[Mesa-dev] [PATCH] ralloc: Omit detailed license information about talloc.

Kenneth Graunke kenneth at whitecape.org
Fri May 9 13:23:38 PDT 2014


On 05/09/2014 12:48 PM, Jose Fonseca wrote:
> 
> 
> ----- Original Message -----
>> On Fri, May 9, 2014 at 2:55 AM,  <jfonseca at vmware.com> wrote:
>>> From: José Fonseca <jfonseca at vmware.com>
>>>
>>> That information misleads source code auditing tools to think that
>>> ralloc itself is released under LGPL v3.
>>>
>>> Instead, simply state talloc is not licensed under a permissive license.
>>> ---
>>>  src/glsl/ralloc.h | 7 +++----
>>>  1 file changed, 3 insertions(+), 4 deletions(-)
>>>
>>> diff --git a/src/glsl/ralloc.h b/src/glsl/ralloc.h
>>> index 4581a7a..70f7d9c 100644
>>> --- a/src/glsl/ralloc.h
>>> +++ b/src/glsl/ralloc.h
>>> @@ -38,10 +38,9 @@
>>>   * Tridgell's talloc, but ralloc is an independent implementation
>>>   * released under the MIT license and tuned for Mesa.
>>>   *
>>> - * The talloc implementation is available under the GNU Lesser
>>> - * General Public License (GNU LGPL), version 3 or later. It is
>>> - * more sophisticated than ralloc in that it includes reference
>>> - * counting and debugging features. See:
>>> https://urldefense.proofpoint.com/v1/url?u=http://talloc.samba.org/&k=oIvRg1%2BdGAgOoM1BIlLLqw%3D%3D%0A&r=NMr9uy2iTjWVixC0wOcYCWEIYhfo80qKwRgdodpoDzA%3D%0A&m=W9QOIrhCEXJz3yIQbKcCXWeb39YRUJRWizIYkimspSE%3D%0A&s=a97e9dd737954d1433e70ee76d13fdc7c1f433e837da4f3c217138445a70320f
>>> + * The talloc implementation is not available under a permissive
>>> + * license. It is more sophisticated than ralloc in that it includes
>>> + * reference counting and debugging features. See:
>>> https://urldefense.proofpoint.com/v1/url?u=http://talloc.samba.org/&k=oIvRg1%2BdGAgOoM1BIlLLqw%3D%3D%0A&r=NMr9uy2iTjWVixC0wOcYCWEIYhfo80qKwRgdodpoDzA%3D%0A&m=W9QOIrhCEXJz3yIQbKcCXWeb39YRUJRWizIYkimspSE%3D%0A&s=a97e9dd737954d1433e70ee76d13fdc7c1f433e837da4f3c217138445a70320f
>>
>> What about this instead?
>>
>> "The talloc implementation is available under a different open source
>> license.
> 
> I'm not sure in what way it is better.  The reason talloc was
> inadequate was not only because it was in a different license but
> because it was a copyleft (non-permissive) license.  (E.g., LLVM has
> a different license from Mesa, but nobody went out its way to replace
> it.)
> 
> That said, I don't care either way.  Provided it doesn't have " GNU
> Lesser General Public License" nor "LPGL" it should be fine.   (I
> also though of saying "copyleft" instead of non-permissive, but
> again, I worry it might cause more confusion.)
> 
> We could also remove this sentence all together.
> 
> Jose

Actually, I don't think copyleft was the problem.  It was because
various VMware and Intel people were paranoid about LGPLv3, and
vehemently opposed to depending on LGPLv3 code.  I think if it was
LGPLv2, nobody would have cared.  (Personally, I think is unwarranted
paranoia, but what can you do?)  It was just easier to reimplement it,
and put it under the same license as the rest of Mesa.

Independently of that, it did give us an opportunity to make different
performance trade-offs: ralloc_parent(x) is O(1) while talloc_parent(x)
is O(n).  Mesa calls parent() all the time, while Samba doesn't, and
benefits more from a different operation being fast.

I'm fine with either sentence, or simply deleting it.  My original
intent was to encourage people who are OK with LGPLv3 to go use the more
robust code that's already available as a reusable shared library.

Perhaps a mix of the two:

"talloc is more sophisticated than ralloc in that it includes reference
counting and useful debugging features.  However, it is released under
a non-permissive open source license."

Whatever you decide to commit is fine by me.

--Ken

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: OpenPGP digital signature
URL: <http://lists.freedesktop.org/archives/mesa-dev/attachments/20140509/a7f4218e/attachment.sig>


More information about the mesa-dev mailing list