Hugh Dickins reported the following
My tmpfs swapping load (tweaked to use huge pages more heavily
than in real life) is far from being a realistic load: but it was
notably slowed down by your throttling mods in 5.16-rc, and this
patch makes it well again - thanks.
But: it very quickly hit NULL pointer until I changed that last
line to
if (first_pgdat)
consider_reclaim_throttle(first_pgdat, sc);
The likely issue is that huge pages are a major component of the test
workload. When this is the case, first_pgdat may never get set if
compaction is ready to continue due to this check
if (IS_ENABLED(CONFIG_COMPACTION) &&
sc->order > PAGE_ALLOC_COSTLY_ORDER &&
compaction_ready(zone, sc)) {
sc->compaction_ready = true;
continue;
}
If this was true for every zone in the zonelist, first_pgdat would never
get set resulting in a NULL pointer exception.
This is a fix to the mmotm patch
mm-vmscan-reduce-throttling-due-to-a-failure-to-make-progress.patch
Reported-by: Hugh Dickins <[email protected]>
Signed-off-by: Mel Gorman <[email protected]>
---
mm/vmscan.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 4c4d5f6cd8a3..700434db5735 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -3530,7 +3530,8 @@ static void shrink_zones(struct zonelist *zonelist, struct scan_control *sc)
shrink_node(zone->zone_pgdat, sc);
}
- consider_reclaim_throttle(first_pgdat, sc);
+ if (first_pgdat)
+ consider_reclaim_throttle(first_pgdat, sc);
/*
* Restore to original mask to avoid the impact on the caller if we