On Mon, Mar 26, 2007 at 01:07:11PM -0700, Paul E. McKenney wrote:
> > > > Does everybody agree on these semantics, though? At least David
> > > > seems to think that mb/rmb/wmb aren't required to order normal
> > > > memory accesses against each other..
> > >
> > > Not on UP. On SMP, ordering
On Mon, Mar 26, 2007 at 10:46:39AM +0200, Lennert Buytenhek wrote:
> On Sun, Mar 25, 2007 at 08:24:18PM -0700, Paul E. McKenney wrote:
>
> > > > > > [ background: On ARM, SMP synchronisation does need barriers but
> > > > > > device
> > > > > > synchronisation does not. The question is that
On Mon, Mar 26, 2007 at 10:46:39AM +0200, Lennert Buytenhek wrote:
On Sun, Mar 25, 2007 at 08:24:18PM -0700, Paul E. McKenney wrote:
[ background: On ARM, SMP synchronisation does need barriers but
device
synchronisation does not. The question is that given this,
On Mon, Mar 26, 2007 at 01:07:11PM -0700, Paul E. McKenney wrote:
Does everybody agree on these semantics, though? At least David
seems to think that mb/rmb/wmb aren't required to order normal
memory accesses against each other..
Not on UP. On SMP, ordering is (almost
Benjamin Herrenschmidt <[EMAIL PROTECTED]> wrote:
> Hrm... I'm not sure I like the io_* name, I think it's even more
> confusing, people will never know when to use what ...
I'd've thought it more obvious, but given there are several types of I/O, some
of which might require different barriering
Lennert Buytenhek <[EMAIL PROTECTED]> wrote:
> Does everybody agree on these semantics, though? At least David seems
> to think that mb/rmb/wmb aren't required to order normal memory accesses
> against each other..
Ummm... I've just realised that your statement here is ambiguous. When you
say
On Sun, Mar 25, 2007 at 08:24:18PM -0700, Paul E. McKenney wrote:
> > > > > [ background: On ARM, SMP synchronisation does need barriers but
> > > > > device
> > > > > synchronisation does not. The question is that given this, whether
> > > > > mb() and friends can be NOPs on ARM or not
On Sun, Mar 25, 2007 at 08:24:18PM -0700, Paul E. McKenney wrote:
[ background: On ARM, SMP synchronisation does need barriers but
device
synchronisation does not. The question is that given this, whether
mb() and friends can be NOPs on ARM or not (i.e. whether mb() is
Lennert Buytenhek [EMAIL PROTECTED] wrote:
Does everybody agree on these semantics, though? At least David seems
to think that mb/rmb/wmb aren't required to order normal memory accesses
against each other..
Ummm... I've just realised that your statement here is ambiguous. When you
say
Benjamin Herrenschmidt [EMAIL PROTECTED] wrote:
Hrm... I'm not sure I like the io_* name, I think it's even more
confusing, people will never know when to use what ...
I'd've thought it more obvious, but given there are several types of I/O, some
of which might require different barriering to
On Sun, Mar 25, 2007 at 11:38:43PM +0200, Lennert Buytenhek wrote:
> On Sun, Mar 25, 2007 at 02:15:42PM -0700, Paul E. McKenney wrote:
>
> > > > [ background: On ARM, SMP synchronisation does need barriers but device
> > > > synchronisation does not. The question is that given this, whether
>
On Fri, Mar 23, 2007 at 01:43:53PM +, David Howells wrote:
>
> [Resend - this time with a comma in the addresses, not a dot]
>
> Lennert Buytenhek <[EMAIL PROTECTED]> wrote:
>
> > [ background: On ARM, SMP synchronisation does need barriers but device
> > synchronisation does not. The
On Sun, Mar 25, 2007 at 02:15:42PM -0700, Paul E. McKenney wrote:
> > > [ background: On ARM, SMP synchronisation does need barriers but device
> > > synchronisation does not. The question is that given this, whether
> > > mb() and friends can be NOPs on ARM or not (i.e. whether mb() is
> >
On Sun, Mar 25, 2007 at 02:15:42PM -0700, Paul E. McKenney wrote:
[ background: On ARM, SMP synchronisation does need barriers but device
synchronisation does not. The question is that given this, whether
mb() and friends can be NOPs on ARM or not (i.e. whether mb() is
supposed
On Fri, Mar 23, 2007 at 01:43:53PM +, David Howells wrote:
[Resend - this time with a comma in the addresses, not a dot]
Lennert Buytenhek [EMAIL PROTECTED] wrote:
[ background: On ARM, SMP synchronisation does need barriers but device
synchronisation does not. The question is
On Sun, Mar 25, 2007 at 11:38:43PM +0200, Lennert Buytenhek wrote:
On Sun, Mar 25, 2007 at 02:15:42PM -0700, Paul E. McKenney wrote:
[ background: On ARM, SMP synchronisation does need barriers but device
synchronisation does not. The question is that given this, whether
mb()
On Fri, 2007-03-23 at 13:43 +, David Howells wrote:
> [Resend - this time with a comma in the addresses, not a dot]
>
> Lennert Buytenhek <[EMAIL PROTECTED]> wrote:
>
> > [ background: On ARM, SMP synchronisation does need barriers but device
> > synchronisation does not. The question is
On Fri, 2007-03-23 at 13:43 +, David Howells wrote:
[Resend - this time with a comma in the addresses, not a dot]
Lennert Buytenhek [EMAIL PROTECTED] wrote:
[ background: On ARM, SMP synchronisation does need barriers but device
synchronisation does not. The question is that given
On Fri, Mar 23, 2007 at 01:43:53PM +, David Howells wrote:
> > [ background: On ARM, SMP synchronisation does need barriers but device
> > synchronisation does not. The question is that given this, whether
> > mb() and friends can be NOPs on ARM or not (i.e. whether mb() is
> >
[Resend - this time with a comma in the addresses, not a dot]
Lennert Buytenhek <[EMAIL PROTECTED]> wrote:
> [ background: On ARM, SMP synchronisation does need barriers but device
> synchronisation does not. The question is that given this, whether
> mb() and friends can be NOPs on ARM or
[Resend - this time with a comma in the addresses, not a dot]
Lennert Buytenhek [EMAIL PROTECTED] wrote:
[ background: On ARM, SMP synchronisation does need barriers but device
synchronisation does not. The question is that given this, whether
mb() and friends can be NOPs on ARM or not
On Fri, Mar 23, 2007 at 01:43:53PM +, David Howells wrote:
[ background: On ARM, SMP synchronisation does need barriers but device
synchronisation does not. The question is that given this, whether
mb() and friends can be NOPs on ARM or not (i.e. whether mb() is
supposed to
22 matches
Mail list logo
