On 06/24/2003 05:51 AM, Donn Washburn wrote:
Well then how come 8.0 with a 2.95.3 compiler worked so well and 8.2 fails so often? INFO: /usr/src/linux is linked correctly and mrproper was run before make cloneconfig. The Makefile option "cloneconfig" is not found in a generic kernel either (so it has been modified also).
cloneconfig is a SuSE addition, not a modification. If i understand correctly, 3.3 is more strict in its checking than 3.2, which was more strict and newer than 2,95.3. Some commands have changed, but programs that worked before need changes to compile under 3.3, because now the programs will fail instead of lots of Warnings. I see this as progress (as was explained here recently regarding the kernel), as the resulting compilation is better. In your case, I would guess, the code for the k6 still needs 'fixed', not that it wasn't broken before, it would just give a warning before, now it will fail and force 'the programmer' to actually fix the problems. It may initially cause more work but will eventually create cleaner code. Not being a programmer, this is just what i have picked up from this list, but I have compiled several programs on 8.2, not all without problems, but so far, all fixable problems that were probably programming errors/shortcuts that are no longer tolerated. Just my 2 cents. ;-) If I am way off, I am sure those more knowledgeable will correct the errors. -- Joe Morris New Tribes Mission Email Address: Joe_Morris@ntm.org Web Address: http://www.mydestiny.net/~joe_morris Registered Linux user 231871 God said, I AM that I AM. I say, by the grace of God, I am what I am.