mm/gup: introduce memfd_pin_folios() for pinning memfd folios

For drivers that would like to longterm-pin the folios associated with a
memfd, the memfd_pin_folios() API provides an option to not only pin the
folios via FOLL_PIN but also to check and migrate them if they reside in
movable zone or CMA block.  This API currently works with memfds but it
should work with any files that belong to either shmemfs or hugetlbfs. 
Files belonging to other filesystems are rejected for now.

The folios need to be located first before pinning them via FOLL_PIN.  If
they are found in the page cache, they can be immediately pinned. 
Otherwise, they need to be allocated using the filesystem specific APIs
and then pinned.

[akpm@linux-foundation.org: improve the CONFIG_MMU=n situation, per SeongJae]
[vivek.kasireddy@intel.com: return -EINVAL if the end offset is greater than the size of memfd]
  Link: https://lkml.kernel.org/r/IA0PR11MB71850525CBC7D541CAB45DF1F8DB2@IA0PR11MB7185.namprd11.prod.outlook.com
Link: https://lkml.kernel.org/r/20240624063952.1572359-4-vivek.kasireddy@intel.com
Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com> (v2)
Reviewed-by: David Hildenbrand <david@redhat.com> (v3)
Reviewed-by: Christoph Hellwig <hch@lst.de> (v6)
Acked-by: Dave Airlie <airlied@redhat.com>
Acked-by: Gerd Hoffmann <kraxel@redhat.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Hugh Dickins <hughd@google.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Dongwon Kim <dongwon.kim@intel.com>
Cc: Junxiao Chang <junxiao.chang@intel.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

2 files changed, 184 insertions, 0 deletions

diff --git a/mm/gup.c b/mm/gup.c
index d98bb199241e..85d45ec57f7c 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -5,6 +5,7 @@
 #include <linux/spinlock.h>
 
 #include <linux/mm.h>
+#include <linux/memfd.h>
 #include <linux/memremap.h>
 #include <linux/pagemap.h>
 #include <linux/rmap.h>
@@ -17,6 +18,7 @@
 #include <linux/hugetlb.h>
 #include <linux/migrate.h>
 #include <linux/mm_inline.h>
+#include <linux/pagevec.h>
 #include <linux/sched/mm.h>
 #include <linux/shmem_fs.h>
 
@@ -3764,3 +3766,140 @@ long pin_user_pages_unlocked(unsigned long start, unsigned long nr_pages,
 				     &locked, gup_flags);
 }
 EXPORT_SYMBOL(pin_user_pages_unlocked);
+
+/**
+ * memfd_pin_folios() - pin folios associated with a memfd
+ * @memfd:      the memfd whose folios are to be pinned
+ * @start:      the first memfd offset
+ * @end:        the last memfd offset (inclusive)
+ * @folios:     array that receives pointers to the folios pinned
+ * @max_folios: maximum number of entries in @folios
+ * @offset:     the offset into the first folio
+ *
+ * Attempt to pin folios associated with a memfd in the contiguous range
+ * [start, end]. Given that a memfd is either backed by shmem or hugetlb,
+ * the folios can either be found in the page cache or need to be allocated
+ * if necessary. Once the folios are located, they are all pinned via
+ * FOLL_PIN and @offset is populatedwith the offset into the first folio.
+ * And, eventually, these pinned folios must be released either using
+ * unpin_folios() or unpin_folio().
+ *
+ * It must be noted that the folios may be pinned for an indefinite amount
+ * of time. And, in most cases, the duration of time they may stay pinned
+ * would be controlled by the userspace. This behavior is effectively the
+ * same as using FOLL_LONGTERM with other GUP APIs.
+ *
+ * Returns number of folios pinned, which could be less than @max_folios
+ * as it depends on the folio sizes that cover the range [start, end].
+ * If no folios were pinned, it returns -errno.
+ */
+long memfd_pin_folios(struct file *memfd, loff_t start, loff_t end,
+		      struct folio **folios, unsigned int max_folios,
+		      pgoff_t *offset)
+{
+	unsigned int flags, nr_folios, nr_found;
+	unsigned int i, pgshift = PAGE_SHIFT;
+	pgoff_t start_idx, end_idx, next_idx;
+	struct folio *folio = NULL;
+	struct folio_batch fbatch;
+	struct hstate *h;
+	long ret = -EINVAL;
+
+	if (start < 0 || start > end || !max_folios)
+		return -EINVAL;
+
+	if (!memfd)
+		return -EINVAL;
+
+	if (!shmem_file(memfd) && !is_file_hugepages(memfd))
+		return -EINVAL;
+
+	if (end >= i_size_read(file_inode(memfd)))
+		return -EINVAL;
+
+	if (is_file_hugepages(memfd)) {
+		h = hstate_file(memfd);
+		pgshift = huge_page_shift(h);
+	}
+
+	flags = memalloc_pin_save();
+	do {
+		nr_folios = 0;
+		start_idx = start >> pgshift;
+		end_idx = end >> pgshift;
+		if (is_file_hugepages(memfd)) {
+			start_idx <<= huge_page_order(h);
+			end_idx <<= huge_page_order(h);
+		}
+
+		folio_batch_init(&fbatch);
+		while (start_idx <= end_idx && nr_folios < max_folios) {
+			/*
+			 * In most cases, we should be able to find the folios
+			 * in the page cache. If we cannot find them for some
+			 * reason, we try to allocate them and add them to the
+			 * page cache.
+			 */
+			nr_found = filemap_get_folios_contig(memfd->f_mapping,
+							     &start_idx,
+							     end_idx,
+							     &fbatch);
+			if (folio) {
+				folio_put(folio);
+				folio = NULL;
+			}
+
+			next_idx = 0;
+			for (i = 0; i < nr_found; i++) {
+				/*
+				 * As there can be multiple entries for a
+				 * given folio in the batch returned by
+				 * filemap_get_folios_contig(), the below
+				 * check is to ensure that we pin and return a
+				 * unique set of folios between start and end.
+				 */
+				if (next_idx &&
+				    next_idx != folio_index(fbatch.folios[i]))
+					continue;
+
+				folio = page_folio(&fbatch.folios[i]->page);
+
+				if (try_grab_folio(folio, 1, FOLL_PIN)) {
+					folio_batch_release(&fbatch);
+					ret = -EINVAL;
+					goto err;
+				}
+
+				if (nr_folios == 0)
+					*offset = offset_in_folio(folio, start);
+
+				folios[nr_folios] = folio;
+				next_idx = folio_next_index(folio);
+				if (++nr_folios == max_folios)
+					break;
+			}
+
+			folio = NULL;
+			folio_batch_release(&fbatch);
+			if (!nr_found) {
+				folio = memfd_alloc_folio(memfd, start_idx);
+				if (IS_ERR(folio)) {
+					ret = PTR_ERR(folio);
+					if (ret != -EEXIST)
+						goto err;
+				}
+			}
+		}
+
+		ret = check_and_migrate_movable_folios(nr_folios, folios);
+	} while (ret == -EAGAIN);
+
+	memalloc_pin_restore(flags);
+	return ret ? ret : nr_folios;
+err:
+	memalloc_pin_restore(flags);
+	unpin_folios(folios, nr_folios);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(memfd_pin_folios);
diff --git a/mm/memfd.c b/mm/memfd.c
index 7d8d3ab3fa37..e7b7c5294d59 100644
--- a/mm/memfd.c
+++ b/mm/memfd.c
@@ -60,6 +60,51 @@ static void memfd_tag_pins(struct xa_state *xas)
 }
 
 /*
+ * This is a helper function used by memfd_pin_user_pages() in GUP (gup.c).
+ * It is mainly called to allocate a folio in a memfd when the caller
+ * (memfd_pin_folios()) cannot find a folio in the page cache at a given
+ * index in the mapping.
+ */
+struct folio *memfd_alloc_folio(struct file *memfd, pgoff_t idx)
+{
+#ifdef CONFIG_HUGETLB_PAGE
+	struct folio *folio;
+	gfp_t gfp_mask;
+	int err;
+
+	if (is_file_hugepages(memfd)) {
+		/*
+		 * The folio would most likely be accessed by a DMA driver,
+		 * therefore, we have zone memory constraints where we can
+		 * alloc from. Also, the folio will be pinned for an indefinite
+		 * amount of time, so it is not expected to be migrated away.
+		 */
+		gfp_mask = htlb_alloc_mask(hstate_file(memfd));
+		gfp_mask &= ~(__GFP_HIGHMEM | __GFP_MOVABLE);
+
+		folio = alloc_hugetlb_folio_nodemask(hstate_file(memfd),
+						     numa_node_id(),
+						     NULL,
+						     gfp_mask,
+						     false);
+		if (folio && folio_try_get(folio)) {
+			err = hugetlb_add_to_page_cache(folio,
+							memfd->f_mapping,
+							idx);
+			if (err) {
+				folio_put(folio);
+				free_huge_folio(folio);
+				return ERR_PTR(err);
+			}
+			return folio;
+		}
+		return ERR_PTR(-ENOMEM);
+	}
+#endif
+	return shmem_read_folio(memfd->f_mapping, idx);
+}
+
+/*
  * Setting SEAL_WRITE requires us to verify there's no pending writer. However,
  * via get_user_pages(), drivers might have some pending I/O without any active
  * user-space mappings (eg., direct-IO, AIO). Therefore, we look at all folios