David, maybe the difference is that when the stack allocation fails, it
doesn't lock up the whole machine in a endless loop? Think about it for a
moment.
And yes, it wasn't "slab" that locked the machine. But it was certainly
slab that made it almost impossible to see from the sources that it was
going to do a multi-page-order allocation and that it was a lot more
dangerous than it used to be in 2.0.x..
That code _worked_ in 2.0.x. Not because the 2.0.x memory management was
any better than the current one, like some people have tried to maintain.
But because of the current slab code.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html