineers,
particularly in France where the taxes for working are very expensive)
the original developper to do its works, and that's all.
what more do you want ???

> > IMHO, a F-CPU would consume too much power and silicon for a digital camera ;-)
> > (in today's standards).
> fcpu will be finish in 2 years, so...
ain't it a bit optimistic ? :-P

> > > In SW, Linux and some software are in GPL but could be used with
> > > proprietary program.
> > and vice versa : Emacs runs under Windows, for example.
> > But concerning our case : it is simply a matter of tools.
> > i don't want to change my mind simply because others don't do their
> > work at making their product work.
i should add : these tools are extremely expensive and they work
only in ideal cases ;-P 

> > > Lot of compagny want to sell solution with linux
> > > for embedded market, for example. This system work well, no code stolen,
> > > but you could sell something, so made a wide spread of the GPL code (Red
> > > hat, Mandrake are responsible to make linux so well known, never forget
> > > that !).
> >
> > What RedHat and Mandrake do (apart from playing with stocks) is service
> > around, not on, what they distribute. They distribute, customize,
> But that the problem ! How would you customise fcpu with GPL ?
open EMACS and a good VHDL book.
don't forget to read the manual, btw.

> For exemple, suse couldn't make a priorietary distribution of the fcpu.
that is their problem.

> If a distrib could be see as a "chip".
btw, their distribution is "proprietary" because they ADDED
a non-gpl part at the critical place. this is what i want to prevent
and avoid, and this is what WILL happen if LGPL is used.

now, apart from being in bad terms with Lionel, does Suse's
strategy help anybody ???

> That's the problem. With the current
> licence system, i'm afraid that no compagny will invest an euro in the
> project. But we need some compagny to produice chips, to make the fcpu
> alive.

don't worry for that.
Japan, India, South America and other countries are already doing their
own projects. I have contacts with several working groups.
i'll let you know when i have news.
My REAL worry is to have the core ready enough before october :-(

> In SW GPl allow to run gpl program and proprietary program. How do you
> do that in HW ?
that is a problem that we have to solve.

> > >  In the case of the fcpu, where a compagny could make money ? By selling
> > > exactly the same chip than the other ? But no system is built like that,
> > > any more (no use in SoC)! So forget the use of the fcpu in wide spread use.
> > please tell me : why are compnies still making 74xxx devices ?
> > look at catalogs ! (ie Radiospares/Arrow/Radioshacks...)
> > Why do companies produce the "same" part numbers ???
> > one will emphasize on the package (SOP, DIL, ...), the other
> > will provide more diversity, another will reduce the power consumption,
> > another will increase the speed, another has more marketing
> > and better fundries... and you know what ? even though they produce
> > "the same thing" these companies are still in business ;-P
> 
> Haven't you seen that that kind of chip are almost always made by the
> same compagny (ST for power product,...). It's old line of product still
> in production, not new one. The catalog that you speak are for PME of 10
> people who produice very few number of pieces for a product, it's not
> serious for "great public" for exemple. (try to buy a big fpga with
> radiospare or Farnell !)

however, if your previous statement was true, there would be only
one single source for the 74xxx series. well, there are FPGAs,
GAL, PAL and even resistors, condensers, connectors....

> > Even with a single source code, you can make a large number of
> > variations. the resulting F-CPU can be customized without even
> > having to redistribute the source files, simply because
> > f-cpu_config.vhdl will have a few changed lines. Use a different
> > compiler/synthesis tool, go to another fundry, hire more or less
> > people and the product will be more or less expensive with more
> > or less performance.
> 
> I don't speak about the fcpu it-self, but all peripheral put beside.
> Imagine an LCD controller for example for PDA. This has nothing to do
> with the fcpu project ! Or imagine a compagny which made a video card,
> and put a 3D accelerator beside the fcpu.

so what ?
do you want short-term  or long-term POV ?

> > > It's still possible to change our licence because few people write
> > > code(whygee and Michael, that's all i think). I think that LGPL are more
> > > appropriate.
> > i am against, you are not. I will follow Michael Riepe's choice.
> > What does he vote for ? i do not want to spoil the situation
> > between you and me so i prefer to listen to others's choices.
> > As you know, i am not the reincarnation of the project.
> 
> The trend is to know how to interest chip maker in our project. That's all !

i don't think that it is a real problem.
for DATE 2002, we'll go together speak to the companies' reprensentors.
you will be surprised. but it's not the "big thing". they are only followers.

> > > > > Today's µp implement semaphore with atomic read_modify_write
> > > > > instructions.
> > > > so what ?
> > > So you don't need special port !
> > And what is your answer to the performance loss ?
> > do you have anything smarter ?
> Memory access in an atomic way like is made by other cpu. That's enought.

i told you : FC0 handles memory accesses with 256-bit lines.
if you do RMW you WASTE a lot of performance.
but i think that you don't care.

ok, i'll try to sleep. i hope that this licence discussion will not
degenerate in a war, because i prefer to code : "make file, not war"
(lame attempt at doing humor at 4am, sorry)

> nicO
WHYGEE
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/