pppd namespace pollution

Till Straumann strauman at SLAC.Stanford.EDU
Fri Mar 21 23:08:00 UTC 2003


Aaron J. Grier wrote:
> On Fri, Mar 21, 2003 at 10:40:56PM +0100, Valette Eric wrote:
> 
>>>So Tills proposal really sounds interesting to me. Although it 
>>>should possibly be avoided to "rename" the offical calls (the 
>>>user's interface) to these libraries...
>>
>>I agree that if we can find a way to :
>>
>>	1) write reversible script to transform the sources from the 
>>	original package to the one used by rtems,
>>	2) write the reverse script,
>>	3) record both scripts in the rtems source tree,
>>	4) put more modules in library,
>>	5) manage the dependencies among libraries,
>>
>>This would be  a nearly perfect solution :-)
> 
> 
> how about weakening the global symbols?  would that be sufficient?
> 
> if it could be applied in a general fashion it could also be useful for
> BSPs.
> 

I had proposed that earlier but I don't really like it. Plus, Ralf
pointed out that it is a GNU/ELF specific solution (OTOH: are not
all BSPs GNU/ELF?). It only would make sense for 'replacement'
implementations, i.e. code that may not be implemented in a standard
library (such as strlcpy). It does not make sense, however for
symbols like 'error' or 'debug'.

-- Till




More information about the users mailing list