Re: [PATCH] docs/memory-barriers.txt: Enforce heavy ordering for port I/O accesses

[Andrea Parri]

On Mon, Nov 26, 2018 at 04:52:14PM +0000, Will Deacon wrote:
> David Laight explains:
> 
>   | A long time ago there was a document from Intel that said that
>   | inb/outb weren't necessarily synchronised wrt memory accesses.
>   | (Might be P-pro era). However no processors actually behaved that
>   | way and more recent docs say that inb/outb are fully ordered.

No intention to diminish David Laight's authority of course, but I would
have really appreciated a reference to these "recent docs" (section, pg.
or the like, especially if a reference manual...) here...


> 
> This also reflects the situation on other architectures, the the port
> accessor macros tend to be implemented in terms of readX/writeX.
> 
> Update Documentation/memory-barriers.txt to reflect reality.

..., IOW, what do you mean by "reality"?

> 
> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx>
> Cc: Arnd Bergmann <arnd@xxxxxxxx>
> Cc: David Laight <David.Laight@xxxxxxxxxx>
> Cc: Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Cc: "Paul E. McKenney" <paulmck@xxxxxxxxxxxxxxxxxx>
> Signed-off-by: Will Deacon <will.deacon@xxxxxxx>

Please Cc me on future patches to memory-barriers.txt (can not speak for
my co-maintainers, but I'm inclined to say that get_maintainers.pl knows
better...).

  Andrea


> ---
> 
> Just remembered I had this patch kicking around in my tree...
> 
>  Documentation/memory-barriers.txt | 6 ++----
>  1 file changed, 2 insertions(+), 4 deletions(-)
> 
> diff --git a/Documentation/memory-barriers.txt b/Documentation/memory-barriers.txt
> index c1d913944ad8..0c34c5dac138 100644
> --- a/Documentation/memory-barriers.txt
> +++ b/Documentation/memory-barriers.txt
> @@ -2619,10 +2619,8 @@ functions:
>       intermediary bridges (such as the PCI host bridge) may not fully honour
>       that.
>  
> -     They are guaranteed to be fully ordered with respect to each other.
> -
> -     They are not guaranteed to be fully ordered with respect to other types of
> -     memory and I/O operation.
> +     They are guaranteed to be fully ordered with respect to each other and
> +     also with respect to other types of memory and I/O operation.
>  
>   (*) readX(), writeX():
>  
> -- 
> 2.1.4
>