ISA vs. patent/trademark

Hi,

Are ISA covered by patents or trademarks? Is it allowed to develop a 
processor core for a popular ISA as long as no reference is made to any 
of the original company trademarks? Many thanks for your comments.

Eric

"Eric DELAGE" <"eric UNDERSCORE delage AT yahoo DOT fr"> schrieb im
Newsbeitrag news:4252c3dc$0$3114$8fcfb975@news.wanadoo.fr...
> Hi,
>
> Are ISA covered by patents or trademarks? Is it allowed to develop a
> processor core for a popular ISA as long as no reference is made to any
> of the original company trademarks? Many thanks for your comments.
>
> Eric

It all depends what you are up to and against whom, some companies are very
quick to get their lawers involved.

However I think it is not actually possible to fully copyright an ISA,
because

1) languages can not be patented, any one can learn another language if he
pleases so. That is covered by basic human rights I hope.
2) an ISA can be considered a description of some formal language, in any
case you yourself can learn to understand the bytecode, etc.. that cant be
prohibied.
3) You can buy a smart-dog and teach that dog that ISA (if you are able
too), its isnt prohibited to try.
4) You can buy and FPGA and teach that FPGA to understand what you learned
in step[2] - that again can not be directly forbidden.

Now, that doesnt mean the lawers want come after you, they can try at least.
But by being VERY VERY Careful it is possible to make the case a 'no case'
for the lawers. But you really have to have a 'clean implementation' - and
you have to withstand any pressure from the lawers who try to prove the
opposite.

ARM shuts down, nnARM was forced underground/dead, picoturbo was closed down
MIPS shoots as well
SPARC is Open for anyone to implement. that the reason why Gaisler did
choose SPARC for their space stuff
Altera tolerates the GPL version of NIOS-I, but is also trying to shut down
OpenSource NIOS-II
Xilinx has said they have nothing against 3rd MicroBlaze implementations

my 2cents
Antti

> Are ISA covered by patents or trademarks? Is it allowed to develop a
> processor core for a popular ISA as long as no reference is made to
any

Depends on the ISA. ARM, for instance, is covered by thick IP
protection layers, and is vigorously defended. 8051, not so much :)

larwe@larwe.com wrote:
> 
>>Are ISA covered by patents or trademarks? Is it allowed to develop a
>>processor core for a popular ISA as long as no reference is made to
> 
> any
> 
> Depends on the ISA. ARM, for instance, is covered by thick IP
> protection layers, and is vigorously defended. 8051, not so much :)
> 
W/ ARM for instance, I understand that they can protect their IP 
implementation w/ patents or copyrights but what would be the lawwhich 
prevent someone from implementing a processor core running an ARMv5?

If someone implements a core able to run any compiled w/ gcc using some 
of the x86 flags, could INTEL prevent him from distributing the core. No 
mention is directly done of the INTEL name; no claim would be done for 
any INTEL compatibility of the processor core. What could happen?

Eric

> Now, that doesnt mean the lawers want come after you, they can try at least.
> But by being VERY VERY Careful it is possible to make the case a 'no case'
> for the lawers. But you really have to have a 'clean implementation' - and
> you have to withstand any pressure from the lawers who try to prove the
> opposite.

The keypoint, I think, is: are we talking about 1) patents (but the ARM 
ISA is full of instructions developped before it appears) 2) trademarks 
(but as long as no mention/no claim regarding the original names are 
done, would it be ok?) 3) copyright (but if an ISA is a language, an 
someone copyright it?)

Eric

> W/ ARM for instance, I understand that they can protect their IP
> implementation w/ patents or copyrights but what would be the
lawwhich
> prevent someone from implementing a processor core running an ARMv5?

Patents relating to specific instruction set features. I don't know
what patents they hold specifically; I'm sure there are some.

> If someone implements a core able to run any compiled w/ gcc using
some
> of the x86 flags, could INTEL prevent him from distributing the core.
No
> mention is directly done of the INTEL name; no claim would be done
for
> any INTEL compatibility of the processor core. What could happen?

You could be sued for distributing a product in violation of another
party's intellectual property rights. For example, suppose Intel has
patents on some aspect of MMX. You implement MMX compatible
instructions in an infringing way. They have a claim against your
product.

Example: I make a device that plays video off optical disks. It just so
happens by accident that my design is exactly the same as, and in fact
compatible with, the DVD standard. The DVD consortium can sue me for
infringement. They could still sue me with equal success if I can PROVE
that I developed my device totally without any reference to their
designs.

Eric DELAGE <"eric UNDERSCORE delage AT yahoo DOT fr"> writes:

>> Now, that doesnt mean the lawers want come after you, they can try at least.
>> But by being VERY VERY Careful it is possible to make the case a 'no case'
>> for the lawers. But you really have to have a 'clean implementation' - and
>> you have to withstand any pressure from the lawers who try to prove the
>> opposite.
>
> The keypoint, I think, is: are we talking about 1) patents (but the
> ARM ISA is full of instructions developped before it appears) 2)
> trademarks (but as long as no mention/no claim regarding the original
> names are done, would it be ok?) 3) copyright (but if an ISA is a
> language, an someone copyright it?)


Point 1: As for MIPS, the unaligned load/store instructions (i think)
are covered by patents. In order to implement these, you must obtain a
license from MIPS technologies.

Point 2: For trademark reasons you cannot call a processor for MIPS
compatible or compliant, without implementing the full ISA.  Goto
point 1.

(Infinite loop exception, program aborted).

That hasn't stopped people from implementing processors with the MIPS
ISA, *except* for the patented instructions.  But the people doing
this are very careful about not calling the processor for anything
related to MIPS...


Kai
-- 
Kai Harrekilde-Petersen <khp(at)harrekilde(dot)dk>

"Eric DELAGE" <"eric UNDERSCORE delage AT yahoo DOT fr"> skrev i meddelandet
news:4252ccf3$0$3139$8fcfb975@news.wanadoo.fr...
> larwe@larwe.com wrote:
> >
> >>Are ISA covered by patents or trademarks? Is it allowed to develop a
> >>processor core for a popular ISA as long as no reference is made to
> >
> > any
> >
> > Depends on the ISA. ARM, for instance, is covered by thick IP
> > protection layers, and is vigorously defended. 8051, not so much :)
> >
> W/ ARM for instance, I understand that they can protect their IP
> implementation w/ patents or copyrights but what would be the lawwhich
> prevent someone from implementing a processor core running an ARMv5?
>
You protect an ISA by patenting some special thing which is required to
implement the ISA.
I happen to think that the ARM Thumb patent is a load of rubbish.

The basis of the patent is the "ARM Ltd discovery" that less code is better
than more code.
Code compression for RISC is mentioned already in the original RISC paper by
Katevenis.



> Eric




-- 
Best Regards
Ulf Samuelsson

Kai Harrekilde-Petersen wrote:
> Eric DELAGE <"eric UNDERSCORE delage AT yahoo DOT fr"> writes:
> 
> 
>>>Now, that doesnt mean the lawers want come after you, they can try at least.
>>>But by being VERY VERY Careful it is possible to make the case a 'no case'
>>>for the lawers. But you really have to have a 'clean implementation' - and
>>>you have to withstand any pressure from the lawers who try to prove the
>>>opposite.
>>
>>The keypoint, I think, is: are we talking about 1) patents (but the
>>ARM ISA is full of instructions developped before it appears) 2)
>>trademarks (but as long as no mention/no claim regarding the original
>>names are done, would it be ok?) 3) copyright (but if an ISA is a
>>language, an someone copyright it?)
> 
> 
> 
> Point 1: As for MIPS, the unaligned load/store instructions (i think)
> are covered by patents. In order to implement these, you must obtain a
> license from MIPS technologies.
> 
> Point 2: For trademark reasons you cannot call a processor for MIPS
> compatible or compliant, without implementing the full ISA.  Goto
> point 1.

See also the ST7 family from ST, and the Motorola HC05 series.....


> (Infinite loop exception, program aborted).
> 
> That hasn't stopped people from implementing processors with the MIPS
> ISA, *except* for the patented instructions.  But the people doing
> this are very careful about not calling the processor for anything
> related to MIPS...

  I heard that the NIOS II is 'very similar' to MIPS - can anyone who 
knows both cores in detail comment ?

-jg

Q
If some processor that is not any way MIPs like can otherwise perform a
register ld/st for a word on any byte boundary is that a problem, or
only if the rest of it is MIPs like too.


If a processor has a general move n bytes ie mv rd,rs,4 and works on
any boundary and any length into memory locations that are also a
register, a problem?.

Am I not mistaken that most all cpus allowed unaligned ld/st till the
RISC religion took over. Guess I will have to look at the patent.

johnjakson at usa dot com