On Jul 9, 2013, at 11:00 , Joel Sherrill <[email protected]> wrote:
> On 7/9/2013 9:50 AM, Sebastian Huber wrote:
>> On 07/09/2013 02:21 PM, Vipul Nayyar wrote:
>>> Seems like a valid point to me, that you got there.
>>>
>>> Sebastian, What do you suggest ? Simply remove the 1 from '#define
>>> USE_INTERRUPTS 1' or change #if defined() to simply #if ?
>> I suggest to use one of the two methods suggested by Peter. The #if one
>> seems
>> to be more bullet proof, but I think in the RTEMS sources the #ifdef ones are
>> prevailing.
>>
> I think we need to be heading to the #if style.
>
> As I mentioned, Chris and I recently discussed some cases where we
> think BSPs may not be testing bspopts.h (e.g. configure.ac) constants
> correctly.
I agree that RTEMS should move toward the method that I prefer.
But to minimize the number of changes that Vipul makes that need to be reviewed
(by both himself and others) I think he should just change the "#define OPTION
1" to "#define OPTION". The RTEMS style guide can be updated to prefer the
"#if" over "#ifdef" directive and additional changes can be made in the future.
If the additional changes fall within the scope of what Vipul is working on now
("Updated Legacy code") then I definitely think it should be changed to be done
the way that I like.
Peter
-----------------
Peter Dufault
HD Associates, Inc. Software and System Engineering
_______________________________________________
rtems-devel mailing list
[email protected]
http://www.rtems.org/mailman/listinfo/rtems-devel
