what happened to -g (4.7)

[Till Straumann]

Ralf Corsepius wrote:
> On Wed, 2005-11-16 at 19:34 -0800, Till Straumann wrote:
> 
>>RTEMS used to be built with debugging sections (-g)
>>-- after my last CVS update this is no longer true.
> 
> Yes, what you currently see is me gradually cleaning up
> configure-scripts and gradually removing historic artifacts having
> remained from RTEMS's makefile fragment-based configuration.
> 
> Though "-g" having gone list is an accident, this all currently is WIP,
> and with me not yet being sure were we will end up. At least, dropping
> it entirely and permanently (i.e. require users to explicitly specify
> it) would be a much cleaner and simpler approach than what we currently
> use.
> 
> 
>>What do I have to do to get -g back?
> 
> ATM, try CFLAGS_FOR_TARGET or adding -g to CFLAGS_OPTIMIZE_V.

Where do I set those (so that it gets added globally, ie, for all BSPs)?

Thanks
T.

> 
> 
>>Can't it be enabled by default (even if fully optimized)?
>>If no - at least in maintainer mode?
> 
> maintainer-mode has nothing to do with CFLAGS.
> 
> maintainer-mode is addressing disabling/enabling generation of generated
> code (Makefile.ins, aclocal.m4, configure, preinstall.am, rpcgen
> generated code).
> 
> Ralf
> 
> 
>