On Thu, May 16, 2024 at 10:56 PM kernel test robot
<oliver.sang@xxxxxxxxx> wrote:
>
>
>
> Hello,
>
> kernel test robot noticed a -11.9% regression of will-it-scale.per_process_ops on:
>
>
> commit: 70a64b7919cbd6c12306051ff2825839a9d65605 ("memcg: dynamically allocate lruvec_stats")
> https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
I think we may want to go back to the approach of reordering the
indices to separate memcg and non-memcg stats. If we really want to
conserve the order in which the stats are exported to userspace, we
can use a translation table on the read path instead of the update
path.
