Pavel Pisa commented on a discussion: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1209#note_149159


I agree with manual fixup. The original generator was horrible and extremely 
tedious work solution (manually assisted OCR from PDF files, formatting and 
text flow has changed from chapter to chapter) because original Ti headers has 
incompatible license so they cannot be used even as the source for the 
generator for RTEMS style aligned headers. The Ti contributed to help situation 
to release headers under acceptable license later. It is seem that in some 
section the first comment character has failed repeatedly from the windows for 
OCR.

>From the original workflow point, the right solution would be to update 
>definitions in the raw OCR generated file

https://github.com/AoLaD/rtems-tms570-utils/tree/master/headers/raw_files

But I expect, that even TMS570LC4357 support has been done by manual header 
files adjustments so reasonable regeneration is not possible anymore.

I think that the right way for today is to update generator to use CMSIS System 
View Description format(SVD) files, these are standard now, license should not 
be problem, they are in another language and are descriptive lists only. The 
example how to use them is for example there

https://github.com/robertlipe/svdtoheaders

But when the headers are regenerated (ideally by some unification from SVDs 
from all RTEMS supported and expected to be supported TMS570 family members) it 
would be necessary to check match with the current files.

So I am for manual concretions for now.

-- 
View it on GitLab: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1209#note_149159
You're receiving this email because of your account on gitlab.rtems.org.


_______________________________________________
bugs mailing list
[email protected]
http://lists.rtems.org/mailman/listinfo/bugs

Reply via email to