Re: test_and_set_bit() not atomic forever? [cli/sti in char/vt.c [patch]]
Alan Cox (alan@lxorguk.ukuu.org.uk)
Mon, 1 Jun 1998 16:16:32 +0100 (BST)
> > processors. With a single processor, the net effect of the extra
> > unneeded lock prefix is a double performance loss: 1) a memory cycle
> > loss to read each lock prefix 2) since locked instruction cycles are
> > noncacheable, this can result in the loss of _many_ clock cycles.
>
> Indeed, especially those aiming to have one kernel run on everything.
>
> Perhaps a fixup table to convert all the `lock's to `nop's at
> initialisation time?
It isnt worth the effort, with all the other SMP entanglements lock is
a minor issue. Just compile the right kernel
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu