From: Gregor M. <Gre...@gm...> - 2002-08-04 12:21:04
|
wael oraiby wrote: > hi , > > I dunno if you have this code , it should compile > fine and run tasking , it'snt much tho... > > hope it's the right file this time ;) > > kintaro > > > > __________________________________________________ > Do You Yahoo!? > Yahoo! Health - Feel better, live better > http://health.yahoo.com It is the right file this time :). The code is newer than anything I've seen so far. The multitasking demo behaves fine in bochs and on real hardware, although the output is a little bit different. I suspect that it's because of different simulated performance. This code actually outdates the CVS version. However, I think that some pieces of code should be reorganized: - The kfree() function currently takes two parameters, a pointer and the size of that memory block. This second parameter shouldn't be there as it requires the other kernel code to keep track of the size of all allocations they made, which is quite a bad thing. The block's size could be stored in a header for each block that is invisible to code outside of the memory manager. However, such a header could easily be corrupted by some misbehaving code. - The multitasking code should be moved out of main.c. task.c could be a new home for all this code. I can do this on my own - and if I do so, I'll send you the new versions of these files. This brings me to a another topic: Do you have trouble accessing the CVS archive at SourceForge? If so, we will need to find another solution to keep our code changes synchronized with less effort. I'm shure that we can move the CVS archive to a server without SSL if that would help you. Gregor -- ***************************************************** * Gregor Mueckl Gre...@gm... * * * * The ChallengeOS project: * * http://challengeos.sourceforge.net * ***************************************************** * Math problems? * * Call 1-800-[(10x)(13i)^2]-[sin(xy)/2.362x]. * ***************************************************** |