2020년 7월 2일 (목) 오전 3:02, Vlastimil Babka <vba...@suse.cz>님이 작성: > > On 6/17/20 7:26 AM, js1...@gmail.com wrote: > > From: Joonsoo Kim <iamjoonsoo....@lge.com> > > Hi, how about a more descriptive subject, such as
Hello, > mm/vmscan: add new anonymous pages to inactive LRU list This patch does two things to implement workingset protection. - add new anonymous pages to inactive LRU list - check two reference to activate So, I think that the current subject is more descriptive for this patch. > > In current implementation, newly created or swap-in anonymous page > > is started on active list. Growing active list results in rebalancing > > active/inactive list so old pages on active list are demoted to inactive > > list. Hence, the page on active list isn't protected at all. > > > > Following is an example of this situation. > > > > Assume that 50 hot pages on active list. Numbers denote the number of > > pages on active/inactive list (active | inactive). > > > > 1. 50 hot pages on active list > > 50(h) | 0 > > > > 2. workload: 50 newly created (used-once) pages > > 50(uo) | 50(h) > > > > 3. workload: another 50 newly created (used-once) pages > > 50(uo) | 50(uo), swap-out 50(h) > > > > This patch tries to fix this issue. > > Like as file LRU, newly created or swap-in anonymous pages will be > > inserted to the inactive list. They are promoted to active list if > > enough reference happens. This simple modification changes the above > > example as following. > > > > 1. 50 hot pages on active list > > 50(h) | 0 > > > > 2. workload: 50 newly created (used-once) pages > > 50(h) | 50(uo) > > > > 3. workload: another 50 newly created (used-once) pages > > 50(h) | 50(uo), swap-out 50(uo) > > > > As you can see, hot pages on active list would be protected. > > > > Note that, this implementation has a drawback that the page cannot > > be promoted and will be swapped-out if re-access interval is greater than > > the size of inactive list but less than the size of total(active+inactive). > > To solve this potential issue, following patch will apply workingset > > detection that is applied to file LRU some day before. > > detection similar to the one that's already applied to file LRU. Will change! > > v6: Before this patch, all anon pages (inactive + active) are considered > > as workingset. However, with this patch, only active pages are considered > > as workingset. So, file refault formula which uses the number of all > > anon pages is changed to use only the number of active anon pages. > > a "v6" note is more suitable for a diffstat area than commit log, but it's > good > to mention this so drop the 'v6:'? Okay. > > Acked-by: Johannes Weiner <han...@cmpxchg.org> > > Signed-off-by: Joonsoo Kim <iamjoonsoo....@lge.com> > > Acked-by: Vlastimil Babka <vba...@suse.cz> Thanks! > One more nit below. > > > --- a/mm/swap.c > > +++ b/mm/swap.c > > @@ -476,23 +476,24 @@ void lru_cache_add(struct page *page) > > EXPORT_SYMBOL(lru_cache_add); > > > > /** > > - * lru_cache_add_active_or_unevictable > > + * lru_cache_add_inactive_or_unevictable > > * @page: the page to be added to LRU > > * @vma: vma in which page is mapped for determining reclaimability > > * > > - * Place @page on the active or unevictable LRU list, depending on its > > + * Place @page on the inactive or unevictable LRU list, depending on its > > * evictability. Note that if the page is not evictable, it goes > > * directly back onto it's zone's unevictable list, it does NOT use a > > * per cpu pagevec. > > */ > > -void lru_cache_add_active_or_unevictable(struct page *page, > > +void lru_cache_add_inactive_or_unevictable(struct page *page, > > struct vm_area_struct *vma) > > { > > + bool unevictable; > > + > > VM_BUG_ON_PAGE(PageLRU(page), page); > > > > - if (likely((vma->vm_flags & (VM_LOCKED | VM_SPECIAL)) != VM_LOCKED)) > > - SetPageActive(page); > > - else if (!TestSetPageMlocked(page)) { > > + unevictable = (vma->vm_flags & (VM_LOCKED | VM_SPECIAL)) == VM_LOCKED; > > + if (unevictable && !TestSetPageMlocked(page)) { > > I guess this could be "if (unlikely(unevictable) && ..." to match the previous > if (likely(evictable)) else ... I will fix it, too. Thanks.