Skip to contentSkip to conversation

Fix two issues about bigalloc feature

PATCH Applied Ye Bin 3 participants 9 messages Dec 8, 2022 – Dec 9, 2022 ext4 file system

lore
Share & tools
lore

Ye Bin

author
#0 lore raw

v4 cover Fix two issues about bigalloc feature

From: Ye Bin <[email protected]>

Diff v4 Vs v3:
1. Fix checkpatch warning.
2. Fix alignment issues about patch [2-3]

Diff v3 Vs v2:
1. Add fixes tag and rename label 'out' for first patch.
2. Do not split string across lines for second patch.
3. Just check pending tree if empty, drop clear code for third patch.

Diff V2 vs V1:
Use ext4_error() when detect 'i_reserved_data_blocks' and pending tree abnormal.

Ye Bin (3):
ext4: fix incorrect calculate 'reserved' in '__es_remove_extent' when
enable bigalloc feature
ext4: record error when detect abnormal 'i_reserved_data_blocks'
ext4: add check pending tree when evict inode

 fs/ext4/extents_status.c |  3 ++-
 fs/ext4/super.c          | 13 +++++++++----
 2 files changed, 11 insertions(+), 5 deletions(-)
Signature · 1 line
2.31.1
author
#1 lore raw

PATCH v4 1/3 ext4: fix incorrect calculate 'reserved' in '__es_remove_extent' when enable bigalloc feature

From: Ye Bin <[email protected]>

Syzbot report issue as follows:
EXT4-fs error (device loop0): ext4_validate_block_bitmap:398: comm rep:
bg 0: block 5: invalid block bitmap
EXT4-fs (loop0): Delayed block allocation failed for inode 18 at logical
offset 0 with max blocks 32 with error 28
EXT4-fs (loop0): This should not happen!! Data will be lost

EXT4-fs (loop0): Total free blocks count 0
EXT4-fs (loop0): Free/Dirty block details
EXT4-fs (loop0): free_blocks=0
EXT4-fs (loop0): dirty_blocks=32
EXT4-fs (loop0): Block reservation details
EXT4-fs (loop0): i_reserved_data_blocks=2
EXT4-fs (loop0): Inode 18 (00000000845cd634):
i_reserved_data_blocks (1) not cleared!

Above issue happens as follows:
Assume:
sbi->s_cluster_ratio = 16
Step1:
Insert delay block [0, 31] -> ei->i_reserved_data_blocks=2 Step2:
ext4_writepages
mpage_map_and_submit_extent -> return failed mpage_release_unused_pages -> to release [0, 30]

    ext4_es_remove_extent -> remove lblk=0 end=30
      __es_remove_extent -> len1=0 len2=31-30=1
 __es_remove_extent:
 ...
 if (len2 > 0) {
  ...
	  if (len1 > 0) {
		  ...
	  } else {
		es->es_lblk = end + 1;
		es->es_len = len2;
		...
	  }
  	if (count_reserved)
		count_rsvd(inode, lblk, ...);
	goto out; -> will return but didn't calculate 'reserved'

...
Step3:
ext4_destroy_inode -> trigger "i_reserved_data_blocks (1) not cleared!"

To solve above issue if 'len2>0' call 'get_rsvd()' before goto out.

Reported-by [email protected]
Fixes 8fcc3a580651 ("ext4: rework reserved cluster accounting when invalidating pages")
---
 fs/ext4/extents_status.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/ext4/extents_status.c +2 -1
index cd0a861853e3..7ada374ff27d 100644
@@ -1371,7 +1371,7 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 		if (count_reserved)
 			count_rsvd(inode, lblk, orig_es.es_len - len1 - len2,
 				   &orig_es, &rc);
-		goto out;
+		goto out_get_reserved;
 	}
 
 	if (len1 > 0) {
@@ -1413,6 +1413,7 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 		}
 	}
 
+out_get_reserved:
 	if (count_reserved)
 		*reserved = get_rsvd(inode, end, es, &rc);
 out:
Signature · 1 line
2.31.1
Reviewed-by on v4 1/3
#2 lore raw

Re: [PATCH v4 1/3] ext4: fix incorrect calculate 'reserved' in '__es_remove_extent' when enable bigalloc feature

* Ye Bin <[email protected]>:

Quoted 71 lines

From: Ye Bin <[email protected]>

Syzbot report issue as follows:
EXT4-fs error (device loop0): ext4_validate_block_bitmap:398: comm rep:
bg 0: block 5: invalid block bitmap
EXT4-fs (loop0): Delayed block allocation failed for inode 18 at logical
offset 0 with max blocks 32 with error 28
EXT4-fs (loop0): This should not happen!! Data will be lost

EXT4-fs (loop0): Total free blocks count 0
EXT4-fs (loop0): Free/Dirty block details
EXT4-fs (loop0): free_blocks=0
EXT4-fs (loop0): dirty_blocks=32
EXT4-fs (loop0): Block reservation details
EXT4-fs (loop0): i_reserved_data_blocks=2
EXT4-fs (loop0): Inode 18 (00000000845cd634):
i_reserved_data_blocks (1) not cleared!

Above issue happens as follows:
Assume:
sbi->s_cluster_ratio = 16
Step1:
Insert delay block [0, 31] -> ei->i_reserved_data_blocks=2 Step2:
ext4_writepages
mpage_map_and_submit_extent -> return failed mpage_release_unused_pages -> to release [0, 30]

    ext4_es_remove_extent -> remove lblk=0 end=30
      __es_remove_extent -> len1=0 len2=31-30=1
 __es_remove_extent:
 ...
 if (len2 > 0) {
  ...
	  if (len1 > 0) {
		  ...
	  } else {
		es->es_lblk = end + 1;
		es->es_len = len2;
		...
	  }
  	if (count_reserved)
		count_rsvd(inode, lblk, ...);
	goto out; -> will return but didn't calculate 'reserved'

...
Step3:
ext4_destroy_inode -> trigger "i_reserved_data_blocks (1) not cleared!"

To solve above issue if 'len2>0' call 'get_rsvd()' before goto out.

Reported-by [email protected]
Fixes 8fcc3a580651 ("ext4: rework reserved cluster accounting when invalidating pages")
---
 fs/ext4/extents_status.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/ext4/extents_status.c +1 -1
index cd0a861853e3..7ada374ff27d 100644
@@ -1371,7 +1371,7 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 		if (count_reserved)
 			count_rsvd(inode, lblk, orig_es.es_len - len1 - len2,
 				   &orig_es, &rc);
-		goto out;
+		goto out_get_reserved;
 	}
 
 	if (len1 > 0) {

fs/ext4/extents_status.c static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk, · line 1419

Quoted 7 lines
fs/ext4/extents_status.c +1 -0
@@ -1413,6 +1413,7 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 		}
 	}
 
+out_get_reserved:
 	if (count_reserved)
 		*reserved = get_rsvd(inode, end, es, &rc);
 out:

OK, this looks good.

As before, feel free to add:

Quoted 2 lines

2.31.1

maintainer Applied
#3 lore raw

Re: [PATCH v4 1/3] ext4: fix incorrect calculate 'reserved' in '__es_remove_extent' when enable bigalloc feature

In reply to Ye Bin
Quoted 2 lines · Ye Bin

From: Ye Bin <[email protected]>

Thanks, applied with an edit commit description to make it clearer what's being fixed.

ext4: fix reserved cluster accounting in __es_remove_extent()

When bigalloc is enabled, reserved cluster accounting for delayed allocation is handled in extent_status.c. With a corrupted file system, it's possible for this accounting to be incorrect, dsicovered by Syzbot:

....

In general, it's better to explain what is being changed and why, and put the big messy Syzbot change after the English description of the change. Remember, what's important is that we make ext4 better --- not that we are getting rid of a Syzbot report. When someone reads the commit description later, what they will care about is how the code has been improved.

Cheers,

- Ted

Quoted 78 lines

Syzbot report issue as follows:
EXT4-fs error (device loop0): ext4_validate_block_bitmap:398: comm rep:
bg 0: block 5: invalid block bitmap
EXT4-fs (loop0): Delayed block allocation failed for inode 18 at logical
offset 0 with max blocks 32 with error 28
EXT4-fs (loop0): This should not happen!! Data will be lost

EXT4-fs (loop0): Total free blocks count 0
EXT4-fs (loop0): Free/Dirty block details
EXT4-fs (loop0): free_blocks=0
EXT4-fs (loop0): dirty_blocks=32
EXT4-fs (loop0): Block reservation details
EXT4-fs (loop0): i_reserved_data_blocks=2
EXT4-fs (loop0): Inode 18 (00000000845cd634):
i_reserved_data_blocks (1) not cleared!

Above issue happens as follows:
Assume:
sbi->s_cluster_ratio = 16
Step1:
Insert delay block [0, 31] -> ei->i_reserved_data_blocks=2 Step2:
ext4_writepages
mpage_map_and_submit_extent -> return failed mpage_release_unused_pages -> to release [0, 30]

    ext4_es_remove_extent -> remove lblk=0 end=30
      __es_remove_extent -> len1=0 len2=31-30=1
 __es_remove_extent:
 ...
 if (len2 > 0) {
  ...
	  if (len1 > 0) {
		  ...
	  } else {
		es->es_lblk = end + 1;
		es->es_len = len2;
		...
	  }
  	if (count_reserved)
		count_rsvd(inode, lblk, ...);
	goto out; -> will return but didn't calculate 'reserved'

...
Step3:
ext4_destroy_inode -> trigger "i_reserved_data_blocks (1) not cleared!"

To solve above issue if 'len2>0' call 'get_rsvd()' before goto out.

Reported-by [email protected]
Fixes 8fcc3a580651 ("ext4: rework reserved cluster accounting when invalidating pages")
---
 fs/ext4/extents_status.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/ext4/extents_status.c +2 -1
index cd0a861853e3..7ada374ff27d 100644
@@ -1371,7 +1371,7 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 		if (count_reserved)
 			count_rsvd(inode, lblk, orig_es.es_len - len1 - len2,
 				   &orig_es, &rc);
-		goto out;
+		goto out_get_reserved;
 	}
 
 	if (len1 > 0) {
@@ -1413,6 +1413,7 @@ static int __es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 		}
 	}
 
+out_get_reserved:
 	if (count_reserved)
 		*reserved = get_rsvd(inode, end, es, &rc);
 out:
Signature · 1 line
2.31.1
author
#4 lore raw

PATCH v4 2/3 ext4: record error when detect abnormal 'i_reserved_data_blocks'

In reply to Ye Bin

From: Ye Bin <[email protected]>

If 'i_reserved_data_blocks' is not cleared which mean something wrong with code, free space accounting is likely wrong, according to Jan Kara's advice use ext4_error() to record this abnormal let fsck to repair and also we can capture this issue.

---
 fs/ext4/super.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
fs/ext4/super.c +4 -4
index 840e0a614959..4b2d257d3845 100644
@@ -1387,10 +1387,10 @@ static void ext4_destroy_inode(struct inode *inode)
 	}
 
 	if (EXT4_I(inode)->i_reserved_data_blocks)
-		ext4_msg(inode->i_sb, KERN_ERR,
-			 "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",
-			 inode->i_ino, EXT4_I(inode),
-			 EXT4_I(inode)->i_reserved_data_blocks);
+		ext4_error(inode->i_sb,
+			   "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",
+			   inode->i_ino, EXT4_I(inode),
+			   EXT4_I(inode)->i_reserved_data_blocks);
 }
 
 static void init_once(void *foo)
Signature · 1 line
2.31.1
maintainer
#5 lore raw

Re: [PATCH v4 2/3] ext4: record error when detect abnormal 'i_reserved_data_blocks'

Quoted 6 lines · Ye Bin

From: Ye Bin <[email protected]>

If 'i_reserved_data_blocks' is not cleared which mean something wrong with code, free space accounting is likely wrong, according to Jan Kara's advice use ext4_error() to record this abnormal let fsck to repair and also we can capture this issue.

If i_reserved_data_block, it means that there is something wrong with our delayed allocation accounting. However, this accounting is usually only an in-memory error. The one exception is if quota is enabled, in which case the space is decremented from the user/group/project quota.

However, if quota is not in use, which is the overwhelmingly common case, there will be nothing for fsck to repair. (It does mean that df will show a slightly smaller free space value due to the messed up delayed allocation accounting, but that disappears after a reboot.)

Marking the file system as in need of repair when nothing is actually wrong with the on-disk file system can backfire, in that it confuses users when they see the ext4 error but then when they run fsck, fsck reports nothing wrong --- at which point they file a bug.

We could use ext4_error if quotas are enabled, and ext4_msg if not, but it might not worth the extra complexity.

Cheers,

- Ted

author
#6 lore raw

PATCH v4 3/3 ext4: add check pending tree when evict inode

In reply to Ye Bin

From: Ye Bin <[email protected]>

Syzbot found the following issue:

BUG: memory leak
unreferenced object 0xffff8881bde17420 (size 32):
  comm "rep", pid 2327, jiffies 4295381963 (age 32.265s)
  hex dump (first 32 bytes):
    01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<00000000ac6d38f8>] __insert_pending+0x13c/0x2d0
    [<00000000d717de3b>] ext4_es_insert_delayed_block+0x399/0x4e0
    [<000000004be03913>] ext4_da_map_blocks.constprop.0+0x739/0xfa0
    [<00000000885a832a>] ext4_da_get_block_prep+0x10c/0x440
    [<0000000029b7f8ef>] __block_write_begin_int+0x28d/0x860
    [<00000000e182ebc3>] ext4_da_write_inline_data_begin+0x2d1/0xf30
    [<00000000ced0c8a2>] ext4_da_write_begin+0x612/0x860
    [<000000008d5f27fa>] generic_perform_write+0x215/0x4d0
    [<00000000552c1cde>] ext4_buffered_write_iter+0x101/0x3b0
    [<0000000052177ae8>] do_iter_readv_writev+0x19f/0x340
    [<000000004b9de834>] do_iter_write+0x13b/0x650
    [<00000000e2401b9b>] iter_file_splice_write+0x5a5/0xab0
    [<0000000023aa5d90>] direct_splice_actor+0x103/0x1e0
    [<0000000089e00fc1>] splice_direct_to_actor+0x2c9/0x7b0
    [<000000004386851e>] do_splice_direct+0x159/0x280
    [<00000000b567e609>] do_sendfile+0x932/0x1200

Above issue fixed by
commit 1b8f787ef547 ("ext4: fix warning in 'ext4_da_release_space'") in this scene. To make things better add check pending tree when evict inode.
According to Eric Whitney's suggestion, bigalloc + inline is still in development so we just add test for this situation, there isn't need to add code to free pending tree entry.

---
 fs/ext4/super.c | 5 +++++
 1 file changed, 5 insertions(+)
fs/ext4/super.c +5 -0
index 4b2d257d3845..15b6634975e7 100644
@@ -1391,6 +1391,11 @@ static void ext4_destroy_inode(struct inode *inode)
 			   "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",
 			   inode->i_ino, EXT4_I(inode),
 			   EXT4_I(inode)->i_reserved_data_blocks);
+
+	if (!RB_EMPTY_ROOT(&EXT4_I(inode)->i_pending_tree.root))
+		ext4_error(inode->i_sb,
+			   "Inode %lu (%p): i_pending_tree not empty!",
+			   inode->i_ino, EXT4_I(inode));
 }
 
 static void init_once(void *foo)
Signature · 1 line
2.31.1
Reviewed-by on v4 3/3
#7 lore raw

Re: [PATCH v4 3/3] ext4: add check pending tree when evict inode

* Ye Bin <[email protected]>:

Quoted 48 lines

From: Ye Bin <[email protected]>

Syzbot found the following issue:

BUG: memory leak
unreferenced object 0xffff8881bde17420 (size 32):
  comm "rep", pid 2327, jiffies 4295381963 (age 32.265s)
  hex dump (first 32 bytes):
    01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<00000000ac6d38f8>] __insert_pending+0x13c/0x2d0
    [<00000000d717de3b>] ext4_es_insert_delayed_block+0x399/0x4e0
    [<000000004be03913>] ext4_da_map_blocks.constprop.0+0x739/0xfa0
    [<00000000885a832a>] ext4_da_get_block_prep+0x10c/0x440
    [<0000000029b7f8ef>] __block_write_begin_int+0x28d/0x860
    [<00000000e182ebc3>] ext4_da_write_inline_data_begin+0x2d1/0xf30
    [<00000000ced0c8a2>] ext4_da_write_begin+0x612/0x860
    [<000000008d5f27fa>] generic_perform_write+0x215/0x4d0
    [<00000000552c1cde>] ext4_buffered_write_iter+0x101/0x3b0
    [<0000000052177ae8>] do_iter_readv_writev+0x19f/0x340
    [<000000004b9de834>] do_iter_write+0x13b/0x650
    [<00000000e2401b9b>] iter_file_splice_write+0x5a5/0xab0
    [<0000000023aa5d90>] direct_splice_actor+0x103/0x1e0
    [<0000000089e00fc1>] splice_direct_to_actor+0x2c9/0x7b0
    [<000000004386851e>] do_splice_direct+0x159/0x280
    [<00000000b567e609>] do_sendfile+0x932/0x1200

Above issue fixed by
commit 1b8f787ef547 ("ext4: fix warning in 'ext4_da_release_space'") in this scene. To make things better add check pending tree when evict inode.
According to Eric Whitney's suggestion, bigalloc + inline is still in development so we just add test for this situation, there isn't need to add code to free pending tree entry.

---
 fs/ext4/super.c | 5 +++++
 1 file changed, 5 insertions(+)
fs/ext4/super.c +0 -0
index 4b2d257d3845..15b6634975e7 100644
@@ -1391,6 +1391,11 @@ static void ext4_destroy_inode(struct inode *inode)
 			   "Inode %lu (%p): i_reserved_data_blocks (%u) not cleared!",

fs/ext4/super.c static void ext4_destroy_inode(struct inode *inode) · line 1401

Quoted 10 lines
fs/ext4/super.c +5 -0
 			   inode->i_ino, EXT4_I(inode),
 			   EXT4_I(inode)->i_reserved_data_blocks);
+
+	if (!RB_EMPTY_ROOT(&EXT4_I(inode)->i_pending_tree.root))
+		ext4_error(inode->i_sb,
+			   "Inode %lu (%p): i_pending_tree not empty!",
+			   inode->i_ino, EXT4_I(inode));
 }
 
 static void init_once(void *foo)

Looks good. Feel free to add:

Quoted 2 lines

2.31.1

maintainer
#8 lore raw

Re: [PATCH v4 3/3] ext4: add check pending tree when evict inode

In reply to Ye Bin
Quoted 8 lines · Ye Bin

Above issue fixed by
commit 1b8f787ef547 ("ext4: fix warning in 'ext4_da_release_space'") in this scene. To make things better add check pending tree when evict inode.
According to Eric Whitney's suggestion, bigalloc + inline is still in development so we just add test for this situation, there isn't need to add code to free pending tree entry.

The i_pending_tree is an in-memory data structure, and so it's not appropriate to call ext4_error(), because there will be nothing for fsck to fix.

If you really want to a bug to be noticed, you could use a ext4_msg plus a WARN_ON(); but an ext4_error() is really not appropriate.

Cheers,

- Ted