Re: Thread-private mappings and graphics (was Re: Per-Processor Data Page)

Ralf Baechle (ralf@uni-koblenz.de)
Mon, 20 Dec 1999 01:24:46 +0100


On Thu, Dec 16, 1999 at 04:40:09PM -0800, Linus Torvalds wrote:

> Note that historically atomic locks have been EXPENSIVE. I think that was
> true on MIPS too, even with the load-locked thing, just because with the
> weak cache coherency the thing was initially done on the _bus_, not inside
> the CPU. But happily that time is long gone, and it's not coming back.

ll/sc instructions on MIPS are as expensive as normal loads / stores.

(The pre-ll/sc thing SGI used in their R3000 based SMPs was horribly
expensive, special purpose hardware that was of course different from other
MIPS vendors' solutions and apparently at times needed *miliseconds* to do
an atomic operation old SGIers told me ...)

Ralf

-
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/