ISA vs. patent/trademark - Page 2

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
Re: ISA vs. patent/trademark

Quoted text here. Click to load it

ARM has an "interesting" way of handling unaligned word addresses: If
the address is not word aligned, it uses the rounded-down address to
load a word but then rotates the word such that the byte at the
unaligned address is the LSB of the resulting word (this is for
little-endian mode).  The behaviour is probably a side-effect of the
byte load instruction (which ANDs with 0xFF after the rotate).  I once
wrote a fast string copier that exploited this behaviour, but I don't
think it makes unaligned word access any faster.

        Torben

Re: ISA vs. patent/trademark
Quoted text here. Click to load it

Everett M. Greene writes:
Quoted text here. Click to load it

Of course you can patent something new that eliminates the need
for something old!

The patent is on the two instructions, and their use in eliminating
the need for the usual hardware that would support unaligned access.

Quoted text here. Click to load it

The use of the two instructions in question seemed like something
that wasn't obvious when I first saw them.  I think it passes the
obviousness test.

However, the patent office has a much lower bar for obviousness than
you or I would have.  Even though I think this patent was reasonable,
they certainly grant many others that I don't think they should.

Eric

Re: ISA vs. patent/trademark
Quoted text here. Click to load it
the




This certainly does make sense. There definitely is prior art for
computers that do not support unaligned access to memory; the
System/360 comes to mind.

And they can handle unaligned operands, but it takes at least four
instructions:

   A     5,ALIGNED

becomes, say

   LH    6,UALIGNED
   SLL   6,16
   IH    6,UALIGNED+2
   A     5,6

or even

   LC    6,UALIGNED
   SLL   6,24
   L     7,UALIGNED+1
   SR    7,8
   N     7,#X'00FFFFFF'
   O     6,7
   A     5,6

so if the MIPS speeds things up by having a "fetch left half of n
bytes" followed by "fetch right half of N-n bytes, then perform the
operation" instructions (as a RISC chip, it might not have the 'perform
the operation' part) it has indeed done something new.

John Savard


Re: ISA vs. patent/trademark
snipped-for-privacy@ecn.ab.ca writes:
Quoted text here. Click to load it

ICM was introduced with 370 ... insert character under mask
     ICM   6,B'1111',UNALIGNED
     ar    5,6

problem with LH was that it was arithmetic (not logical) and
propogated the sign bit (and it required half-word alignment).

--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn /

Re: ISA vs. patent/trademark


Quoted text here. Click to load it
any

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


Re: ISA vs. patent/trademark
Quoted text here. Click to load it
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

Re: ISA vs. patent/trademark
Quoted text here. Click to load it
lawwhich

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

Quoted text here. Click to load it
some
No
for

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.


Re: ISA vs. patent/trademark

"Eric DELAGE" <"eric UNDERSCORE delage AT yahoo DOT fr"> skrev i meddelandet
Quoted text here. Click to load it
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.



Quoted text here. Click to load it




--
Best Regards
Ulf Samuelsson



Re: ISA vs. patent/trademark
Quoted text here. Click to load it

Original RISC paper by Katevenis?  While I was able to find a 1983
paper by him, near as I can tell the original RISC paper is still the
one by Patterson amd Ditzel in 1980.
--
Joseph J. Pfeiffer, Jr., Ph.D.       Phone -- (505) 646-1605
Department of Computer Science       FAX   -- (505) 646-1002
We've slightly trimmed the long signature. Click to see the full one.
Re: ISA vs. patent/trademark

|> >
|> > 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.
|>
|> Original RISC paper by Katevenis?  While I was able to find a 1983
|> paper by him, near as I can tell the original RISC paper is still the
|> one by Patterson amd Ditzel in 1980.

There were papers containing much of the technical content in the
1960s.


Regards,
Nick Maclaren.

Re: ISA vs. patent/trademark
snipped-for-privacy@cus.cam.ac.uk (Nick Maclaren) writes:

Quoted text here. Click to load it
better
Quoted text here. Click to load it
by
Quoted text here. Click to load it

Of course.  But the one that tied it all together in a single,
coherent bundle was Patterson and Ditzel.
--
Joseph J. Pfeiffer, Jr., Ph.D.       Phone -- (505) 646-1605
Department of Computer Science       FAX   -- (505) 646-1002
We've slightly trimmed the long signature. Click to see the full one.
Re: ISA vs. patent/trademark
Quoted text here. Click to load it

Are you asserting that the IBM 801 papers didn't tie it all together
in a single coherent bundle?

Re: ISA vs. patent/trademark

Quoted text here. Click to load it

The earliest 801 paper I'm familiar with was published in 1982.  While
the project certainly tied things together in a coherent RISC bundle
before the RISC and MIPS projects (as both Hennessy and Patterson
acknowledge),  Patterson and Ditzel is the first publication I'm aware
of that does.
--
Joseph J. Pfeiffer, Jr., Ph.D.       Phone -- (505) 646-1605
Department of Computer Science       FAX   -- (505) 646-1002
We've slightly trimmed the long signature. Click to see the full one.
Re: ISA vs. patent/trademark
Quoted text here. Click to load it
better
paper by
OK, but both preceeded the ARM chip and code compression for RISC
is therefore not a new thing discovered by ARM Ltd.
They base their Thumb patent on the claim that architectures have only been
developed
to increase the performance, not to reduce code size, and that they
discovered the need
for code space reduction for RISC.
I believe that the width of datapaths has been driven mostly by the need to
increase addressspace.
If you do not accept the ARM claim, then the Thumb patent becomes really
weak.

Quoted text here. Click to load it

--
Best Regards
Ulf Samuelsson                 snipped-for-privacy@atmel.com
We've slightly trimmed the long signature. Click to see the full one.
Re: ISA vs. patent/trademark
+---------------
| I happen to think that the ARM Thumb patent is a load of rubbish.
+---------------

Yup. See my previous posting about the LINC/LINC-8/PDP-12 machines...


-Rob

-----
627 26th Avenue            <URL:http://rpw3.org/
San Mateo, CA 94403        (650)572-2607


Re: ISA vs. patent/trademark

Quoted text here. Click to load it
to

better
paper by

I note that the Itanium ISA is covered by a patent on its unique method
of explicitly indicating parallelism. The more conventional way of
indicating parallelism, by a 'parallel' bit on each instruction, was
used on TI signal processing chips.

I'm sad to hear that there is a patent on the ARM Thumb instruction set
that extends to the general principle, because something like it is
what the PowerPC architecture desperately needs - so that people can
use it the way IBM wants, without compromising the architecture.

John Savard


Re: ISA vs. patent/trademark

Quoted text here. Click to load it

There are Freescale (formerly Motorola) PowerPC microcontrollers with code
compression (look for the MPC562 on the Freescale website, for example).
The do not use an alternative instruction set like the Thumb - instead, it
is a more dynamic compression.  Software utilities compress your
executable and generate tables which you must load into the decompressor's
lookup table ram on startup.  The decompressor is part of the instruction
pre-fetch burst controller, which passes fully decompressed instructions
on to the main cpu core.  This makes it more flexibile than the Thumb
technique, and you get the full power of the core rather than a subset,
but it's more of a hassle to configure (and probably awkward for debugging).


Re: ISA vs. patent/trademark
Quoted text here. Click to load it

I didn't think it was on Thumb instructions, but on the method of
switching between ARM and Thumb execution (using the LSB of the PC to
indicate which mode you're in).

Forget Thumb anyway, it's rubbish. What patents there are on Thumb2
should be more interesting.

Cheers,
Jon

Re: ISA vs. patent/trademark

snipped-for-privacy@beniston.com (Jon Beniston) writes:
|>
|> > I'm sad to hear that there is a patent on the ARM Thumb instruction set
|> > that extends to the general principle, because something like it is
|> > what the PowerPC architecture desperately needs - so that people can
|> > use it the way IBM wants, without compromising the architecture.
|>
|> I didn't think it was on Thumb instructions, but on the method of
|> switching between ARM and Thumb execution (using the LSB of the PC to
|> indicate which mode you're in).

Totally different from using the top bit, as was done on the IBM 370
range and others, of course.


Regards,
Nick Maclaren.

Re: ISA vs. patent/trademark
Quoted text here. Click to load it

Are you sure? A lot of the layers appear to be camoflage and not actual
protection. The ARM business case appears to be:
    1) milk money from people who buy ARM IP
    2) if you see somebody else making ARMs, sue
    3) try to get an out of court settlement at all costs
    4) including by paying the "offender" lots of cash to go away,
       stop making ARMs and being quiet about the whole business


--
    Sander

+++ Out of cheese error +++

Site Timeline