[En-Nut-Discussion] test cases for Nut/OS

[Michael Gong]

Hi, Matthias,

----- Original Message ----- 
From: "Matthias Ringwald" <mringwal at inf.ethz.ch>
To: "Ethernut User Chat (English)" <en-nut-discussion at egnite.de>
Sent: Tuesday, February 06, 2007 8:54 AM
Subject: Re: [En-Nut-Discussion] test cases for Nut/OS


> Hi
>
>>> Michael Gong schrieb:
>>>> Is there a test suite existed for Nut/OS so that I could run it
>>>> before
>>>> and after any changes to Nut/OS, in order to make sure the change
>>>> will
>>>> not cause regression ?
>>>>
>>>
>>> Unfortunately not. I discussed this long time ago with Matthias
>>> Ringwald
>>> and he suggested to create a testsuite, which will at least check 
>>> all
>>> kernel functions. No idea, whether the ETH Zurich has something done
>>> here.
>>>
>>
>> Does the test suite not exist at all or it is not publicly available 
>> ?
>
> There is none. We do have an automatic build system which builds
> nut/os and some apps after each change in our or nut's CVS.
>
> some while ago we added the nut tracer, some macros which allow
> to do very precise (some uS) timestamping of core nut/os functions
> and even measure latencies. but this is probably only for special
> testing of the os and not really suited for app/os regression test

Thanks for the reply.

Since you mentioned nut tracer, you might be the right person to ask. My 
research project involves trying to refactor the Nut tracing capability 
by using AOP (aspect-oriented-programming). But before doing that, I 
need to know how to use the Nut tracer.

So my question is what are the steps for a user application to invoke 
the tracing functionality inside Nut/OS ? The user application might be 
just a simple program, like the helloworld.


>> I
>> am wondering how Nut/OS developers test their code .
>
> printf.. :)

Cool :-)

Thanks.

Mike


>
>
> cheers
> Matthias
> _______________________________________________
> http://lists.egnite.de/mailman/listinfo/en-nut-discussion
>