On Tue, Nov 26, 2013 at 11:55:58PM -0500, Jon Masters wrote:
> On 11/11/2013 12:13 PM, Paul E. McKenney wrote:
> > Hello, Tony,
> >
> > Does Itanium permit speculative stores? For example, on Itanium what are
> > the permitted outcomes of the following litm
On Tue, Nov 26, 2013 at 11:55:58PM -0500, Jon Masters wrote:
On 11/11/2013 12:13 PM, Paul E. McKenney wrote:
Hello, Tony,
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero
On 11/11/2013 12:13 PM, Paul E. McKenney wrote:
> Hello, Tony,
>
> Does Itanium permit speculative stores? For example, on Itanium what are
> the permitted outcomes of the following litmus test, where both x and y
> are initially zero?
>
> CPU 0
On 11/11/2013 12:13 PM, Paul E. McKenney wrote:
Hello, Tony,
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero?
CPU 0 CPU 1
r1
On Tue, Nov 12, 2013 at 07:26:17PM +0100, Peter Zijlstra wrote:
> On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
> > > Does Itanium permit speculative stores? For example, on Itanium what are
> > > the permitted outcomes of the following litmus tes
On Tue, Nov 12, 2013 at 06:46:20PM +, Luck, Tony wrote:
> > So the point we're having a discussion on is if any architecture has
> > visible speculative STORES and if there's an architecture that doesn't
> > have control dependencies.
> >
> > On the visible speculative STORES; can, if in the
On Tue, Nov 12, 2013 at 06:46:20PM +, Luck, Tony wrote:
> > So the point we're having a discussion on is if any architecture has
> > visible speculative STORES and if there's an architecture that doesn't
> > have control dependencies.
> >
> > On the visible speculative STORES; can, if in the
> So the point we're having a discussion on is if any architecture has
> visible speculative STORES and if there's an architecture that doesn't
> have control dependencies.
>
> On the visible speculative STORES; can, if in the above example we have
> regular loads/stores:
>
> LOAD r1, x
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
> The "ACCESS_ONCE" macro casts to volatile - which will make gcc generate
> ordered "ld.acq" and "st.rel" instructions for your code snippets. So I think
> you should be fine.
Hurm.. so:
+#define smp_store_release(p, v)
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
> > Does Itanium permit speculative stores? For example, on Itanium what are
> > the permitted outcomes of the following litmus test, where both x and y
> > are initially zero?
>
> We have a complier visi
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
> > Does Itanium permit speculative stores? For example, on Itanium what are
> > the permitted outcomes of the following litmus test, where both x and y
> > are initially zero?
>
> We have a complier visi
> Does Itanium permit speculative stores? For example, on Itanium what are
> the permitted outcomes of the following litmus test, where both x and y
> are initially zero?
We have a complier visible speculative read via the "ld.s" and "chk"
instructions. But
there i
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero?
We have a complier visible speculative read via the ld.s and chk
instructions. But
there is no speculative write (st.s
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero?
We have a complier visible speculative read via the ld.s
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero?
We have a complier visible speculative read via the ld.s
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
The ACCESS_ONCE macro casts to volatile - which will make gcc generate
ordered ld.acq and st.rel instructions for your code snippets. So I think
you should be fine.
Hurm.. so:
+#define smp_store_release(p, v)
So the point we're having a discussion on is if any architecture has
visible speculative STORES and if there's an architecture that doesn't
have control dependencies.
On the visible speculative STORES; can, if in the above example we have
regular loads/stores:
LOAD r1, x
On Tue, Nov 12, 2013 at 06:46:20PM +, Luck, Tony wrote:
So the point we're having a discussion on is if any architecture has
visible speculative STORES and if there's an architecture that doesn't
have control dependencies.
On the visible speculative STORES; can, if in the above
On Tue, Nov 12, 2013 at 06:46:20PM +, Luck, Tony wrote:
So the point we're having a discussion on is if any architecture has
visible speculative STORES and if there's an architecture that doesn't
have control dependencies.
On the visible speculative STORES; can, if in the above
On Tue, Nov 12, 2013 at 07:26:17PM +0100, Peter Zijlstra wrote:
On Tue, Nov 12, 2013 at 06:00:26PM +, Luck, Tony wrote:
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero
Hello, Tony,
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero?
CPU 0 CPU 1
r1 = ACCESS_ONCE(x);r2 = ACCESS_ONCE(y
Hello, Tony,
Does Itanium permit speculative stores? For example, on Itanium what are
the permitted outcomes of the following litmus test, where both x and y
are initially zero?
CPU 0 CPU 1
r1 = ACCESS_ONCE(x);r2 = ACCESS_ONCE(y
22 matches
Mail list logo