[RFC PATCH 4/4] arm64:numa: adding numa support for arm64 platforms.

On Mon, Oct 20, 2014 at 8:00 PM, Steve Capper [off-list ref] wrote:

On Mon, Oct 20, 2014 at 03:25:56PM +0100, Steve Capper wrote:

quoted

On Fri, Oct 17, 2014 at 10:49:56PM +0530, Ganapatrao Kulkarni wrote:

[...]

quoted

         /*Proximity Distance matrix for 4Node system

quoted

                       <from-node to-node distance>
        */
         node-matrix=    <0 0 10>,
                               <0 1 20>,
                               <0 2 30>,
                               <0 3 10>,
                               <1 0 20>,
                                <1 1 10>,
                               <1 2 30>,
                              <1 3 10>,
                              <2 0 30>,
                              <2 1 20>,
                              <2 2 10>,
                             <2 3 10>,
                            <3 0 10>,
                            <3 1 20>,
                            <3 2 30>,
                            <3 3 10>,
 }

Hi Ganapat,
The above caught my attention.

For a 4-node system do we not need 16 distances; the implication of that
would be that the distance between node A-B could be different from the
distance between B-A? Also the distance from a node to itself could be
safely assumed to be zero?

I think we should have a symmetric matrix with zero-diagonals so strictly
only seven values would need specifying for a 4-node system.

Thanks Stave for the comments.
I too thought initially to take the assumption like
distance B-A is same as A-B and A-A is always defined to LOCAL.
However this example is an ideal case which will define all 16
distances for 4 node system.
In actual DT, we can skip to provide A-A and B-A, In kernel
implementation we will use these common/generic assumptions
to derive missing distances.

s/seven/six/

I really need to learn how to count.... :-/

quoted

Cheers,
--
Steve

thanks
Ganapat