EDK : FSL macros defined by Xilinx are wrong

[glen herrmannsfeldt]

rickman <gnuarm@gmail.com> wrote:

(snip, someone wrote)

As far as the OP is concerned, my only point was that the definition of
"work" is a legal one, and that - in general - attempts to circumvent
the GPL typically imply that the GPL does apply to the whole you're
trying to split up. This is hard enough to work through for software,
adding a hardware element to it makes it much harder.

Ok, this is where we started I believe. You have stated that it is
"hard" to apply GPL to hardware. But I don't follow your reasoning in
that regard. I don't see how it is any different from software. The
fact that it costs more to replicate hardware than it does to replicate
software doesn't seem relevant to me.

Well, one reason relates to what copyright protects. It protects
the expression of the idea. In both cases, one can make a new expression
and avoid the copyright. In the case of a large software project, that
isn't likely. It would take a very long time to generate a different
expression of linux to get around its copyright.

But in the case of hardware, reasonably often that isn't true.
Consider a PC board as an example hardware. I can redo the layout,
with exactly the same components wired up the same way, but with
completely different placement. IANAL, but I suspect that would get
around any copyright. Seems to me that if you take the netlist and
run it through a different PC board router, that would probably
be enough.

That makes me wonder how hard it would be to write a program that
would generate a new expression of a software idea. Change it in
enough ways to avoid copyright, but such that the function was
guaranteed (as well as software can) to be the same. That is,
no human in the loop to make human errors.

Not quite the same, but the usual way to get around drug patents
is to modify the molecule in some way. Add a methyl group onto
one ond, or hydroxyl on the other. Away from the active site, but
that is usually enough to get around a patent.

You can trademark the name, and often enough that will discourage
any copying. People who like Coke won't necessarily switch to Pepsi,
even if they did believe it was pretty much the same formula.
(Well, Pepsi is doing well enough, but even so, some like Coke more.)

I don't see any reason why "applying a copyright license to physical
hardware is both legally and philosophically difficult." It is done all
the time. All of my products are copyrighted hardware. A recipient of
GPLed hardware has the same rights as a recipient of a GPLed binary with
the same level of effort required to enforce the GPL on those products.

Have you tried suing anyone for illegally copying it?

-- glen

 

[rickman]

On 11/12/2014 5:49 PM, DJ Delorie wrote:

rickman <gnuarm@gmail.com> writes:
I really have no idea what point you are trying to make with all
this. You go around and around without making anything clear. Instead
of discussing this in tiny paragraphs, perhaps you could make a single
post that actually explains your thoughts?

Because each time I make a statement, someone picks it apart from a
different direction. That led to "clarifications" that weren't relevent
to my original point.

My point is that the FSF chose not to try to cover hardware with the
GPL, because hardware is a tangible good that costs time and money to
replicate, whereas software (either in source or binary format) is
intangible and thus effectively cost-free to copy.

A correlary to this point is that you can't say "the GPL only applies to
the design files" because if the GPL doesn't apply to the physical
hardware, there's no way to ensure the recipient of that hardware has
the rights to its design files. But, applying a copyright license to
physical hardware is both legally and philosophically difficult.

As far as the OP is concerned, my only point was that the definition of
"work" is a legal one, and that - in general - attempts to circumvent
the GPL typically imply that the GPL does apply to the whole you're
trying to split up. This is hard enough to work through for software,
adding a hardware element to it makes it much harder.

Ok, this is where we started I believe. You have stated that it is
"hard" to apply GPL to hardware. But I don't follow your reasoning in
that regard. I don't see how it is any different from software. The
fact that it costs more to replicate hardware than it does to replicate
software doesn't seem relevant to me.

I don't see any reason why "applying a copyright license to physical
hardware is both legally and philosophically difficult." It is done all
the time. All of my products are copyrighted hardware. A recipient of
GPLed hardware has the same rights as a recipient of a GPLed binary with
the same level of effort required to enforce the GPL on those products.

--

Rick

 

[rickman]

On 11/13/2014 12:00 AM, glen herrmannsfeldt wrote:

rickman <gnuarm@gmail.com> wrote:

(snip, someone wrote)

As far as the OP is concerned, my only point was that the definition of
"work" is a legal one, and that - in general - attempts to circumvent
the GPL typically imply that the GPL does apply to the whole you're
trying to split up. This is hard enough to work through for software,
adding a hardware element to it makes it much harder.

Ok, this is where we started I believe. You have stated that it is
"hard" to apply GPL to hardware. But I don't follow your reasoning in
that regard. I don't see how it is any different from software. The
fact that it costs more to replicate hardware than it does to replicate
software doesn't seem relevant to me.

Well, one reason relates to what copyright protects. It protects
the expression of the idea. In both cases, one can make a new expression
and avoid the copyright. In the case of a large software project, that
isn't likely. It would take a very long time to generate a different
expression of linux to get around its copyright.

But in the case of hardware, reasonably often that isn't true.
Consider a PC board as an example hardware. I can redo the layout,
with exactly the same components wired up the same way, but with
completely different placement. IANAL, but I suspect that would get
around any copyright. Seems to me that if you take the netlist and
run it through a different PC board router, that would probably
be enough.

You have to start with a schematic which is copyrighted. So this would
need to be redone. The PCB layout would need to be redone from scratch.
If you lift any portion of the PCB layout verbatim that is the same as
plagiarizing a line or paragraph from a book.

Yes, unless a design feature is patented, you can reuse the design
concept. But that is not unique to hardware. I can rewrite software
and I am no longer in violation of copyright.

I don't see a significant difference.

That makes me wonder how hard it would be to write a program that
would generate a new expression of a software idea. Change it in
enough ways to avoid copyright, but such that the function was
guaranteed (as well as software can) to be the same. That is,
no human in the loop to make human errors.

I don't know how copyright protects software. I mean I don't know just
what constitutes violations and what does not. That is an issue for the
courts to decide.

Not quite the same, but the usual way to get around drug patents
is to modify the molecule in some way. Add a methyl group onto
one ond, or hydroxyl on the other. Away from the active site, but
that is usually enough to get around a patent.

With drugs changing one portion of the molecule *will* change the way it
works 999 times out of 1000. That is why the various drugs for a single
purpose all are qualified separately and have different side effects, etc.

I don't see any reason why "applying a copyright license to physical
hardware is both legally and philosophically difficult." It is done all
the time. All of my products are copyrighted hardware. A recipient of
GPLed hardware has the same rights as a recipient of a GPLed binary with
the same level of effort required to enforce the GPL on those products.

Have you tried suing anyone for illegally copying it?

That is not an unusual occurrence. Here is even an example of *real*
hardware being copied. Interesting that they are being sued for
trademark infringement. I'm surprised that they can trademark a chair.
Maybe that is over the name, "Navy" chair vs. "Naval" chair.

http://www.woodworkingnetwork.com/wood-market-trends/woodworking-industry-news/production-woodworking-news/Restoration-Hardware-Sued-for-Chair-Design-Copies-174032731.html#sthash.2GAPnw9Y.dpbs

--

Rick

 

[alb]

Hi Glen,

glen herrmannsfeldt <gah@ugcs.caltech.edu> wrote:
[]

I must admit that I've always only considered piracy as the act of
'copying/reproducing' without permission, hence not really applicable to
what the OP was doing.

Not sure about the OP, but what do you call the case when you have
a license, but flexlm mistakenly believes that all are in use?

These types of 'malfunctions' will never be accepted as a justification
for cracking the license and/or the license server. It would be
equivalent to stealing a car because yours is temporary out of gas!

In nearly every license there's a clause of NO WARRANTY that you accept
when using the software and while the software provider is relatively
keen to solve 'major' problems, they might not be equally interested to
get you out of the mud because you screwed up.

We all know that s**t happens and when it happens your only legal choice
is to call them and try to get an answer.

License managers are far away from perfection as any other artifact.
Nevertheless they provide a sufficiently detailed debugging
infrastructure to spot the issue rather soon. In your bug report provide
the log files of your lmgrd (is *always* good practice to store logging
information from the server to a file).

Often the issue is on your side and it's handled by some call center
delocalized in Sumatra. Some times though your question might trigger a
patch.

I have in the past known license managers to get confused, for
example when a machine crashes without releasing one.

with flexlm you can use lmremove in those cases, there's still an issue
of 'linger time' of 30 minutes which may affect you, but depending on
how many of you are using the license server you can simply think about
restarting the flexlm which will spawn the vendor daemons and clear
their internal list of used licenses, causing all users to lose their
grants. Usually the license grant is regained as soon as the lmgrd is
restarted, but it may happen that the license module in the client
application has caused the application to quit in the meantime.

The FLEXlm UM is IMO quite easy to follow:
http://www.vcpc.univie.ac.at/information/software/pgi/flexuser/TOC.htm

If the software product and their licensing schemes do not suit you
well, instead of breaking the law and forging the contract you signed
when you bought the product, put more efforts in alternative products
with a licensing philosophy that suit you more.

As a customer we always have a choice and we should use it to send clear
messages to vendors. Unfortunately the EDA tools empire is highly
fortified and cluttered by millions of patents which eventually minimize
your available choice.

Don't forget though that one of the most powerful means the EDA vendors
have to stay afloat is through 'lock ins'. Even if you crack the license
but *use* their software, you'll spread it within your organization and
create consense, expertise and an infrastructure around that product
that would be difficult to get away from. EDA vendors know that and can
push this philosophy to the extreme, i.e. as long as you use their
software they won't care much if you've paid for it.

Al

 

[Tomas D.]

Hello Mr. alb,

Imagine you do not have the means to protect your thoughts/ideas in a
safe manner and somebody strong enough and powerful enough can spoil
your weaknesses and use your thoughts/ideas to make profits.

If someone would tell me, that there's a weakness in my safe, I would do my
best to make it safer. You wouldn't do the same?

Anybody doesn't have to say the open code of my safe, as in this case, I
didn't tell the methods how to get the software hacked, but the word has
been spread for a long time now.

You don't think so?

 

rickman <gnuarm@gmail.com> wrote:

I can't remember the exact context, but I recall a landmark case
(possibly in Europe) where the user was not able to copy or use licensed
software for some reason. The company was not able to help, possibly
because it was out of business or maybe the product was no longer
supported. The person cracked the key and used or copied the product as
he was otherwise allowed to do. Somehow this ended up in the courts and
they upheld his right to do what was needed in the way of reverse
engineering and work around to use the legally obtained product.

Finnish copyright law explicitly allows you to make modifications to
legally obtained computer software that are necessary to make it work
for its intended purpose. The legislation also explicitly states that
the prohibitions on circumventing effective technological copy
protection measures are not applied to computer software.

-a

 

[alb]

Hi Tomas,

Tomas D. <mailsoc@gmial.com> wrote:
[]

If someone would tell me, that there's a weakness in my safe, I would do my
best to make it safer. You wouldn't do the same?

I'd appreciate if somebody tells me that I have a weakness in my safer,
on the contrary I would certainly not be happy if the same guy posts a
message wide open on newsserver mirrored a gazillion times with my code
in big letters.

Anybody doesn't have to say the open code of my safe, as in this case, I
didn't tell the methods how to get the software hacked, but the word has
been spread for a long time now.

This is actually what you did. You provided the exact sequence on how to
*crack* (not *hack*, see the difference here on the 4th paragraph:
http://www.catb.org/esr/faqs/hacker-howto.html#what_is)

> You don't think so?

No I don't.

Al

 

[alb]

Hi Tomas,

alb <al.basili@gmail.com> wrote:

If someone would tell me, that there's a weakness in my safe, I would
do my best to make it safer. You wouldn't do the same?

I'd appreciate if somebody tells me that I have a weakness in my
safer, on the contrary I would certainly not be happy if the same guy
posts a message wide open on newsserver mirrored a gazillion times
with my code in big letters.

please disregard the previous message since for a threading issue I
thought I was replying to the OP. My apologies, I'll reply to your
message separately.

Al

 

[alb]

Hi Tomas,

Tomas D. <mailsoc@gmial.com> wrote:

Imagine you do not have the means to protect your thoughts/ideas in a
safe manner and somebody strong enough and powerful enough can spoil
your weaknesses and use your thoughts/ideas to make profits.

If someone would tell me, that there's a weakness in my safe, I would do my
best to make it safer. You wouldn't do the same?

If I find a weakness in somebody's safe I'd tell him. It would be up to
him to take the counter measures to make it safer.

Anybody doesn't have to say the open code of my safe, as in this case, I
didn't tell the methods how to get the software hacked, but the word has
been spread for a long time now.

When it comes to security, the tipical approach is to reveal the bug
once is *fixed* in order not to expose the affected victim to a massive
attack. If you found a breach, by chance or because of your skills or
job, you should report it to Xilinx before the breach can be exploited
further.

> You don't think so?

I do not think that forging licenses, cracking software and stealing
artwork is ethical, but, as somebody pointed out earlier, ethics may
vary according to culture/beliefs/social environment. I am certainly
sure that those actions are *illegal*, more or less everywhere.

Al

 

[rickman]

On 11/26/2014 5:12 AM, alb wrote:

Hi Tomas,

Tomas D. <mailsoc@gmial.com> wrote:
Imagine you do not have the means to protect your thoughts/ideas in a
safe manner and somebody strong enough and powerful enough can spoil
your weaknesses and use your thoughts/ideas to make profits.

If someone would tell me, that there's a weakness in my safe, I would do my
best to make it safer. You wouldn't do the same?

If I find a weakness in somebody's safe I'd tell him. It would be up to
him to take the counter measures to make it safer.

Anybody doesn't have to say the open code of my safe, as in this case, I
didn't tell the methods how to get the software hacked, but the word has
been spread for a long time now.

When it comes to security, the tipical approach is to reveal the bug
once is *fixed* in order not to expose the affected victim to a massive
attack. If you found a breach, by chance or because of your skills or
job, you should report it to Xilinx before the breach can be exploited
further.

You don't think so?

I do not think that forging licenses, cracking software and stealing
artwork is ethical, but, as somebody pointed out earlier, ethics may
vary according to culture/beliefs/social environment. I am certainly
sure that those actions are *illegal*, more or less everywhere.

I can't remember the exact context, but I recall a landmark case
(possibly in Europe) where the user was not able to copy or use licensed
software for some reason. The company was not able to help, possibly
because it was out of business or maybe the product was no longer
supported. The person cracked the key and used or copied the product as
he was otherwise allowed to do. Somehow this ended up in the courts and
they upheld his right to do what was needed in the way of reverse
engineering and work around to use the legally obtained product.

In the US there is a law specifically making it a crime to break
"encryption" on any product. It can be the most simple of cyphers but
if you break the cypher, you can go to jail, no matter if there was any
gain or even if the product was used improperly in any way. Pretty
bizarre if you ask me.

--

Rick

 

[Tomas D.]

Hello alb,

When it comes to security, the tipical approach is to reveal the bug
once is *fixed* in order not to expose the affected victim to a massive
attack. If you found a breach, by chance or because of your skills or
job, you should report it to Xilinx before the breach can be exploited
further.

they already know that for years, believe me. IP cores are not their
business. They're selling FPGAs.
OTOH looking at the source code is sometimes useful.

 

On Friday, June 16, 2000 2:00:00 AM UTC-5, rob_di...@my-deja.com wrote:

This is a serious question, because I often wonder why people ask such
questions:-

was it really quicker to post this question to a news group and wait a
day or so for someone to give you a one liner answer than to read page
1 of any EPLD datasheet and page 1 of any FPGA datasheet?

Rob

PS the answer you were given was correct as far as it went but a full
answer would take at least 200 words and did you really expect someone
to bother?

Dude do you really have to be an @$$ about it? Either help him or don't but don't chastise someone for asking what to YOU seems like a dumb question.

 

[Nicolas Matringe]

Le 11/12/2014 18:40, gentrysm1@gmail.com a ĂŠcrit :

On Friday, June 16, 2000 2:00:00 AM UTC-5, rob_di...@my-deja.com wrote:
[...]
Dude do you really have to be an @$$ about it? Either help him or don't but don't chastise someone for asking what to YOU seems like a dumb question.

Do you realize the post you're answering to is 14 years old ?

Nicolas

 

[Erik Chalupa]

As for recent update (version 1.2) IDE comes with a huge content patch, bug fixes and overall quality improvements. Basic noncommercial license is still free of charge and until 31.1. 2015 the other prices are with 80%+ discounts.

Regards
MMSystems team

 

On January 9,Allan Herriman wrote:

...take an existing synchronous design and turn it
into a N-way hyperthreaded design by replacing all
the flip flops with N chained flip flops.

I'd like to look at those papers again, but I can't
remember what it was called, and my google-fu is weak.

I think "C-Slow retiming" is the academic term for the transformation.

The earliest moniker I've heard as applied to a processor is "barrel processor" [CDC], with something like "N-way sequential multithreaded" being the modern terminology.

IIRC, Tobias at edaptix had written some articles about his automated implementation of the technique on the late Programmable Planet website, see:
http://www.edaptix.com/coremultiplier.htm

-Brian

 

[HT-Lab]

On 10/01/2015 03:45, Allan Herriman wrote:

Hi,

I vaguely recall reading papers that described an automated pipelining
technique that could take an existing synchronous design and turn it into
a N-way hyperthreaded design by replacing all the flip flops with N
chained flip flops.

I'd like to look at those papers again, but I can't remember what it was
called, and my google-fu is weak.

Any ideas?

Thanks,
Allan

Perhaps PicoPIPE as used(patented) by Archronix?

Hans
www.ht-lab.com

 

[Allan Herriman]

On Sat, 10 Jan 2015 07:57:59 -0800, brimdavis wrote:

On January 9,Allan Herriman wrote:

...take an existing synchronous design and turn it into a N-way
hyperthreaded design by replacing all the flip flops with N chained
flip flops.

I'd like to look at those papers again, but I can't remember what it
was called, and my google-fu is weak.

I think "C-Slow retiming" is the academic term for the transformation.

The earliest moniker I've heard as applied to a processor is "barrel
processor" [CDC], with something like "N-way sequential multithreaded"
being the modern terminology.

IIRC, Tobias at edaptix had written some articles about his automated
implementation of the technique on the late Programmable Planet website,
see: http://www.edaptix.com/coremultiplier.htm

-Brian

C-Slow it was. Thanks very much.

Regards,
Allan

 

[Mike Field]

On Saturday, 10 January 2015 16:45:13 UTC+13, Allan Herriman wrote:

I'd like to look at those papers again, but I can't remember what it was
called, and my google-fu is weak.

Any ideas?

I've seen it termed 'System Hyper Pipelining'.

 

[princesse91]

Hi Ahmed,
Can you tell me how did you generate VHDL from Handel-C?? I'm working on the conversion from c++ to vhdl with handel-c but i didn't know how to do it..
Thanks

 

VHDL is leads to verbose designs which are slow to write and hard to visualise.

AHDL is elegant and specifically deigned fpga type architectures.

It is as to 'C' whereas VHDL is like Perl - write once and read never.

The vast majority of fpga designs use synchronous logic. Fighting with archaic ideas of processes with endless 'global' signals passing between them is like returning to the dark ages of BASIC before structured programming caught on.

Most of us think about the change events and not the hold events in a design. The unassigned state evaluating as false (or 0) is a key the the simplicity of AHDL. This is why AHDL code usually compiles and works first time.
A simple missing else clause in VHDL will double the amount of logic you end up with - for no gain.