Filipe Manana [Sun, 7 Dec 2014 21:31:47 +0000 (21:31 +0000)]
Btrfs: fix fs corruption on transaction abort if device supports discard
When we abort a transaction we iterate over all the ranges marked as dirty
in fs_info->freed_extents[0] and fs_info->freed_extents[1], clear them
from those trees, add them back (unpin) to the free space caches and, if
the fs was mounted with "-o discard", perform a discard on those regions.
Also, after adding the regions to the free space caches, a fitrim ioctl call
can see those ranges in a block group's free space cache and perform a discard
on the ranges, so the same issue can happen without "-o discard" as well.
This causes corruption, affecting one or multiple btree nodes (in the worst
case leaving the fs unmountable) because some of those ranges (the ones in
the fs_info->pinned_extents tree) correspond to btree nodes/leafs that are
referred by the last committed super block - breaking the rule that anything
that was committed by a transaction is untouched until the next transaction
commits successfully.
I ran into this while running in a loop (for several hours) the fstest that
I recently submitted:
[PATCH] fstests: add btrfs test to stress chunk allocation/removal and fstrim
The corruption always happened when a transaction aborted and then fsck complained
like this:
_check_btrfs_filesystem: filesystem on /dev/sdc is inconsistent
*** fsck.btrfs output ***
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
Check tree block failed, want=94945280, have=0
read block failed check_tree_block
Couldn't open file system
In this case 94945280 corresponded to the root of a tree.
Using frace what I observed was the following sequence of steps happened:
1) transaction N started, fs_info->pinned_extents pointed to
fs_info->freed_extents[0];
4) transaction N commit starts, fs_info->pinned_extents now points to
fs_info->freed_extents[1], and transaction N completes;
5) transaction N + 1 starts;
6) eb is COWed, and btrfs_free_tree_block() called for this eb;
7) eb range (94945280 to 94945280 + 16Kb) is added to
fs_info->pinned_extents (fs_info->freed_extents[1]);
8) Something goes wrong in transaction N + 1, like hitting ENOSPC
for example, and the transaction is aborted, turning the fs into
readonly mode. The stack trace I got for example:
[112065.253935] [<ffffffff8140c7b6>] dump_stack+0x4d/0x66
[112065.254271] [<ffffffff81042984>] warn_slowpath_common+0x7f/0x98
[112065.254567] [<ffffffffa0325990>] ? __btrfs_abort_transaction+0x50/0x10b [btrfs]
[112065.261674] [<ffffffff810429e5>] warn_slowpath_fmt+0x48/0x50
[112065.261922] [<ffffffffa032949e>] ? btrfs_free_path+0x26/0x29 [btrfs]
[112065.262211] [<ffffffffa0325990>] __btrfs_abort_transaction+0x50/0x10b [btrfs]
[112065.262545] [<ffffffffa036b1d6>] btrfs_remove_chunk+0x537/0x58b [btrfs]
[112065.262771] [<ffffffffa033840f>] btrfs_delete_unused_bgs+0x1de/0x21b [btrfs]
[112065.263105] [<ffffffffa0343106>] cleaner_kthread+0x100/0x12f [btrfs]
(...)
[112065.264493] ---[ end trace dd7903a975a31a08 ]---
[112065.264673] BTRFS: error (device sdc) in btrfs_remove_chunk:2625: errno=-28 No space left
[112065.264997] BTRFS info (device sdc): forced readonly
9) The clear kthread sees that the BTRFS_FS_STATE_ERROR bit is set in
fs_info->fs_state and calls btrfs_cleanup_transaction(), which in
turn calls btrfs_destroy_pinned_extent();
10) Then btrfs_destroy_pinned_extent() iterates over all the ranges
marked as dirty in fs_info->freed_extents[], and for each one
it calls discard, if the fs was mounted with "-o discard", and
adds the range to the free space cache of the respective block
group;
11) btrfs_trim_block_group(), invoked from the fitrim ioctl code path,
sees the free space entries and performs a discard;
12) After an umount and mount (or fsck), our eb's location on disk was full
of zeroes, and it should have been untouched, because it was marked as
dirty in the fs_info->pinned_extents tree, and therefore used by the
trees that the last committed superblock points to.
Fix this by not performing a discard and not adding the ranges to the free space
caches - it's useless from this point since the fs is now in readonly mode and
we won't write free space caches to disk anymore (otherwise we would leak space)
nor any new superblock. By not adding the ranges to the free space caches, it
prevents other code paths from allocating that space and write to it as well,
therefore being safer and simpler.
Cc: stable@vger.kernel.org # any kernel released after 2011-01-06 Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 678886bdc6378c1cbd5072da2c5a3035000214e3)
Filipe Manana [Thu, 4 Dec 2014 18:38:30 +0000 (18:38 +0000)]
Btrfs: always clear a block group node when removing it from the tree
Always clear a block group's rbnode after removing it from the rbtree to
ensure that any tasks that might be holding a reference on the block group
don't end up accessing stale rbnode left and right child pointers through
next_block_group().
This is a leftover from the change titled:
"Btrfs: fix invalid block group rbtree access after bg is removed"
Filipe Manana [Thu, 4 Dec 2014 15:31:01 +0000 (15:31 +0000)]
Btrfs: ensure deletion from pinned_chunks list is protected
The call to remove_extent_mapping() actually deletes the extent map
from the list it's included in - fs_info->pinned_chunks - and that
list is protected by the chunk mutex. Therefore make that call
while holding the chunk mutex and remove the redundant list delete
call because it's a noop.
This fixes an overlook of the patch titled
"Btrfs: fix race between fs trimming and block group remove/allocation"
following the same obvervation from the patch titled
"Btrfs: fix unprotected deletion from pending_chunks list".
Josef Bacik [Wed, 26 Nov 2014 16:52:54 +0000 (11:52 -0500)]
Btrfs: make get_caching_control unconditionally return the ctl
This was written when we didn't do a caching control for the fast free space
cache loading. However we started doing that a long time ago, and there is
still a small window of time that we could be caching the block group the fast
way, so if there is a caching_ctl at all on the block group just return it, the
callers all wait properly for what they want. Thanks,
Filipe Manana [Tue, 2 Dec 2014 18:07:49 +0000 (18:07 +0000)]
Btrfs: fix unprotected deletion from pending_chunks list
On block group remove if the corresponding extent map was on the
transaction->pending_chunks list, we were deleting the extent map
from that list, through remove_extent_mapping(), without any
synchronization with chunk allocation (which iterates that list
and adds new elements to it). Fix this by ensure that this is done
while the chunk mutex is held, since that's the mutex that protects
the list in the chunk allocation code path.
This applies on top (depends on) of my previous patch titled:
"Btrfs: fix race between fs trimming and block group remove/allocation"
But the issue in fact was already present before that change, it only
became easier to hit after Josef's 3.18 patch that added automatic
removal of empty block groups.
Filipe Manana [Tue, 2 Dec 2014 18:07:30 +0000 (18:07 +0000)]
Btrfs: fix fs mapping extent map leak
On chunk allocation error (label "error_del_extent"), after adding the
extent map to the tree and to the pending chunks list, we would leave
decrementing the extent map's refcount by 2 instead of 3 (our allocation
+ tree reference + list reference).
Also, on chunk/block group removal, if the block group was on the list
pending_chunks we weren't decrementing the respective list reference.
This applies on top (depends on) of my previous patch titled:
"Btrfs: fix race between fs trimming and block group remove/allocation"
But the issue in fact was already present before that change, it only
became easier to hit after Josef's 3.18 patch that added automatic
removal of empty block groups.
Filipe Manana [Mon, 1 Dec 2014 17:04:40 +0000 (17:04 +0000)]
Btrfs: fix memory leak after block remove + trimming
There was a free space entry structure memeory leak if a block
group is remove while a free space entry is being trimmed, which
the following diagram explains:
CPU 1 CPU 2
btrfs_trim_block_group()
trim_no_bitmap()
remove free space entry from
block group cache's rbtree
do_trimming()
The free space entry added after doing the discard of its respective
range ends up never being freed.
Detected after doing an "rmmod btrfs" after running the stress test
recently submitted for fstests:
Filipe Manana [Wed, 26 Nov 2014 15:28:55 +0000 (15:28 +0000)]
Btrfs: make btrfs_abort_transaction consider existence of new block groups
If the transaction handle doesn't have used blocks but has created new block
groups make sure we turn the fs into readonly mode too. This is because the
new block groups didn't get all their metadata persisted into the chunk and
device trees, and therefore if a subsequent transaction starts, allocates
space from the new block groups, writes data or metadata into that space,
commits successfully and then after we unmount and mount the filesystem
again, the same space can be allocated again for a new block group,
resulting in file data or metadata corruption.
Example where we don't abort the transaction when we fail to finish the
chunk allocation (add items to the chunk and device trees) and later a
future transaction where the block group is removed fails because it can't
find the chunk item in the chunk tree:
Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit c92f6be34c501406daf5e61f3569a1813f985393)
Filipe Manana [Mon, 1 Dec 2014 17:04:09 +0000 (17:04 +0000)]
Btrfs: fix race between writing free space cache and trimming
Trimming is completely transactionless, and the way it operates consists
of hiding free space entries from a block group, perform the trim/discard
and then make the free space entries visible again.
Therefore while a free space entry is being trimmed, we can have free space
cache writing running in parallel (as part of a transaction commit) which
will miss the free space entry. This means that an unmount (or crash/reboot)
after that transaction commit and mount again before another transaction
starts/commits after the discard finishes, we will have some free space
that won't be used again unless the free space cache is rebuilt. After the
unmount, fsck (btrfsck, btrfs check) reports the issue like the following
example:
*** fsck.btrfs output ***
checking extents
checking free space cache
There is no free space entry for 521764864-521781248
There is no free space entry for 521764864-1103101952
cache appears valid but isnt 29360128
Checking filesystem on /dev/sdc
UUID: b4789e27-4774-4626-98e9-ae8dfbfb0fb5
found 1235681286 bytes used err is -22
(...)
Another issue caused by this race is a crash while writing bitmap entries
to the cache, because while the cache writeout task accesses the bitmaps,
the trim task can be concurrently modifying the bitmap or worse might
be freeing the bitmap. The later case results in the following crash:
Fix this by serializing both tasks in such a way that cache writeout
doesn't wait for the trim/discard of free space entries to finish and
doesn't miss any free space entry.
Filipe Manana [Thu, 27 Nov 2014 21:14:15 +0000 (21:14 +0000)]
Btrfs: fix race between fs trimming and block group remove/allocation
Our fs trim operation, which is completely transactionless (doesn't start
or joins an existing transaction) consists of visiting all block groups
and then for each one to iterate its free space entries and perform a
discard operation against the space range represented by the free space
entries. However before performing a discard, the corresponding free space
entry is removed from the free space rbtree, and when the discard completes
it is added back to the free space rbtree.
If a block group remove operation happens while the discard is ongoing (or
before it starts and after a free space entry is hidden), we end up not
waiting for the discard to complete, remove the extent map that maps
logical address to physical addresses and the corresponding chunk metadata
from the the chunk and device trees. After that and before the discard
completes, the current running transaction can finish and a new one start,
allowing for new block groups that map to the same physical addresses to
be allocated and written to.
So fix this by keeping the extent map in memory until the discard completes
so that the same physical addresses aren't reused before it completes.
If the physical locations that are under a discard operation end up being
used for a new metadata block group for example, and dirty metadata extents
are written before the discard finishes (the VM might call writepages() of
our btree inode's i_mapping for example, or an fsync log commit happens) we
end up overwriting metadata with zeroes, which leads to errors from fsck
like the following:
checking extents
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
read block failed check_tree_block
owner ref check failed [833912832 16384]
Errors found in extent allocation tree or chunk allocation
checking free space cache
checking fs roots
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
Check tree block failed, want=833912832, have=0
read block failed check_tree_block
root 5 root dir 256 error
root 5 inode 260 errors 2001, no inode item, link count wrong
unresolved ref dir 256 index 0 namelen 8 name foobar_3 filetype 1 errors 6, no dir index, no inode ref
root 5 inode 262 errors 2001, no inode item, link count wrong
unresolved ref dir 256 index 0 namelen 8 name foobar_5 filetype 1 errors 6, no dir index, no inode ref
root 5 inode 263 errors 2001, no inode item, link count wrong
(...)
Filipe Manana [Wed, 26 Nov 2014 15:28:52 +0000 (15:28 +0000)]
Btrfs: fix freeing used extents after removing empty block group
There's a race between adding a block group to the list of the unused
block groups and removing an unused block group (cleaner kthread) that
leads to freeing extents that are in use or a crash during transaction
commmit. Basically the cleaner kthread, when executing
btrfs_delete_unused_bgs(), might catch the newly added block group to
the list fs_info->unused_bgs and clear the range representing the whole
group from fs_info->freed_extents[] before the task that added the block
group to the list (running update_block_group()) marked the last freed
extent as dirty in fs_info->freed_extents (pinned_extents).
That is:
CPU 1 CPU 2
btrfs_delete_unused_bgs()
update_block_group()
add block group to
fs_info->unused_bgs
got block group from the list
clear_extent_bits for the whole
block group range in freed_extents[]
set_extent_dirty for the
range covering the freed
extent in freed_extents[]
(fs_info->pinned_extents)
block group deleted, and a new block
group with the same logical address is
created
reserve space from the new block group
for new data or metadata - the reserved
space overlaps the range specified by
CPU 1 for set_extent_dirty()
commit transaction
find all ranges marked as dirty in
fs_info->pinned_extents, clear them
and add them to the free space cache
Alternatively, if CPU 2 doesn't create a new block group with the same
logical address, we get a crash/BUG_ON at transaction commit when unpining
extent ranges because we can't find a block group for the range marked as
dirty by CPU 1. Sample trace:
So fix this by making update_block_group() first set the range as dirty
in pinned_extents before adding the block group to the unused_bgs list.
Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit ae0ab003f2840e7ed66ed762ec0d4e298ad1e725)
Filipe Manana [Wed, 26 Nov 2014 15:28:51 +0000 (15:28 +0000)]
Btrfs: fix crash caused by block group removal
If we remove a block group (because it became empty), we might have left
a caching_ctl structure in fs_info->caching_block_groups that points to
the block group and is accessed at transaction commit time. This results
in accessing an invalid or incorrect block group. This issue became visible
after Josef's patch "Btrfs: remove empty block groups automatically".
So if the block group is removed make sure we don't leave a dangling
caching_ctl in caching_block_groups.
Filipe Manana [Wed, 26 Nov 2014 15:28:50 +0000 (15:28 +0000)]
Btrfs: fix invalid block group rbtree access after bg is removed
If we grab a block group, for example in btrfs_trim_fs(), we will be holding
a reference on it but the block group can be removed after we got it (via
btrfs_remove_block_group), which means it will no longer be part of the
rbtree.
However, btrfs_remove_block_group() was only calling rb_erase() which leaves
the block group's rb_node left and right child pointers with the same content
they had before calling rb_erase. This was dangerous because a call to
next_block_group() would access the node's left and right child pointers (via
rb_next), which can be no longer valid.
Fix this by clearing a block group's node after removing it from the tree,
and have next_block_group() do a tree search to get the next block group
instead of using rb_next() if our block group was removed.
Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 292cbd51ecf85d73195a3e3193937fa770f6ea71)
Miao Xie [Tue, 25 Nov 2014 08:39:28 +0000 (16:39 +0800)]
Btrfs, raid56: fix use-after-free problem in the final device replace procedure on raid56
The commit c404e0dc (Btrfs: fix use-after-free in the finishing
procedure of the device replace) fixed a use-after-free problem
which happened when removing the source device at the end of device
replace, but at that time, btrfs didn't support device replace
on raid56, so we didn't fix the problem on the raid56 profile.
Currently, we implemented device replace for raid56, so we need
kick that problem out before we enable that function for raid56.
The fix method is very simple, we just increase the bio per-cpu
counter before we submit a raid56 io, and decrease the counter
when the raid56 io ends.
Miao Xie [Fri, 14 Nov 2014 09:45:42 +0000 (17:45 +0800)]
Btrfs, replace: write raid56 parity into the replace target device
This function reused the code of parity scrub, and we just write
the right parity or corrected parity into the target device before
the parity scrub end.
Miao Xie [Fri, 14 Nov 2014 08:06:25 +0000 (16:06 +0800)]
Btrfs, replace: write dirty pages into the replace target device
The implementation is simple:
- In order to avoid changing the code logic of btrfs_map_bio and
RAID56, we add the stripes of the replace target devices at the
end of the stripe array in btrfs bio, and we sort those target
device stripes in the array. And we keep the number of the target
device stripes in the btrfs bio.
- Except write operation on RAID56, all the other operation don't
take the target device stripes into account.
- When we do write operation, we read the data from the common devices
and calculate the parity. Then write the dirty data and new parity
out, at this time, we will find the relative replace target stripes
and wirte the relative data into it.
Note: The function that copying old data on the source device to
the target device was implemented in the past, it is similar to
the other RAID type.
Miao Xie [Thu, 6 Nov 2014 09:20:58 +0000 (17:20 +0800)]
Btrfs, raid56: support parity scrub on raid56
The implementation is:
- Read and check all the data with checksum in the same stripe.
All the data which has checksum is COW data, and we are sure
that it is not changed though we don't lock the stripe. because
the space of that data just can be reclaimed after the current
transction is committed, and then the fs can use it to store the
other data, but when doing scrub, we hold the current transaction,
that is that data can not be recovered, it is safe that read and check
it out of the stripe lock.
- Lock the stripe
- Read out all the data without checksum and parity
The data without checksum and the parity may be changed if we don't
lock the stripe, so we need read it in the stripe lock context.
- Check the parity
- Re-calculate the new parity and write back it if the old parity
is not right
- Unlock the stripe
If we can not read out the data or the data we read is corrupted,
we will try to repair it. If the repair fails. we will mark the
horizontal sub-stripe(pages on the same horizontal) as corrupted
sub-stripe, and we will skip the parity check and repair of that
horizontal sub-stripe.
And in order to skip the horizontal sub-stripe that has no data, we
introduce a bitmap. If there is some data on the horizontal sub-stripe,
we will the relative bit to 1, and when we check and repair the
parity, we will skip those horizontal sub-stripes that the relative
bits is 0.
Miao Xie [Thu, 6 Nov 2014 08:14:21 +0000 (16:14 +0800)]
Btrfs, raid56: use a variant to record the operation type
We will introduce new operation type later, if we still use integer
variant as bool variant to record the operation type, we would add new
variant and increase the size of raid bio structure. It is not good,
by this patch, we define different number for different operation,
and we can just use a variant to record the operation type.
Miao Xie [Thu, 23 Oct 2014 06:42:50 +0000 (14:42 +0800)]
Btrfs, scrub: repair the common data on RAID5/6 if it is corrupted
This patch implement the RAID5/6 common data repair function, the
implementation is similar to the scrub on the other RAID such as
RAID1, the differentia is that we don't read the data from the
mirror, we use the data repair function of RAID5/6.
Miao Xie [Wed, 15 Oct 2014 03:18:44 +0000 (11:18 +0800)]
Btrfs, raid56: don't change bbio and raid_map
Because we will reuse bbio and raid_map during the scrub later, it is
better that we don't change any variant of bbio and don't free it at
the end of IO request. So we introduced similar variants into the raid
bio, and don't access those bbio's variants any more.
Filipe Manana [Wed, 29 Oct 2014 11:57:59 +0000 (11:57 +0000)]
Btrfs: fix snapshot inconsistency after a file write followed by truncate
If right after starting the snapshot creation ioctl we perform a write against a
file followed by a truncate, with both operations increasing the file's size, we
can get a snapshot tree that reflects a state of the source subvolume's tree where
the file truncation happened but the write operation didn't. This leaves a gap
between 2 file extent items of the inode, which makes btrfs' fsck complain about it.
For example, if we perform the following file operations:
and the snapshot creation ioctl was just called before the second write, we often
can get the following inode items in the snapshot's btree:
item 120 key (257 INODE_ITEM 0) itemoff 7987 itemsize 160
inode generation 146 transid 7 size 90123 block group 0 mode 100600 links 1 uid 0 gid 0 rdev 0 flags 0x0
item 121 key (257 INODE_REF 256) itemoff 7967 itemsize 20
inode ref index 282 namelen 10 name: foobar
item 122 key (257 EXTENT_DATA 0) itemoff 7914 itemsize 53
extent data disk byte 1104855040 nr 32768
extent data offset 0 nr 32768 ram 32768
extent compression 0
item 123 key (257 EXTENT_DATA 53248) itemoff 7861 itemsize 53
extent data disk byte 0 nr 0
extent data offset 0 nr 40960 ram 40960
extent compression 0
There's a file range, corresponding to the interval [32K; ALIGN(16K + 32770, 4096)[
for which there's no file extent item covering it. This is because the file write
and file truncate operations happened both right after the snapshot creation ioctl
called btrfs_start_delalloc_inodes(), which means we didn't start and wait for the
ordered extent that matches the write and, in btrfs_setsize(), we were able to call
btrfs_cont_expand() before being able to commit the current transaction in the
snapshot creation ioctl. So this made it possibe to insert the hole file extent
item in the source subvolume (which represents the region added by the truncate)
right before the transaction commit from the snapshot creation ioctl.
Btrfs' fsck tool complains about such cases with a message like the following:
>From a user perspective, the expectation when a snapshot is created while those
file operations are being performed is that the snapshot will have a file that
either:
1) is empty
2) only the first write was captured
3) only the 2 writes were captured
4) both writes and the truncation were captured
But never capture a state where only the first write and the truncation were
captured (since the second write was performed before the truncation).
Filipe Manana [Tue, 21 Oct 2014 10:11:41 +0000 (11:11 +0100)]
Btrfs: ensure send always works on roots without orphans
Move the logic from the snapshot creation ioctl into send. This avoids
doing the transaction commit if send isn't used, and ensures that if
a crash/reboot happens after the transaction commit that created the
snapshot and before the transaction commit that switched the commit
root, send will not get a commit root that differs from the main root
(that has orphan items).
Filipe Manana [Mon, 3 Nov 2014 14:08:39 +0000 (14:08 +0000)]
Btrfs: fix freeing used extent after removing empty block group
Due to ignoring errors returned by clear_extent_bits (at the moment only
-ENOMEM is possible), we can end up freeing an extent that is actually in
use (i.e. return the extent to the free space cache).
The sequence of steps that lead to this:
1) Cleaner thread starts execution and calls btrfs_delete_unused_bgs(), with
the goal of freeing empty block groups;
2) btrfs_delete_unused_bgs() finds an empty block group, joins the current
transaction (or starts a new one if none is running) and attempts to
clear the EXTENT_DIRTY bit for the block group's range from freed_extents[0]
and freed_extents[1] (of which one corresponds to fs_info->pinned_extents);
3) Clearing the EXTENT_DIRTY bit (via clear_extent_bits()) fails with
-ENOMEM, but such error is ignored and btrfs_delete_unused_bgs() proceeds
to delete the block group and the respective chunk, while pinned_extents
remains with that bit set for the whole (or a part of the) range covered
by the block group;
4) Later while the transaction is still running, the chunk ends up being reused
for a new block group (maybe for different purpose, data or metadata), and
extents belonging to the new block group are allocated for file data or btree
nodes/leafs;
5) The current transaction is committed, meaning that we unpinned one or more
extents from the new block group (through btrfs_finish_extent_commit() and
unpin_extent_range()) which are now being used for new file data or new
metadata (through btrfs_finish_extent_commit() and unpin_extent_range()).
And unpinning means we returned the extents to the free space cache of the
new block group, which implies those extents can be used for future allocations
while they're still in use.
Alternatively, we can hit a BUG_ON() when doing a lookup for a block group's cache
object in unpin_extent_range() if a new block group didn't end up being allocated for
the same chunk (step 4 above).
Fix this by not freeing the block group and chunk if we fail to clear the dirty bit.
Qu Wenruo [Thu, 30 Oct 2014 08:52:31 +0000 (16:52 +0800)]
btrfs: Fix a lockdep warning when running xfstest.
The following lockdep warning is triggered during xfstests:
[ 1702.980872] =========================================================
[ 1702.981181] [ INFO: possible irq lock inversion dependency detected ]
[ 1702.981482] 3.18.0-rc1 #27 Not tainted
[ 1702.981781] ---------------------------------------------------------
[ 1702.982095] kswapd0/77 just changed the state of lock:
[ 1702.982415] (&delayed_node->mutex){+.+.-.}, at: [<ffffffffa03b0b51>] __btrfs_release_delayed_node+0x41/0x1f0 [btrfs]
[ 1702.982794] but this lock took another, RECLAIM_FS-unsafe lock in the past:
[ 1702.983160] (&fs_info->dev_replace.lock){+.+.+.}
and interrupts could create inverse lock ordering between them.
[ 1702.984675]
other info that might help us debug this:
[ 1702.985524] Chain exists of:
&delayed_node->mutex --> &found->groups_sem --> &fs_info->dev_replace.lock
[ 1702.986799] Possible interrupt unsafe locking scenario:
It is because the btrfs_kobj_{add/rm}_device() will call memory
allocation with GFP_KERNEL,
which may flush fs page cache to free space, waiting for it self to do
the commit, causing the deadlock.
To solve the problem, move btrfs_kobj_{add/rm}_device() out of the
dev_replace lock range, also involing split the
btrfs_rm_dev_replace_srcdev() function into remove and free parts.
Now only btrfs_rm_dev_replace_remove_srcdev() is called in dev_replace
lock range, and kobj_{add/rm} and btrfs_rm_dev_replace_free_srcdev() are
called out of the lock range.
Filipe Manana [Thu, 13 Nov 2014 17:01:45 +0000 (17:01 +0000)]
Btrfs: ensure ordered extent errors aren't missed on fsync
When doing a fsync with a fast path we have a time window where we can miss
the fact that writeback of some file data failed, and therefore we endup
returning success (0) from fsync when we should return an error.
The steps that lead to this are the following:
1) We start all ordered extents by calling filemap_fdatawrite_range();
2) We do some other work like locking the inode's i_mutex, start a transaction,
start a log transaction, etc;
3) We enter btrfs_log_inode(), acquire the inode's log_mutex and collect all the
ordered extents from inode's ordered tree into a list;
4) But by the time we do ordered extent collection, some ordered extents we started
at step 1) might have already completed with an error, and therefore we didn't
found them in the ordered tree and had no idea they finished with an error. This
makes our fsync return success (0) to userspace, but has no bad effects on the log
like for example insertion of file extent items into the log that point to unwritten
extents, because the invalid extent maps were removed before the ordered extent
completed (in inode.c:btrfs_finish_ordered_io).
So after collecting the ordered extents just check if the inode's i_mapping has any
error flags set (AS_EIO or AS_ENOSPC) and leave with an error if it does. Whenever
writeback fails for a page of an ordered extent, we call mapping_set_error (done in
extent_io.c:end_extent_writepage, called by extent_io.c:end_bio_extent_writepage)
that sets one of those error flags in the inode's i_mapping flags.
This change also has the side effect of fixing the issue where for fast fsyncs we
never checked/cleared the error flags from the inode's i_mapping flags, which means
that a full fsync performed after a fast fsync could get such errors that belonged
to the fast fsync - because the full fsync calls btrfs_wait_ordered_range() which
calls filemap_fdatawait_range(), and the later checks for and clears those flags,
while for fast fsyncs we never call filemap_fdatawait_range() or anything else
that checks for and clears the error flags from the inode's i_mapping.
Filipe Manana [Thu, 13 Nov 2014 17:00:35 +0000 (17:00 +0000)]
Btrfs: collect only the necessary ordered extents on ranged fsync
Instead of collecting all ordered extents from the inode's ordered tree
and then wait for all of them to complete, just collect the ones that
overlap the fsync range.
Filipe Manana [Thu, 13 Nov 2014 16:59:53 +0000 (16:59 +0000)]
Btrfs: don't ignore log btree writeback errors
If an error happens during writeback of log btree extents, make sure the
error is returned to the caller (fsync), so that it takes proper action
(commit current transaction) instead of writing a superblock that points
to log btrees with all or some nodes that weren't durably persisted.
Josef Bacik [Fri, 14 Nov 2014 21:16:30 +0000 (16:16 -0500)]
Btrfs: do not move em to modified list when unpinning
We use the modified list to keep track of which extents have been modified so we
know which ones are candidates for logging at fsync() time. Newly modified
extents are added to the list at modification time, around the same time the
ordered extent is created. We do this so that we don't have to wait for ordered
extents to complete before we know what we need to log. The problem is when
something like this happens
log extent 0-4k on inode 1
copy csum for 0-4k from ordered extent into log
sync log
commit transaction
log some other extent on inode 1
ordered extent for 0-4k completes and adds itself onto modified list again
log changed extents
see ordered extent for 0-4k has already been logged
at this point we assume the csum has been copied
sync log
crash
On replay we will see the extent 0-4k in the log, drop the original 0-4k extent
which is the same one that we are replaying which also drops the csum, and then
we won't find the csum in the log for that bytenr. This of course causes us to
have errors about not having csums for certain ranges of our inode. So remove
the modified list manipulation in unpin_extent_cache, any modified extents
should have been added well before now, and we don't want them re-logged. This
fixes my test that I could reliably reproduce this problem with. Thanks,
cc: stable@vger.kernel.org Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit a28046956c71985046474283fa3bcd256915fb72)
Josef Bacik [Fri, 21 Nov 2014 19:52:38 +0000 (14:52 -0500)]
Btrfs: make sure logged extents complete in the current transaction V3
Liu Bo pointed out that my previous fix would lose the generation update in the
scenario I described. It is actually much worse than that, we could lose the
entire extent if we lose power right after the transaction commits. Consider
the following
write extent 0-4k
log extent in log tree
commit transaction
< power fail happens here
ordered extent completes
We would lose the 0-4k extent because it hasn't updated the actual fs tree, and
the transaction commit will reset the log so it isn't replayed. If we lose
power before the transaction commit we are save, otherwise we are not.
Fix this by keeping track of all extents we logged in this transaction. Then
when we go to commit the transaction make sure we wait for all of those ordered
extents to complete before proceeding. This will make sure that if we lose
power after the transaction commit we still have our data. This also fixes the
problem of the improperly updated extent generation. Thanks,
cc: stable@vger.kernel.org Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 50d9aa99bd35c77200e0e3dd7a72274f8304701f)
Josef Bacik [Thu, 6 Nov 2014 15:19:54 +0000 (10:19 -0500)]
Btrfs: make sure we wait on logged extents when fsycning two subvols
If we have two fsync()'s race on different subvols one will do all of its work
to get into the log_tree, wait on it's outstanding IO, and then allow the
log_tree to finish it's commit. The problem is we were just free'ing that
subvols logged extents instead of waiting on them, so whoever lost the race
wouldn't really have their data on disk. Fix this by waiting properly instead
of freeing the logged extents. Thanks,
cc: stable@vger.kernel.org Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 9dba8cf128ef98257ca719722280c9634e7e9dc7)
David Sterba [Fri, 14 Nov 2014 14:05:06 +0000 (15:05 +0100)]
btrfs: fix wrong accounting of raid1 data profile in statfs
The sizes that are obtained from space infos are in raw units and have
to be adjusted according to the raid factor. This was missing for
f_bavail and df reported doubled size for raid1.
Reported-by: Martin Steigerwald <Martin@lichtvoll.de> Fixes: ba7b6e62f420 ("btrfs: adjust statfs calculations according to raid profiles") CC: stable@vger.kernel.org Signed-off-by: David Sterba <dsterba@suse.cz> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 0d95c1bec906dd1ad951c9c001e798ca52baeb0f)
This leads to an ABBA pattern deadlock. To fix it,
o we just change it to an AABB pattern which means to @unlock_extent_bits()
before we @lock_page(), and in this way the @extent_read_full_page_nolock()
is no longer in an locked context, so change it back to @extent_read_full_page()
to regain protection.
o Since we @unlock_extent_bits() earlier, then before @write_page_nocow(),
the extent may not really point at the physical block we want, so we
have to check it before write.
Signed-off-by: Gui Hecheng <guihc.fnst@cn.fujitsu.com> Tested-by: David Sterba <dsterba@suse.cz> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 321592427c0146126aadfab8a9b663de1875c9f4)
Filipe Manana [Sun, 9 Nov 2014 08:38:39 +0000 (08:38 +0000)]
Btrfs: make xattr replace operations atomic
Replacing a xattr consists of doing a lookup for its existing value, delete
the current value from the respective leaf, release the search path and then
finally insert the new value. This leaves a time window where readers (getxattr,
listxattrs) won't see any value for the xattr. Xattrs are used to store ACLs,
so this has security implications.
This change also fixes 2 other existing issues which were:
*) Deleting the old xattr value without verifying first if the new xattr will
fit in the existing leaf item (in case multiple xattrs are packed in the
same item due to name hash collision);
*) Returning -EEXIST when the flag XATTR_CREATE is given and the xattr doesn't
exist but we have have an existing item that packs muliple xattrs with
the same name hash as the input xattr. In this case we should return ENOSPC.
A test case for xfstests follows soon.
Thanks to Alexandre Oliva for reporting the non-atomicity of the xattr replace
implementation.
Reported-by: Alexandre Oliva <oliva@gnu.org> Signed-off-by: Filipe Manana <fdmanana@suse.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 5f5bc6b1e2d5a6f827bc860ef2dc5b6f365d1339)
Filipe Manana [Mon, 3 Nov 2014 14:12:57 +0000 (14:12 +0000)]
Btrfs: avoid premature -ENOMEM in clear_extent_bit()
We try to allocate an extent state structure before acquiring the extent
state tree's spinlock as we might need a new one later and therefore avoid
doing later an atomic allocation while holding the tree's spinlock. However
we returned -ENOMEM if that initial non-atomic allocation failed, which is
a bit excessive since we might end up not needing the pre-allocated extent
state at all - for the case where the tree doesn't have any extent states
that cover the input range and cover too any other range. Therefore don't
return -ENOMEM if that pre-allocation fails.
Josef Bacik [Mon, 3 Nov 2014 13:56:50 +0000 (08:56 -0500)]
Btrfs: don't take the chunk_mutex/dev_list mutex in statfs V2
Our gluster boxes get several thousand statfs() calls per second, which begins
to suck hardcore with all of the lock contention on the chunk mutex and dev list
mutex. We don't really need to hold these things, if we have transient
weirdness with statfs() because of the chunk allocator we don't care, so remove
this locking.
We still need the dev_list lock if you mount with -o alloc_start however, which
is a good argument for nuking that thing from orbit, but that's a patch for
another day. Thanks,
Josef Bacik [Fri, 31 Oct 2014 13:49:34 +0000 (09:49 -0400)]
Btrfs: move read only block groups onto their own list V2
Our gluster boxes were spending lots of time in statfs because our fs'es are
huge. The problem is statfs loops through all of the block groups looking for
read only block groups, and when you have several terabytes worth of data that
ends up being a lot of block groups. Move the read only block groups onto a
read only list and only proces that list in
btrfs_account_ro_block_groups_free_space to reduce the amount of churn. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 633c0aad4c0243a506a3e8590551085ad78af82d)
Stefan Behrens [Fri, 17 Oct 2014 12:10:10 +0000 (14:10 +0200)]
Btrfs: check-int: don't complain about balanced blocks
The xfstest btrfs/014 which tests the balance operation caused that the
check_int module complained that known blocks changed their physical
location. Since this is not an error in this case, only print such
message if the verbose mode was enabled.
Reported-by: Wang Shilong <wangshilong1991@gmail.com> Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de> Tested-by: Wang Shilong <wangshilong1991@gmail.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit cf90c59e680ee83c263a60ee2cd464a7d96cc6d5)
Stefan Behrens [Thu, 16 Oct 2014 15:48:48 +0000 (17:48 +0200)]
Btrfs: check_int: use the known block location
The xfstest btrfs/014 which tests the balance operation caused issues with
the check_int module. The attempt was made to use btrfs_map_block() to
find the physical location for a written block. However, this was not
at all needed since the location of the written block was known since
a hook to submit_bio() was the reason for entering the check_int module.
Additionally, after a block relocation it happened that btrfs_map_block()
failed causing misleading error messages afterwards.
This patch changes the check_int module to use the known information of
the physical location from the bio.
Reported-by: Wang Shilong <wangshilong1991@gmail.com> Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de> Tested-by: Wang Shilong <wangshilong1991@gmail.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit f382e4653f06f4e285b7800f80e6548e026f90a8)
Filipe Manana [Mon, 13 Oct 2014 11:28:39 +0000 (12:28 +0100)]
Btrfs: avoid returning -ENOMEM in convert_extent_bit() too early
We try to allocate an extent state before acquiring the tree's spinlock
just in case we end up needing to split an existing extent state into two.
If that allocation failed, we would return -ENOMEM.
However, our only single caller (transaction/log commit code), passes in
an extent state that was cached from a call to find_first_extent_bit() and
that has a very high chance to match exactly the input range (always true
for a transaction commit and very often, but not always, true for a log
commit) - in this case we end up not needing at all that initial extent
state used for an eventual split. Therefore just don't return -ENOMEM if
we can't allocate the temporary extent state, since we might not need it
at all, and if we end up needing one, we'll do it later anyway.
Filipe Manana [Mon, 13 Oct 2014 11:28:38 +0000 (12:28 +0100)]
Btrfs: make find_first_extent_bit be able to cache any state
Right now the only caller of find_first_extent_bit() that is interested
in caching extent states (transaction or log commit), never gets an extent
state cached. This is because find_first_extent_bit() only caches states
that have at least one of the flags EXTENT_IOBITS or EXTENT_BOUNDARY, and
the transaction/log commit caller always passes a tree that doesn't have
ever extent states with any of those flags (they can only have one of the
following flags: EXTENT_DIRTY, EXTENT_NEW or EXTENT_NEED_WAIT).
This change together with the following one in the patch series (titled
"Btrfs: avoid returning -ENOMEM in convert_extent_bit() too early") will
help reduce significantly the chances of calls to convert_extent_bit()
fail with -ENOMEM when called from the transaction/log commit code.
Filipe Manana [Mon, 13 Oct 2014 11:28:37 +0000 (12:28 +0100)]
Btrfs: deal with convert_extent_bit errors to avoid fs corruption
When committing a transaction or a log, we look for btree extents that
need to be durably persisted by searching for ranges in a io tree that
have some bits set (EXTENT_DIRTY or EXTENT_NEW). We then attempt to clear
those bits and set the EXTENT_NEED_WAIT bit, with calls to the function
convert_extent_bit, and then start writeback for the extents.
That function however can return an error (at the moment only -ENOMEM
is possible, specially when it does GFP_ATOMIC allocation requests
through alloc_extent_state_atomic) - that means the ranges didn't got
the EXTENT_NEED_WAIT bit set (or at least not for the whole range),
which in turn means a call to btrfs_wait_marked_extents() won't find
those ranges for which we started writeback, causing a transaction
commit or a log commit to persist a new superblock without waiting
for the writeback of extents in that range to finish first.
Therefore if a crash happens after persisting the new superblock and
before writeback finishes, we have a superblock pointing to roots that
weren't fully persisted or roots that point to nodes or leafs that weren't
fully persisted, causing all sorts of unexpected/bad behaviour as we endup
reading garbage from disk or the content of some node/leaf from a past
generation that got cowed or deleted and is no longer valid (for this later
case we end up getting error messages like "parent transid verify failed on
X wanted Y found Z" when reading btree nodes/leafs from disk).
Eryu Guan [Mon, 13 Oct 2014 04:42:12 +0000 (12:42 +0800)]
Btrfs: return failure if btrfs_dev_replace_finishing() failed
device replace could fail due to another running scrub process or any
other errors btrfs_scrub_dev() may hit, but this failure doesn't get
returned to userspace.
The following steps could reproduce this issue
mkfs -t btrfs -f /dev/sdb1 /dev/sdb2
mount /dev/sdb1 /mnt/btrfs
while true; do btrfs scrub start -B /mnt/btrfs >/dev/null 2>&1; done &
btrfs replace start -Bf /dev/sdb2 /dev/sdb3 /mnt/btrfs
# if this replace succeeded, do the following and repeat until
# you see this log in dmesg
# BTRFS: btrfs_scrub_dev(/dev/sdb2, 2, /dev/sdb3) failed -115
#btrfs replace start -Bf /dev/sdb3 /dev/sdb2 /mnt/btrfs
# once you see the error log in dmesg, check return value of
# replace
echo $?
Introduce a new dev replace result
BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS
to catch -EINPROGRESS explicitly and return other errors directly to
userspace.
Since we are allocating memory for hash table array, so
it will be good if we could allocate continuous pages here.
Fix this problem by firstly trying kzalloc(), if we fail,
use vzalloc() instead.
Signed-off-by: Wang Shilong <wangshilong1991@gmail.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 6b3a4d60dbf7a60b91d810498e8c212f26718efd)
Filipe Manana [Fri, 10 Oct 2014 09:45:12 +0000 (10:45 +0100)]
Btrfs: report error after failure inlining extent in compressed write path
If cow_file_range_inline() failed, when called from compress_file_range(),
we were tagging the locked page for writeback, end its writeback and unlock it,
but not marking it with an error nor setting AS_EIO in inode's mapping flags.
This made it impossible for a caller of filemap_fdatawrite_range (writepages)
or filemap_fdatawait_range() to know that an error happened. And the return
value of compress_file_range() is useless because it's returned to a workqueue
task and not to the task calling filemap_fdatawrite_range (writepages).
This change applies on top of the previous patchset starting at the patch
titled:
"[1/5] Btrfs: set page and mapping error on compressed write failure"
Which changed extent_clear_unlock_delalloc() to use SetPageError and
mapping_set_error().
Filipe Manana [Thu, 9 Oct 2014 20:18:55 +0000 (21:18 +0100)]
Btrfs: correctly flush compressed data before/after direct IO
For compressed writes, after doing the first filemap_fdatawrite_range() we
don't get the pages tagged for writeback immediately. Instead we create
a workqueue task, which is run by other kthread, and keep the pages locked.
That other kthread compresses data, creates the respective ordered extent/s,
tags the pages for writeback and unlocks them. Therefore we need a second
call to filemap_fdatawrite_range() if we have compressed writes, as this
second call will wait for the pages to become unlocked, then see they became
tagged for writeback and finally wait for the writeback to finish.
Filipe Manana [Thu, 9 Oct 2014 20:15:44 +0000 (21:15 +0100)]
Btrfs: make inode.c:compress_file_range() return void
Its return value is useless, its single caller ignores it and can't do
anything with it anyway, since it's a workqueue task and not the task
calling filemap_fdatawrite_range (writepages) nor filemap_fdatawait_range().
Failure is communicated to such functions via start and end of writeback
with the respective pages tagged with an error and AS_EIO flag set in the
inode's imapping.
after previous steps, inode will be detected as bad compression ratio,
and NOCOMPRESS flag will be set for that inode.
Reason is that compress have a max limit pages every time(128K), if a
132k write in, it will be splitted into two write(128k+4k), this bug
is a leftover for commit 68bb462d42a(Btrfs: don't compress for a small write)
Fix this problem by checking every time before compression, if it is a
small write(<=blocksize), we bail out and fall into nocompression directly.
Signed-off-by: Wang Shilong <wangshilong1991@gmail.com> Reviewed-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 4bcbb33255131adbe481c0467df26d654ce3bc78)
Filipe Manana [Tue, 7 Oct 2014 00:48:26 +0000 (01:48 +0100)]
Btrfs: don't ignore compressed bio write errors
Our compressed bio write end callback was essentially ignoring the error
parameter. When a write error happens, it must pass a value of 0 to the
inode's write_page_end_io_hook callback, SetPageError on the respective
pages and set AS_EIO in the inode's mapping flags, so that a call to
filemap_fdatawait_range() / filemap_fdatawait() can find out that errors
happened (we surely don't want silent failures on fsync for example).
Filipe Manana [Mon, 6 Oct 2014 21:14:26 +0000 (22:14 +0100)]
Btrfs: make inode.c:submit_compressed_extents() return void
Its return value is completely ignored by its single caller and it's
useless anyway, since errors are indicated through SetPageError and
the bit AS_EIO set in the flags of the inode's mapping. The caller
can't do anything with the value, as it's invoked from a workqueue
task and not by the task calling filemap_fdatawrite_range (which calls
the writepages address space callback, which in turn calls the inode's
fill_delalloc callback).
Filipe Manana [Mon, 6 Oct 2014 21:14:25 +0000 (22:14 +0100)]
Btrfs: process all async extents on compressed write failure
If we had an error when processing one of the async extents from our list,
we were not processing the remaining async extents, meaning we would leak
those async_extent structs, never release the pages with the compressed
data and never unlock and clear the dirty flag from the inode's pages (those
that correspond to the uncompressed content).
Filipe Manana [Mon, 6 Oct 2014 21:14:24 +0000 (22:14 +0100)]
Btrfs: don't leak pages and memory on compressed write error
In inode.c:submit_compressed_extents(), if we fail before calling
btrfs_submit_compressed_write(), or when that function fails, we
were freeing the async_extent structure without releasing its pages
and freeing the pages array.
Filipe Manana [Mon, 6 Oct 2014 21:14:23 +0000 (22:14 +0100)]
Btrfs: fix hang on compressed write error
In inode.c:submit_compressed_extents(), before calling btrfs_submit_compressed_write()
we start writeback for all pages, clear their dirty flag, unlock them, etc, but if
btrfs_submit_compressed_write() fails (at the moment it can only fail with -ENOMEM),
we never end the writeback on the pages, so any filemap_fdatawait_range() call will
hang forever. We were also not calling the writepage end io hook, which means the
corresponding ordered extent will never complete and all its waiters will block
forever, such as a full fsync (via btrfs_wait_ordered_range()).
Filipe Manana [Mon, 6 Oct 2014 21:14:22 +0000 (22:14 +0100)]
Btrfs: set page and mapping error on compressed write failure
If we fail in submit_compressed_extents() before calling btrfs_submit_compressed_write(),
we start and end the writeback for the pages (clear their dirty flag, unlock them, etc)
but we don't tag the pages, nor the inode's mapping, with an error. This makes it
impossible for a caller of filemap_fdatawait_range() (fsync, or transaction commit
for e.g.) know that there was an error.
Note that the return value of submit_compressed_extents() is useless, as that function
is executed by a workqueue task and not directly by the fill_delalloc callback. This
means the writepage/s callbacks of the inode's address space operations don't get that
return value.
David Sterba [Wed, 12 Nov 2014 13:24:35 +0000 (14:24 +0100)]
btrfs: introduce pending action: commit
In some contexts, like in sysfs handlers, we don't want to trigger a
transaction commit. It's a heavy operation, we don't know what external
locks may be taken. Instead, make it possible to finish the operation
through sync syscall or SYNC_FS ioctl.
David Sterba [Wed, 5 Feb 2014 14:26:17 +0000 (15:26 +0100)]
btrfs: switch inode_cache option handling to pending changes
The pending mount option(s) now share namespace and bits with the normal
options, and the existing one for (inode_cache) is unset unconditionally
at each transaction commit.
Introduce a separate namespace for pending changes and enhance the
descriptions of the intended change to use separate bits for each
action.
David Sterba [Fri, 28 Mar 2014 16:38:48 +0000 (17:38 +0100)]
btrfs: do commit in sync_fs if there are pending changes
If a pending change is requested, it's not processed unless there is a
transaction commit about to happen, not even after sync or SYNC_FS
ioctl. For example a remount that toggles the inode_cache option will
not take effect after sync on a quiescent filesystem.
David Sterba [Wed, 5 Feb 2014 14:26:17 +0000 (15:26 +0100)]
btrfs: add support for processing pending changes
There are some actions that modify global filesystem state but cannot be
performed at the time of request, but later at the transaction commit
time when the filesystem is in a known state.
For example enabling new incompat features on-the-fly or issuing
transaction commit from unsafe contexts (sysfs handlers).
In snd_usbmidi_error_timer(), the driver tries to resubmit MIDI input
URBs to reactivate the MIDI stream, but this causes the error when
some of URBs are still pending like:
... by not hitting rename_retry for reasons other than rename having
happened. In other words, do _not_ restart when finding that
between unlocking the child and locking the parent the former got
into __dentry_kill(). Skip the killed siblings instead...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Proper operation with the rewritten PCI mini driver requires that a flag be set
when interrupts are enabled. This flag was missed. This patch is one of three needed to
fix the kernel regression reported at https://bugzilla.kernel.org/show_bug.cgi?id=88951.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Tested-by: Catalin Iacob <iacobcatalin@gmail.com> Cc: Catalin Iacob <iacobcatalin@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the major update of the rtlwifi-family of drivers, one of the callback entries
was missed, which leads to memory corruption. Unfortunately, this corruption
never caused a kernel oops, but showed up in other parts of the system.
This patch is one of three needed to fix the kernel regression reported at
https://bugzilla.kernel.org/show_bug.cgi?id=88951.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Tested-by: Catalin Iacob <iacobcatalin@gmail.com> Cc: Catalin Iacob <iacobcatalin@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In the major update of the rtlwifi-family of drivers, there was an editing
mistake. Unfortunately, this particular error leads to memory corruption that
silently leads to failure of the system. This patch is one of three needed to
fix the kernel regression reported at https://bugzilla.kernel.org/show_bug.cgi?id=88951.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net> Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Tested-by: Catalin Iacob <iacobcatalin@gmail.com> Cc: Catalin Iacob <iacobcatalin@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For netlink, we shouldn't be using arch_fast_hash() as a hashing
discipline, but rather jhash() instead.
Since netlink sockets can be opened by any user, a local attacker
would be able to easily create collisions with the DPDK-derived
arch_fast_hash(), which trades off performance for security by
using crc32 CPU instructions on x86_64.
While it might have a legimite use case in other places, it should
be avoided in netlink context, though. As rhashtable's API is very
flexible, we could later on still decide on other hashing disciplines,
if legitimate.
Reference: http://thread.gmane.org/gmane.linux.kernel/1844123 Fixes: e341694e3eb5 ("netlink: Convert netlink_lookup() to use RCU protected hash table") Cc: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Thomas Graf <tgraf@suug.ch> Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The code should be using the _bh variant of rcu_dereference.
Signed-off-by: Valdis Kletnieks <valdis.kletnieks@vt.edu> Acked-by: Eric Dumazet <edumazet@google.com> Acked-by: John Fastabend <john.r.fastabend@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 97a6d1bb2b658ac85ed88205ccd1ab809899884d (xen-netfront: Fix
handling packets on compound pages with skb_linearize) attempted to
fix a problem where an skb that would have required too many slots
would be dropped causing TCP connections to stall.
However, it filled in the first slot using the original buffer and not
the new one and would use the wrong offset and grant access to the
wrong page.
Netback would notice the malformed request and stop all traffic on the
VIF, reporting:
Reported-by: Anthony Wright <anthony@overnetdata.com> Tested-by: Anthony Wright <anthony@overnetdata.com> Signed-off-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When I cooked commit c3658e8d0f1 ("tcp: fix possible NULL dereference in
tcp_vX_send_reset()") I missed other spots we could deref a NULL
skb_dst(skb)
Again, if a socket is provided, we do not need skb_dst() to get a
pointer to network namespace : sock_net(sk) is good enough.
Reported-by: Dann Frazier <dann.frazier@canonical.com> Bisected-by: Dann Frazier <dann.frazier@canonical.com> Tested-by: Dann Frazier <dann.frazier@canonical.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Fixes: ca777eff51f7 ("tcp: remove dst refcount false sharing for prequeue mode") Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To accomodate for enough headroom for tunnels, use MAX_HEADER instead
of LL_MAX_HEADER. Robert reported that he has hit after roughly 40hrs
of trinity an skb_under_panic() via SCTP output path (see reference).
I couldn't reproduce it from here, but not using MAX_HEADER as elsewhere
in other protocols might be one possible cause for this.
In any case, it looks like accounting on chunks themself seems to look
good as the skb already passed the SCTP output path and did not hit
any skb_over_panic(). Given tunneling was enabled in his .config, the
headroom would have been expanded by MAX_HEADER in this case.
Reported-by: Robert Święcki <robert@swiecki.net>
Reference: https://lkml.org/lkml/2014/12/1/507 Fixes: 594ccc14dfe4d ("[SCTP] Replace incorrect use of dev_alloc_skb with alloc_skb in sctp_packet_transmit().") Signed-off-by: Daniel Borkmann <dborkman@redhat.com> Acked-by: Vlad Yasevich <vyasevich@gmail.com> Acked-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mvneta_tx() dereferences skb to get skb->len too late,
as hardware might have completed the transmit and TX completion
could have freed the skb from another cpu.
Fixes: 71f6d1b31fb1 ("net: mvneta: replace Tx timer with a real interrupt") Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The mvneta driver sets the amount of Tx coalesce packets to 16 by
default. Normally that does not cause any trouble since the driver
uses a much larger Tx ring size (532 packets). But some sockets
might run with very small buffers, much smaller than the equivalent
of 16 packets. This is what ping is doing for example, by setting
SNDBUF to 324 bytes rounded up to 2kB by the kernel.
The problem is that there is no documented method to force a specific
packet to emit an interrupt (eg: the last of the ring) nor is it
possible to make the NIC emit an interrupt after a given delay.
In this case, it causes trouble, because when ping sends packets over
its raw socket, the few first packets leave the system, and the first
15 packets will be emitted without an IRQ being generated, so without
the skbs being freed. And since the socket's buffer is small, there's
no way to reach that amount of packets, and the ping ends up with
"send: no buffer available" after sending 6 packets. Running with 3
instances of ping in parallel is enough to hide the problem, because
with 6 packets per instance, that's 18 packets total, which is enough
to grant a Tx interrupt before all are sent.
The original driver in the LSP kernel worked around this design flaw
by using a software timer to clean up the Tx descriptors. This timer
was slow and caused terrible network performance on some Tx-bound
workloads (such as routing) but was enough to make tools like ping
work correctly.
Instead here, we simply set the packet counts before interrupt to 1.
This ensures that each packet sent will produce an interrupt. NAPI
takes care of coalescing interrupts since the interrupt is disabled
once generated.
No measurable performance impact nor CPU usage were observed on small
nor large packets, including when saturating the link on Tx, and this
fixes tools like ping which rely on too small a send buffer. If one
wants to increase this value for certain workloads where it is safe
to do so, "ethtool -C $dev tx-frames" will override this default
setting.
This fix needs to be applied to stable kernels starting with 3.10.
Tested-By: Maggie Mae Roxas <maggie.mae.roxas@gmail.com> Signed-off-by: Willy Tarreau <w@1wt.eu> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove optimize_div() from BPF_MOD | BPF_K case
since we don't know the dividend and fix the
emit_mod() by reading the mod operation result from HI register
Signed-off-by: Denis Kirjanov <kda@linux-powerpc.org> Reviewed-by: Markos Chandras <markos.chandras@imgtec.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Set the inner mac header to point to the GRE payload when
doing GRO. This is needed if we proceed to send the packet
through GRE GSO which now uses the inner mac header instead
of inner network header to determine the length of encapsulation
headers.
Fixes: 14051f0452a2 ("gre: Use inner mac length when computing tunnel length") Reported-by: Wolfgang Walter <linux@stwm.de> Signed-off-by: Tom Herbert <therbert@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, when trying to reuse a socket, vxlan_sock_add will grab
vn->sock_lock, locate a reusable socket, inc refcount and release
vn->sock_lock.
But vxlan_sock_release() will first decrement refcount, and then grab
that lock. refcnt operations are atomic but as currently we have
deferred works which hold vs->refcnt each, this might happen, leading to
a use after free (specially after vxlan_igmp_leave):
CPU 1 CPU 2
deferred work vxlan_sock_add
... ...
spin_lock(&vn->sock_lock)
vs = vxlan_find_sock();
vxlan_sock_release
dec vs->refcnt, reaches 0
spin_lock(&vn->sock_lock)
vxlan_sock_hold(vs), refcnt=1
spin_unlock(&vn->sock_lock)
hlist_del_rcu(&vs->hlist);
vxlan_notify_del_rx_port(vs)
spin_unlock(&vn->sock_lock)
So when we look for a reusable socket, we check if it wasn't freed
already before reusing it.
Signed-off-by: Marcelo Ricardo Leitner <mleitner@redhat.com> Fixes: 7c47cedf43a8b3 ("vxlan: move IGMP join/leave to work queue") Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Linus Torvalds [Sun, 7 Dec 2014 20:00:14 +0000 (12:00 -0800)]
Merge branch 'for-3.18-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata
Pull libata fixes from Tejun Heo:
"Three libata fixes for v3.18. Nothing too interesting. PCI ID ID and
quirk additions to ahci and an error handling path fix in sata_fsl"
* 'for-3.18-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:
ahci: disable MSI on SAMSUNG 0xa800 SSD
sata_fsl: fix error handling of irq_of_parse_and_map
AHCI: Add DeviceIDs for Sunrise Point-LP SATA controller
Linus Torvalds [Sat, 6 Dec 2014 19:26:01 +0000 (11:26 -0800)]
Merge branch 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux
Pull i2c fixes from Wolfram Sang:
"Here are two more driver bugfixes for I2C which would be good to have"
* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
i2c: cadence: Set the hardware time-out register to maximum value
i2c: davinci: generate STP always when NACK is received
Linus Torvalds [Fri, 5 Dec 2014 18:47:19 +0000 (10:47 -0800)]
Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Thomas Gleixner:
"Two final fixlets for 3.18:
- Prevent microcode reload wreckage on 32bit
- Unbreak cross compilation"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, microcode: Limit the microcode reloading to 64-bit for now
x86: Use $(OBJDUMP) instead of plain objdump
Linus Torvalds [Fri, 5 Dec 2014 06:39:37 +0000 (22:39 -0800)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull drm intel fixes from Dave Airlie:
"Two intel stable fixes, that should be it from me for this round"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/i915: Unlock panel even when LVDS is disabled
drm/i915: More cautious with pch fifo underruns
projects / linux / log