Re: ordered memory access

Jes Sorensen (Jes.Sorensen@cern.ch)
30 Sep 1999 14:39:59 +0200


>>>>> "Manfred" == Manfred Spraul <manfreds@colorfullife.com> writes:

Manfred> Jes Sorensen wrote:
>> atomic_foo() should be ordered, at least that the idea behind it as
>> far as I understand.

Manfred> That might be the idea, but that's not the implementation:
Manfred> Pentium III documentation: (24319201.pdf)

Right, intuitively I would expect an atomic operation to guarantee
read/write ordering though, however if this is not what we want, we
should at least document it.

Manfred> I've read that rmb() is expensive on RISC architectures, what
Manfred> about adding "get_?mb()" functions?

rmb() is still less expensive than plain mb().

Jes

-
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/