Chris Johns commented on a discussion: 
https://gitlab.rtems.org/rtems/rtos/rtems-examples/-/issues/2#note_135020


Is the code generated with coverage instrumented to handle sampling or is the 
generated code same as the code without the option present? If it is different 
I can see a situation where a user is building RTEMS for coverage testing and 
that setting is accidentally merged and production code is not what it should 
be. At the moment there is a clear indication coverage is enabled, a linker 
error. I think it is right to look for a solution and fixing this situation 
however we should address in some way how a user can see the build is not a 
production standard build. If we capture and hold the build state somewhere it 
then becomes the user's responsibility to deal with managing that data.

I am also looking ahead a little to RTEMS dependent libraries, for example an 
external HAL package. In this case there will be a library that needs to flow 
through transparently to user builds. May be we can avoid a fragmented approach.

-- 
View it on GitLab: 
https://gitlab.rtems.org/rtems/rtos/rtems-examples/-/issues/2#note_135020
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