On 12/17/2013 7:27 AM, Sebastian Huber wrote:
> On 2013-12-17 14:20, Jennifer Averett wrote:
>> I started out with it as one, but because of the way the tests run
>> a lot of data comes out (one line for each comparison).  I split them
>> up so each test doesn't output so many screens of data.  If I put them
>> back as one, would it be better only print at the top of the test (ex.
>> AND test) then print details upon a failure within that (this doesn't
>> match the way other tests work though)
> 
> The only interesting output is the "END OF TEST" message, since this checked 
> by 
> the RTEMS testing script.  I case of failure you get an assert message, this 
> should be enough.
> 

Depends on what you mean by interesting. The intent
was to note what the test was doing before the asserts.

Combining the tests also has a tendency to give us huge tests
which don't run on all targets.  I don't know in this case
how large the resulting combined test would be.

Since puts/printf are already wrapped by macros to control
whether the output is buffered or not, it would be easy to
make all the tests silent except for begin/end messages. I
was already beginning to consider a macro for "begin" and
"end" test rather than depend on people attempting to manually
make them consistent.


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherr...@oarcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985
_______________________________________________
rtems-devel mailing list
rtems-devel@rtems.org
http://www.rtems.org/mailman/listinfo/rtems-devel

Reply via email to