Re: [PATCH 1/2] sched/wait: Break up long wake list walk

Linus Torvalds writes: (Summary) On Fri, Aug 18, 2017 at 5:23 AM, Mel Gorman <mgorman@techsingularity.net> wrote:

[...]

HPAGE_PMD_ORDER);
That can't make any difference. We already have:
That can't make any difference. We already have:

  #define GFP_TRANSHUGE_LIGHT     ((GFP_HIGHUSER_MOVABLE | __GFP_COMP | \
                         __GFP_NOMEMALLOC |

It *might* be simply that we have a shit-ton of threads, and the
thread that holds the page lock for migration is just preempted out.