Re: [PATCH net-next 0/9] mlx4: order-0 allocations and page recycling

On Tue, 2017-02-07 at 08:26 -0800, Eric Dumazet wrote:

In most cases, using exactly 12 bytes per slot would allow better

typo : 24 bytes on 64bit arches.

(Instead of 128 bytes with the current implementation)

packing. Only one cpu is using this area, no need to force strange
alignments, for the sake of avoiding a multiply !