Skip to contentSkip to conversation

bio: check return values of bio_add_page

PATCH In review Johannes Thumshirn 2 participants 24 messages Apr 20, 2023 – Apr 21, 2023

Subsystems (14)
lore
Share & tools
lore

Johannes Thumshirn

author reviewer
#0 lore raw

v4 cover bio: check return values of bio_add_page

We have two functions for adding a page to a bio, __bio_add_page() which is used to add a single page to a freshly created bio and bio_add_page() which is used to add a page to an existing bio.

While __bio_add_page() is expected to succeed, bio_add_page() can fail.

This series converts the callers of bio_add_page() which can easily use __bio_add_page() to using it and checks the return of bio_add_page() for callers that don't work on a freshly created bio.

Lastly it marks bio_add_page() as __must_check so we don't have to go again and audit all callers.

NOTE: David already applied the two btrfs patches to his tree but I've left them in the series to make the build bot happy.

Changes to v3:
- Added __bio_add_folio and use it in iomap (Willy)
- Mark bio_add_folio must check (Willy)
- s/GFS/GFS2/ (Andreas)

Changes to v2:
- Removed 'wont fail' comments pointed out by Song

Changes to v1:
- Removed pointless comment pointed out by Willy
- Changed commit messages pointed out by Damien
- Colledted Damien's Reviews and Acks

Johannes Thumshirn (22):
swap: use __bio_add_page to add page to bio
drbd: use __bio_add_page to add page to bio
dm: dm-zoned: use __bio_add_page for adding single metadata page
fs: buffer: use __bio_add_page to add single page to bio
md: use __bio_add_page to add single page
md: raid5-log: use __bio_add_page to add single page
md: raid5: use __bio_add_page to add single page to new bio
btrfs: repair: use __bio_add_page for adding single page
btrfs: raid56: use __bio_add_page to add single page
jfs: logmgr: use __bio_add_page to add single page to bio
gfs2: use __bio_add_page for adding single page to bio
zonefs: use __bio_add_page for adding single page to bio
zram: use __bio_add_page for adding single page to bio
floppy: use __bio_add_page for adding single page to bio
md: check for failure when adding pages in alloc_behind_master_bio
md: raid1: use __bio_add_page for adding single page to bio
md: raid1: check if adding pages to resync bio fails
dm-crypt: check if adding pages to clone bio fails
block: mark bio_add_page as __must_check
block: add __bio_add_folio
fs: iomap: use __bio_add_folio where possible
block: mark bio_add_folio as __must_check

 block/bio.c                      |  8 ++++++++
 drivers/block/drbd/drbd_bitmap.c |  4 +---
 drivers/block/floppy.c           |  2 +-
 drivers/block/zram/zram_drv.c    |  2 +-
 drivers/md/dm-crypt.c            |  9 ++++++++-
 drivers/md/dm-zoned-metadata.c   |  6 +++---
 drivers/md/md.c                  |  4 ++--
 drivers/md/raid1-10.c            | 11 ++++++-----
 drivers/md/raid1.c               |  7 +++++--
 drivers/md/raid10.c              | 20 ++++++++++----------
 drivers/md/raid5-cache.c         |  2 +-
 drivers/md/raid5-ppl.c           |  4 ++--
 fs/btrfs/bio.c                   |  2 +-
 fs/btrfs/raid56.c                |  2 +-
 fs/buffer.c                      |  2 +-
 fs/gfs2/ops_fstype.c             |  2 +-
 fs/iomap/buffered-io.c           |  6 +++---
 fs/jfs/jfs_logmgr.c              |  4 ++--
 fs/zonefs/super.c                |  2 +-
 include/linux/bio.h              |  5 +++--
 mm/page_io.c                     |  8 ++++----
 21 files changed, 65 insertions(+), 47 deletions(-)

base-commit: af67688dca57999fd848f051eeea1d375ba546b2

Signature · 1 line
2.39.2
author reviewer
#1 lore raw

PATCH v4 1/22 swap: use __bio_add_page to add page to bio

From: Johannes Thumshirn <[email protected]>

The swap code only adds a single page to a newly created bio. So use __bio_add_page() to add the page which is guaranteed to succeed in this case.

This brings us closer to marking bio_add_page() as __must_check.

---
 mm/page_io.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
mm/page_io.c +4 -4
index 87b682d18850..684cd3c7b59b 100644
@@ -338,7 +338,7 @@ static void swap_writepage_bdev_sync(struct page *page,
 	bio_init(&bio, sis->bdev, &bv, 1,
 		 REQ_OP_WRITE | REQ_SWAP | wbc_to_write_flags(wbc));
 	bio.bi_iter.bi_sector = swap_page_sector(page);
-	bio_add_page(&bio, page, thp_size(page), 0);
+	__bio_add_page(&bio, page, thp_size(page), 0);
 
 	bio_associate_blkg_from_page(&bio, page);
 	count_swpout_vm_event(page);
@@ -360,7 +360,7 @@ static void swap_writepage_bdev_async(struct page *page,
 			GFP_NOIO);
 	bio->bi_iter.bi_sector = swap_page_sector(page);
 	bio->bi_end_io = end_swap_bio_write;
-	bio_add_page(bio, page, thp_size(page), 0);
+	__bio_add_page(bio, page, thp_size(page), 0);
 
 	bio_associate_blkg_from_page(bio, page);
 	count_swpout_vm_event(page);
@@ -468,7 +468,7 @@ static void swap_readpage_bdev_sync(struct page *page,
 
 	bio_init(&bio, sis->bdev, &bv, 1, REQ_OP_READ);
 	bio.bi_iter.bi_sector = swap_page_sector(page);
-	bio_add_page(&bio, page, thp_size(page), 0);
+	__bio_add_page(&bio, page, thp_size(page), 0);
 	/*
 	 * Keep this task valid during swap readpage because the oom killer may
 	 * attempt to access it in the page fault retry time check.
@@ -488,7 +488,7 @@ static void swap_readpage_bdev_async(struct page *page,
 	bio = bio_alloc(sis->bdev, 1, REQ_OP_READ, GFP_KERNEL);
 	bio->bi_iter.bi_sector = swap_page_sector(page);
 	bio->bi_end_io = end_swap_bio_read;
-	bio_add_page(bio, page, thp_size(page), 0);
+	__bio_add_page(bio, page, thp_size(page), 0);
 	count_vm_event(PSWPIN);
 	submit_bio(bio);
 }
Signature · 1 line
2.39.2
author reviewer
#2 lore raw

PATCH v4 2/22 drbd: use __bio_add_page to add page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The drbd code only adds a single page to a newly created bio. So use __bio_add_page() to add the page which is guaranteed to succeed in this case.

This brings us closer to marking bio_add_page() as __must_check.

---
 drivers/block/drbd/drbd_bitmap.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/block/drbd/drbd_bitmap.c +1 -3
index 289876ffbc31..050154eb963d 100644
@@ -1043,9 +1043,7 @@ static void bm_page_io_async(struct drbd_bm_aio_ctx *ctx, int page_nr) __must_ho
 	bio = bio_alloc_bioset(device->ldev->md_bdev, 1, op, GFP_NOIO,
 			&drbd_md_io_bio_set);
 	bio->bi_iter.bi_sector = on_disk_sector;
-	/* bio_add_page of a single page to an empty bio will always succeed,
-	 * according to api.  Do we want to assert that? */
-	bio_add_page(bio, page, len, 0);
+	__bio_add_page(bio, page, len, 0);
 	bio->bi_private = ctx;
 	bio->bi_end_io = drbd_bm_endio;
 
Signature · 1 line
2.39.2
author reviewer
#3 lore raw

PATCH v4 3/22 dm: dm-zoned: use __bio_add_page for adding single metadata page

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

dm-zoned uses bio_add_page() for adding a single page to a freshly created metadata bio.

Use __bio_add_page() instead as adding a single page to a new bio is always guaranteed to succeed.

This brings us a step closer to marking bio_add_page() __must_check

---
 drivers/md/dm-zoned-metadata.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/md/dm-zoned-metadata.c +3 -3
index cf9402064aba..8dbe102ab271 100644
@@ -577,7 +577,7 @@ static struct dmz_mblock *dmz_get_mblock_slow(struct dmz_metadata *zmd,
 	bio->bi_iter.bi_sector = dmz_blk2sect(block);
 	bio->bi_private = mblk;
 	bio->bi_end_io = dmz_mblock_bio_end_io;
-	bio_add_page(bio, mblk->page, DMZ_BLOCK_SIZE, 0);
+	__bio_add_page(bio, mblk->page, DMZ_BLOCK_SIZE, 0);
 	submit_bio(bio);
 
 	return mblk;
@@ -728,7 +728,7 @@ static int dmz_write_mblock(struct dmz_metadata *zmd, struct dmz_mblock *mblk,
 	bio->bi_iter.bi_sector = dmz_blk2sect(block);
 	bio->bi_private = mblk;
 	bio->bi_end_io = dmz_mblock_bio_end_io;
-	bio_add_page(bio, mblk->page, DMZ_BLOCK_SIZE, 0);
+	__bio_add_page(bio, mblk->page, DMZ_BLOCK_SIZE, 0);
 	submit_bio(bio);
 
 	return 0;
@@ -752,7 +752,7 @@ static int dmz_rdwr_block(struct dmz_dev *dev, enum req_op op,
 	bio = bio_alloc(dev->bdev, 1, op | REQ_SYNC | REQ_META | REQ_PRIO,
 			GFP_NOIO);
 	bio->bi_iter.bi_sector = dmz_blk2sect(block);
-	bio_add_page(bio, page, DMZ_BLOCK_SIZE, 0);
+	__bio_add_page(bio, page, DMZ_BLOCK_SIZE, 0);
 	ret = submit_bio_wait(bio);
 	bio_put(bio);
 
Signature · 1 line
2.39.2
author reviewer
#4 lore raw

PATCH v4 4/22 fs: buffer: use __bio_add_page to add single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The buffer_head submission code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 fs/buffer.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/buffer.c +1 -1
index 9e1e2add541e..855dc41fe162 100644
@@ -2733,7 +2733,7 @@ static void submit_bh_wbc(blk_opf_t opf, struct buffer_head *bh,
 
 	bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9);
 
-	bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh));
+	__bio_add_page(bio, bh->b_page, bh->b_size, bh_offset(bh));
 	BUG_ON(bio->bi_iter.bi_size != bh->b_size);
 
 	bio->bi_end_io = end_bio_bh_io_sync;
Signature · 1 line
2.39.2
author reviewer
#5 lore raw

PATCH v4 5/22 md: use __bio_add_page to add single page

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The md-raid superblock writing code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 drivers/md/md.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/md.c +2 -2
index 13321dbb5fbc..20b9cd7c2f39 100644
@@ -958,7 +958,7 @@ void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
 	atomic_inc(&rdev->nr_pending);
 
 	bio->bi_iter.bi_sector = sector;
-	bio_add_page(bio, page, size, 0);
+	__bio_add_page(bio, page, size, 0);
 	bio->bi_private = rdev;
 	bio->bi_end_io = super_written;
 
@@ -999,7 +999,7 @@ int sync_page_io(struct md_rdev *rdev, sector_t sector, int size,
 		bio.bi_iter.bi_sector = sector + rdev->new_data_offset;
 	else
 		bio.bi_iter.bi_sector = sector + rdev->data_offset;
-	bio_add_page(&bio, page, size, 0);
+	__bio_add_page(&bio, page, size, 0);
 
 	submit_bio_wait(&bio);
 
Signature · 1 line
2.39.2
author reviewer
#6 lore raw

PATCH v4 6/22 md: raid5-log: use __bio_add_page to add single page

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The raid5 log metadata submission code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 drivers/md/raid5-cache.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/md/raid5-cache.c +1 -1
index 46182b955aef..852b265c5db4 100644
@@ -792,7 +792,7 @@ static struct r5l_io_unit *r5l_new_meta(struct r5l_log *log)
 	io->current_bio = r5l_bio_alloc(log);
 	io->current_bio->bi_end_io = r5l_log_endio;
 	io->current_bio->bi_private = io;
-	bio_add_page(io->current_bio, io->meta_page, PAGE_SIZE, 0);
+	__bio_add_page(io->current_bio, io->meta_page, PAGE_SIZE, 0);
 
 	r5_reserve_log_entry(log, io);
 
Signature · 1 line
2.39.2
author reviewer
#7 lore raw

PATCH v4 7/22 md: raid5: use __bio_add_page to add single page to new bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The raid5-ppl submission code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked. For adding consecutive pages, the return is actually checked and a new bio is allocated if adding the page fails.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 drivers/md/raid5-ppl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/md/raid5-ppl.c +2 -2
index e495939bb3e0..eaea57aee602 100644
@@ -465,7 +465,7 @@ static void ppl_submit_iounit(struct ppl_io_unit *io)
 
 	bio->bi_end_io = ppl_log_endio;
 	bio->bi_iter.bi_sector = log->next_io_sector;
-	bio_add_page(bio, io->header_page, PAGE_SIZE, 0);
+	__bio_add_page(bio, io->header_page, PAGE_SIZE, 0);
 
 	pr_debug("%s: log->current_io_sector: %llu\n", __func__,
 	    (unsigned long long)log->next_io_sector);
@@ -496,7 +496,7 @@ static void ppl_submit_iounit(struct ppl_io_unit *io)
 					       prev->bi_opf, GFP_NOIO,
 					       &ppl_conf->bs);
 			bio->bi_iter.bi_sector = bio_end_sector(prev);
-			bio_add_page(bio, sh->ppl_page, PAGE_SIZE, 0);
+			__bio_add_page(bio, sh->ppl_page, PAGE_SIZE, 0);
 
 			bio_chain(bio, prev);
 			ppl_submit_iounit_bio(io, prev);
Signature · 1 line
2.39.2
author reviewer
#8 lore raw

PATCH v4 8/22 btrfs: repair: use __bio_add_page for adding single page

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The btrfs repair bio submission code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 fs/btrfs/bio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/bio.c +1 -1
index 726592868e9c..73220a219c91 100644
@@ -224,7 +224,7 @@ static struct btrfs_failed_bio *repair_one_sector(struct btrfs_bio *failed_bbio,
 	repair_bio = bio_alloc_bioset(NULL, 1, REQ_OP_READ, GFP_NOFS,
 				      &btrfs_repair_bioset);
 	repair_bio->bi_iter.bi_sector = failed_bbio->saved_iter.bi_sector;
-	bio_add_page(repair_bio, bv->bv_page, bv->bv_len, bv->bv_offset);
+	__bio_add_page(repair_bio, bv->bv_page, bv->bv_len, bv->bv_offset);
 
 	repair_bbio = btrfs_bio(repair_bio);
 	btrfs_bio_init(repair_bbio, failed_bbio->inode, NULL, fbio);
Signature · 1 line
2.39.2
author reviewer
#9 lore raw

PATCH v4 9/22 btrfs: raid56: use __bio_add_page to add single page

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The btrfs raid58 sector submission code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 fs/btrfs/raid56.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/raid56.c +1 -1
index 642828c1b299..c8173e003df6 100644
@@ -1108,7 +1108,7 @@ static int rbio_add_io_sector(struct btrfs_raid_bio *rbio,
 	bio->bi_iter.bi_sector = disk_start >> 9;
 	bio->bi_private = rbio;
 
-	bio_add_page(bio, sector->page, sectorsize, sector->pgoff);
+	__bio_add_page(bio, sector->page, sectorsize, sector->pgoff);
 	bio_list_add(bio_list, bio);
 	return 0;
 }
Signature · 1 line
2.39.2
author reviewer
#10 lore raw

PATCH v4 10/22 jfs: logmgr: use __bio_add_page to add single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The JFS IO code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 fs/jfs/jfs_logmgr.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/jfs/jfs_logmgr.c +2 -2
index 695415cbfe98..15c645827dec 100644
@@ -1974,7 +1974,7 @@ static int lbmRead(struct jfs_log * log, int pn, struct lbuf ** bpp)
 
 	bio = bio_alloc(log->bdev, 1, REQ_OP_READ, GFP_NOFS);
 	bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
-	bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
+	__bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
 	BUG_ON(bio->bi_iter.bi_size != LOGPSIZE);
 
 	bio->bi_end_io = lbmIODone;
@@ -2115,7 +2115,7 @@ static void lbmStartIO(struct lbuf * bp)
 
 	bio = bio_alloc(log->bdev, 1, REQ_OP_WRITE | REQ_SYNC, GFP_NOFS);
 	bio->bi_iter.bi_sector = bp->l_blkno << (log->l2bsize - 9);
-	bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
+	__bio_add_page(bio, bp->l_page, LOGPSIZE, bp->l_offset);
 	BUG_ON(bio->bi_iter.bi_size != LOGPSIZE);
 
 	bio->bi_end_io = lbmIODone;
Signature · 1 line
2.39.2
author reviewer
#11 lore raw

PATCH v4 11/22 gfs2: use __bio_add_page for adding single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The GFS2 superblock reading code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 fs/gfs2/ops_fstype.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/gfs2/ops_fstype.c +1 -1
index 6de901c3b89b..e0cd0d43b12f 100644
@@ -254,7 +254,7 @@ static int gfs2_read_super(struct gfs2_sbd *sdp, sector_t sector, int silent)
 
 	bio = bio_alloc(sb->s_bdev, 1, REQ_OP_READ | REQ_META, GFP_NOFS);
 	bio->bi_iter.bi_sector = sector * (sb->s_blocksize >> 9);
-	bio_add_page(bio, page, PAGE_SIZE, 0);
+	__bio_add_page(bio, page, PAGE_SIZE, 0);
 
 	bio->bi_end_io = end_bio_io_page;
 	bio->bi_private = page;
Signature · 1 line
2.39.2
author reviewer
#12 lore raw

PATCH v4 12/22 zonefs: use __bio_add_page for adding single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The zonefs superblock reading code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 fs/zonefs/super.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/zonefs/super.c +1 -1
index 23b8b299c64e..9350221abfc5 100644
@@ -1128,7 +1128,7 @@ static int zonefs_read_super(struct super_block *sb)
 
 	bio_init(&bio, sb->s_bdev, &bio_vec, 1, REQ_OP_READ);
 	bio.bi_iter.bi_sector = 0;
-	bio_add_page(&bio, page, PAGE_SIZE, 0);
+	__bio_add_page(&bio, page, PAGE_SIZE, 0);
 
 	ret = submit_bio_wait(&bio);
 	if (ret)
Signature · 1 line
2.39.2
author reviewer
#13 lore raw

PATCH v4 13/22 zram: use __bio_add_page for adding single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The zram writeback code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 drivers/block/zram/zram_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/block/zram/zram_drv.c +1 -1
index aa490da3cef2..9179bd0f248c 100644
@@ -760,7 +760,7 @@ static ssize_t writeback_store(struct device *dev,
 			 REQ_OP_WRITE | REQ_SYNC);
 		bio.bi_iter.bi_sector = blk_idx * (PAGE_SIZE >> 9);
 
-		bio_add_page(&bio, bvec.bv_page, bvec.bv_len,
+		__bio_add_page(&bio, bvec.bv_page, bvec.bv_len,
 				bvec.bv_offset);
 		/*
 		 * XXX: A single page IO would be inefficient for write
Signature · 1 line
2.39.2
maintainer Reviewed-by on v4 13/22
#14 lore raw

Re: [PATCH v4 13/22] zram: use __bio_add_page for adding single page to bio

+ Minchan

Quoted 7 lines · Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The zram writeback code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Quoted 7 lines

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

author reviewer
#15 lore raw

PATCH v4 14/22 floppy: use __bio_add_page for adding single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The floppy code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 drivers/block/floppy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/block/floppy.c +1 -1
index 487840e3564d..6f46a30f7c36 100644
@@ -4147,7 +4147,7 @@ static int __floppy_read_block_0(struct block_device *bdev, int drive)
 	cbdata.drive = drive;
 
 	bio_init(&bio, bdev, &bio_vec, 1, REQ_OP_READ);
-	bio_add_page(&bio, page, block_size(bdev), 0);
+	__bio_add_page(&bio, page, block_size(bdev), 0);
 
 	bio.bi_iter.bi_sector = 0;
 	bio.bi_flags |= (1 << BIO_QUIET);
Signature · 1 line
2.39.2
author reviewer
#16 lore raw

PATCH v4 15/22 md: check for failure when adding pages in alloc_behind_master_bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

alloc_behind_master_bio() can possibly add multiple pages to a bio, but it is not checking for the return value of bio_add_page() if adding really succeeded.

Check if the page adding succeeded and if not bail out.

---
 drivers/md/raid1.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/md/raid1.c +4 -1
index 68a9e2d9985b..8283ef177f6c 100644
@@ -1147,7 +1147,10 @@ static void alloc_behind_master_bio(struct r1bio *r1_bio,
 		if (unlikely(!page))
 			goto free_pages;
 
-		bio_add_page(behind_bio, page, len, 0);
+		if (!bio_add_page(behind_bio, page, len, 0)) {
+			free_page(page);
+			goto free_pages;
+		}
 
 		size -= len;
 		i++;
Signature · 1 line
2.39.2
author reviewer
#17 lore raw

PATCH v4 16/22 md: raid1: use __bio_add_page for adding single page to bio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

The sync request code uses bio_add_page() to add a page to a newly created bio. bio_add_page() can fail, but the return value is never checked.

Use __bio_add_page() as adding a single page to a newly created bio is guaranteed to succeed.

This brings us a step closer to marking bio_add_page() as __must_check.

---
 drivers/md/raid1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/md/raid1.c +1 -1
index 8283ef177f6c..ff89839455ec 100644
@@ -2917,7 +2917,7 @@ static sector_t raid1_sync_request(struct mddev *mddev, sector_t sector_nr,
 				 * won't fail because the vec table is big
 				 * enough to hold all these pages
 				 */
-				bio_add_page(bio, page, len, 0);
+				__bio_add_page(bio, page, len, 0);
 			}
 		}
 		nr_sectors += len>>9;
Signature · 1 line
2.39.2
author reviewer
#18 lore raw

PATCH v4 17/22 md: raid1: check if adding pages to resync bio fails

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

Check if adding pages to resync bio fails and if bail out.

As the comment above suggests this cannot happen, WARN if it actually happens.

This way we can mark bio_add_pages as __must_check.

---
 drivers/md/raid1-10.c | 11 ++++++-----
 drivers/md/raid10.c   | 20 ++++++++++----------
 2 files changed, 16 insertions(+), 15 deletions(-)
drivers/md/raid1-10.c +6 -5
index e61f6cad4e08..cd349e69ed77 100644
@@ -101,11 +101,12 @@ static void md_bio_reset_resync_pages(struct bio *bio, struct resync_pages *rp,
 		struct page *page = resync_fetch_page(rp, idx);
 		int len = min_t(int, size, PAGE_SIZE);
 
-		/*
-		 * won't fail because the vec table is big
-		 * enough to hold all these pages
-		 */
-		bio_add_page(bio, page, len, 0);
+		if (WARN_ON(!bio_add_page(bio, page, len, 0))) {
+			bio->bi_status = BLK_STS_RESOURCE;
+			bio_endio(bio);
+			return;
+		}
+
 		size -= len;
 	} while (idx++ < RESYNC_PAGES && size > 0);
 }
drivers/md/raid10.c +10 -10
index 6c66357f92f5..59e52cf01569 100644
@@ -3804,11 +3804,11 @@ static sector_t raid10_sync_request(struct mddev *mddev, sector_t sector_nr,
 		for (bio= biolist ; bio ; bio=bio->bi_next) {
 			struct resync_pages *rp = get_resync_pages(bio);
 			page = resync_fetch_page(rp, page_idx);
-			/*
-			 * won't fail because the vec table is big enough
-			 * to hold all these pages
-			 */
-			bio_add_page(bio, page, len, 0);
+			if (WARN_ON(!bio_add_page(bio, page, len, 0))) {
+				bio->bi_status = BLK_STS_RESOURCE;
+				bio_endio(bio);
+				goto giveup;
+			}
 		}
 		nr_sectors += len>>9;
 		sector_nr += len>>9;
@@ -4985,11 +4985,11 @@ static sector_t reshape_request(struct mddev *mddev, sector_t sector_nr,
 		if (len > PAGE_SIZE)
 			len = PAGE_SIZE;
 		for (bio = blist; bio ; bio = bio->bi_next) {
-			/*
-			 * won't fail because the vec table is big enough
-			 * to hold all these pages
-			 */
-			bio_add_page(bio, page, len, 0);
+			if (WARN_ON(!bio_add_page(bio, page, len, 0))) {
+				bio->bi_status = BLK_STS_RESOURCE;
+				bio_endio(bio);
+				return sectors_done;
+			}
 		}
 		sector_nr += len >> 9;
 		nr_sectors += len >> 9;
Signature · 1 line
2.39.2
author reviewer
#19 lore raw

PATCH v4 18/22 dm-crypt: check if adding pages to clone bio fails

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

Check if adding pages to clone bio fails and if it does retry with reclaim. This mirrors the behaviour of page allocation in crypt_alloc_buffer().

This way we can mark bio_add_pages as __must_check.

---
 drivers/md/dm-crypt.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/md/dm-crypt.c +8 -1
index 3ba53dc3cc3f..19f7e087c6df 100644
@@ -1693,7 +1693,14 @@ static struct bio *crypt_alloc_buffer(struct dm_crypt_io *io, unsigned int size)
 
 		len = (remaining_size > PAGE_SIZE) ? PAGE_SIZE : remaining_size;
 
-		bio_add_page(clone, page, len, 0);
+		if (!bio_add_page(clone, page, len, 0)) {
+			mempool_free(page, &cc->page_pool);
+			crypt_free_buffer_pages(cc, clone);
+			bio_put(clone);
+			gfp_mask |= __GFP_DIRECT_RECLAIM;
+			goto retry;
+
+		}
 
 		remaining_size -= len;
 	}
Signature · 1 line
2.39.2
author reviewer
#20 lore raw

PATCH v4 19/22 block: mark bio_add_page as __must_check

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

Now that all users of bio_add_page check for the return value, mark bio_add_page as __must_check.

---
 include/linux/bio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/bio.h +1 -1
index d766be7152e1..0f8a8d7a6384 100644
@@ -465,7 +465,7 @@ extern void bio_uninit(struct bio *);
 void bio_reset(struct bio *bio, struct block_device *bdev, blk_opf_t opf);
 void bio_chain(struct bio *, struct bio *);
 
-int bio_add_page(struct bio *, struct page *, unsigned len, unsigned off);
+int __must_check bio_add_page(struct bio *, struct page *, unsigned len, unsigned off);
 bool bio_add_folio(struct bio *, struct folio *, size_t len, size_t off);
 extern int bio_add_pc_page(struct request_queue *, struct bio *, struct page *,
 			   unsigned int, unsigned int);
Signature · 1 line
2.39.2
author reviewer
#21 lore raw

PATCH v4 20/22 block: add __bio_add_folio

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

Just like for bio_add_pages() add a no-fail variant for bio_add_folio().

---
 block/bio.c         | 8 ++++++++
 include/linux/bio.h | 1 +
 2 files changed, 9 insertions(+)
block/bio.c +8 -0
index fd11614bba4d..f3a3524b53e4 100644
@@ -1138,6 +1138,14 @@ int bio_add_page(struct bio *bio, struct page *page,
 }
 EXPORT_SYMBOL(bio_add_page);
 
+void __bio_add_folio(struct bio *bio, struct folio *folio, size_t len,
+		     size_t off)
+{
+	WARN_ON_ONCE(len > UINT_MAX);
+	WARN_ON_ONCE(off > UINT_MAX);
+	__bio_add_page(bio, &folio->page, len, off);
+}
+
 /**
  * bio_add_folio - Attempt to add part of a folio to a bio.
  * @bio: BIO to add to.
include/linux/bio.h +1 -0
index 0f8a8d7a6384..99fa832db836 100644
@@ -473,6 +473,7 @@ int bio_add_zone_append_page(struct bio *bio, struct page *page,
 			     unsigned int len, unsigned int offset);
 void __bio_add_page(struct bio *bio, struct page *page,
 		unsigned int len, unsigned int off);
+void __bio_add_folio(struct bio *, struct folio *, size_t len, size_t off);
 int bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter);
 void bio_iov_bvec_set(struct bio *bio, struct iov_iter *iter);
 void __bio_release_pages(struct bio *bio, bool mark_dirty);
Signature · 1 line
2.39.2
author reviewer
#22 lore raw

PATCH v4 21/22 fs: iomap: use __bio_add_folio where possible

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

When the iomap buffered-io code can't add a folio to a bio, it allocates a new bio and adds the folio to that one. This is done using bio_add_folio(), but doesn't check for errors.

As adding a folio to a newly created bio can't fail, use the newly introduced __bio_add_folio() function.

---
 fs/iomap/buffered-io.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
fs/iomap/buffered-io.c +3 -3
index 6f4c97a6d7e9..473598b68067 100644
@@ -312,7 +312,7 @@ static loff_t iomap_readpage_iter(const struct iomap_iter *iter,
 			ctx->bio->bi_opf |= REQ_RAHEAD;
 		ctx->bio->bi_iter.bi_sector = sector;
 		ctx->bio->bi_end_io = iomap_read_end_io;
-		bio_add_folio(ctx->bio, folio, plen, poff);
+		__bio_add_folio(ctx->bio, folio, plen, poff);
 	}
 
 done:
@@ -546,7 +546,7 @@ static int iomap_read_folio_sync(loff_t block_start, struct folio *folio,
 
 	bio_init(&bio, iomap->bdev, &bvec, 1, REQ_OP_READ);
 	bio.bi_iter.bi_sector = iomap_sector(iomap, block_start);
-	bio_add_folio(&bio, folio, plen, poff);
+	__bio_add_folio(&bio, folio, plen, poff);
 	return submit_bio_wait(&bio);
 }
 
@@ -1589,7 +1589,7 @@ iomap_add_to_ioend(struct inode *inode, loff_t pos, struct folio *folio,
 
 	if (!bio_add_folio(wpc->ioend->io_bio, folio, len, poff)) {
 		wpc->ioend->io_bio = iomap_chain_bio(wpc->ioend->io_bio);
-		bio_add_folio(wpc->ioend->io_bio, folio, len, poff);
+		__bio_add_folio(wpc->ioend->io_bio, folio, len, poff);
 	}
 
 	if (iop)
Signature · 1 line
2.39.2
author reviewer
#23 lore raw

PATCH v4 22/22 block: mark bio_add_folio as __must_check

In reply to Johannes Thumshirn

From: Johannes Thumshirn <[email protected]>

Now that all callers of bio_add_folio() check the return value, mark it as __must_check.

---
 include/linux/bio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/bio.h +1 -1
index 99fa832db836..36cfc091caed 100644
@@ -466,7 +466,7 @@ void bio_reset(struct bio *bio, struct block_device *bdev, blk_opf_t opf);
 void bio_chain(struct bio *, struct bio *);
 
 int __must_check bio_add_page(struct bio *, struct page *, unsigned len, unsigned off);
-bool bio_add_folio(struct bio *, struct folio *, size_t len, size_t off);
+bool __must_check bio_add_folio(struct bio *, struct folio *, size_t len, size_t off);
 extern int bio_add_pc_page(struct request_queue *, struct bio *, struct page *,
 			   unsigned int, unsigned int);
 int bio_add_zone_append_page(struct bio *bio, struct page *page,
Signature · 1 line
2.39.2