> > > > You already must not place any data structures into WC memory --- for
> > > > example, spinlocks wouldn't work there.
> > >
> > > What do you mean "already"?
> >
> > I mean "in current kernel" (I checked it in 2.6.22)
>
> Ahh, that's not "current kernel", though ;)
>
>
You already must not place any data structures into WC memory --- for
example, spinlocks wouldn't work there.
What do you mean already?
I mean in current kernel (I checked it in 2.6.22)
Ahh, that's not current kernel, though ;)
4071c718555d955a35e9651f77086096ad87d498
On Wed, Oct 17, 2007 at 01:51:17PM +0800, Herbert Xu wrote:
> Nick Piggin <[EMAIL PROTECTED]> wrote:
> >
> > Also, for non-wb memory. I don't think the Intel document referenced
> > says anything about this, but the AMD document says that loads can pass
> > loads (page 8, rule b).
> >
> > This is
On Wed, Oct 17, 2007 at 02:30:32AM +0200, Mikulas Patocka wrote:
> > > You already must not place any data structures into WC memory --- for
> > > example, spinlocks wouldn't work there.
> >
> > What do you mean "already"?
>
> I mean "in current kernel" (I checked it in 2.6.22)
Ahh, that's not
On Wed, Oct 17, 2007 at 02:30:32AM +0200, Mikulas Patocka wrote:
You already must not place any data structures into WC memory --- for
example, spinlocks wouldn't work there.
What do you mean already?
I mean in current kernel (I checked it in 2.6.22)
Ahh, that's not current kernel,
On Wed, Oct 17, 2007 at 01:51:17PM +0800, Herbert Xu wrote:
Nick Piggin [EMAIL PROTECTED] wrote:
Also, for non-wb memory. I don't think the Intel document referenced
says anything about this, but the AMD document says that loads can pass
loads (page 8, rule b).
This is why our rmb()
Nick Piggin <[EMAIL PROTECTED]> wrote:
>
> Also, for non-wb memory. I don't think the Intel document referenced
> says anything about this, but the AMD document says that loads can pass
> loads (page 8, rule b).
>
> This is why our rmb() is still an lfence.
BTW, Xen (in particular, the code in
> > You already must not place any data structures into WC memory --- for
> > example, spinlocks wouldn't work there.
>
> What do you mean "already"?
I mean "in current kernel" (I checked it in 2.6.22)
> If we already have drivers loading data from
> WC memory, then rmb() needs to order them,
On Wed, Oct 17, 2007 at 01:05:16AM +0200, Mikulas Patocka wrote:
> > > I see, AMD says that WC memory loads can be out-of-order.
> > >
> > > There is very little usability to it --- framebuffer and AGP aperture is
> > > the only piece of memory that is WC and no kernel structures are placed
> >
> > I see, AMD says that WC memory loads can be out-of-order.
> >
> > There is very little usability to it --- framebuffer and AGP aperture is
> > the only piece of memory that is WC and no kernel structures are placed
> > there, so it is possible to remove that lfence.
>
> No. In Linux
On Tue, Oct 16, 2007 at 12:33:54PM +0200, Mikulas Patocka wrote:
>
>
> On Tue, 16 Oct 2007, Nick Piggin wrote:
>
> > > > The cpus also have an explicit set of instructions that deliberately do
> > > > unordered stores/loads, and s/lfence etc are mostly designed for those.
> > >
> > > I know
On Tue, 16 Oct 2007, Nick Piggin wrote:
> > > The cpus also have an explicit set of instructions that deliberately do
> > > unordered stores/loads, and s/lfence etc are mostly designed for those.
> >
> > I know about unordered stores (movnti & similar) --- they basically use
> >
On Mon, 15 Oct 2007, H. Peter Anvin wrote:
> Mikulas Patocka wrote:
> >
> > I know about unordered stores (movnti & similar) --- they basically use
> > write-combining method on memory that is normally write-back --- and they
> > need sfence. But which one instruction does unordered load and
On Mon, 15 Oct 2007, H. Peter Anvin wrote:
Mikulas Patocka wrote:
I know about unordered stores (movnti similar) --- they basically use
write-combining method on memory that is normally write-back --- and they
need sfence. But which one instruction does unordered load and needs
On Tue, 16 Oct 2007, Nick Piggin wrote:
The cpus also have an explicit set of instructions that deliberately do
unordered stores/loads, and s/lfence etc are mostly designed for those.
I know about unordered stores (movnti similar) --- they basically use
write-combining method on
On Tue, Oct 16, 2007 at 12:33:54PM +0200, Mikulas Patocka wrote:
On Tue, 16 Oct 2007, Nick Piggin wrote:
The cpus also have an explicit set of instructions that deliberately do
unordered stores/loads, and s/lfence etc are mostly designed for those.
I know about unordered
I see, AMD says that WC memory loads can be out-of-order.
There is very little usability to it --- framebuffer and AGP aperture is
the only piece of memory that is WC and no kernel structures are placed
there, so it is possible to remove that lfence.
No. In Linux kernel, rmb()
On Wed, Oct 17, 2007 at 01:05:16AM +0200, Mikulas Patocka wrote:
I see, AMD says that WC memory loads can be out-of-order.
There is very little usability to it --- framebuffer and AGP aperture is
the only piece of memory that is WC and no kernel structures are placed
there, so it
You already must not place any data structures into WC memory --- for
example, spinlocks wouldn't work there.
What do you mean already?
I mean in current kernel (I checked it in 2.6.22)
If we already have drivers loading data from
WC memory, then rmb() needs to order them, whether or
Nick Piggin [EMAIL PROTECTED] wrote:
Also, for non-wb memory. I don't think the Intel document referenced
says anything about this, but the AMD document says that loads can pass
loads (page 8, rule b).
This is why our rmb() is still an lfence.
BTW, Xen (in particular, the code in
On Tue, Oct 16, 2007 at 12:08:01AM +0200, Mikulas Patocka wrote:
> > On Mon, 15 Oct 2007 22:47:42 +0200 (CEST)
> > Mikulas Patocka <[EMAIL PROTECTED]> wrote:
> >
> > > > According to latest memory ordering specification documents from
> > > > Intel and AMD, both manufacturers are committed to
Mikulas Patocka wrote:
I know about unordered stores (movnti & similar) --- they basically use
write-combining method on memory that is normally write-back --- and they
need sfence. But which one instruction does unordered load and needs
lefence?
PREFETCHNTA.
-hpa
-
To
> On Mon, 15 Oct 2007 22:47:42 +0200 (CEST)
> Mikulas Patocka <[EMAIL PROTECTED]> wrote:
>
> > > According to latest memory ordering specification documents from
> > > Intel and AMD, both manufacturers are committed to in-order loads
> > > from cacheable memory for the x86 architecture. Hence,
On Mon, 15 Oct 2007 22:47:42 +0200 (CEST)
Mikulas Patocka <[EMAIL PROTECTED]> wrote:
> > According to latest memory ordering specification documents from
> > Intel and AMD, both manufacturers are committed to in-order loads
> > from cacheable memory for the x86 architecture. Hence, smp_rmb()
> >
> According to latest memory ordering specification documents from Intel
> and AMD, both manufacturers are committed to in-order loads from
> cacheable memory for the x86 architecture. Hence, smp_rmb() may be a
> simple barrier.
>
>
According to latest memory ordering specification documents from Intel
and AMD, both manufacturers are committed to in-order loads from
cacheable memory for the x86 architecture. Hence, smp_rmb() may be a
simple barrier.
http://developer.intel.com/products/processor/manuals/318147.pdf
On Mon, 15 Oct 2007 22:47:42 +0200 (CEST)
Mikulas Patocka [EMAIL PROTECTED] wrote:
According to latest memory ordering specification documents from
Intel and AMD, both manufacturers are committed to in-order loads
from cacheable memory for the x86 architecture. Hence, smp_rmb()
may be a
On Mon, 15 Oct 2007 22:47:42 +0200 (CEST)
Mikulas Patocka [EMAIL PROTECTED] wrote:
According to latest memory ordering specification documents from
Intel and AMD, both manufacturers are committed to in-order loads
from cacheable memory for the x86 architecture. Hence, smp_rmb()
may
Mikulas Patocka wrote:
I know about unordered stores (movnti similar) --- they basically use
write-combining method on memory that is normally write-back --- and they
need sfence. But which one instruction does unordered load and needs
lefence?
PREFETCHNTA.
-hpa
-
To
On Tue, Oct 16, 2007 at 12:08:01AM +0200, Mikulas Patocka wrote:
On Mon, 15 Oct 2007 22:47:42 +0200 (CEST)
Mikulas Patocka [EMAIL PROTECTED] wrote:
According to latest memory ordering specification documents from
Intel and AMD, both manufacturers are committed to in-order loads
30 matches
Mail list logo