On Mon 05-11-18 19:13:48, Kuo-Hsin Yang wrote: > The i915 driver uses shmemfs to allocate backing storage for gem > objects. These shmemfs pages can be pinned (increased ref count) by > shmem_read_mapping_page_gfp(). When a lot of pages are pinned, vmscan > wastes a lot of time scanning these pinned pages. In some extreme case, > all pages in the inactive anon lru are pinned, and only the inactive > anon lru is scanned due to inactive_ratio, the system cannot swap and > invokes the oom-killer. Mark these pinned pages as unevictable to speed > up vmscan. > > Export pagevec API check_move_unevictable_pages(). Thanks for reworking the patch. This looks much more to my taste. At least the mm part. I haven't really looked at the the drm part. Just a nit below > This patch was inspired by Chris Wilson's change [1]. > > [1]: https://patchwork.kernel.org/patch/9768741/ I would recommend using msg-id based url. > Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> > Cc: Michal Hocko <mhocko@xxxxxxxx> > Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx> > Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx> > Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> > Cc: Dave Hansen <dave.hansen@xxxxxxxxx> > Signed-off-by: Kuo-Hsin Yang <vovoy@xxxxxxxxxxxx> other than that Acked-by: Michal Hocko <mhocko@xxxxxxxx> [...] > @@ -4184,15 +4185,13 @@ int page_evictable(struct page *page) > > #ifdef CONFIG_SHMEM > /** > - * check_move_unevictable_pages - check pages for evictability and move to appropriate zone lru list > - * @pages: array of pages to check > - * @nr_pages: number of pages to check > + * check_move_unevictable_pages - move evictable pages to appropriate evictable > + * lru lists I am not sure this is an improvement. I would just keep the original wording. It is not great either but the explicit note about check for evictability sounds like a better fit to me. > + * @pvec: pagevec with pages to check > * > - * Checks pages for evictability and moves them to the appropriate lru list. > - * > - * This function is only used for SysV IPC SHM_UNLOCK. > + * This function is only used to move shmem pages. I do not really see anything that would be shmem specific here. We can use this function for any LRU pages unless I am missing something obscure. I would just drop the last sentence. A note that this function should be only used for LRU pages would be nice. > */ > -void check_move_unevictable_pages(struct page **pages, int nr_pages) > +void check_move_unevictable_pages(struct pagevec *pvec) > { > struct lruvec *lruvec; > struct pglist_data *pgdat = NULL; > @@ -4200,8 +4199,8 @@ void check_move_unevictable_pages(struct page **pages, int nr_pages) > int pgrescued = 0; > int i; > > - for (i = 0; i < nr_pages; i++) { > - struct page *page = pages[i]; > + for (i = 0; i < pvec->nr; i++) { > + struct page *page = pvec->pages[i]; > struct pglist_data *pagepgdat = page_pgdat(page); > > pgscanned++; > @@ -4233,4 +4232,5 @@ void check_move_unevictable_pages(struct page **pages, int nr_pages) > spin_unlock_irq(&pgdat->lru_lock); > } > } > +EXPORT_SYMBOL(check_move_unevictable_pages); > #endif /* CONFIG_SHMEM */ > -- > 2.19.1.930.g4563a0d9d0-goog > -- Michal Hocko SUSE Labs _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
