(Yet another) RTEMS License Question

gds gds at chartertn.net
Fri Jul 8 14:19:38 UTC 2011

On 07/08/2011 09:40 AM, Joel Sherrill wrote:
> On 07/07/2011 11:32 PM, Gene Smith wrote:
>> I understand that any product sold or distributed containing an
>> unmodified RTEMS release does not have to make available the source code
>> of the application. This is clear from reading the license. However,
>> what is not clear is if there is a requirement to inform the customer
>> that RTEMS is present and make available on demand the unmodified RTEMS
>> code used in the product?
>> Assuming that the answer to the above question is that the unmodified
>> RTEMS code tree doesn't have to be provided, what if creation of the
>> RTEMS application also requires a new or modified BSP, libcpu or other
>> change to internal GPL licensed RTEMS files that are not yet or never
>> become a part of an official RTEMS release or tree? Does this possibly
>> trigger a requirement to make available to customers this modified RTEMS
>> tree used in the product?
> No code in the RTEMS source tree (e.g. RTEMS CVS module)
> is supposed to licensed in a way that would impose any
> obligations or restrictions on an end user. No advertising,
> offering source, relinking kit, etc.
> This also applies to newlib and the gcc language support
> libraries shipped with the RPMs.
> This does NOT necessarily apply to code. For example,
> if you used the RTEMS port of GNU readline or the GNU
> Scientific Library then you would have to abide by the
> GPL because of that.

Thanks for the quick answer. By "code" above I think you are saying that linking 
in pure GPL (v2 or v3) such as the RTEMS add-on of GNU readline would make the 
whole package subject to GPL which is obviously true.

> I hope that is clear.
>> Thanks,
>> -gene

More information about the users mailing list