Skip to contentSkip to conversation

btrfs: lzo: reject compressed segment that overflows the compressed input

PATCH Superseded Weiming Shi 2 participants 3 messages Jun 6 – Jun 7 Btrfs file system

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 btrfs: lzo: reject compressed segment that overflows the compressed input

lzo_decompress_bio() validates each on-disk segment length seg_len only against the workspace cbuf size, not against the compressed input size (compressed_len, the total folio bytes of the bio). A crafted extent can carry a segment whose seg_len passes the cbuf check but runs past the end of the bio, so copy_compressed_segment() walks off the last folio:
get_current_folio() then returns the NULL folio from bio_next_folio(), and with CONFIG_BTRFS_ASSERT disabled (default) folio_size(NULL) faults.

 BUG: KASAN: null-ptr-deref in lzo_decompress_bio (fs/btrfs/lzo.c:383)
 Read of size 8 at addr 0000000000000000 by task kworker/u8:1/29
 Workqueue: btrfs-endio simple_end_io_work
  kasan_report (mm/kasan/report.c:590)
  lzo_decompress_bio (fs/btrfs/lzo.c:383)
  end_bbio_compressed_read (fs/btrfs/compression.c:1065)
  btrfs_bio_end_io (fs/btrfs/bio.c:135)
  btrfs_check_read_bio (fs/btrfs/bio.c:180 fs/btrfs/bio.c:285)
  simple_end_io_work
  process_one_work
  worker_thread

Reject any segment whose payload would extend beyond compressed_len before copying it.

Fixes a6e66e6f8c1b ("btrfs: rework lzo_decompress_bio() to make it subpage compatible")
Assisted-by Claude:claude-opus-4-8
---
 fs/btrfs/lzo.c | 4 ++++
 1 file changed, 4 insertions(+)
fs/btrfs/lzo.c +4 -0
index 2de18c7b563a..887d740b27aa 100644
@@ -491,6 +491,10 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
 			return -EIO;
 		}
 
+		/* The segment must not extend beyond the compressed input. */
+		if (unlikely(cur_in + seg_len > compressed_len))
+			return -EIO;
+
 		/* Copy the compressed segment payload into workspace */
 		copy_compressed_segment(cb, &fi, &cur_folio_index, workspace->cbuf,
 					seg_len, &cur_in);
Signature · 1 line
2.43.0
#1 lore raw
Quoted 36 lines · Weiming Shi

lzo_decompress_bio() validates each on-disk segment length seg_len only against the workspace cbuf size, not against the compressed input size (compressed_len, the total folio bytes of the bio). A crafted extent can carry a segment whose seg_len passes the cbuf check but runs past the end of the bio, so copy_compressed_segment() walks off the last folio:
get_current_folio() then returns the NULL folio from bio_next_folio(), and with CONFIG_BTRFS_ASSERT disabled (default) folio_size(NULL) faults.

  BUG: KASAN: null-ptr-deref in lzo_decompress_bio (fs/btrfs/lzo.c:383)
  Read of size 8 at addr 0000000000000000 by task kworker/u8:1/29
  Workqueue: btrfs-endio simple_end_io_work
   kasan_report (mm/kasan/report.c:590)
   lzo_decompress_bio (fs/btrfs/lzo.c:383)
   end_bbio_compressed_read (fs/btrfs/compression.c:1065)
   btrfs_bio_end_io (fs/btrfs/bio.c:135)
   btrfs_check_read_bio (fs/btrfs/bio.c:180 fs/btrfs/bio.c:285)
   simple_end_io_work
   process_one_work
   worker_thread

Reject any segment whose payload would extend beyond compressed_len before copying it.

Fixes a6e66e6f8c1b ("btrfs: rework lzo_decompress_bio() to make it subpage compatible")
Assisted-by Claude:claude-opus-4-8
---
  fs/btrfs/lzo.c | 4 ++++
  1 file changed, 4 insertions(+)
fs/btrfs/lzo.c +0 -0
index 2de18c7b563a..887d740b27aa 100644

fs/btrfs/lzo.c int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) · line 497

Quoted 7 lines
fs/btrfs/lzo.c +4 -0
@@ -491,6 +491,10 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
  			return -EIO;
  		}
 
+		/* The segment must not extend beyond the compressed input. */
+		if (unlikely(cur_in + seg_len > compressed_len))
+			return -EIO;
+

In lzo_decompress() we return -EUCLEAN, please follow that pattern.

Although it would be better to output a message when such -EUCLEAN is returned.

Otherwise looks good to me.

Thanks,
Qu

Quoted 3 lines
fs/btrfs/lzo.c +0 -0
  		/* Copy the compressed segment payload into workspace */
  		copy_compressed_segment(cb, &fi, &cur_folio_index, workspace->cbuf,
  					seg_len, &cur_in);
#2 lore raw
Quoted 48 lines · Qu Wenruo
Quoted 36 lines · Weiming Shi

lzo_decompress_bio() validates each on-disk segment length seg_len only against the workspace cbuf size, not against the compressed input size (compressed_len, the total folio bytes of the bio). A crafted extent can carry a segment whose seg_len passes the cbuf check but runs past the end of the bio, so copy_compressed_segment() walks off the last folio:
get_current_folio() then returns the NULL folio from bio_next_folio(), and with CONFIG_BTRFS_ASSERT disabled (default) folio_size(NULL) faults.

  BUG: KASAN: null-ptr-deref in lzo_decompress_bio (fs/btrfs/lzo.c:383)
  Read of size 8 at addr 0000000000000000 by task kworker/u8:1/29
  Workqueue: btrfs-endio simple_end_io_work
   kasan_report (mm/kasan/report.c:590)
   lzo_decompress_bio (fs/btrfs/lzo.c:383)
   end_bbio_compressed_read (fs/btrfs/compression.c:1065)
   btrfs_bio_end_io (fs/btrfs/bio.c:135)
   btrfs_check_read_bio (fs/btrfs/bio.c:180 fs/btrfs/bio.c:285)
   simple_end_io_work
   process_one_work
   worker_thread

Reject any segment whose payload would extend beyond compressed_len before copying it.

Fixes a6e66e6f8c1b ("btrfs: rework lzo_decompress_bio() to make it subpage compatible")
Assisted-by Claude:claude-opus-4-8
---
  fs/btrfs/lzo.c | 4 ++++
  1 file changed, 4 insertions(+)
fs/btrfs/lzo.c +0 -0
index 2de18c7b563a..887d740b27aa 100644

fs/btrfs/lzo.c int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) · line 496

Quoted 6 lines
fs/btrfs/lzo.c +4 -0
@@ -491,6 +491,10 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
  			return -EIO;
  		}
+		/* The segment must not extend beyond the compressed input. */
+		if (unlikely(cur_in + seg_len > compressed_len))
+			return -EIO;
+

In lzo_decompress() we return -EUCLEAN, please follow that pattern.

Quoted 12 lines

Although it would be better to output a message when such -EUCLEAN is returned.

Otherwise looks good to me.

Thanks,
Qu

Quoted 3 lines
fs/btrfs/lzo.c +0 -0
  		/* Copy the compressed segment payload into workspace */
  		copy_compressed_segment(cb, &fi, &cur_folio_index, workspace->cbuf,
  					seg_len, &cur_in);

Thanks for the review. v2 sent.

Thanks,
Weiming Shi