btrfs: do proper folio cleanup when run_delalloc_nocow() failed

[BUG]
With CONFIG_DEBUG_VM set, test case generic/476 has some chance to crash
with the following VM_BUG_ON_FOLIO():

 BTRFS error (device dm-3): cow_file_range failed, start 1146880 end 1253375 len 106496 ret -28
 BTRFS error (device dm-3): run_delalloc_nocow failed, start 1146880 end 1253375 len 106496 ret -28
 page: refcount:4 mapcount:0 mapping:00000000592787cc index:0x12 pfn:0x10664
 aops:btrfs_aops [btrfs] ino:101 dentry name(?):"f1774"
 flags: 0x2fffff80004028(uptodate|lru|private|node=0|zone=2|lastcpupid=0xfffff)
 page dumped because: VM_BUG_ON_FOLIO(!folio_test_locked(folio))
 ------------[ cut here ]------------
 kernel BUG at mm/page-writeback.c:2992!
 Internal error: Oops - BUG: 00000000f2000800 [#1] SMP
 CPU: 2 UID: 0 PID: 3943513 Comm: kworker/u24:15 Tainted: G           OE      6.12.0-rc7-custom+ torvalds#87
 Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
 Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022
 Workqueue: events_unbound btrfs_async_reclaim_data_space [btrfs]
 pc : folio_clear_dirty_for_io+0x128/0x258
 lr : folio_clear_dirty_for_io+0x128/0x258
 Call trace:
  folio_clear_dirty_for_io+0x128/0x258
  btrfs_folio_clamp_clear_dirty+0x80/0xd0 [btrfs]
  __process_folios_contig+0x154/0x268 [btrfs]
  extent_clear_unlock_delalloc+0x5c/0x80 [btrfs]
  run_delalloc_nocow+0x5f8/0x760 [btrfs]
  btrfs_run_delalloc_range+0xa8/0x220 [btrfs]
  writepage_delalloc+0x230/0x4c8 [btrfs]
  extent_writepage+0xb8/0x358 [btrfs]
  extent_write_cache_pages+0x21c/0x4e8 [btrfs]
  btrfs_writepages+0x94/0x150 [btrfs]
  do_writepages+0x74/0x190
  filemap_fdatawrite_wbc+0x88/0xc8
  start_delalloc_inodes+0x178/0x3a8 [btrfs]
  btrfs_start_delalloc_roots+0x174/0x280 [btrfs]
  shrink_delalloc+0x114/0x280 [btrfs]
  flush_space+0x250/0x2f8 [btrfs]
  btrfs_async_reclaim_data_space+0x180/0x228 [btrfs]
  process_one_work+0x164/0x408
  worker_thread+0x25c/0x388
  kthread+0x100/0x118
  ret_from_fork+0x10/0x20
 Code: 910a8021 a90363f7 a9046bf9 94012379 (d4210000)
 ---[ end trace 0000000000000000 ]---

[CAUSE]
The first two lines of extra debug messages show the problem is caused
by the error handling of run_delalloc_nocow().

E.g. we have the following dirtied range (4K blocksize 4K page size):

    0                 16K                  32K
    |//////////////////////////////////////|
    |  Pre-allocated  |

And the range [0, 16K) has a preallocated extent.

- Enter run_delalloc_nocow() for range [0, 16K)
  Which found range [0, 16K) is preallocated, can do the proper NOCOW
  write.

- Enter fallback_to_fow() for range [16K, 32K)
  Since the range [16K, 32K) is not backed by preallocated extent, we
  have to go COW.

- cow_file_range() failed for range [16K, 32K)
  So cow_file_range() will do the clean up by clearing folio dirty,
  unlock the folios.

  Now the folios in range [16K, 32K) is unlocked.

- Enter extent_clear_unlock_delalloc() from run_delalloc_nocow()
  Which is called with PAGE_START_WRITEBACK to start page writeback.
  But folios can only be marked writeback when it's properly locked,
  thus this triggered the VM_BUG_ON_FOLIO().

Furthermore there is another hidden but common bug that
run_delalloc_nocow() is not clearing the folio dirty flags in its error
handling path.
This is the common bug shared between run_delalloc_nocow() and
cow_file_range().

[FIX]
- Clear folio dirty for range [@start, @cur_offset)
  Introduce a helper, cleanup_dirty_folios(), which
  will find and lock the folio in the range, clear the dirty flag and
  start/end the writeback, with the extra handling for the
  @locked_folio.

- Introduce a helper to record the last failed COW range end
  This is to trace which range we should skip, to avoid double
  unlocking.

- Skip the failed COW range for the error handling

Cc: stable@vger.kernel.org
Signed-off-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>

fs/btrfs/inode.c

@@ -1958,6 +1958,48 @@ static int can_nocow_file_extent(struct btrfs_path *path,
 	return ret < 0 ? ret : can_nocow;
 }
 
+static void cleanup_dirty_folios(struct btrfs_inode *inode,
+				 struct folio *locked_folio,
+				 u64 start, u64 end, int error)
+{
+	struct btrfs_fs_info *fs_info = inode->root->fs_info;
+	struct address_space *mapping = inode->vfs_inode.i_mapping;
+	pgoff_t start_index = start >> PAGE_SHIFT;
+	pgoff_t end_index = end >> PAGE_SHIFT;
+	u32 len;
+
+	ASSERT(end + 1 - start < U32_MAX);
+	ASSERT(IS_ALIGNED(start, fs_info->sectorsize) &&
+	       IS_ALIGNED(end + 1, fs_info->sectorsize));
+	len = end + 1 - start;
+
+	/*
+	 * Handle the locked folio first.
+	 * btrfs_folio_clamp_*() helpers can handle range out of the folio case.
+	 */
+	btrfs_folio_clamp_clear_dirty(fs_info, locked_folio, start, len);
+	btrfs_folio_clamp_set_writeback(fs_info, locked_folio, start, len);
+	btrfs_folio_clamp_clear_writeback(fs_info, locked_folio, start, len);
+
+	for (pgoff_t index = start_index; index <= end_index; index++) {
+		struct folio *folio;
+
+		/* Already handled at the beginning. */
+		if (index == locked_folio->index)
+			continue;
+		folio = __filemap_get_folio(mapping, index, FGP_LOCK, GFP_NOFS);
+		/* Cache already dropped, no need to do any cleanup. */
+		if (IS_ERR(folio))
+			continue;
+		btrfs_folio_clamp_clear_dirty(fs_info, folio, start, len);
+		btrfs_folio_clamp_set_writeback(fs_info, folio, start, len);
+		btrfs_folio_clamp_clear_writeback(fs_info, folio, start, len);
+		folio_unlock(folio);
+		folio_put(folio);
+	}
+	mapping_set_error(mapping, error);
+}
+
 /*
  * when nowcow writeback call back.  This checks for snapshots or COW copies
  * of the extents that exist in the file, and COWs the file as required.
@@ -1973,6 +2015,11 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 	struct btrfs_root *root = inode->root;
 	struct btrfs_path *path;
 	u64 cow_start = (u64)-1;
+	/*
+	 * If not 0, represents the inclusive end of the last fallback_to_cow()
+	 * range. Only for error handling.
+	 */
+	u64 cow_end = 0;
 	u64 cur_offset = start;
 	int ret;
 	bool check_prev = true;
@@ -2133,6 +2180,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 					      found_key.offset - 1);
 			cow_start = (u64)-1;
 			if (ret) {
+				cow_end = found_key.offset - 1;
 				btrfs_dec_nocow_writers(nocow_bg);
 				goto error;
 			}
@@ -2206,24 +2254,55 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode,
 		cow_start = cur_offset;
 
 	if (cow_start != (u64)-1) {
-		cur_offset = end;
 		ret = fallback_to_cow(inode, locked_folio, cow_start, end);
 		cow_start = (u64)-1;
-		if (ret)
+		if (ret) {
+			cow_end = end;
 			goto error;
+		}
 	}
 
 	btrfs_free_path(path);
 	return 0;
 
 error:
+	/*
+	 * There are several error cases:
+	 *
+	 * 1) Failed without falling back to COW
+	 *    start         cur_start              end
+	 *    |/////////////|                      |
+	 *
+	 *    For range [start, cur_start) the folios are already unlocked (except
+	 *    @locked_folio), EXTENT_DELALLOC already removed.
+	 *    Only need to clear the dirty flag as they will never be submitted.
+	 *    Ordered extent and extent maps are handled by
+	 *    btrfs_mark_ordered_io_finished() inside run_delalloc_range().
+	 *
+	 * 2) Failed with error from fallback_to_cow()
+	 *    start         cur_start   cow_end    end
+	 *    |/////////////|-----------|          |
+	 *
+	 *    For range [start, cur_start) it's the same as case 1).
+	 *    But for range [cur_start, cow_end), the folios have dirty flag
+	 *    cleared and unlocked, EXTENT_DEALLLOC cleared.
+	 *    There may or may not be any ordered extents/extent maps allocated.
+	 *
+	 *    We should not call extent_clear_unlock_delalloc() on range [cur_start,
+	 *    cow_end), as the folios are already unlocked.
+	 *
+	 * So clear the folio dirty flags for [start, cur_offset) first.
+	 */
+	if (cur_offset > start)
+		cleanup_dirty_folios(inode, locked_folio, start, cur_offset - 1, ret);
+
 	/*
 	 * If an error happened while a COW region is outstanding, cur_offset
-	 * needs to be reset to cow_start to ensure the COW region is unlocked
-	 * as well.
+	 * needs to be reset to @cow_end + 1 to skip the COW range, as
+	 * cow_file_range() will do the proper cleanup at error.
 	 */
-	if (cow_start != (u64)-1)
-		cur_offset = cow_start;
+	if (cow_end)
+		cur_offset = cow_end + 1;
 
 	/*
 	 * We need to lock the extent here because we're clearing DELALLOC and