FAILED: patch "[PATCH] f2fs: fix submit accounting for IPU writeback" failed to apply to 5.10-stable tree

FAILED: patch "[PATCH] f2fs: fix submit accounting for IPU writeback" failed to apply to 5.10-stable tree

[Sasha Levin]

The patch below does not apply to the 5.10-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y
git checkout FETCH_HEAD
git cherry-pick -x 6d1f622b9c069378cdb83088be2b15a9b480b8a3
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<stable@vger.kernel.org>' --in-reply-to '20261107062516-failed-5.10-6d1f622b9c06@kernel.org' --subject-prefix 'PATCH 5.10.y' 'HEAD^..'

Thanks,
Sasha

------------------ original commit in Linus's tree ------------------

From 6d1f622b9c069378cdb83088be2b15a9b480b8a3 Mon Sep 17 00:00:00 2001
From: Seongjae Jeong <jsjlee1020@gmail.com>
Date: Wed, 14 Oct 2026 11:26:29 +0000
Subject: [PATCH] f2fs: fix submit accounting for IPU writeback

f2fs_write_cache_pages() uses the submitted count, which is set to
fio.submitted in f2fs_write_single_data_page(), to update nwritten
and decrement wbc->nr_to_write.

In the OPU path, f2fs_submit_page_write() sets fio->submitted to 1.
However, in the IPU path, fio->submitted is never set and remains zero.
As a result, successful IPU writes are not accounted for.
This can cause asynchronous writeback to continue beyond its budget.

Set fio->submitted in f2fs_inplace_write_data() after a page is
successfully merged into a bio or submitted, so that successful
IPU writes are included in the writeback accounting.

Cc: stable@kernel.org
Fixes: d68f735b3bc9 ("f2fs: check io submission more precisely")
Signed-off-by: Seongjae Jeong <jsjlee1020@gmail.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index f6e1cb1c0b71d..92c3f255ea8fa 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -4292,6 +4292,7 @@ int f2fs_inplace_write_data(struct f2fs_io_info *fio)
 	else
 		err = f2fs_submit_page_bio(fio);
 	if (!err) {
+		fio->submitted = 1;
 		f2fs_update_device_state(fio->sbi, fio->ino,
 						fio->new_blkaddr, 1);
 		f2fs_update_iostat(fio->sbi, fio_inode(fio),
-- 
2.53.0

[PATCH 5.10.y 1/4] f2fs: rename logical_to_blk and blk_to_logical

[Sasha Levin]

From: Jaegeuk Kim <jaegeuk@kernel.org>

[ Upstream commit 6cbfcab5ff51491e67c75fd8d672cff97127fd55 ]

This patch renames two functions like below having u64.
 - logical_to_blk to bytes_to_blks
 - blk_to_logical to blks_to_bytes

Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Stable-dep-of: 6d1f622b9c06 ("f2fs: fix submit accounting for IPU writeback")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/data.c | 40 ++++++++++++++++++++--------------------
 1 file changed, 20 insertions(+), 20 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 2933b32032e79..e681ff67d63ef 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -1860,14 +1860,14 @@ static int get_data_block_bmap(struct inode *inode, sector_t iblock,
 						NO_CHECK_TYPE, create);
 }
 
-static inline sector_t logical_to_blk(struct inode *inode, loff_t offset)
+static inline u64 bytes_to_blks(struct inode *inode, u64 bytes)
 {
-	return (offset >> inode->i_blkbits);
+	return (bytes >> inode->i_blkbits);
 }
 
-static inline loff_t blk_to_logical(struct inode *inode, sector_t blk)
+static inline u64 blks_to_bytes(struct inode *inode, u64 blks)
 {
-	return (blk << inode->i_blkbits);
+	return (blks << inode->i_blkbits);
 }
 
 static int f2fs_xattr_fiemap(struct inode *inode,
@@ -1895,7 +1895,7 @@ static int f2fs_xattr_fiemap(struct inode *inode,
 			return err;
 		}
 
-		phys = (__u64)blk_to_logical(inode, ni.blk_addr);
+		phys = blks_to_bytes(inode, ni.blk_addr);
 		offset = offsetof(struct f2fs_inode, i_addr) +
 					sizeof(__le32) * (DEF_ADDRS_PER_INODE -
 					get_inline_xattr_addrs(inode));
@@ -1927,7 +1927,7 @@ static int f2fs_xattr_fiemap(struct inode *inode,
 			return err;
 		}
 
-		phys = (__u64)blk_to_logical(inode, ni.blk_addr);
+		phys = blks_to_bytes(inode, ni.blk_addr);
 		len = inode->i_sb->s_blocksize;
 
 		f2fs_put_page(page, 1);
@@ -1997,18 +1997,18 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			goto out;
 	}
 
-	if (logical_to_blk(inode, len) == 0)
-		len = blk_to_logical(inode, 1);
+	if (bytes_to_blks(inode, len) == 0)
+		len = blks_to_bytes(inode, 1);
 
-	start_blk = logical_to_blk(inode, start);
-	last_blk = logical_to_blk(inode, start + len - 1);
+	start_blk = bytes_to_blks(inode, start);
+	last_blk = bytes_to_blks(inode, start + len - 1);
 
 next:
 	memset(&map_bh, 0, sizeof(struct buffer_head));
 	map_bh.b_size = len;
 
 	if (compr_cluster)
-		map_bh.b_size = blk_to_logical(inode, cluster_size - 1);
+		map_bh.b_size = blks_to_bytes(inode, cluster_size - 1);
 
 	ret = get_data_block(inode, start_blk, &map_bh, 0,
 					F2FS_GET_BLOCK_FIEMAP, &next_pgofs);
@@ -2019,7 +2019,7 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	if (!buffer_mapped(&map_bh)) {
 		start_blk = next_pgofs;
 
-		if (blk_to_logical(inode, start_blk) < blk_to_logical(inode,
+		if (blks_to_bytes(inode, start_blk) < blks_to_bytes(inode,
 						max_inode_blocks(inode)))
 			goto prep_next;
 
@@ -2045,9 +2045,9 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		compr_cluster = false;
 
 
-		logical = blk_to_logical(inode, start_blk - 1);
-		phys = blk_to_logical(inode, map_bh.b_blocknr);
-		size = blk_to_logical(inode, cluster_size);
+		logical = blks_to_bytes(inode, start_blk - 1);
+		phys = blks_to_bytes(inode, map_bh.b_blocknr);
+		size = blks_to_bytes(inode, cluster_size);
 
 		flags |= FIEMAP_EXTENT_ENCODED;
 
@@ -2065,14 +2065,14 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		goto prep_next;
 	}
 
-	logical = blk_to_logical(inode, start_blk);
-	phys = blk_to_logical(inode, map_bh.b_blocknr);
+	logical = blks_to_bytes(inode, start_blk);
+	phys = blks_to_bytes(inode, map_bh.b_blocknr);
 	size = map_bh.b_size;
 	flags = 0;
 	if (buffer_unwritten(&map_bh))
 		flags = FIEMAP_EXTENT_UNWRITTEN;
 
-	start_blk += logical_to_blk(inode, size);
+	start_blk += bytes_to_blks(inode, size);
 
 prep_next:
 	cond_resched();
@@ -4003,7 +4003,7 @@ static int check_swap_activate_fast(struct swap_info_struct *sis,
 	 * to be very smart.
 	 */
 	cur_lblock = 0;
-	last_lblock = logical_to_blk(inode, i_size_read(inode));
+	last_lblock = bytes_to_blks(inode, i_size_read(inode));
 	len = i_size_read(inode);
 
 	while (cur_lblock <= last_lblock && cur_lblock < sis->max) {
@@ -4025,7 +4025,7 @@ static int check_swap_activate_fast(struct swap_info_struct *sis,
 			goto err_out;
 
 		pblock = map_bh.b_blocknr;
-		nr_pblocks = logical_to_blk(inode, map_bh.b_size);
+		nr_pblocks = bytes_to_blks(inode, map_bh.b_size);
 
 		if (cur_lblock + nr_pblocks >= sis->max)
 			nr_pblocks = sis->max - cur_lblock;
-- 
2.53.0

[PATCH 5.10.y 2/4] f2fs: use new conversion functions between blks and bytes

[Sasha Levin]

From: Jaegeuk Kim <jaegeuk@kernel.org>

[ Upstream commit 43b9d4b4d91c02ca1159507d46e3687d98b4cb8f ]

This patch cleans up blks and bytes conversions.

Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Stable-dep-of: 6d1f622b9c06 ("f2fs: fix submit accounting for IPU writeback")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/data.c | 46 +++++++++++++++++++++-------------------------
 1 file changed, 21 insertions(+), 25 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index e681ff67d63ef..e033c8dfb0348 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -1802,6 +1802,16 @@ bool f2fs_overwrite_io(struct inode *inode, loff_t pos, size_t len)
 	return true;
 }
 
+static inline u64 bytes_to_blks(struct inode *inode, u64 bytes)
+{
+	return (bytes >> inode->i_blkbits);
+}
+
+static inline u64 blks_to_bytes(struct inode *inode, u64 blks)
+{
+	return (blks << inode->i_blkbits);
+}
+
 static int __get_data_block(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh, int create, int flag,
 			pgoff_t *next_pgofs, int seg_type, bool may_write)
@@ -1810,7 +1820,7 @@ static int __get_data_block(struct inode *inode, sector_t iblock,
 	int err;
 
 	map.m_lblk = iblock;
-	map.m_len = bh->b_size >> inode->i_blkbits;
+	map.m_len = bytes_to_blks(inode, bh->b_size);
 	map.m_next_pgofs = next_pgofs;
 	map.m_next_extent = NULL;
 	map.m_seg_type = seg_type;
@@ -1820,7 +1830,7 @@ static int __get_data_block(struct inode *inode, sector_t iblock,
 	if (!err) {
 		map_bh(bh, inode->i_sb, map.m_pblk);
 		bh->b_state = (bh->b_state & ~F2FS_MAP_FLAGS) | map.m_flags;
-		bh->b_size = (u64)map.m_len << inode->i_blkbits;
+		bh->b_size = blks_to_bytes(inode, map.m_len);
 	}
 	return err;
 }
@@ -1860,16 +1870,6 @@ static int get_data_block_bmap(struct inode *inode, sector_t iblock,
 						NO_CHECK_TYPE, create);
 }
 
-static inline u64 bytes_to_blks(struct inode *inode, u64 bytes)
-{
-	return (bytes >> inode->i_blkbits);
-}
-
-static inline u64 blks_to_bytes(struct inode *inode, u64 blks)
-{
-	return (blks << inode->i_blkbits);
-}
-
 static int f2fs_xattr_fiemap(struct inode *inode,
 				struct fiemap_extent_info *fieinfo)
 {
@@ -2105,8 +2105,7 @@ static int f2fs_read_single_page(struct inode *inode, struct page *page,
 					bool is_readahead)
 {
 	struct bio *bio = *bio_ret;
-	const unsigned blkbits = inode->i_blkbits;
-	const unsigned blocksize = 1 << blkbits;
+	const unsigned blocksize = blks_to_bytes(inode, 1);
 	sector_t block_in_file;
 	sector_t last_block;
 	sector_t last_block_in_file;
@@ -2115,8 +2114,8 @@ static int f2fs_read_single_page(struct inode *inode, struct page *page,
 
 	block_in_file = (sector_t)page_index(page);
 	last_block = block_in_file + nr_pages;
-	last_block_in_file = (f2fs_readpage_limit(inode) + blocksize - 1) >>
-							blkbits;
+	last_block_in_file = bytes_to_blks(inode,
+			f2fs_readpage_limit(inode) + blocksize - 1);
 	if (last_block > last_block_in_file)
 		last_block = last_block_in_file;
 
@@ -2229,8 +2228,7 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret,
 	struct bio *bio = *bio_ret;
 	unsigned int start_idx = cc->cluster_idx << cc->log_cluster_size;
 	sector_t last_block_in_file;
-	const unsigned blkbits = inode->i_blkbits;
-	const unsigned blocksize = 1 << blkbits;
+	const unsigned blocksize = blks_to_bytes(inode, 1);
 	struct decompress_io_ctx *dic = NULL;
 	struct bio_post_read_ctx *ctx;
 	bool for_verity = false;
@@ -2239,8 +2237,8 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret,
 
 	f2fs_bug_on(sbi, f2fs_cluster_is_empty(cc));
 
-	last_block_in_file = (f2fs_readpage_limit(inode) +
-					blocksize - 1) >> blkbits;
+	last_block_in_file = bytes_to_blks(inode,
+			f2fs_readpage_limit(inode) + blocksize - 1);
 
 	/* get rid of pages beyond EOF */
 	for (i = 0; i < cc->cluster_size; i++) {
@@ -4068,7 +4066,6 @@ static int check_swap_activate(struct swap_info_struct *sis,
 	struct inode *inode = mapping->host;
 	unsigned blocks_per_page;
 	unsigned long page_no;
-	unsigned blkbits;
 	sector_t probe_block;
 	sector_t last_block;
 	sector_t lowest_block = -1;
@@ -4079,8 +4076,7 @@ static int check_swap_activate(struct swap_info_struct *sis,
 	if (PAGE_SIZE == F2FS_BLKSIZE)
 		return check_swap_activate_fast(sis, swap_file, span);
 
-	blkbits = inode->i_blkbits;
-	blocks_per_page = PAGE_SIZE >> blkbits;
+	blocks_per_page = bytes_to_blks(inode, PAGE_SIZE);
 
 	/*
 	 * Map all the blocks into the extent list.  This code doesn't try
@@ -4088,7 +4084,7 @@ static int check_swap_activate(struct swap_info_struct *sis,
 	 */
 	probe_block = 0;
 	page_no = 0;
-	last_block = i_size_read(inode) >> blkbits;
+	last_block = bytes_to_blks(inode, i_size_read(inode));
 	while ((probe_block + blocks_per_page) <= last_block &&
 			page_no < sis->max) {
 		unsigned block_in_page;
@@ -4128,7 +4124,7 @@ static int check_swap_activate(struct swap_info_struct *sis,
 			}
 		}
 
-		first_block >>= (PAGE_SHIFT - blkbits);
+		first_block >>= (PAGE_SHIFT - inode->i_blkbits);
 		if (page_no) {	/* exclude the header page */
 			if (first_block < lowest_block)
 				lowest_block = first_block;
-- 
2.53.0

[PATCH 5.10.y 3/4] f2fs: remove allow_outplace_dio()

[Sasha Levin]

From: Eric Biggers <ebiggers@google.com>

[ Upstream commit 6de8687ccdefed40d617492f4e1b3962eb577b6b ]

We can just check f2fs_lfs_mode() directly.  The block_unaligned_IO()
check is redundant because in LFS mode, f2fs doesn't do direct I/O
writes that aren't block-aligned (due to f2fs_force_buffered_io()
returning true in this case, triggering the fallback to buffered I/O).

Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Stable-dep-of: 6d1f622b9c06 ("f2fs: fix submit accounting for IPU writeback")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/data.c |  2 +-
 fs/f2fs/f2fs.h | 10 ----------
 fs/f2fs/file.c |  2 +-
 3 files changed, 2 insertions(+), 12 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index e033c8dfb0348..3f7056f9971f5 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -3725,7 +3725,7 @@ static ssize_t f2fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
 	if (f2fs_force_buffered_io(inode, iocb, iter))
 		return 0;
 
-	do_opu = allow_outplace_dio(inode, iocb, iter);
+	do_opu = rw == WRITE && f2fs_lfs_mode(sbi);
 
 	trace_f2fs_direct_IO_enter(inode, offset, count, rw);
 
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index cead30cab94bc..43f14bbfef49b 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -4135,16 +4135,6 @@ static inline int block_unaligned_IO(struct inode *inode,
 	return align & blocksize_mask;
 }
 
-static inline int allow_outplace_dio(struct inode *inode,
-				struct kiocb *iocb, struct iov_iter *iter)
-{
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	int rw = iov_iter_rw(iter);
-
-	return (f2fs_lfs_mode(sbi) && (rw == WRITE) &&
-				!block_unaligned_IO(inode, iocb, iter));
-}
-
 static inline bool f2fs_force_buffered_io(struct inode *inode,
 				struct kiocb *iocb, struct iov_iter *iter)
 {
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 9c12691ace0b6..f43cc0ba83837 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -4250,7 +4250,7 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 			 * back to buffered IO.
 			 */
 			if (!f2fs_force_buffered_io(inode, iocb, from) &&
-					allow_outplace_dio(inode, iocb, from))
+					f2fs_lfs_mode(F2FS_I_SB(inode)))
 				goto write;
 		}
 		preallocated = true;
-- 
2.53.0

[PATCH 5.10.y 4/4] f2fs: fix submit accounting for IPU writeback

[Sasha Levin]

From: Seongjae Jeong <jsjlee1020@gmail.com>

[ Upstream commit 6d1f622b9c069378cdb83088be2b15a9b480b8a3 ]

f2fs_write_cache_pages() uses the submitted count, which is set to
fio.submitted in f2fs_write_single_data_page(), to update nwritten
and decrement wbc->nr_to_write.

In the OPU path, f2fs_submit_page_write() sets fio->submitted to 1.
However, in the IPU path, fio->submitted is never set and remains zero.
As a result, successful IPU writes are not accounted for.
This can cause asynchronous writeback to continue beyond its budget.

Set fio->submitted in f2fs_inplace_write_data() after a page is
successfully merged into a bio or submitted, so that successful
IPU writes are included in the writeback accounting.

Cc: stable@kernel.org
Fixes: d68f735b3bc9 ("f2fs: check io submission more precisely")
Signed-off-by: Seongjae Jeong <jsjlee1020@gmail.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
[ sashal: Adjusted context to preserve the older device-state and I/O-statistics calls. ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/f2fs/segment.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index fec3b4a9e9542..4d759917c2849 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -3563,6 +3563,7 @@ int f2fs_inplace_write_data(struct f2fs_io_info *fio)
 	else
 		err = f2fs_submit_page_bio(fio);
 	if (!err) {
+		fio->submitted = 1;
 		update_device_state(fio);
 		f2fs_update_iostat(fio->sbi, fio->io_type, F2FS_BLKSIZE);
 	}
-- 
2.53.0