It doesn't happen in any performance critical path that
I've seen.
Sure it does, watch with strace what running a fairly involved GNU
configure script does. It also happens to be one of the best tests
for getting all the tlb and cache flushing correct during fork and
exit as well.
If you do fork and exit stupidly from a architecture specific MM layer
perspective, fork+exec("anything") will look even more stupid. To me
doing the former correctly goes hand in hand with doing the latter
correctly. Any time I've improved fork/exit numbers, the exec tests
showed improvements as well.
Later,
David S. Miller
davem@dm.cobaltmicro.com
-
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.tux.org/lkml/