Darren Hart wrote:
> Sebastien Dugue wrote:
>>   Maybe a lot of the realtime tests could benefit from such an addition.
>> The criterion for PASS/FAIL should IMHO be configurable, it won't be the
>> same for an embedded 57 MHz ARM7 or a 10.25GHz Power9. It will also 
>> depend
>> on the RT goals the system has been designed for.
> 
> 
> Completely agree.  In fact, I would prefer that the tests don't even 
> have a criteria unless one is passed to them.  Or perhaps none at all, 
> and they should only report results (unless they are a functional sort 
> of test).  Then parsers can be written with a variety of user-generated 
> profiles that report on if the results meet the requirements of the 
> profile.
> 
> This would be a significant effort, and we should get as many people to 
> chime in as possible before we start any implementation in this direction.

I would imagine for all tests, a string of common args/criteria  to be 
used in a sigle option (ex: -o arg1=xx,arg2=yy,crit1=zz,...) or stored 
in a file. Each test would pick the args it needs.

Gilles.

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Reply via email to