Using memfd_pin_folios() will ensure that the pages are pinned
correctly using FOLL_PIN. And, this also ensures that we don't
accidentally break features such as memory hotunplug as it would
not allow pinning pages in the movable zone.
Using this new API also simplifies the code as we no longer have
to deal with extracting individual pages from their mappings or
handle shmem and hugetlb cases separately.
v2:
- Adjust to the change in signature of pin_user_pages_fd() by
passing in file * instead of fd.
v3:
- Limit the changes in this patch only to those that are required
for using pin_user_pages_fd()
- Slightly improve the commit message
v4:
- Adjust to the change in name of the API (memfd_pin_user_pages)
v5:
- Adjust to the changes in memfd_pin_folios which now populates
a list of folios and offsets
v6:
- Don't unnecessarily use folio_page() (Matthew)
- Pass [start, end] and max_folios to memfd_pin_folios()
- Create another temporary array to hold the folios returned by
memfd_pin_folios() as we populate ubuf->folios.
- Unpin the folios only once as memfd_pin_folios pins them once
v7:
- Use a list to track the folios that need to be unpinned
v8:
- Use unpin_folio() instead of unpin_user_page() to unpin folio
Cc: David Hildenbrand <da...@redhat.com>
Cc: Matthew Wilcox <wi...@infradead.org>
Cc: Daniel Vetter <daniel.vet...@ffwll.ch>
Cc: Mike Kravetz <mike.krav...@oracle.com>
Cc: Hugh Dickins <hu...@google.com>
Cc: Peter Xu <pet...@redhat.com>
Cc: Jason Gunthorpe <j...@nvidia.com>
Cc: Gerd Hoffmann <kra...@redhat.com>
Cc: Dongwon Kim <dongwon....@intel.com>
Cc: Junxiao Chang <junxiao.ch...@intel.com>
Signed-off-by: Vivek Kasireddy <vivek.kasire...@intel.com>
---
drivers/dma-buf/udmabuf.c | 153 +++++++++++++++++++-------------------
1 file changed, 78 insertions(+), 75 deletions(-)
diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c
index a8f3af61f7f2..afa8bfd2a2a9 100644
--- a/drivers/dma-buf/udmabuf.c
+++ b/drivers/dma-buf/udmabuf.c
@@ -30,6 +30,12 @@ struct udmabuf {
struct sg_table *sg;
struct miscdevice *device;
pgoff_t *offsets;
+ struct list_head unpin_list;
+};
+
+struct udmabuf_folio {
+ struct folio *folio;
+ struct list_head list;
};
static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf)
@@ -153,17 +159,43 @@ static void unmap_udmabuf(struct dma_buf_attachment *at,
return put_sg_table(at->dev, sg, direction);
}
+static void unpin_all_folios(struct list_head *unpin_list)
+{
+ struct udmabuf_folio *ubuf_folio;
+
+ while (!list_empty(unpin_list)) {
+ ubuf_folio = list_first_entry(unpin_list,
+ struct udmabuf_folio, list);
+ unpin_folio(ubuf_folio->folio);
+
+ list_del(&ubuf_folio->list);
+ kfree(ubuf_folio);
+ }
+}
+
+static int add_to_unpin_list(struct list_head *unpin_list,
+ struct folio *folio)
+{
+ struct udmabuf_folio *ubuf_folio;
+
+ ubuf_folio = kzalloc(sizeof(*ubuf_folio), GFP_KERNEL);
+ if (!ubuf_folio)
+ return -ENOMEM;
+
+ ubuf_folio->folio = folio;
+ list_add_tail(&ubuf_folio->list, unpin_list);
+ return 0;
+}
+
static void release_udmabuf(struct dma_buf *buf)
{
struct udmabuf *ubuf = buf->priv;
struct device *dev = ubuf->device->this_device;
- pgoff_t pg;
if (ubuf->sg)
put_sg_table(dev, ubuf->sg, DMA_BIDIRECTIONAL);
- for (pg = 0; pg < ubuf->pagecount; pg++)
- folio_put(ubuf->folios[pg]);
+ unpin_all_folios(&ubuf->unpin_list);
kfree(ubuf->offsets);
kfree(ubuf->folios);
kfree(ubuf);
@@ -218,64 +250,6 @@ static const struct dma_buf_ops udmabuf_ops = {
#define SEALS_WANTED (F_SEAL_SHRINK)
#define SEALS_DENIED (F_SEAL_WRITE)
-static int handle_hugetlb_pages(struct udmabuf *ubuf, struct file *memfd,
- pgoff_t offset, pgoff_t pgcnt,
- pgoff_t *pgbuf)
-{
- struct hstate *hpstate = hstate_file(memfd);
- pgoff_t mapidx = offset >> huge_page_shift(hpstate);
- pgoff_t subpgoff = (offset & ~huge_page_mask(hpstate)) >> PAGE_SHIFT;
- pgoff_t maxsubpgs = huge_page_size(hpstate) >> PAGE_SHIFT;
- struct folio *folio = NULL;
- pgoff_t pgidx;
-
- mapidx <<= huge_page_order(hpstate);
- for (pgidx = 0; pgidx < pgcnt; pgidx++) {
- if (!folio) {
- folio = __filemap_get_folio(memfd->f_mapping,
- mapidx,
- FGP_ACCESSED, 0);
- if (IS_ERR(folio))
- return PTR_ERR(folio);
- }
-
- folio_get(folio);
- ubuf->folios[*pgbuf] = folio;
- ubuf->offsets[*pgbuf] = subpgoff << PAGE_SHIFT;
- (*pgbuf)++;
- if (++subpgoff == maxsubpgs) {
- folio_put(folio);
- folio = NULL;
- subpgoff = 0;
- mapidx += pages_per_huge_page(hpstate);
- }
- }
-
- if (folio)
- folio_put(folio);
-
- return 0;
-}
-
-static int handle_shmem_pages(struct udmabuf *ubuf, struct file *memfd,
- pgoff_t offset, pgoff_t pgcnt,
- pgoff_t *pgbuf)
-{
- pgoff_t pgidx, pgoff = offset >> PAGE_SHIFT;
- struct folio *folio = NULL;
-
- for (pgidx = 0; pgidx < pgcnt; pgidx++) {
- folio = shmem_read_folio(memfd->f_mapping, pgoff + pgidx);
- if (IS_ERR(folio))
- return PTR_ERR(folio);
-
- ubuf->folios[*pgbuf] = folio;
- (*pgbuf)++;
- }
-
- return 0;
-}
-
static int check_memfd_seals(struct file *memfd)
{
int seals;
@@ -321,16 +295,19 @@ static long udmabuf_create(struct miscdevice *device,
struct udmabuf_create_list *head,
struct udmabuf_create_item *list)
{
- pgoff_t pgcnt, pgbuf = 0, pglimit;
+ pgoff_t pgoff, pgcnt, pglimit, pgbuf = 0;
+ long nr_folios, ret = -EINVAL;
struct file *memfd = NULL;
+ struct folio **folios;
struct udmabuf *ubuf;
- int ret = -EINVAL;
- u32 i, flags;
+ u32 i, j, k, flags;
+ loff_t end;
ubuf = kzalloc(sizeof(*ubuf), GFP_KERNEL);
if (!ubuf)
return -ENOMEM;
+ INIT_LIST_HEAD(&ubuf->unpin_list);
pglimit = (size_limit_mb * 1024 * 1024) >> PAGE_SHIFT;
for (i = 0; i < head->count; i++) {
if (!IS_ALIGNED(list[i].offset, PAGE_SIZE))
@@ -366,17 +343,44 @@ static long udmabuf_create(struct miscdevice *device,
goto err;
pgcnt = list[i].size >> PAGE_SHIFT;
- if (is_file_hugepages(memfd))
- ret = handle_hugetlb_pages(ubuf, memfd,
- list[i].offset,
- pgcnt, &pgbuf);
- else
- ret = handle_shmem_pages(ubuf, memfd,
- list[i].offset,
- pgcnt, &pgbuf);
- if (ret < 0)
+ folios = kmalloc_array(pgcnt, sizeof(*folios), GFP_KERNEL);
+ if (!folios) {
+ ret = -ENOMEM;
goto err;
+ }
+ end = list[i].offset + (pgcnt << PAGE_SHIFT) - 1;
+ ret = memfd_pin_folios(memfd, list[i].offset, end,
+ folios, pgcnt, &pgoff);
+ if (ret < 0) {
+ kfree(folios);
+ goto err;
+ }
+
+ nr_folios = ret;
+ pgoff >>= PAGE_SHIFT;
+ for (j = 0, k = 0; j < pgcnt; j++) {
+ ubuf->folios[pgbuf] = folios[k];
+ ubuf->offsets[pgbuf] = pgoff << PAGE_SHIFT;
+
+ if (j == 0 || ubuf->folios[pgbuf-1] != folios[k]) {
+ ret = add_to_unpin_list(&ubuf->unpin_list,
+ folios[k]);
+ if (ret < 0) {
+ kfree(folios);
+ goto err;
+ }
+ }
+
+ pgbuf++;
+ if (++pgoff == folio_nr_pages(folios[k])) {
+ pgoff = 0;
+ if (++k == nr_folios)
+ break;
+ }
+ }
+
+ kfree(folios);
fput(memfd);
}
@@ -388,10 +392,9 @@ static long udmabuf_create(struct miscdevice *device,
return ret;
err:
- while (pgbuf > 0)
- folio_put(ubuf->folios[--pgbuf]);
if (memfd)
fput(memfd);
+ unpin_all_folios(&ubuf->unpin_list);
kfree(ubuf->offsets);
kfree(ubuf->folios);
kfree(ubuf);
--
2.39.2
