[En-Nut-Discussion] Stellaris Licences, was: Re: Ethernut on TI's ...

Uwe Bonnes bon at elektron.ikp.physik.tu-darmstadt.de
Mon Oct 29 21:15:01 CET 2012

>>>>> "Philipp" == Philipp Burch <phip at hb9etc.ch> writes:

    Philipp> Hi Uwe, On 10/28/2012 06:27 PM, Uwe Bonnes wrote:
    >>>>>>> "Philipp" == Philipp Burch <phip at hb9etc.ch> writes:
    >>  ...
    Philipp> Then, another concern is the whole licensing shit. While
    Philipp> checking the patch above, I skimmed through TI's license file
    Philipp> for the CMSIS, located in nut/include/arch/cm3/lm3/vendor/. I
    Philipp> don't really understand it, but is it possible, that TI
    Philipp> basically prohibits the redistribution of these files? If so,
    Philipp> we would need to add another annoying hurdle for possible users
    Philipp> and require them to download and unpack the headers
    Philipp> themselves. I really hate it.
    >>  Huch,
    >> anything licensed under
    >> (lm3)nut/include/arch/cm3/lm3/vendor/license.html should never and
    >> probably will never go into Ethernut. It is in total contrast to
    >> (lm3)nut/include/arch/cm3/lm3/vendor/lm3s9b96.h Does above license
    >> apply to another group of files? Does one hand at TI eventually not
    >> know what the other hands does?

    Philipp> ok, I've had another look at all this. The situation is as
    Philipp> follows:

    Philipp> TI provides their "StellarisWare" distribution which contains -
    Philipp> Lots of examples - Their "driverlib" called collection of
    Philipp> high-level functions to access the processor peripherals - Some
    Philipp> third-party components - The "raw" header files, one of those
    Philipp> is lm3s9b96.h

    Philipp> Additionally, there are three license-related files which I've
    Philipp> uploaded to http://hb9etc.ch/ethernut/licenses/ for reference:
    Philipp> - license.html, this is the one also contained in my copy of
    Philipp> the source tree - MANIFEST.TXT containing a list of the
    Philipp> included components and their respective licenses -
    Philipp> manifest.html, which looks to have the same content as
    Philipp> MANIFEST.TXT

    Philipp> When scrolling to the "Stellaris Device Header Files", it reads
    Philipp> this:

    Philipp> -------- 8< --------- 8< ----------- Software Name: Stellaris
    Philipp> Device Header Files Version: 9107 License Type: TI BSD
    Philipp> Delivered As: Source Modified by TI: N/A Location: ./inc
    Philipp> Obtained From: TI -------- 8< --------- 8< -----------

    Philipp> Whatever "TI BSD" means...

I also just had another look in the Stellarisware directories and especially

For files in./driverlib things are consistant. The file talk are labelled
// This is part of revision 9107 of the Stellaris Peripheral Driver Library.
and ./MANIFEST.TXT are grants license to "Stellaris Peripheral Driver
Library" another time. However the headers under ./inc, like e.g.
inc/lm3s9b96.h talk about
// This is part of revision 9107 of the Stellaris Firmware Development Package.
and "Stellaris Firmware Development Package"  never mentions that name in
./MANIFEST.TXT. But in my opinion the disclaimer in the header files itsself
is enough to be sufficent liberal for inclusion.

    Philipp> Additionally, I've found TI's CMSIS files for their devices in
    Philipp> a different package. This is where the lm3s9b96_cmsis.h file
    Philipp> comes from (originally named LM3S9B96.h). This package does
    Philipp> only contain one licensing file, named "LICENSE.txt", which you
    Philipp> can find in the same folder as the others above. As far as I
    Philipp> can tell, this contains the same as the license.html file. So
    Philipp> this would probably render their CMSIS files useless as they
    Philipp> might not be used for almost anything.

Is that "other" source also "revision 9107". That would be strange!.


Uwe Bonnes                bon at elektron.ikp.physik.tu-darmstadt.de

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------

More information about the En-Nut-Discussion mailing list