!14823 mm：some shmem bugfix (ed17ddff) · Commits · EulixOS / Software / Kernel

Documentation/admin-guide/mm/transhuge.rst

+1 −1

Original line number	Diff line number	Diff line
		@@ -459,7 +459,7 @@ AnonHugePmdMapped).
		The number of file transparent huge pages mapped to userspace is available
		by reading ShmemPmdMapped and ShmemHugePages fields in ``/proc/meminfo``.
		To identify what applications are mapping file transparent huge pages, it
		is necessary to read ``/proc/PID/smaps`` and count the FileHugeMapped fields
		is necessary to read ``/proc/PID/smaps`` and count the FilePmdMapped fields
		for each mapping.

		Note that reading the smaps file is expensive and reading it

mm/shmem.c

+37 −39

Original line number	Diff line number	Diff line
		@@ -635,15 +635,14 @@ static const char *shmem_format_huge(int huge)
		#endif

		static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo,
		struct shrink_control *sc, unsigned long nr_to_split)
		struct shrink_control *sc, unsigned long nr_to_free)
		{
		LIST_HEAD(list), pos, next;
		LIST_HEAD(to_remove);
		struct inode *inode;
		struct shmem_inode_info *info;
		struct folio *folio;
		unsigned long batch = sc ? sc->nr_to_scan : 128;
		int split = 0;
		unsigned long split = 0, freed = 0;

		if (list_empty(&sbinfo->shrinklist))
		return SHRINK_STOP;
		@@ -661,13 +660,6 @@ static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo,
		goto next;
		}

		/* Check if there's anything to gain */
		if (round_up(inode->i_size, PAGE_SIZE) ==
		round_up(inode->i_size, HPAGE_PMD_SIZE)) {
		list_move(&info->shrinklist, &to_remove);
		goto next;
		}

		list_move(&info->shrinklist, &list);
		next:
		sbinfo->shrinklist_len--;
		@@ -676,34 +668,36 @@ static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo,
		}
		spin_unlock(&sbinfo->shrinklist_lock);

		list_for_each_safe(pos, next, &to_remove) {
		info = list_entry(pos, struct shmem_inode_info, shrinklist);
		inode = &info->vfs_inode;
		list_del_init(&info->shrinklist);
		iput(inode);
		}

		list_for_each_safe(pos, next, &list) {
		pgoff_t next, end;
		loff_t i_size;
		int ret;
		pgoff_t index;

		info = list_entry(pos, struct shmem_inode_info, shrinklist);
		inode = &info->vfs_inode;

		if (nr_to_split && split >= nr_to_split)
		if (nr_to_free && freed >= nr_to_free)
		goto move_back;

		index = (inode->i_size & HPAGE_PMD_MASK) >> PAGE_SHIFT;
		folio = filemap_get_folio(inode->i_mapping, index);
		if (IS_ERR(folio))
		i_size = i_size_read(inode);
		folio = filemap_get_entry(inode->i_mapping, i_size / PAGE_SIZE);
		if (!folio \|\| xa_is_value(folio))
		goto drop;

		/* No huge page at the end of the file: nothing to split */
		/* No large folio at the end of the file: nothing to split */
		if (!folio_test_large(folio)) {
		folio_put(folio);
		goto drop;
		}

		/* Check if there is anything to gain from splitting */
		next = folio_next_index(folio);
		end = shmem_fallocend(inode, DIV_ROUND_UP(i_size, PAGE_SIZE));
		if (end <= folio->index \|\| end >= next) {
		folio_put(folio);
		goto drop;
		}

		/*
		* Move the inode on the list back to shrinklist if we failed
		* to lock the page at this time.
		@@ -724,6 +718,7 @@ static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo,
		if (ret)
		goto move_back;

		freed += next - end;
		split++;
		drop:
		list_del_init(&info->shrinklist);
		@@ -768,7 +763,7 @@ static long shmem_unused_huge_count(struct super_block *sb,
		#define shmem_huge SHMEM_HUGE_DENY

		static unsigned long shmem_unused_huge_shrink(struct shmem_sb_info *sbinfo,
		struct shrink_control *sc, unsigned long nr_to_split)
		struct shrink_control *sc, unsigned long nr_to_free)
		{
		return 0;
		}
		@@ -781,6 +776,14 @@ static bool shmem_huge_global_enabled(struct inode *inode, pgoff_t index,
		}
		#endif /* CONFIG_TRANSPARENT_HUGEPAGE */

		static void shmem_update_stats(struct folio *folio, int nr_pages)
		{
		if (folio_test_pmd_mappable(folio))
		__lruvec_stat_mod_folio(folio, NR_SHMEM_THPS, nr_pages);
		__lruvec_stat_mod_folio(folio, NR_FILE_PAGES, nr_pages);
		__lruvec_stat_mod_folio(folio, NR_SHMEM, nr_pages);
		}

		/*
		* Somewhat like filemap_add_folio, but error if expected item has gone.
		*/
		@@ -815,10 +818,7 @@ static int shmem_add_to_page_cache(struct folio *folio,
		xas_store(&xas, folio);
		if (xas_error(&xas))
		goto unlock;
		if (folio_test_pmd_mappable(folio))
		__lruvec_stat_mod_folio(folio, NR_SHMEM_THPS, nr);
		__lruvec_stat_mod_folio(folio, NR_FILE_PAGES, nr);
		__lruvec_stat_mod_folio(folio, NR_SHMEM, nr);
		shmem_update_stats(folio, nr);
		shmem_reliable_folio_add(folio, nr);
		mapping->nrpages += nr;
		unlock:
		@@ -847,8 +847,7 @@ static void shmem_delete_from_page_cache(struct folio folio, void radswap)
		error = shmem_replace_entry(mapping, folio->index, folio, radswap);
		folio->mapping = NULL;
		mapping->nrpages -= nr;
		__lruvec_stat_mod_folio(folio, NR_FILE_PAGES, -nr);
		__lruvec_stat_mod_folio(folio, NR_SHMEM, -nr);
		shmem_update_stats(folio, -nr);
		shmem_reliable_folio_add(folio, -nr);
		xa_unlock_irq(&mapping->i_pages);
		folio_put_refs(folio, nr);
		@@ -1527,7 +1526,7 @@ static int shmem_writepage(struct page page, struct writeback_control wbc)
		!shmem_falloc->waitq &&
		index >= shmem_falloc->start &&
		index < shmem_falloc->next)
		shmem_falloc->nr_unswapped++;
		shmem_falloc->nr_unswapped += nr_pages;
		else
		shmem_falloc = NULL;
		spin_unlock(&inode->i_lock);
		@@ -1685,6 +1684,7 @@ unsigned long shmem_allowable_huge_orders(struct inode *inode,
		unsigned long mask = READ_ONCE(huge_shmem_orders_always);
		unsigned long within_size_orders = READ_ONCE(huge_shmem_orders_within_size);
		unsigned long vm_flags = vma ? vma->vm_flags : 0;
		pgoff_t aligned_index;
		bool global_huge;
		loff_t i_size;
		int order;
		@@ -1719,9 +1719,9 @@ unsigned long shmem_allowable_huge_orders(struct inode *inode,
		/* Allow mTHP that will be fully within i_size. */
		order = highest_order(within_size_orders);
		while (within_size_orders) {
		index = round_up(index + 1, order);
		aligned_index = round_up(index + 1, 1 << order);
		i_size = round_up(i_size_read(inode), PAGE_SIZE);
		if (i_size >> PAGE_SHIFT >= index) {
		if (i_size >> PAGE_SHIFT >= aligned_index) {
		mask \|= within_size_orders;
		break;
		}
		@@ -1872,7 +1872,7 @@ static struct folio shmem_alloc_and_add_folio(struct vm_fault vmf,
		* Try to reclaim some space by splitting a few
		* large folios beyond i_size on the filesystem.
		*/
		shmem_unused_huge_shrink(sbinfo, NULL, 2);
		shmem_unused_huge_shrink(sbinfo, NULL, pages);
		/*
		* And do a shmem_recalc_inode() to account for freed pages:
		* except our folio is there in cache, so not quite balanced.
		@@ -1973,11 +1973,9 @@ static int shmem_replace_folio(struct folio **foliop, gfp_t gfp,
		}
		if (!error) {
		mem_cgroup_replace_folio(old, new);
		__lruvec_stat_mod_folio(new, NR_FILE_PAGES, nr_pages);
		__lruvec_stat_mod_folio(new, NR_SHMEM, nr_pages);
		shmem_update_stats(new, nr_pages);
		shmem_reliable_folio_add(new, nr_pages);
		__lruvec_stat_mod_folio(old, NR_FILE_PAGES, -nr_pages);
		__lruvec_stat_mod_folio(old, NR_SHMEM, -nr_pages);
		shmem_update_stats(old, -nr_pages);
		shmem_reliable_folio_add(old, -nr_pages);
		}
		xa_unlock_irq(&swap_mapping->i_pages);
		@@ -2372,7 +2370,7 @@ static int shmem_get_folio_gfp(struct inode *inode, pgoff_t index,
		alloced = true;
		if (folio_test_large(folio) &&
		DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE) <
		folio_next_index(folio) - 1) {
		folio_next_index(folio)) {
		struct shmem_sb_info *sbinfo = SHMEM_SB(inode->i_sb);
		struct shmem_inode_info *info = SHMEM_I(inode);
		/*