Re: [patch] Re: spin_unlock optimization(i386)
Alan Cox (alan@lxorguk.ukuu.org.uk)
Mon, 22 Nov 1999 09:54:36 +0000 (GMT)
> An rmb() will prevent reads that occur before it from being
> speculatively executed after it, and vice versa. rmb() expands to asm
> volatile ("lock; addl $0,0(%esp)" : : : "memory"). Being a locked
> operation it serves as a processor barrier for both reads and writes;
> being in cache, it is fast on a PPro. The "memory" constraint makes it
> a compiler memory barrier; register operations can still be reordered
> around it in theory. That's ok.
Unfortunately the lock addl turns into as big an overhead as the rest of the
spinlock clearing
> rmb() then "movl $0,%0".
We do seem to have some cases where we have implicit synchronization points
before an unlock where a movl would work (ie would could use some kind of
__spin_unlock(). Several drivers end outl(value, lp->base+offset); spin_unlock
-
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/