Merge request https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1208 
was reviewed by Joel Sherrill

--
  
Joel Sherrill started a new discussion on 
cpukit/include/rtems/confdefs/inittask.h: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1208#note_148622

 > - * warning: comparison of function 'Init' not equal to a null pointer is 
 > always
 > - * true [-Wtautological-pointer-compare]
 > +  * _CONFIGURE_ASSERT_NOT_NULL is used to validate function pointers at 
 > compile time.

Your added comments are WAY over the accepted line limit.

--
  
Joel Sherrill started a new discussion on 
cpukit/include/rtems/confdefs/inittask.h: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1208#note_148623

 > +  * Those warnings are therefore suppressed locally because they are a 
 > by-product of the assertion, not a real defect.
 > +  * Additionally, -Wpragmas is suppressed since some toolchains may not 
 > support every warning option named below,
 > +  * which maintains compatibility across different toolchains without 
 > causing unnecessary warnings.

There is also a check in _cpukit/include/rtems/confdefs/threads.h_ to ensure 
that at least one style of initialisation task/thread is defined. You have 
three options and you must use one of them.

--
  
Joel Sherrill started a new discussion on 
cpukit/include/rtems/confdefs/inittask.h: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1208#note_148624

 > - * _CONFIGURE_ASSERT_NOT_NULL() below:
 > - *
 > - * warning: the address of 'void Init()' will never be NULL [-Waddress]

The same set of checks need to be done for the POSIX Initialization thread 
entry point. See _cpukit/include/rtems/confdefs/initthreads.h_ and add a 
similar set of checks.


-- 
View it on GitLab: 
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/1208
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