Fortran - Intel Fortran licensing issues

Author

Intel Fortran licensing issues

Mark Mackey

2005-08-17, 9:14 am

Hi all.

The FAQ for the QMD extension to folding@home (at
http://folding.stanford.edu/QMD.html) states:

> What about AMD support? We are currently using the Intel Fortran
> compiler and libraries. This software explicitly prohibits our use of
> this code on AMD CPUs in the license. At FAH, we follow software EULAs
> and are thus bound to this limitation imposed by Intel. We are currently
> investigating other possibilities and do not have much hope that Intel
> will budge on this.

Are they correct? I can't see any such restriction on a quick perusal of
the license, but I'm not a lawyer. Can anyone from Intel comment?

--
Mark Mackey http://www.swallowtail.org/
code code code code code code code code code code code code code bug code co
de code code code bug code code code code code code code code code code code
code code code code code code code code code code code code code code code c

Tim Prince

2005-08-17, 9:15 am

Mark Mackey wrote:
> Hi all.
>
> The FAQ for the QMD extension to folding@home (at
> http://folding.stanford.edu/QMD.html) states:
>
>
>
> Are they correct?
Did you check what Steve Lionel has said about AMD support on the Intel
forum http://softwareforums.intel.com/ids/board?board.id=5 ? There are
architecture options specifically intended to support both Intel and AMD
CPUs. Generally speaking, bug reports would have to be reproduced on
an Intel platform before getting attention, but that shouldn't stop you
from submitting them.

Steve Lionel

2005-08-17, 5:04 pm

On 17 Aug 2005 11:11:58 +0100 (BST), Mark Mackey
<markm@chiark.greenend.org.uk> wrote:

>Hi all.
>
>The FAQ for the QMD extension to folding@home (at
>http://folding.stanford.edu/QMD.html) states:
>
>
>Are they correct? I can't see any such restriction on a quick perusal of
>the license, but I'm not a lawyer. Can anyone from Intel comment?

No, they are not correct. The Intel compilers license has never, to my
knowledge, contained language restricting use of the compiler to processors
made by Intel. (My experience in this goes back four years.) In fact, we
want everyone to use our compilers and, if you'll look at SPEC submissions,
you'll see that AMD has used our compilers in many of its submissions over the
years.

What is new as of our 9.0 compilers is that we explicitly call out AMD Opteron
and Athlon processors as supported in our System Requirements. Previously, we
listed only Intel processors, but that did not mean that the compiler did not
work or was not allowed to be used with non-Intel processors.

Tim Prince writes that problems must be reproducible on Intel processors to be
accepted by Intel Support. That was the policy a couple of years ago, but is
no longer the case. We have AMD systems in-house for our regular testing and
support investigation, as needed, though I can't recall when I last saw a
problem report coming from an AMD user that required such. We welcome AMD
users - of course, we'd prefer that they use Intel processors, but we feel
we're better off to have people use compilers that make the best use of Intel
processors while providing competitive performance on AMD as well. We're
aware that we're not always the best on AMD today, but we're working on that.

Further reading:

Intel Fortran System Requirements:
http://www.intel.com/cd/software/pr...flin/219993.htm

Intel Compilers End-User License Agreement:
http://www.intel.com/cd/software/pr...flin/219715.htm

(These are for the Linux compilers, but the wording is similar for Windows.)

Steve Lionel
Software Products Division
Intel Corporation
Nashua, NH

User communities for Intel Software Development Products
http://softwareforums.intel.com/
Intel Fortran Support
http://developer.intel.com/software/products/support/

Brooks Moses

2005-08-18, 4:03 am

Steve Lionel wrote:
> On 17 Aug 2005 11:11:58 +0100 (BST), Mark Mackey
> <markm@chiark.greenend.org.uk> wrote:
>
> No, they are not correct. The Intel compilers license has never, to my
> knowledge, contained language restricting use of the compiler to processors
> made by Intel. (My experience in this goes back four years.) In fact, we
> want everyone to use our compilers and, if you'll look at SPEC submissions,
> you'll see that AMD has used our compilers in many of its submissions over the
> years.
[...]

I have forwarded this to the people in charge of Folding@Home -- I'm
noting that, so that other people won't also email them and thereby
swamp them with letters. (Since I'm also at Stanford, I could use the
campus directory to look up their email addresses; I'm not sure how
public they are.)

- Brooks

--
The "bmoses-nospam" address is valid; no unmunging needed.

Brooks Moses

2005-08-18, 4:03 am

Brooks Moses wrote:
> Steve Lionel wrote:
>
> I have forwarded this to the people in charge of Folding@Home -- I'm
> noting that, so that other people won't also email them and thereby
> swamp them with letters. (Since I'm also at Stanford, I could use the
> campus directory to look up their email addresses; I'm not sure how
> public they are.)

And I just heard back from the programmer, who says he'll update the
FAQ. Wow, that was prompt! :)

- Brooks

--
The "bmoses-nospam" address is valid; no unmunging needed.

Mark Mackey

2005-08-18, 4:04 am

In article <43041691.5070406@cits1.stanford.edu>,
Brooks Moses <bmoses-nospam@cits1.stanford.edu> wrote:
>
>And I just heard back from the programmer, who says he'll update the
>FAQ. Wow, that was prompt! :)

Thanks for sorting that out. I was almost certain that the original
statement in the FAQ was wrong, but I needed to check :).

--
Mark Mackey http://www.swallowtail.org/
code code code code code code code code code code code code code bug code co
de code code code bug code code code code code code code code code code code
code code code code code code code code code code code code code code code c

Sponsored Links