Merge tag 'vfs-6.11-rc6.fixes' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs

Pull vfs fixes from Christian Brauner:
 "VFS:

   - Ensure that backing files uses file->f_ops->splice_write() for
     splice

  netfs:

   - Revert the removal of PG_private_2 from netfs_release_folio() as
     cephfs still relies on this

   - When AS_RELEASE_ALWAYS is set on a mapping the folio needs to
     always be invalidated during truncation

   - Fix losing untruncated data in a folio by making letting
     netfs_release_folio() return false if the folio is dirty

   - Fix trimming of streaming-write folios in netfs_inval_folio()

   - Reset iterator before retrying a short read

   - Fix interaction of streaming writes with zero-point tracker

  afs:

   - During truncation afs currently calls truncate_setsize() which sets
     i_size, expands the pagecache and truncates it. The first two
     operations aren't needed because they will have already been done.
     So call truncate_pagecache() instead and skip the redundant parts

  overlayfs:

   - Fix checking of the number of allowed lower layers so 500 layers
     can actually be used instead of just 499

   - Add missing '\n' to pr_err() output

   - Pass string to ovl_parse_layer() and thus allow it to be used for
     Opt_lowerdir as well

  pidfd:

   - Revert blocking the creation of pidfds for kthread as apparently
     userspace relies on this. Specifically, it breaks systemd during
     shutdown

  romfs:

   - Fix romfs_read_folio() to use the correct offset with
     folio_zero_tail()"

* tag 'vfs-6.11-rc6.fixes' of gitolite.kernel.org:pub/scm/linux/kernel/git/vfs/vfs:
  netfs: Fix interaction of streaming writes with zero-point tracker
  netfs: Fix missing iterator reset on retry of short read
  netfs: Fix trimming of streaming-write folios in netfs_inval_folio()
  netfs: Fix netfs_release_folio() to say no if folio dirty
  afs: Fix post-setattr file edit to do truncation correctly
  mm: Fix missing folio invalidation calls during truncation
  ovl: ovl_parse_param_lowerdir: Add missed '\n' for pr_err
  ovl: fix wrong lowerdir number check for parameter Opt_lowerdir
  ovl: pass string to ovl_parse_layer()
  backing-file: convert to using fops->splice_write
  Revert "pidfd: prevent creation of pidfds for kthreads"
  romfs: fix romfs_read_folio()
  netfs, ceph: Partially revert "netfs: Replace PG_fscache by setting folio->private and marking dirty"

1 files changed, 45 insertions, 15 deletions

diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c
index 83e644bd518f..c1f321cf5999 100644
--- a/fs/netfs/misc.c
+++ b/fs/netfs/misc.c
@@ -97,10 +97,22 @@ EXPORT_SYMBOL(netfs_clear_inode_writeback);
 void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length)
 {
 	struct netfs_folio *finfo;
+	struct netfs_inode *ctx = netfs_inode(folio_inode(folio));
 	size_t flen = folio_size(folio);
 
 	_enter("{%lx},%zx,%zx", folio->index, offset, length);
 
+	if (offset == 0 && length == flen) {
+		unsigned long long i_size = i_size_read(&ctx->inode);
+		unsigned long long fpos = folio_pos(folio), end;
+
+		end = umin(fpos + flen, i_size);
+		if (fpos < i_size && end > ctx->zero_point)
+			ctx->zero_point = end;
+	}
+
+	folio_wait_private_2(folio); /* [DEPRECATED] */
+
 	if (!folio_test_private(folio))
 		return;
 
@@ -113,18 +125,34 @@ void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length)
 		/* We have a partially uptodate page from a streaming write. */
 		unsigned int fstart = finfo->dirty_offset;
 		unsigned int fend = fstart + finfo->dirty_len;
-		unsigned int end = offset + length;
+		unsigned int iend = offset + length;
 
 		if (offset >= fend)
 			return;
-		if (end <= fstart)
+		if (iend <= fstart)
+			return;
+
+		/* The invalidation region overlaps the data.  If the region
+		 * covers the start of the data, we either move along the start
+		 * or just erase the data entirely.
+		 */
+		if (offset <= fstart) {
+			if (iend >= fend)
+				goto erase_completely;
+			/* Move the start of the data. */
+			finfo->dirty_len = fend - iend;
+			finfo->dirty_offset = offset;
+			return;
+		}
+
+		/* Reduce the length of the data if the invalidation region
+		 * covers the tail part.
+		 */
+		if (iend >= fend) {
+			finfo->dirty_len = offset - fstart;
 			return;
-		if (offset <= fstart && end >= fend)
-			goto erase_completely;
-		if (offset <= fstart && end > fstart)
-			goto reduce_len;
-		if (offset > fstart && end >= fend)
-			goto move_start;
+		}
+
 		/* A partial write was split.  The caller has already zeroed
 		 * it, so just absorb the hole.
 		 */
@@ -137,12 +165,6 @@ erase_completely:
 	folio_clear_uptodate(folio);
 	kfree(finfo);
 	return;
-reduce_len:
-	finfo->dirty_len = offset + length - finfo->dirty_offset;
-	return;
-move_start:
-	finfo->dirty_len -= offset - finfo->dirty_offset;
-	finfo->dirty_offset = offset;
 }
 EXPORT_SYMBOL(netfs_invalidate_folio);
 
@@ -159,12 +181,20 @@ bool netfs_release_folio(struct folio *folio, gfp_t gfp)
 	struct netfs_inode *ctx = netfs_inode(folio_inode(folio));
 	unsigned long long end;
 
-	end = folio_pos(folio) + folio_size(folio);
+	if (folio_test_dirty(folio))
+		return false;
+
+	end = umin(folio_pos(folio) + folio_size(folio), i_size_read(&ctx->inode));
 	if (end > ctx->zero_point)
 		ctx->zero_point = end;
 
 	if (folio_test_private(folio))
 		return false;
+	if (unlikely(folio_test_private_2(folio))) { /* [DEPRECATED] */
+		if (current_is_kswapd() || !(gfp & __GFP_FS))
+			return false;
+		folio_wait_private_2(folio);
+	}
 	fscache_note_page_release(netfs_i_cookie(ctx));
 	return true;
 }