Btrfs: heuristic: add bucket and sample counters and other defines
Add basic defines and structures for data sampling.
Added macros:
- For future sampling algo
- For bucket size
Heuristic workspace:
- Add bucket for storing byte type counters
- Add sample array for storing partial copy of input data range
- Add counter for store current sample size to workspace
Signed-off-by: Timofey Titovets <nefelim4ag@gmail.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ minor coding style fixes, comments updated ] Signed-off-by: David Sterba <dsterba@suse.com>
Btrfs: compression: separate heuristic/compression workspaces
Compression heuristic itself is not a compression type, as current
infrastructure provides workspaces for several compression types, it's
difficult to just add heuristic workspace.
Just refactor the code to support compression/heuristic workspaces with
maximum code sharing and minimum changes in it.
Signed-off-by: Timofey Titovets <nefelim4ag@gmail.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ coding style fixes ] Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:16:02 +0000 (14:16 -0400)]
btrfs: move btrfs_truncate_block out of trans handle
Since we do a delalloc reserve in btrfs_truncate_block we can deadlock
with freeze. If somebody else is trying to allocate metadata for this
inode and it gets stuck in start_delalloc_inodes because of freeze we
will deadlock. Be safe and move this outside of a trans handle. This
also has a side-effect of making sure that we're not leaving stale data
behind in the other_encoding or encryption case. Not an issue now since
nobody uses it, but it would be a problem in the future.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:16:01 +0000 (14:16 -0400)]
btrfs: don't call btrfs_start_delalloc_roots in flushoncommit
We're holding the sb_start_intwrite lock at this point, and doing async
filemap_flush of the inodes will result in a deadlock if we freeze the
fs during this operation. This is because we could do a
btrfs_join_transaction() in the thread we are waiting on which would
block at sb_start_intwrite, and thus deadlock. Using
writeback_inodes_sb() side steps the problem by not introducing all of
these extra locking dependencies.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:16:00 +0000 (14:16 -0400)]
btrfs: track refs in a rb_tree instead of a list
If we get a significant amount of delayed refs for a single block (think
modifying multiple snapshots) we can end up spending an ungodly amount
of time looping through all of the entries trying to see if they can be
merged. This is because we only add them to a list, so we have O(2n)
for every ref head. This doesn't make any sense as we likely have refs
for different roots, and so they cannot be merged. Tracking in a tree
will allow us to break as soon as we hit an entry that doesn't match,
making our worst case O(n).
With this we can also merge entries more easily. Before we had to hope
that matching refs were on the ends of our list, but with the tree we
can search down to exact matches and merge them at insert time.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:15:59 +0000 (14:15 -0400)]
btrfs: add a comp_refs() helper
Instead of open-coding the delayed ref comparisons, add a helper to do
the comparisons generically and use that everywhere. We compare
sequence numbers last for following patches.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:15:58 +0000 (14:15 -0400)]
btrfs: switch args for comp_*_refs
Make it more consistent, we want the inserted ref to be compared against
what's already in there. This will make the order go from lowest seq ->
highest seq, which will make us more likely to make forward progress if
there's a seqlock currently held.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:15:57 +0000 (14:15 -0400)]
btrfs: make the delalloc block rsv per inode
The way we handle delalloc metadata reservations has gotten
progressively more complicated over the years. There is so much cruft
and weirdness around keeping the reserved count and outstanding counters
consistent and handling the error cases that it's impossible to
understand.
Fix this by making the delalloc block rsv per-inode. This way we can
calculate the actual size of the outstanding metadata reservations every
time we make a change, and then reserve the delta based on that amount.
This greatly simplifies the code everywhere, and makes the error
handling in btrfs_delalloc_reserve_metadata far less terrifying.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Thu, 19 Oct 2017 18:15:55 +0000 (14:15 -0400)]
Btrfs: rework outstanding_extents
Right now we do a lot of weird hoops around outstanding_extents in order
to keep the extent count consistent. This is because we logically
transfer the outstanding_extent count from the initial reservation
through the set_delalloc_bits. This makes it pretty difficult to get a
handle on how and when we need to mess with outstanding_extents.
Fix this by revamping the rules of how we deal with outstanding_extents.
Now instead everybody that is holding on to a delalloc extent is
required to increase the outstanding extents count for itself. This
means we'll have something like this
In order to make the ordered extent transition we of course must now
make ordered extents carry their own outstanding_extent reservation, so
for cow_file_range we end up with
This makes all manipulations of outstanding_extents much more explicit.
Every successful call to btrfs_delalloc_reserve_metadata _must_ now be
combined with btrfs_release_delalloc_extents, even in the error case, as
that is the only function that actually modifies the
outstanding_extents counter.
The drawback to this is now we are much more likely to have transient
cases where outstanding_extents is much larger than it actually should
be. This could happen before as we manipulated the delalloc bits, but
now it happens basically at every write. This may put more pressure on
the ENOSPC flushing code, but I think making this code simpler is worth
the cost. I have another change coming to mitigate this side-effect
somewhat.
I also added trace points for the counter manipulation. These were used
by a bpf script I wrote to help track down leak issues.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
btrfs: increase output size for LOGICAL_INO_V2 ioctl
Build-server workloads have hundreds of references per file after dedup.
Multiply by a few snapshots and we quickly exhaust the limit of 2730
references per extent that can fit into a 64K buffer.
Raise the limit to 16M to be consistent with other btrfs ioctls
(e.g. TREE_SEARCH_V2, FILE_EXTENT_SAME).
To minimize surprising userspace behavior, apply this change only to
the LOGICAL_INO_V2 ioctl.
Signed-off-by: Zygo Blaxell <ce3g8jdj@umail.furryterror.org> Reviewed-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com> Tested-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com> Signed-off-by: David Sterba <dsterba@suse.com>
btrfs: add a flags argument to LOGICAL_INO and call it LOGICAL_INO_V2
Now that check_extent_in_eb()'s extent offset filter can be turned off,
we need a way to do it from userspace.
Add a 'flags' field to the btrfs_logical_ino_args structure to disable
extent offset filtering, taking the place of one of the existing
reserved[] fields.
Previous versions of LOGICAL_INO neglected to check whether any of the
reserved fields have non-zero values. Assigning meaning to those fields
now may change the behavior of existing programs that left these fields
uninitialized. The lack of a zero check also means that new programs
have no way to know whether the kernel is honoring the flags field.
To avoid these problems, define a new ioctl LOGICAL_INO_V2. We can
use the same argument layout as LOGICAL_INO, but shorten the reserved[]
array by one element and turn it into the 'flags' field. The V2 ioctl
explicitly checks that reserved fields and unsupported flag bits are zero
so that userspace can negotiate future feature bits as they are defined.
Since the memory layouts of the two ioctls' arguments are compatible,
there is no need for a separate function for logical_to_ino_v2 (contrast
with tree_search_v2 vs tree_search where the layout and code are quite
different). A version parameter and an 'if' statement will suffice.
Now that we have a flags field in logical_ino_args, add a flag
BTRFS_LOGICAL_INO_ARGS_IGNORE_OFFSET to get the behavior we want,
and pass it down the stack to iterate_inodes_from_logical.
Motivation and background, copied from the patchset cover letter:
root@tester:~# btrfs-debug-tree /dev/vdc -t 5
[...]
item 6 key (261 EXTENT_DATA 0) itemoff 15825 itemsize 53
extent data disk byte 1103101952 nr 73728
extent data offset 0 nr 8192 ram 73728
extent compression(none)
item 7 key (261 EXTENT_DATA 8192) itemoff 15772 itemsize 53
extent data disk byte 1103175680 nr 4096
extent data offset 0 nr 4096 ram 4096
extent compression(none)
item 8 key (261 EXTENT_DATA 12288) itemoff 15719 itemsize 53
extent data disk byte 1103101952 nr 73728
extent data offset 12288 nr 61440 ram 73728
extent compression(none)
[...]
There are two references to the same extent with different, non-overlapping
byte offsets:
[------------------72K extent at 1103101952----------------------]
[--8K----------------|--4K unreachable----|--60K-----------------]
^ ^
| |
[--8K ref offset 0--][--4K ref offset 0--][--60K ref offset 12K--]
|
v
[-----4K extent-----] at 1103175680
We want to find all of the references to extent bytenr 1103101952.
Without the patch (and without running btrfs-debug-tree), we have to
do it with 18 LOGICAL_INO calls:
root@tester:~# btrfs ins log 1103101952 -P /test/
Using LOGICAL_INO
inode 261 offset 0 root 5
root@tester:~# for x in $(seq 0 17); do btrfs ins log $((1103101952 + x * 4096)) -P /test/; done 2>&1 | grep inode
inode 261 offset 0 root 5
inode 261 offset 4096 root 5 <- same extent ref as offset 0
(offset 8192 returns empty set, not reachable)
inode 261 offset 12288 root 5
inode 261 offset 16384 root 5 \
inode 261 offset 20480 root 5 |
inode 261 offset 24576 root 5 |
inode 261 offset 28672 root 5 |
inode 261 offset 32768 root 5 |
inode 261 offset 36864 root 5 \
inode 261 offset 40960 root 5 > all the same extent ref as offset 12288.
inode 261 offset 45056 root 5 / More processing required in userspace
inode 261 offset 49152 root 5 | to figure out these are all duplicates.
inode 261 offset 53248 root 5 |
inode 261 offset 57344 root 5 |
inode 261 offset 61440 root 5 |
inode 261 offset 65536 root 5 |
inode 261 offset 69632 root 5 /
In the worst case the extents are 128MB long, and we have to do 32768
iterations of the loop to find one 4K extent ref.
With the patch, we just use one call to map all refs to the extent at once:
root@tester:~# btrfs ins log 1103101952 -P /test/
Using LOGICAL_INO_V2
inode 261 offset 0 root 5
inode 261 offset 12288 root 5
The TREE_SEARCH ioctl allows userspace to retrieve the offset and
extent bytenr fields easily once the root, inode and offset are known.
This is sufficient information to build a complete map of the extent
and all of its references. Userspace can use this information to make
better choices to dedup or defrag.
Signed-off-by: Zygo Blaxell <ce3g8jdj@umail.furryterror.org> Reviewed-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com> Tested-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com>
[ copy background and motivation from cover letter ] Signed-off-by: David Sterba <dsterba@suse.com>
btrfs: add a flag to iterate_inodes_from_logical to find all extent refs for uncompressed extents
The LOGICAL_INO ioctl provides a backward mapping from extent bytenr and
offset (encoded as a single logical address) to a list of extent refs.
LOGICAL_INO complements TREE_SEARCH, which provides the forward mapping
(extent ref -> extent bytenr and offset, or logical address). These are
useful capabilities for programs that manipulate extents and extent
references from userspace (e.g. dedup and defrag utilities).
When the extents are uncompressed (and not encrypted and not other),
check_extent_in_eb performs filtering of the extent refs to remove any
extent refs which do not contain the same extent offset as the 'logical'
parameter's extent offset. This prevents LOGICAL_INO from returning
references to more than a single block.
To find the set of extent references to an uncompressed extent from [a, b),
userspace has to run a loop like this pseudocode:
for (i = a; i < b; ++i)
extent_ref_set += LOGICAL_INO(i);
At each iteration of the loop (up to 32768 iterations for a 128M extent),
data we are interested in is collected in the kernel, then deleted by
the filter in check_extent_in_eb.
When the extents are compressed (or encrypted or other), the 'logical'
parameter must be an extent bytenr (the 'a' parameter in the loop).
No filtering by extent offset is done (or possible?) so the result is
the complete set of extent refs for the entire extent. This removes
the need for the loop, since we get all the extent refs in one call.
Add an 'ignore_offset' argument to iterate_inodes_from_logical,
[...several levels of function call graph...], and check_extent_in_eb, so
that we can disable the extent offset filtering for uncompressed extents.
This flag can be set by an improved version of the LOGICAL_INO ioctl to
get either behavior as desired.
There is no functional change in this patch. The new flag is always
false.
Signed-off-by: Zygo Blaxell <ce3g8jdj@umail.furryterror.org> Reviewed-by: David Sterba <dsterba@suse.com>
[ minor coding style fixes ] Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 16 Oct 2017 13:29:23 +0000 (16:29 +0300)]
btrfs: send: remove unused code
This code was first introduced in 31db9f7c23fb ("Btrfs: introduce
BTRFS_IOC_SEND for btrfs send/receive") and it was not functional, then
it got slightly refactored in e938c8ad543c ("Btrfs: code cleanups for
send/receive"), alas it was still dead. So let's remove it for good!
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Adam Borowski [Fri, 15 Sep 2017 15:36:58 +0000 (17:36 +0200)]
btrfs: allow setting zlib compression level via :9
This is bikeshedding, but it seems people are drastically more likely to
understand "zlib:9" as compression level rather than an algorithm
version compared to "zlib9".
Based on feedback on the mailinglist, the ":9" will be the only accepted
syntax. The level must be a single digit. Unrecognized format will
result to the default, for forward compatibility in a similar way the
compression algorithm specifier was relaxed in commit a7164fa4e055daf6368c ("btrfs: prepare for extensions in compression
options").
Signed-off-by: Adam Borowski <kilobyte@angband.pl> Reviewed-by: David Sterba <dsterba@suse.com>
[ tighten the accepted format ] Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Fri, 15 Sep 2017 15:36:57 +0000 (17:36 +0200)]
btrfs: allow to set compression level for zlib
Preliminary support for setting compression level for zlib, the
following works:
$ mount -o compess=zlib # default
$ mount -o compess=zlib0 # same
$ mount -o compess=zlib9 # level 9, slower sync, less data
$ mount -o compess=zlib1 # level 1, faster sync, more data
$ mount -o remount,compress=zlib3 # level set by remount
The compress-force works the same as compress'. The level is visible in
the same format in /proc/mounts. Level set via file property does not
work yet.
Required patch: "btrfs: prepare for extensions in compression options"
Nikolay Borisov [Mon, 16 Oct 2017 13:48:40 +0000 (16:48 +0300)]
btrfs: Replace opencoded sizes with their symbolic constants
Currently btrfs' code uses a mix of opencoded sizes and defines from sizes.h.
Let's unifiy the code base to always use the symbolic constants. No functional
changes
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Gu JinXiang [Wed, 11 Oct 2017 08:44:27 +0000 (16:44 +0800)]
btrfs: Use bd_dev to generate index when dev_state_hashtable add items.
Fix missing change from commit f8f84b2dfda5
("btrfs: index check-integrity state hash by a dev_t").
Function btrfsic_dev_state_hashtable_lookup uses dev_t to generate hashval
when look in up a btrfsic_dev_state in hash table. So when we add a
btrfsic_dev_state into the hash table, it should also use dev_t.
Reproducer of this bug:
Use MOUNT_OPTIONS="-o check_int" when running xfstest, device can not be
mounted successfully. So xfstest can not run.
Signed-off-by: Gu JinXiang <gujx@cn.fujitsu.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Anand Jain [Sat, 14 Oct 2017 00:34:02 +0000 (08:34 +0800)]
btrfs: fix false EIO for missing device
When one of the device is missing, bbio_error() takes care of setting
the error status. And if its only IO that is pending in that stripe, it
fails to check the status of the other IO at %bbio_error before setting
the error %bi_status for the %orig_bio. Fix this by checking if
%bbio->error has exceeded the %bbio->max_errors.
Reproducer as below fdatasync error is seen intermittently.
dd: fdatasync failed for ‘/btrfs/LSe’: Input/output error
The reason for the intermittences of the problem is because
the following conditions have to be met, which depends on timing:
In btrfs_map_bio()
- the RAID1 the missing device has to be at %dev_nr = 1
In bbio_error()
. before bbio_error() is called the bio of the not-missing
device at %dev_nr = 0 must be completed so that the below
condition is true
if (atomic_dec_and_test(&bbio->stripes_pending)) {
Signed-off-by: Anand Jain <anand.jain@oracle.com> Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: David Sterba <dsterba@suse.com>
Code cleanup for better understanding:
Variable needs_unlock to be called extent_locked to show state as
opposed to action. Changed the type to int, to reduce code in the
critical path.
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:57 +0000 (15:43 -0400)]
btrfs: remove delayed_ref_node from ref_head
This is just excessive information in the ref_head, and makes the code
complicated. It is a relic from when we had the heads and the refs in
the same tree, which is no longer the case. With this removal I've
cleaned up a bunch of the cruft around this old assumption as well.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:56 +0000 (15:43 -0400)]
btrfs: move all ref head cleanup to the helper function
We do a couple different cleanup operations on the ref head. We adjust
counters, we'll free any reserved space if we didn't end up using the
ref, and we clear the pending csum bytes. Move all these disparate
things into cleanup_ref_head and clean up the logic in
__btrfs_run_delayed_refs so that it handles the !ref case a lot cleaner,
as well as making run_one_delayed_ref() only deal with real refs and not
the ref head.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:55 +0000 (15:43 -0400)]
btrfs: move ref_mod modification into the if (ref) logic
We only use this logic if our ref isn't a ref_head, so move it up into
the if (ref) case since we know that this is a normal ref and not a
delayed ref head.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:52 +0000 (15:43 -0400)]
btrfs: add a helper to return a head ref
Simplify the error handling in __btrfs_run_delayed_refs by breaking out
the code used to return a head back to the delayed_refs tree for
processing into a helper function.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:51 +0000 (15:43 -0400)]
Btrfs: only check delayed ref usage in should_end_transaction
We were only doing btrfs_check_space_for_delayed_refs() if the metadata
space was full, ie we couldn't allocate chunks. This assumes we'll be
able to allocate chunks during transaction commit, but since nothing
does a LIMIT flush during the transaction commit this won't actually
happen unless we happen to run shy of actual space. We already take
into account a full fs in btrfs_check_space_for_delayed_refs() so just
kill this extra check to make sure we're ending the transaction when we
need to.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:50 +0000 (15:43 -0400)]
Btrfs: add a extent ref verify tool
We were having corruption issues that were tied back to problems with
the extent tree. In order to track them down I built this tool to try
and find the culprit, which was pretty successful. If you compile with
this tool on it will live verify every ref update that the fs makes and
make sure it is consistent and valid. I've run this through with
xfstests and haven't gotten any false positives. Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ update error messages, add fixup from Dan Carpenter to handle errors
of read_tree_block ] Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:49 +0000 (15:43 -0400)]
btrfs: pass root to various extent ref mod functions
We need the actual root for the ref verifier tool to work, so change
these functions to pass the root around instead. This will be used in
a subsequent patch.
Signed-off-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Fri, 29 Sep 2017 19:43:48 +0000 (15:43 -0400)]
btrfs: add ref-verify mount option
This adds the infrastructure for turning ref verify on and off for a
mount, to be used by a later patch.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ enhnance btrfs_print_mod_info to print if ref-verify is compiled in ] Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 4 Oct 2017 15:30:11 +0000 (17:30 +0200)]
btrfs: get rid of sector_t and use u64 offset in submit_extent_page
The use of sector_t in the callchain of submit_extent_page is not
necessary. Switch to u64 and rename the variable and use byte units
instead of 512b, ie. dropping the >> 9 shifts and avoiding the
con(tro)versions of sector_t.
Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: David Sterba <dsterba@suse.com>
David Sterba [Wed, 4 Oct 2017 15:07:07 +0000 (17:07 +0200)]
btrfs: scrub: get rid of sector_t
The use of sector_t is not necessry, it's just for a warning. Switch to
u64 and rename the variable and use byte units instead of 512b, ie.
dropping the >> 9 shifts. The messages are adjusted as well.
Reviewed-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Wed, 27 Sep 2017 14:43:13 +0000 (10:43 -0400)]
btrfs: fix send ioctl on 32bit with 64bit kernel
We pass in a pointer in our send arg struct, this means the struct size
doesn't match with 32bit user space and 64bit kernel space. Fix this by
adding a compat mode and doing the appropriate conversion.
Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ move structure to the beginning, next to receive 32bit compat ] Signed-off-by: David Sterba <dsterba@suse.com>
Anand Jain [Mon, 9 Oct 2017 03:07:46 +0000 (11:07 +0800)]
btrfs: fix use of error or warning for missing device
When device is missing without the -o degraded option then its an error
so report it as an error instead of a warning. And when -o degraded
option is provided, log the missing device as warning.
Signed-off-by: Anand Jain <anand.jain@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ switch error to bool ] Signed-off-by: David Sterba <dsterba@suse.com>
Anand Jain [Wed, 11 Oct 2017 04:46:18 +0000 (12:46 +0800)]
btrfs: add_missing_dev() should return the actual error
add_missing_dev() can return device pointer so that IS_ERR/PTR_ERR can
be used to check for the actual error that occurred in the function.
Signed-off-by: Anand Jain <anand.jain@oracle.com> Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
[ minor error message adjustment ] Signed-off-by: David Sterba <dsterba@suse.com>
Arnd Bergmann [Fri, 13 Oct 2017 09:27:35 +0000 (11:27 +0200)]
btrfs: tree-checker: use %zu format string for size_t
We now get a harmless compile-time on 32-bit architectures:
fs/btrfs/tree-checker.c: In function 'check_extent_data_item':
fs/btrfs/tree-checker.c:189:70: error: format '%lu' expects argument of type 'long unsigned int', but argument 6 has type 'unsigned int' [-Werror=format=]
This changes the format string to use %zu instead of %lu for size_t.
Fixes: c1f6520bf360 ("btrfs: tree-checker: Enhance output for check_extent_data_item") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Thu, 7 Sep 2017 17:22:22 +0000 (11:22 -0600)]
Btrfs: remove nr_async_submits and async_submit_draining
Now that we have the combo of flushing twice, which can make sure IO
have started since the second flush will wait for page lock which
won't be unlocked unless setting page writeback and queuing ordered
extents, we don't need %async_submit_draining, %async_delalloc_pages
and %nr_async_submits to tell whether the IO has actually started.
Moreover, all the flushers in use are followed by functions that wait
for ordered extents to complete, so %nr_async_submits, which tracks
whether bio's async submit has made progress, doesn't really make
sense.
However, %async_delalloc_pages is still required by shrink_delalloc()
as that function doesn't flush twice in the normal case (just issues a
writeback with WB_REASON_FS_FREE_SPACE).
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Thu, 7 Sep 2017 17:22:21 +0000 (11:22 -0600)]
Btrfs: do not make defrag wait on async_delalloc_pages
By setting compression for a defrag task, the task will start IO at
the end of defrag.
After the combo of filemap_flush(), we've already made sure that
dirty pages have made progress via async compress thread because the
second filemap_flush() will wait for page lock, which won't be
unlocked until those pages have been marked as writeback and ordered
extents have been queued.
And this is for per-inode defrag, it's not helpful to wait on a global
%async_delalloc_pages and %nr_async_submits from fs_info.
Although waiting on %nr_async_submits means that all bios are
submitted down to per-device schedule IO lists, it doesn't wait for
their completions, thus users still need to do fsync/sync to make sure
the data is on disk. While with this change, it makes sure that pages
are marked with writeback bits and will be submitted asynchronously
shortly, therefore, the behavior of defrag option '-c' remains unchanged.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Thu, 7 Sep 2017 17:22:20 +0000 (11:22 -0600)]
Btrfs: remove nr_async_bios
This was intended to congest higher layers to not send bios, but as
1) the congested bit has been taken by writeback
Async bios come from buffered writes and DIO writes.
For DIO writes, we want to submit them ASAP, while for buffered writes,
writeback uses balance_dirty_pages() to throttle how much dirty pages we
can have.
2) and no one is waiting for %nr_async_bios down to zero,
Historically, it was introduced along with changes which let
checksumming workload spread accross different cpus. And at that time,
pdflush was used instead of per-bdi flushing, perhaps pdflush did not
have the necessary information for writeback to do throttling.
We can safely remove them now.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
[ additional explanation from mails, removed unused variable 'limit' ] Signed-off-by: David Sterba <dsterba@suse.com>
Rakesh Pandit [Tue, 10 Oct 2017 10:48:05 +0000 (13:48 +0300)]
btrfs: use appropriate replacements for __sb_{start,end}_write calls
Commit a53f4f8e9c8eb ("btrfs: Don't call btrfs_start_transaction() on
frozen fs to avoid deadlock.") started using internal calls and we
replace them with more suitable ones.
Signed-off-by: Rakesh Pandit <rakesh@tuxera.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Currently struct names for sysfs are generated only based on the
attribute names. This means that attribute names cannot be reused in
multiple places throughout the complete btrfs sysfs hierarchy.
E.g. allocation/data/total_bytes and allocation/data/single/total_bytes
result in the same struct name btrfs_attr_total_bytes. A workaround for
this case was made in the past by ad hoc creating an extra macro
wrapper, BTRFS_RAID_ATTR, that inserts some extra text in the struct
name.
Instead of polluting sysfs.h with such kind of extra macro definitions,
and only doing so when there are collisions, use a prefix which gets
inserted in the struct name, so we keep everything nicely grouped
together by default.
Current collections of attributes are:
* (the toplevel, empty prefix)
* allocation
* space_info
* raid
* features
Signed-off-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Thu, 28 Sep 2017 08:45:26 +0000 (11:45 +0300)]
btrfs: Refactor transaction handling in received subvolume ioctl
If btrfs_transaction_commit fails it will proceed to call
cleanup_transaction, which in turn already does btrfs_abort_transaction.
So let's remove the unnecessary code duplication. Also let's be explicit
about handling failure of btrfs_uuid_tree_add by calling
btrfs_end_transaction.
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_udpate_root can fail and it aborts the transaction, the correct
way to handle an aborted transaction is to explicitly end with
btrfs_end_transaction. Even now the code is correct since
btrfs_commit_transaction would handle an aborted transaction but this is
more of an implementation detail. So let's be explicit in handling
failure in btrfs_update_root.
Furthermore btrfs_commit_transaction can also fail and by ignoring it's
return value we could have left the in-memory copy of the root item in
an inconsistent state. So capture the error value which allows us to
correctly revert the RO/RW flags in case of commit failure.
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
btrfs: undo writable superblocke when sprouting fails
When new device is being added to seed FS, seed FS is marked writable,
but when we fail to bring in the new device, we missed to undo the
writable part. This patch fixes it.
Signed-off-by: Anand Jain <anand.jain@oracle.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Qu Wenruo [Wed, 23 Aug 2017 07:57:58 +0000 (16:57 +0900)]
btrfs: Add sanity check for EXTENT_DATA when reading out leaf
Add extra checks for item with EXTENT_DATA type. This checks the
following thing:
0) Key offset
All key offsets must be aligned to sectorsize.
Inline extent must have 0 for key offset.
1) Item size
Uncompressed inline file extent size must match item size.
(Compressed inline file extent has no information about its on-disk size.)
Regular/preallocated file extent size must be a fixed value.
2) Every member of regular file extent item
Including alignment for bytenr and offset, possible value for
compression/encryption/type.
3) Type/compression/encode must be one of the valid values.
This should be the most comprehensive and strict check in the context
of btrfs_item for EXTENT_DATA.
Signed-off-by: Qu Wenruo <quwenruo.btrfs@gmx.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ switch to BTRFS_FILE_EXTENT_TYPES, similar to what
BTRFS_COMPRESS_TYPES does ] Signed-off-by: David Sterba <dsterba@suse.com>
Qu Wenruo [Wed, 23 Aug 2017 07:57:56 +0000 (16:57 +0900)]
btrfs: Refactor check_leaf function for later expansion
Current check_leaf() function does a good job checking key order and
item offset/size.
However it only checks from slot 0 to the last but one slot, this is
good but makes later expansion hard.
So this refactoring iterates from slot 0 to the last slot.
For key comparison, it uses a key with all 0 as initial key, so all
valid keys should be larger than that.
And for item size/offset checks, it compares current item end with
previous item offset.
For slot 0, use leaf end as a special case.
This makes later item/key offset checks and item size checks easier to
be implemented.
Also, makes check_leaf() to return -EUCLEAN other than -EIO to indicate
error.
Signed-off-by: Qu Wenruo <quwenruo.btrfs@gmx.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Josef Bacik [Tue, 22 Aug 2017 20:00:39 +0000 (16:00 -0400)]
btrfs: change how we decide to commit transactions during flushing
Nikolay reported that generic/273 was failing currently with ENOSPC.
Turns out this is because we get to the point where the outstanding
reservations are greater than the pinned space on the fs. This is a
mistake, previously we used the current reservation amount in
may_commit_transaction, not the entire outstanding reservation amount.
Fix this to find the minimum byte size needed to make progress in
flushing, and pass that into may_commit_transaction. From there we can
make a smarter decision on whether to commit the transaction or not.
This fixes the failure in generic/273.
From Nikolai, IOW: when we go to the final stage of deciding whether to
do trans commit, instead of passing all the reservations from all
tickets we just pass the reservation for the current ticket. Otherwise,
in case all reservations exceed pinned space, then we don't commit
transaction and fail prematurely. Before we passed num_bytes from
flush_space, where num_bytes was the sum of all pending reserverations,
but now all we do is take the first ticket and commit the trans if we
can satisfy that.
Fixes: 957780eb2788 ("Btrfs: introduce ticketed enospc infrastructure") Cc: stable@vger.kernel.org # 4.8 Reported-by: Nikolay Borisov <nborisov@suse.com> Signed-off-by: Josef Bacik <jbacik@fb.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Tested-by: Nikolay Borisov <nborisov@suse.com>
[ added Nikolai's comment ] Signed-off-by: David Sterba <dsterba@suse.com>
By analyzing the perf on btrfs send, we found it take large amount of
cpu time on page_cache_sync_readahead. This effort can be reduced after
switching to asynchronous one. Overall performance gain on HDD and SSD
were 9 and 15 percent if simply send a large file.
Signed-off-by: Kuanling Huang <peterh@synology.com> Reviewed-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Allen Pais [Wed, 20 Sep 2017 06:17:46 +0000 (11:47 +0530)]
btrfs: return -ENOMEM on allocation failure in btrfsic
Forward the correct return value -ENOMEM from btrfsic_dev_state_alloc()
too.
Signed-off-by: Allen Pais <allen.lkml@gmail.com> Reviewed-by: Anand Jain <anand.jain@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ adjust changelog ] Signed-off-by: David Sterba <dsterba@suse.com>
btrfs_freespace_write_helper is actually calling normal_worker_helper
instead of btrfs_scrubparity_helper, so somehow kernel has parsed the
incorrect function address while unwinding the stack,
btrfs_scrubparity_helper really shouldn't be shown up.
It's caused by compiler doing inline for our helper function, adding a
noinline tag can fix that.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ use noinline_for_stack ] Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Wed, 13 Sep 2017 18:18:22 +0000 (12:18 -0600)]
Btrfs: remove bio_flags which indicates a meta block of log-tree
Since both committing transaction and writing log-tree are doing
plugging on metadata IO, we can unify to use %sync_writers to benefit
both cases, instead of checking bio_flags while writing meta blocks of
log-tree.
We can remove this bio_flags because in order to write dirty blocks,
log tree also uses btrfs_write_marked_extents(), inside which we
have enabled %sync_writers, therefore, every write goes in a
synchronous way, so does checksuming.
Please also note that, bio_flags is applied per-context while
%sync_writers is applied per-inode, so this might incur some overhead, ie.
1) while log tree is flushing its dirty blocks via
btrfs_write_marked_extents(), in which %sync_writers is increased
by one.
2) in the meantime, some writeback operations may happen upon btrfs's
metadata inode, so these writes go synchronously, too.
However, AFAICS, the overhead is not a big one while the win is that
we unify the two places that needs synchronous way and remove a
special hack/flag.
This removes the bio_flags related stuff for writing log-tree.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Mon, 21 Aug 2017 21:49:59 +0000 (15:49 -0600)]
Btrfs: make plug in writing meta blocks really work
We have started plug in btrfs_write_and_wait_marked_extents() but the
generated IOs actually go to device's schedule IO list where the work
is doing in another task, thus the started plug doesn't make any
sense.
And since we wait for IOs immediately after writing meta blocks, it's
the same case as writing log tree, doing sync submit can merge more
IOs.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Colin Ian King [Mon, 11 Sep 2017 15:15:28 +0000 (16:15 +0100)]
btrfs: avoid null pointer dereference on fs_info when calling btrfs_crit
There are checks on fs_info in __btrfs_panic to avoid dereferencing a
null fs_info, however, there is a call to btrfs_crit that may also
dereference a null fs_info. Fix this by adding a check to see if fs_info
is null and only print the s_id if fs_info is non-null.
Detected by CoverityScan CID#401973 ("Dereference after null check")
Fixes: efe120a067c8 ("Btrfs: convert printk to btrfs_ and fix BTRFS prefix") Signed-off-by: Colin Ian King <colin.king@canonical.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
The value of variable 'can_recover' is never used after being set, thus
it should be removed, as it was never used since the first commit 68a7342c51c95042 ("Btrfs: cleanup orphaned root orphan item").
Signed-off-by: Christos Gkekas <chris.gekas@gmail.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 21 Aug 2017 09:43:50 +0000 (12:43 +0300)]
btrfs: Remove redundant argument of __link_block_group
__link_block_group is called from only 2 places and at each call site the
space_info being passed is the same as the space info assigned to the passed
cache struct. Let's remove the redundant argument and make the function
reference the space_info from the passed block_group_cache. No functional
changes
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ renamed to link_block_group ] Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 21 Aug 2017 09:43:49 +0000 (12:43 +0300)]
btrfs: Rework error handling of add_extent_mapping in __btrfs_alloc_chunk
Currently the code executes add_extent_mapping and if it is successful
it links the new mapping, it then proceeds to unlock the extent mapping
tree and check for failure and handle them. Instead, rework the code to
only perform a single check if add_extent_mapping has failed and handle
it, otherwise the code continues in a linear fashion. No functional
changes
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 21 Aug 2017 09:43:46 +0000 (12:43 +0300)]
btrfs: Remove unused parameter from check_direct_IO
Introduced by 5a5f79b57069 ("Btrfs: allow unaligned DIO") and never
used. The buffered fallback from unaligned DIO works as expected.
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Timofey Titovets <nefelim4ag@gmail.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 21 Aug 2017 09:43:45 +0000 (12:43 +0300)]
btrfs: Remove unused arguments from btrfs_changed_cb_t
btrfs_changed_cb_t represents the signature of the callback being passed
to btrfs_compare_trees. Currently there is only one such callback,
namely changed_cb in send.c. This function doesn't really uses the first
2 parameters, i.e. the roots. Since there are not other functions
implementing the btrfs_changed_cb_t let's remove the unused parameters
from the prototype and implementation.
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: David Sterba <dsterba@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 21 Aug 2017 09:43:44 +0000 (12:43 +0300)]
btrfs: Remove unused parameters from various functions
iterate_dir_item:found_key - introduced in 31db9f7c23fb ("Btrfs:
introduce BTRFS_IOC_SEND for btrfs send/receive"), yet never used.
record_ref:num - ditto
This is a first pass with the low-hanging fruit. There are still quite a
few unsued parameters in some function which have to abide by a callback
interface.
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Nikolay Borisov [Mon, 21 Aug 2017 09:43:43 +0000 (12:43 +0300)]
btrfs: Remove unused variable
Src was initially part of 31ff1cd25d37 ("Btrfs: Copy into the log tree in
big batches"), however 16e7549f045d ("Btrfs: incompatible format change
to remove hole extents") changed parameters passed to copy_items which
made the src variable redundant.
Signed-off-by: Nikolay Borisov <nborisov@suse.com> Reviewed-by: Timofey Titovets <nefelim4ag@gmail.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Fri, 18 Aug 2017 17:54:02 +0000 (11:54 -0600)]
Btrfs: do not async submit for nodatasum inodes
While we submit direct writes, if the inode is flagged with nodatasum,
there's no benefit to submit asynchronously, because
a) we don't have to calculate checksum across processors,
b) and direct IO has started a plug, but async submit makes us queue
IO on each device's scheduled IO list instead of DIO's plug list, so
that IOs get much less merges in general.
Lets use sync submit for nodatasum inodes.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: Josef Bacik <jbacik@fb.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Thu, 3 Aug 2017 19:53:31 +0000 (13:53 -0600)]
Btrfs: search parity device wisely
After mapping block with BTRFS_MAP_WRITE, parities have been sorted to
the end position, so this search can start from the first parity
stripe.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com>
[ copied changelog as a comment ] Signed-off-by: David Sterba <dsterba@suse.com>
Anand Jain [Tue, 1 Aug 2017 10:35:08 +0000 (18:35 +0800)]
btrfs: copy fsid to super_block s_uuid
We didn't copy fsid to struct super_block.s_uuid so Overlay disables
index feature with btrfs as the lower FS.
kernel: overlayfs: fs on '/lower' does not support file handles, falling back to index=off.
Fix this by publishing the fsid through struct super_block.s_uuid.
[ dsterba: I think that setting s_uuid is the last missing bit. Overlay
needs the file handle encoding support from the lower filesystem, which
is supported. Filling the whole filesystem id is correct, the subvolume
id is encoded in the file handle buffer from inside btrfs_encode_fh. ]
Signed-off-by: Anand Jain <anand.jain@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
Liu Bo [Fri, 1 Sep 2017 22:14:28 +0000 (16:14 -0600)]
Btrfs: move finish_wait out of the loop
If we're still going to wait after schedule(), we don't have to do
finish_wait() to remove our %wait_queue_entry since prepare_to_wait()
won't add the same %wait_queue_entry twice.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: David Sterba <dsterba@suse.com> Signed-off-by: David Sterba <dsterba@suse.com>
2) In mac80211, validate user rate mask before configuring it. From
Johannes Berg.
3) Properly enforce memory limits in fair queueing code, from Toke
Hoiland-Jorgensen.
4) Fix lockdep splat in inet_csk_route_req(), from Eric Dumazet.
5) Fix TSO header allocation and management in mvpp2 driver, from Yan
Markman.
6) Don't take socket lock in BH handler in strparser code, from Tom
Herbert.
7) Don't show sockets from other namespaces in AF_UNIX code, from
Andrei Vagin.
8) Fix double free in error path of tap_open(), from Girish Moodalbail.
9) Fix TX map failure path in igb and ixgbe, from Jean-Philippe Brucker
and Alexander Duyck.
10) Fix DCB mode programming in stmmac driver, from Jose Abreu.
11) Fix err_count handling in various tunnels (ipip, ip6_gre). From Xin
Long.
12) Properly align SKB head before building SKB in tuntap, from Jason
Wang.
13) Avoid matching qdiscs with a zero handle during lookups, from Cong
Wang.
14) Fix various endianness bugs in sctp, from Xin Long.
15) Fix tc filter callback races and add selftests which trigger the
problem, from Cong Wang.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (73 commits)
selftests: Introduce a new test case to tc testsuite
selftests: Introduce a new script to generate tc batch file
net_sched: fix call_rcu() race on act_sample module removal
net_sched: add rtnl assertion to tcf_exts_destroy()
net_sched: use tcf_queue_work() in tcindex filter
net_sched: use tcf_queue_work() in rsvp filter
net_sched: use tcf_queue_work() in route filter
net_sched: use tcf_queue_work() in u32 filter
net_sched: use tcf_queue_work() in matchall filter
net_sched: use tcf_queue_work() in fw filter
net_sched: use tcf_queue_work() in flower filter
net_sched: use tcf_queue_work() in flow filter
net_sched: use tcf_queue_work() in cgroup filter
net_sched: use tcf_queue_work() in bpf filter
net_sched: use tcf_queue_work() in basic filter
net_sched: introduce a workqueue for RCU callbacks of tc filter
sctp: fix some type cast warnings introduced since very beginning
sctp: fix a type cast warnings that causes a_rwnd gets the wrong value
sctp: fix some type cast warnings introduced by transport rhashtable
sctp: fix some type cast warnings introduced by stream reconf
...
3. Two more bugs found by Chris:
https://patchwork.ozlabs.org/patch/826696/
https://patchwork.ozlabs.org/patch/826695/
Usually RCU callbacks are simple, however for TC filters and actions,
they are complex because at least TC actions could be destroyed
together with the TC filter in one callback. And RCU callbacks are
invoked in BH context, without locking they are parallel too. All of
these contribute to the cause of these nasty bugs.
Alternatively, we could also:
a) Introduce a spinlock to serialize these RCU callbacks. But as I
said in commit 1697c4bb5245 ("net_sched: carefully handle
tcf_block_put()"), it is very hard to do because of tcf_chain_dump().
Potentially we need to do a lot of work to make it possible (if not
impossible).
b) Just get rid of these RCU callbacks, because they are not
necessary at all, callers of these call_rcu() are all on slow paths
and holding RTNL lock, so blocking is allowed in their contexts.
However, David and Eric dislike adding synchronize_rcu() here.
As suggested by Paul, we could defer the work to a workqueue and
gain the permission of holding RTNL again without any performance
impact, however, in tcf_block_put() we could have a deadlock when
flushing workqueue while hodling RTNL lock, the trick here is to
defer the work itself in workqueue and make it queued after all
other works so that we keep the same ordering to avoid any
use-after-free. Please see the first patch for details.
Patch 1 introduces the infrastructure, patch 2~12 move each
tc filter to the new tc filter workqueue, patch 13 adds
an assertion to catch potential bugs like this, patch 14
closes another rcu callback race, patch 15 and patch 16 add
new test cases.
====================
Reported-by: Chris Mi <chrism@mellanox.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Jiri Pirko <jiri@resnulli.us> Cc: John Fastabend <john.fastabend@gmail.com> Cc: Jamal Hadi Salim <jhs@mojatatu.com> Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / mirror_ubuntu-focal-kernel.git / log