Re: [RFC][PATCH 2/5] [PATCH 2/5] proc: introduce /proc/PID/idle_bitmap

[Dave Hansen]

On 09/01/2018 04:28 AM, Fengguang Wu wrote:
> To walk 1TB memory of 4k active pages, it costs 2s vs 15s system
> time to scan the per-task/global idle bitmaps.

To me, that says this interface simply won't work on large systems.  2s
and 15s are both simply unacceptably long.

> OTOH, the per-task idle bitmap is not suitable in some situations:
> 
> - not accurate for shared pages
> - don't work with non-mapped file pages
> - don't perform well for sparse page tables (pointed out by Huang Ying)

OK, so we have a new ABI that doesn't work on large systems, consumes
lots of time and resources to query and isn't suitable in quite a few
situations.