Re: [patch 2/2] mm, compaction: persistently skip hugetlbfs pageblocks

[David Rientjes]

On Fri, 1 Sep 2017, Vlastimil Babka wrote:

> The pageblock_skip_persistent() function checks for HugeTLB pages of pageblock
> order. When clearing pageblock skip bits for compaction, the bits are not
> cleared for such pageblocks, because they cannot contain base pages suitable
> for migration, nor free pages to use as migration targets.
> 
> This optimization can be simply extended to all compound pages of order equal
> or larger than pageblock order, because migrating such pages (if they support
> it) cannot help sub-pageblock fragmentation. This includes THP's and also
> gigantic HugeTLB pages, which the current implementation doesn't persistently
> skip due to a strict pageblock_order equality check and not recognizing tail
> pages.
> 
> Additionally, this patch removes the pageblock_skip_persistent() calls from
> migration and free scanner, since the generic compound page treatment together
> with update_pageblock_skip() call will also lead to pageblocks starting with a
> large enough compound page being immediately marked for skipping, which then
> becomes persistent.
> 

As mentioned in my other two emails, I'm not sure that persistently 
skipping thp memory is necessary and I disagree that we should not be 
persistently skipping pageblocks when cc->ignore_skip_hint is true.