Re: MADV_HUGEPAGE vs. NUMA semantic (was: Re: [LKP] [mm] ac5b2c1891: vm-scalability.throughput -61.3% regression)

[David Rientjes]

On Fri, 7 Dec 2018, Vlastimil Babka wrote:

> >> But *that* in turn makes for other possible questions:
> >>
> >>  - if the reason we couldn't get a local hugepage is that we're simply
> >> out of local memory (huge *or* small), then maybe a remote hugepage is
> >> better.
> >>
> >>    Note that this now implies that the choice can be an issue of "did
> >> the hugepage allocation fail due to fragmentation, or due to the node
> >> being low of memory"
> > How exactly do you tell? Many systems are simply low on memory due to
> > caching. A clean pagecache is quite cheap to reclaim but it can be more
> > expensive to fault in. Do we consider it to be a viable target?
> 
> Compaction can report if it failed (more precisely: was skipped) due to
> low memory, or for other reasons. It doesn't distinguish how easily
> reclaimable is the memory, but I don't think we should reclaim anything
> (see below).
> 

Note that just reclaiming when the order-0 watermark in 
__compaction_suitable() fails is unfortunately not always sufficient: it 
needs to be accessible to isolate_freepages().  For order-9 memory, it's 
possible for isolate_migratepages_block() to skip over a top of free pages 
that were just reclaimed if there are unmovable pages preventing the 
entire pageblock from being freed.