It appears in 2.2 (from looking at the code in kernel/softirq.c and
includ/asm-alpha, anyway) that bottom-halves are run with a global lock
such that any CPU may get the lock to run the bh's and no other CPU may run
bottom halves until the first has finished and released the lock. If this
is true, it would imply to mean that if I have queue-management code which
is known to only run from bottom-halves (in net_bh() and from a timer, for
example), I do not need any spinlocks... Of course they're always going to
be cheap if they aren't being used, so for future kernel evolution it might
make sense to use them, but I'd like to know if my current understanding is
correct.
-bp
-- # bryan at terran dot org # http://www.terran.org/~bryan
- 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/