Re: [PATCH 6/7] mm/page_alloc: Remotely drain per-cpu lists

[Vlastimil Babka]

On 6/13/22 14:56, Mel Gorman wrote:
> From: Nicolas Saenz Julienne <nsaenzju@xxxxxxxxxx>
> 
> Some setups, notably NOHZ_FULL CPUs, are too busy to handle the per-cpu
> drain work queued by __drain_all_pages().  So introduce a new mechanism to
> remotely drain the per-cpu lists.  It is made possible by remotely locking
> 'struct per_cpu_pages' new per-cpu spinlocks.  A benefit of this new
> scheme is that drain operations are now migration safe.
> 
> There was no observed performance degradation vs.  the previous scheme.
> Both netperf and hackbench were run in parallel to triggering the
> __drain_all_pages(NULL, true) code path around ~100 times per second.  The
> new scheme performs a bit better (~5%), although the important point here
> is there are no performance regressions vs.  the previous mechanism.
> Per-cpu lists draining happens only in slow paths.
> 
> Minchan Kim tested this independently and reported;
> 
> 	My workload is not NOHZ CPUs but run apps under heavy memory
> 	pressure so they goes to direct reclaim and be stuck on
> 	drain_all_pages until work on workqueue run.
> 
> 	unit: nanosecond
> 	max(dur)        avg(dur)                count(dur)
> 	166713013       487511.77786438033      1283
> 
> 	From traces, system encountered the drain_all_pages 1283 times and
> 	worst case was 166ms and avg was 487us.
> 
> 	The other problem was alloc_contig_range in CMA. The PCP draining
> 	takes several hundred millisecond sometimes though there is no
> 	memory pressure or a few of pages to be migrated out but CPU were
> 	fully booked.
> 
> 	Your patch perfectly removed those wasted time.
> 
> Signed-off-by: Nicolas Saenz Julienne <nsaenzju@xxxxxxxxxx>
> Signed-off-by: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
> Tested-by: Minchan Kim <minchan@xxxxxxxxxx>
> Acked-by: Minchan Kim <minchan@xxxxxxxxxx>

Acked-by: Vlastimil Babka <vbabka@xxxxxxx>