page_writeback: put account_page_redirty() after set_page_dirty()
Helper account_page_redirty() fixes dirty pages counter for redirtied
pages. This patch puts it after dirtying and prevents temporary
underflows of dirtied pages counters on zone/bdi and current->nr_dirtied.
Signed-off-by: Konstantin Khebnikov <khlebnikov@yandex-team.ru> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit 8d38633c3b4093aca7524945f1e9249d7d3a44da)
Shaohua Li [Fri, 9 Jan 2015 18:40:15 +0000 (10:40 -0800)]
btrfs: delete chunk allocation attemp when setting block group ro
Below test will fail currently:
mkfs.ext4 -F /dev/sda
btrfs-convert /dev/sda
mount /dev/sda /mnt
btrfs device add -f /dev/sdb /mnt
btrfs balance start -v -dconvert=raid1 -mconvert=raid1 /mnt
The reason is there are some block groups with usage 0, but the whole
disk hasn't free space to allocate new chunk, so we even can't set such
block group readonly. This patch deletes the chunk allocation when
setting block group ro. For META, we already have reserve. But for
SYSTEM, we don't have, so the check_system_chunk is still required.
Naohiro Aota [Mon, 5 Jan 2015 16:01:03 +0000 (01:01 +0900)]
btrfs: clear bio reference after submit_one_bio()
After submit_one_bio(), `bio' can go away. However submit_extent_page()
leave `bio' referable if submit_one_bio() failed (e.g. -ENOMEM on OOM).
It will cause invalid paging request when submit_extent_page() is called
next time.
I reproduced ENOMEM case with the following script (need
CONFIG_FAIL_PAGE_ALLOC, and CONFIG_FAULT_INJECTION_DEBUG_FS).
Filipe Manana [Fri, 23 Jan 2015 18:27:00 +0000 (18:27 +0000)]
Btrfs: add missing cleanup on sysfs init failure
If we failed during initialization of sysfs, we weren't unregistering the
top level btrfs sysfs entry nor the debugfs stuff.
Not unregistering the top level sysfs entry makes future attempts to reload
the btrfs module impossible and the following is reported in dmesg:
Filipe Manana [Thu, 29 Jan 2015 19:18:25 +0000 (19:18 +0000)]
Btrfs: fix race between transaction commit and empty block group removal
Committing a transaction can race with automatic removal of empty block
groups (cleaner kthread), leading to a BUG_ON() in the transaction
commit code while running btrfs_finish_extent_commit(). The following
sequence diagram shows how it can happen:
CPU 1 CPU 2
btrfs_commit_transaction()
fs_info->running_transaction = NULL
btrfs_finish_extent_commit()
find_first_extent_bit()
-> found range for block group X
in fs_info->freed_extents[]
btrfs_delete_unused_bgs()
-> found block group X
Removed block group X's range
from fs_info->freed_extents[]
Fix this by ensuring that unpining the block group's range in
btrfs_finish_extent_commit() is done in a synchronized fashion
with removing the block group's range from freed_extents[]
in btrfs_delete_unused_bgs()
David Sterba [Fri, 31 Oct 2014 16:18:08 +0000 (17:18 +0100)]
btrfs: more superblock checks, lower bounds on devices and sectorsize/nodesize
I received a few crafted images from Jiri, all got through the recently
added superblock checks. The lower bounds checks for num_devices and
sector/node -sizes were missing and caused a crash during mount.
Tools for symbolic code execution were used to prepare the images
contents.
Reported-by: Jiri Slaby <jslaby@suse.cz> Signed-off-by: David Sterba <dsterba@suse.cz> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 75d6ad382bb91f363452119d34238e156589ca2d)
Yang Dongsheng [Tue, 6 Jan 2015 12:54:42 +0000 (20:54 +0800)]
btrfs: qgroup: move WARN_ON() to the correct location.
In function qgroup_excl_accounting(), we need to WARN when
qg->excl is less than what we want to free, same to child
and parents. But currently, for parent qgroup, the WARN_ON()
is located after freeing qg->excl. It will WARN out even we
free it normally.
This patch move this WARN_ON() before freeing qg->excl.
Signed-off-by: Dongsheng Yang <yangds.fnst@cn.fujitsu.com> Reviewed-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 0ee13fe28ce387864c0d2b1e8c52b64abe2fcd02)
Liu Bo [Wed, 17 Dec 2014 08:14:09 +0000 (16:14 +0800)]
Btrfs: cleanup unused run_most
"run_most" is not used anymore.
Signed-off-by: Liu Bo <bo.li.liu@oracle.com> Reviewed-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 26455d3318a1e2a38f783db07981e3ed67de40ed)
Zhao Lei [Tue, 20 Jan 2015 07:11:45 +0000 (15:11 +0800)]
Rename all ref_count to refs in struct
refs is better than ref_count to record a struct's ref count.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Suggested-by: David Sterba <dsterba@suse.cz> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 570193454a5841a9fa477ce1bcc6c30ca6fcf552)
Zhao Lei [Tue, 20 Jan 2015 07:11:44 +0000 (15:11 +0800)]
Btrfs: Introduce BTRFS_BLOCK_GROUP_RAID56_MASK to check raid56 simply
So we can check raid56 with:
(map->type & BTRFS_BLOCK_GROUP_RAID56_MASK)
instead of long:
(map->type & (BTRFS_BLOCK_GROUP_RAID5 | BTRFS_BLOCK_GROUP_RAID6))
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit ffe2d2034bbb34f49f76c808550fdfbea2ea1659)
scrub_setup_recheck_block() have many arguments but most of them
can be get from one of them, we can remove them to make code clean.
Some other cleanup for that function also included in this patch.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit be50a8ddaae1d07135fd7e1c7017c1611075a560)
Zhao Lei [Tue, 20 Jan 2015 07:11:41 +0000 (15:11 +0800)]
Btrfs: Combine per-page recover in dev-replace and scrub
The code are similar, combine them to make code clean and easy to maintenance.
Some lost condition are also completed with benefit of this combination.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit b968fed1c3810a0a0d575cab8a72431fbc807615)
Zhao Lei [Tue, 20 Jan 2015 07:11:40 +0000 (15:11 +0800)]
Btrfs: Separate finding-right-mirror and writing-to-target's process in scrub_handle_errored_block()
In corrent code, code of finding-right-mirror and writing-to-target
are mixed in logic, if we find a right mirror but failed in writing
to target, it will treat as "hadn't found right block", and fill the
target with sblock_bad.
Actually, "failed in writing to target" does not mean "source
block is wrong", this patch separate above two condition in logic,
and do some cleanup to make code clean.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 8d6738c1bd74a27ff6a5043c5211c7bff7745420)
Zhao Lei [Tue, 20 Jan 2015 07:11:34 +0000 (15:11 +0800)]
Btrfs: add ref_count and free function for btrfs_bio
1: ref_count is simple than current RBIO_HOLD_BBIO_MAP_BIT flag
to keep btrfs_bio's memory in raid56 recovery implement.
2: free function for bbio will make code clean and flexible, plus
forced data type checking in compile.
Changelog v1->v2:
Rename following by David Sterba's suggestion:
put_btrfs_bio() -> btrfs_put_bio()
get_btrfs_bio() -> btrfs_get_bio()
bbio->ref_count -> bbio->refs
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit 6e9606d2a2dce098c1739fb3cd82a1c34fd73d3a)
Zhao Lei [Tue, 20 Jan 2015 07:11:31 +0000 (15:11 +0800)]
Btrfs: fix a out-of-bound access of raid_map
We add the number of stripes on target devices into bbio->num_stripes
if we are under device replacement, and we just sort the raid_map of
those stripes that not on the target devices, so if when we need
real raid_map, we need skip the stripes on the target devices.
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com> Signed-off-by: Miao Xie <miaox@cn.fujitsu.com> Signed-off-by: Chris Mason <clm@fb.com>
(cherry picked from commit e34c330d639177bbb345bf2bde16613b00cc6e6b)
Filipe Manana [Wed, 14 Jan 2015 01:52:25 +0000 (01:52 +0000)]
Btrfs: fix fsync log replay for inodes with a mix of regular refs and extrefs
If we have an inode with a large number of hard links, some of which may
be extrefs, turn a regular ref into an extref, fsync the inode and then
replay the fsync log (after a crash/reboot), we can endup with an fsync
log that makes the replay code always fail with -EOVERFLOW when processing
the inode's references.
This is easy to reproduce with the test case I made for xfstests. Its steps
are the following:
# Create a test file with 3001 hard links. This number is large enough to
# make btrfs start using extrefs at some point even if the fs has the maximum
# possible leaf/node size (64Kb).
echo "hello world" > $SCRATCH_MNT/foo
for i in `seq 1 3000`; do
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_`printf "%04d" $i`
done
# Make sure all metadata and data are durably persisted.
sync
# Now remove one link, add a new one with a new name, add another new one with
# the same name as the one we just removed and fsync the inode.
rm -f $SCRATCH_MNT/foo_link_0001
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3001
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_0001
rm -f $SCRATCH_MNT/foo_link_0002
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3002
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3003
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Simulate a crash/power loss. This makes sure the next mount
# will see an fsync log and will replay that log.
# Check that the number of hard links is correct, we are able to remove all
# the hard links and read the file's data. This is just to verify we don't
# get stale file handle errors (due to dangling directory index entries that
# point to inodes that no longer exist).
echo "Link count: $(stat --format=%h $SCRATCH_MNT/foo)"
[ -f $SCRATCH_MNT/foo ] || echo "Link foo is missing"
for ((i = 1; i <= 3003; i++)); do
name=foo_link_`printf "%04d" $i`
if [ $i -eq 2 ]; then
[ -f $SCRATCH_MNT/$name ] && echo "Link $name found"
else
[ -f $SCRATCH_MNT/$name ] || echo "Link $name is missing"
fi
done
rm -f $SCRATCH_MNT/foo_link_*
cat $SCRATCH_MNT/foo
rm -f $SCRATCH_MNT/foo
status=0
exit
The fix is simply to correct the overflow condition when overwriting a
reference item because it was wrong, trying to increase the item in the
fs/subvol tree by an impossible amount. Also ensure that we don't insert
one normal ref and one ext ref for the same dentry - this happened because
processing a dir index entry from the parent in the log happened when
the normal ref item was full, which made the logic insert an extref and
later when the normal ref had enough room, it would be inserted again
when processing the ref item from the child inode in the log.
This issue has been present since the introduction of the extrefs feature
(2012).
A test case for xfstests follows soon. This test only passes if the previous
patch titled "Btrfs: fix fsync when extend references are added to an inode"
is applied too.
Filipe Manana [Tue, 13 Jan 2015 16:40:04 +0000 (16:40 +0000)]
Btrfs: fix fsync when extend references are added to an inode
If we added an extended reference to an inode and fsync'ed it, the log
replay code would make our inode have an incorrect link count, which
was lower then the expected/correct count.
This resulted in stale directory index entries after deleting some of
the hard links, and any access to the dangling directory entries resulted
in -ESTALE errors because the entries pointed to inode items that don't
exist anymore.
This is easy to reproduce with the test case I made for xfstests, and
the bulk of that test is:
# Create a test file with 3001 hard links. This number is large enough to
# make btrfs start using extrefs at some point even if the fs has the maximum
# possible leaf/node size (64Kb).
echo "hello world" > $SCRATCH_MNT/foo
for i in `seq 1 3000`; do
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_`printf "%04d" $i`
done
# Make sure all metadata and data are durably persisted.
sync
# Add one more link to the inode that ends up being a btrfs extref and fsync
# the inode.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link_3001
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Simulate a crash/power loss. This makes sure the next mount
# will see an fsync log and will replay that log.
# Now after the fsync log replay btrfs left our inode with a wrong link count N,
# which was smaller than the correct link count M (N < M).
# So after removing N hard links, the remaining M - N directory entries were
# still visible to user space but it was impossible to do anything with them
# because they pointed to an inode that didn't exist anymore. This resulted in
# stale file handle errors (-ESTALE) when accessing those dentries for example.
#
# So remove all hard links except the first one and then attempt to read the
# file, to verify we don't get an -ESTALE error when accessing the inodel
#
# The btrfs fsck tool also detected the incorrect inode link count and it
# reported an error message like the following:
#
# root 5 inode 257 errors 2001, no inode item, link count wrong
# unresolved ref dir 256 index 2978 namelen 13 name foo_link_2976 filetype 1 errors 4, no inode ref
#
# The fstests framework automatically calls fsck after a test is run, so we
# don't need to call fsck explicitly here.
So make sure an fsync always flushes the delayed inode item, so that the
fsync log contains it (needed in order to trigger the link count fixup
code) and fix the extref counting function, which always return -ENOENT
to its caller (and made it assume there were always 0 extrefs).
This issue has been present since the introduction of the extrefs feature
(2012).
Filipe Manana [Sat, 10 Jan 2015 10:56:48 +0000 (10:56 +0000)]
Btrfs: fix directory inconsistency after fsync log replay
If we have an inode (file) with a link count greater than 1, remove
one of its hard links, fsync the inode, power fail/crash and then
replay the fsync log on the next mount, we end up getting the parent
directory's metadata inconsistent - its i_size still reflects the
deleted hard link and has dangling index entries (with no matching
inode reference entries). This prevents the directory from ever being
deletable, as its i_size can never decrease to BTRFS_EMPTY_DIR_SIZE
even if all of its children inodes are deleted, and the dangling index
entries can never be removed (as they point to an inode that does not
exist anymore).
This is easy to reproduce with the following excerpt from the test case
for xfstests that I just made:
_scratch_mkfs >> $seqres.full 2>&1
_init_flakey
_mount_flakey
# Create a test file with 2 hard links in the same directory.
mkdir -p $SCRATCH_MNT/a/b
echo "hello world" > $SCRATCH_MNT/a/b/foo
ln $SCRATCH_MNT/a/b/foo $SCRATCH_MNT/a/b/bar
# Make sure all metadata and data are durably persisted.
sync
# Now remove one of the hard links and fsync the inode.
rm -f $SCRATCH_MNT/a/b/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/a/b/foo
# Simulate a crash/power loss. This makes sure the next mount
# will see an fsync log and will replay that log.
# Remove the last hard link of the file and attempt to remove its parent
# directory - this failed in btrfs because the fsync log and replay code
# didn't decrement the parent directory's i_size and left dangling directory
# index entries - this made the btrfs rmdir implementation always fail with
# the error -ENOTEMPTY.
#
# The dangling directory index entries were visible to user space, but it was
# impossible to do anything on them (unlink, open, read, write, stat, etc)
# because the inode they pointed to did not exist anymore.
#
# The parent directory's metadata inconsistency (stale index entries) was
# also detected by btrfs' fsck tool, which is run automatically by the fstests
# framework when the test finishes. The error message reported by fsck was:
#
# root 5 inode 259 errors 2001, no inode item, link count wrong
# unresolved ref dir 258 index 3 namelen 3 name bar filetype 1 errors 4, no inode ref
#
rm -f $SCRATCH_MNT/a/b/*
rmdir $SCRATCH_MNT/a/b
rmdir $SCRATCH_MNT/a
To fix this just make sure that after an unlink, if the inode is fsync'ed,
he parent inode is fully logged in the fsync log.
Filipe Manana [Tue, 6 Jan 2015 20:18:45 +0000 (20:18 +0000)]
Btrfs: lookup for block group only if needed when freeing a tree block
Very often our extent buffer's header generation doesn't match the current
transaction's id or it is also referenced by other trees (snapshots), so
we don't need the corresponding block group cache object. Therefore only
search for it if we are going to use it, so we avoid an unnecessary search
in the block groups rbtree (and acquiring and releasing its spinlock).
Freeing a tree block is performed when COWing or deleting a node/leaf,
which implies we are holding the node/leaf's parent node lock, therefore
reducing the amount of time spent when freeing a tree block helps reducing
the amount of time we are holding the parent node's lock.
For example, for a run of xfstests/generic/083, the block group cache
object was needed only 682 times for a total of 226691 calls to free
a tree block.
David Sterba [Wed, 14 Jan 2015 18:52:13 +0000 (19:52 +0100)]
btrfs: switch extent_state state to unsigned
Currently there's a 4B hole in the structure between refs and state and there
are only 16 bits used so we can make it unsigned. This will get a better
packing and may save some stack space for local variables.
The size of extent_state gets reduced by 8B and there are usually a lot
of slab objects.
Filipe Manana [Tue, 20 Jan 2015 12:40:53 +0000 (12:40 +0000)]
Btrfs: fix setup_leaf_for_split() to avoid leaf corruption
We were incorrectly detecting when the target key didn't exist anymore
after releasing the path and re-searching the tree. This could make
us split or duplicate (btrfs_split_item() and btrfs_duplicate_item()
are its only callers at the moment) an item when we should not.
For the case of duplicating an item, we currently only duplicate
checksum items (csum tree) and file extent items (fs/subvol trees).
For the checksum items we end up overriding the item completely,
but for file extent items we update only some of their fields in
the copy (done in __btrfs_drop_extents), which means we can end up
having a logical corruption for some values.
Also for the case where we duplicate a file extent item it will make
us produce a leaf with a wrong key order, as btrfs_duplicate_item()
advances us to the next slot and then its caller sets a smaller key
on the new item at that slot (like in __btrfs_drop_extents() e.g.).
Alternatively if the tree search in setup_leaf_for_split() leaves
with path->slots[0] == btrfs_header_nritems(path->nodes[0]), we end
up accessing beyond the leaf's end (when we check if the item's size
has changed) and make our caller insert an item at the invalid slot
btrfs_header_nritems(path->nodes[0]) + 1, causing an invalid memory
access if the leaf is full or nearly full.
This issue has been present since the introduction of this function
in 2009:
Josef Bacik [Mon, 17 Nov 2014 20:45:48 +0000 (15:45 -0500)]
Btrfs: track dirty block groups on their own list
Currently any time we try to update the block groups on disk we will walk _all_
block groups and check for the ->dirty flag to see if it is set. This function
can get called several times during a commit. So if you have several terabytes
of data you will be a very sad panda as we will loop through _all_ of the block
groups several times, which makes the commit take a while which slows down the
rest of the file system operations.
This patch introduces a dirty list for the block groups that we get added to
when we dirty the block group for the first time. Then we simply update any
block groups that have been dirtied since the last time we called
btrfs_write_dirty_block_groups. This allows us to clean up how we write the
free space cache out so it is much cleaner. Thanks,
fs: export inode_to_bdi and use it in favor of mapping->backing_dev_info
Now that we got rid of the bdi abuse on character devices we can always use
sb->s_bdi to get at the backing_dev_info for a file, except for the block
device special case. Export inode_to_bdi and replace uses of
mapping->backing_dev_info with it to prepare for the removal of
mapping->backing_dev_info.
Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Tejun Heo <tj@kernel.org> Reviewed-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@fb.com>
(cherry picked from commit de1414a654e66b81b5348dbc5259ecf2fb61655e)
Pranith Kumar [Fri, 5 Dec 2014 16:24:45 +0000 (11:24 -0500)]
rcu: Make SRCU optional by using CONFIG_SRCU
SRCU is not necessary to be compiled by default in all cases. For tinification
efforts not compiling SRCU unless necessary is desirable.
The current patch tries to make compiling SRCU optional by introducing a new
Kconfig option CONFIG_SRCU which is selected when any of the components making
use of SRCU are selected.
If we do not select CONFIG_SRCU, srcu.o will not be compiled at all.
text data bss dec hex filename
2007 0 0 2007 7d7 kernel/rcu/srcu.o
Size of arch/powerpc/boot/zImage changes from
text data bss dec hex filename
831552 64180 23944 919676 e087c arch/powerpc/boot/zImage : before
829504 64180 23952 917636 e0084 arch/powerpc/boot/zImage : after
so the savings are about ~2000 bytes.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com> CC: Paul E. McKenney <paulmck@linux.vnet.ibm.com> CC: Josh Triplett <josh@joshtriplett.org> CC: Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
[ paulmck: resolve conflict due to removal of arch/ia64/kvm/Kconfig. ]
(cherry picked from commit 83fe27ea531161a655f02dc7732d14cfaa27fd5d)
Josef Bacik [Tue, 16 Dec 2014 16:54:43 +0000 (08:54 -0800)]
Btrfs: change how we track dirty roots
I've been overloading root->dirty_list to keep track of dirty roots and which
roots need to have their commit roots switched at transaction commit time. This
could cause us to lose an update to the root which could corrupt the file
system. To fix this use a state bit to know if the root is dirty, and if it
isn't set we go ahead and move the root to the dirty list. This way if we
re-dirty the root after adding it to the switch_commit list we make sure to
update it. This also makes it so that the extent root is always the last root
on the dirty list to try and keep the amount of churn down at this point in the
commit. Thanks,
David Sterba [Sun, 15 Jun 2014 01:20:26 +0000 (03:20 +0200)]
btrfs: unify extent buffer allocation api
Make the extent buffer allocation interface consistent. Cloned eb will
set a valid fs_info. For dummy eb, we can drop the length parameter and
set it from fs_info.
The built-in sanity checks may pass a NULL fs_info that's queried for
nodesize, but we know it's 4096.
David Sterba [Sun, 15 Jun 2014 00:39:54 +0000 (02:39 +0200)]
btrfs: sink blocksize parameter to btrfs_find_create_tree_block
Finally it's clear that the requested blocksize is always equal to
nodesize, with one exception, the superblock.
Superblock has fixed size regardless of the metadata block size, but
uses the same helpers to initialize sys array/chunk tree and to work
with the chunk items. So it pretends to be an extent_buffer for a
moment, btrfs_read_sys_array is full of special cases, we're adding one
more.
ACCESS_ONCE does not work reliably on non-scalar types. For
example gcc 4.6 and 4.7 might remove the volatile tag for such
accesses during the SRA (scalar replacement of aggregates) step
(https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145)
Change the ppc/kvm code to replace ACCESS_ONCE with READ_ONCE.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Alexander Graf <agraf@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ACCESS_ONCE does not work reliably on non-scalar types. For
example gcc 4.6 and 4.7 might remove the volatile tag for such
accesses during the SRA (scalar replacement of aggregates) step
(https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145)
Change the ppc/hugetlbfs code to replace ACCESS_ONCE with READ_ONCE.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ACCESS_ONCE does not work reliably on non-scalar types. For
example gcc 4.6 and 4.7 might remove the volatile tag for such
accesses during the SRA (scalar replacement of aggregates) step
(https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145)
Fixup gup_pmd_range.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 927609d622a3 ("kernel: tighten rules for ACCESS ONCE") results in a
compile failure for sh builds with CONFIG_X2TLB enabled.
arch/sh/mm/gup.c: In function 'gup_get_pte':
arch/sh/mm/gup.c:20:2: error: invalid initializer
make[1]: *** [arch/sh/mm/gup.o] Error 1
Replace ACCESS_ONCE with READ_ONCE to fix the problem.
Fixes: 927609d622a3 ("kernel: tighten rules for ACCESS ONCE") Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Guenter Roeck <linux@roeck-us.net> Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently maximum space limit quota format supports is in blocks however
since we store space limits in bytes, this is somewhat confusing. So
store the maximum limit in bytes as well. Also rename the field to match
the new unit and related inode field to match the new naming scheme.
Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ACCESS_ONCE does not work reliably on non-scalar types. For
example gcc 4.6 and 4.7 might remove the volatile tag for such
accesses during the SRA (scalar replacement of aggregates) step
(https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58145)
Change the p2m code to replace ACCESS_ONCE with READ_ONCE.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Reviewed-by: Juergen Gross <jgross@suse.com> Acked-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
if (unlikely(lock->tickets.tail & TICKET_SLOWPATH_FLAG))
__ticket_unlock_slowpath(lock, prev);
which is *exactly* the kind of things you cannot do with spinlocks,
because after you've done the "add_smp()" and released the spinlock
for the fast-path, you can't access the spinlock any more. Exactly
because a fast-path lock might come in, and release the whole data
structure.
Linus suggested that we should not do any writes to lock after unlock(),
and we can move slowpath clearing to fastpath lock.
So this patch implements the fix with:
1. Moving slowpath flag to head (Oleg):
Unlocked locks don't care about the slowpath flag; therefore we can keep
it set after the last unlock, and clear it again on the first (try)lock.
-- this removes the write after unlock. note that keeping slowpath flag would
result in unnecessary kicks.
By moving the slowpath flag from the tail to the head ticket we also avoid
the need to access both the head and tail tickets on unlock.
2. use xadd to avoid read/write after unlock that checks the need for
unlock_kick (Linus):
We further avoid the need for a read-after-release by using xadd;
the prev head value will include the slowpath flag and indicate if we
need to do PV kicking of suspended spinners -- on modern chips xadd
isn't (much) more expensive than an add + load.
The use of READ_ONCE() causes lots of warnings witht he pending paravirt
spinlock fixes, because those ends up having passing a member to a
'const' structure to READ_ONCE().
There should certainly be nothing wrong with using READ_ONCE() with a
const source, but the helper function __read_once_size() would cause
warnings because it would drop the 'const' qualifier, but also because
the destination would be marked 'const' too due to the use of 'typeof'.
Use a union of types in READ_ONCE() to avoid this issue.
Also make sure to use parenthesis around the macro arguments to avoid
possible operator precedence issues.
Commit 927609d622a3 ("kernel: tighten rules for ACCESS ONCE") results in
sparse warnings like "Using plain integer as NULL pointer" - Let's add a
type cast to the dummy assignment.
To avoid warnings lik "sparse: warning: cast to restricted __hc32" we also
use __force on that cast.
Fixes: 927609d622a3 ("kernel: tighten rules for ACCESS ONCE") Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Now that all non-scalar users of ACCESS_ONCE have been converted
to READ_ONCE or ASSIGN once, lets tighten ACCESS_ONCE to only
work on scalar types.
This variant was proposed by Alexei Starovoitov.
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
pmc_dbgfs_unregister() will be called when pmc->dbgfs_dir is unconditionally
NULL on error path in pmc_dbgfs_register(). To prevent this we move the
assignment to where is should be.
Fixes: f855911c1f48 (x86/pmc_atom: Expose PMC device state and platform sleep state) Reported-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Aubrey Li <aubrey.li@linux.intel.com> Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Cc: Kumar P. Mahesh <mahesh.kumar.p@intel.com> Link: http://lkml.kernel.org/r/1421253575-22509-2-git-send-email-andriy.shevchenko@linux.intel.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit b568b8601f05 ("Treat SCI interrupt as normal GSI interrupt")
accidently removes support of legacy PIC interrupt when fixing a
regression for Xen, which causes a nasty regression on HP/Compaq
nc6000 where we fail to register the ACPI interrupt, and thus
lose eg. thermal notifications leading a potentially overheated
machine.
So reintroduce support of legacy PIC based ACPI SCI interrupt.
Reported-by: Ville Syrjälä <syrjala@sci.fi> Tested-by: Ville Syrjälä <syrjala@sci.fi> Signed-off-by: Jiang Liu <jiang.liu@linux.intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Len Brown <len.brown@intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Rafael J. Wysocki <rjw@rjwysocki.net> Cc: Sander Eikelenboom <linux@eikelenboom.it> Cc: linux-pm@vger.kernel.org Link: http://lkml.kernel.org/r/1424052673-22974-1-git-send-email-jiang.liu@linux.intel.com Signed-off-by: Ingo Molnar <mingo@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Note that, it declares the "random_variable" variable as "unsigned int".
Since the result of the shifting operation between STACK_RND_MASK (which
is 0x3fffff on x86_64, 22 bits) and PAGE_SHIFT (which is 12 on x86_64):
random_variable <<= PAGE_SHIFT;
then the two leftmost bits are dropped when storing the result in the
"random_variable". This variable shall be at least 34 bits long to hold
the (22+12) result.
These two dropped bits have an impact on the entropy of process stack.
Concretely, the total stack entropy is reduced by four: from 2^28 to
2^30 (One fourth of expected entropy).
This patch restores back the entropy by correcting the types involved
in the operations in the functions randomize_stack_top() and
stack_maxrandom_size().
Andy pointed out that if an NMI or MCE is received while we're in the
middle of an EFI mixed mode call a triple fault will occur. This can
happen, for example, when issuing an EFI mixed mode call while running
perf.
The reason for the triple fault is that we execute the mixed mode call
in 32-bit mode with paging disabled but with 64-bit kernel IDT handlers
installed throughout the call.
At Andy's suggestion, stop playing the games we currently do at runtime,
such as disabling paging and installing a 32-bit GDT for __KERNEL_CS. We
can simply switch to the __KERNEL32_CS descriptor before invoking
firmware services, and run in compatibility mode. This way, if an
NMI/MCE does occur the kernel IDT handler will execute correctly, since
it'll jump to __KERNEL_CS automatically.
However, this change is only possible post-ExitBootServices(). Before
then the firmware "owns" the machine and expects for its 32-bit IDT
handlers to be left intact to service interrupts, etc.
So, we now need to distinguish between early boot and runtime
invocations of EFI services. During early boot, we need to restore the
GDT that the firmware expects to be present. We can only jump to the
__KERNEL32_CS code segment for mixed mode calls after ExitBootServices()
has been invoked.
A liberal sprinkling of comments in the thunking code should make the
differences in early and late environments more apparent.
Reported-by: Andy Lutomirski <luto@amacapital.net> Tested-by: Borislav Petkov <bp@suse.de> Signed-off-by: Matt Fleming <matt.fleming@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When reading blkio.throttle.io_serviced in a recently created blkio
cgroup, it's possible to race against the creation of a throttle policy,
which delays the allocation of stats_cpu.
Like other functions in the throttle code, just checking for a NULL
stats_cpu prevents the following oops caused by that race.
We have a scenario where after the fsync log replay we can lose file data
that had been previously fsync'ed if we added an hard link for our inode
and after that we sync'ed the fsync log (for example by fsync'ing some
other file or directory).
This is because when adding an hard link we updated the inode item in the
log tree with an i_size value of 0. At that point the new inode item was
in memory only and a subsequent fsync log replay would not make us lose
the file data. However if after adding the hard link we sync the log tree
to disk, by fsync'ing some other file or directory for example, we ended
up losing the file data after log replay, because the inode item in the
persisted log tree had an an i_size of zero.
This is easy to reproduce, and the following excerpt from my test for
xfstests shows this:
# Create one file with data and fsync it.
# This made the btrfs fsync log persist the data and the inode metadata with
# a correct inode->i_size (4096 bytes).
$XFS_IO_PROG -f -c "pwrite -S 0xaa -b 4K 0 4K" -c "fsync" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Now add one hard link to our file. This made the btrfs code update the fsync
# log, in memory only, with an inode metadata having a size of 0.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link
# Now force persistence of the fsync log to disk, for example, by fsyncing some
# other file.
touch $SCRATCH_MNT/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/bar
# Before a power loss or crash, we could read the 4Kb of data from our file as
# expected.
echo "File content before:"
od -t x1 $SCRATCH_MNT/foo
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# After the fsync log replay, because the fsync log had a value of 0 for our
# inode's i_size, we couldn't read anymore the 4Kb of data that we previously
# wrote and fsync'ed. The size of the file became 0 after the fsync log replay.
echo "File content after:"
od -t x1 $SCRATCH_MNT/foo
Another alternative test, that doesn't need to fsync an inode in the same
transaction it was created, is:
# Create our test file with some data.
$XFS_IO_PROG -f -c "pwrite -S 0xaa -b 8K 0 8K" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Make sure the file is durably persisted.
sync
# Append some data to our file, to increase its size.
$XFS_IO_PROG -f -c "pwrite -S 0xcc -b 4K 8K 4K" \
$SCRATCH_MNT/foo | _filter_xfs_io
# Fsync the file, so from this point on if a crash/power failure happens, our
# new data is guaranteed to be there next time the fs is mounted.
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/foo
# Add one hard link to our file. This made btrfs write into the in memory fsync
# log a special inode with generation 0 and an i_size of 0 too. Note that this
# didn't update the inode in the fsync log on disk.
ln $SCRATCH_MNT/foo $SCRATCH_MNT/foo_link
# Now make sure the in memory fsync log is durably persisted.
# Creating and fsync'ing another file will do it.
touch $SCRATCH_MNT/bar
$XFS_IO_PROG -c "fsync" $SCRATCH_MNT/bar
# As expected, before the crash/power failure, we should be able to read the
# 12Kb of file data.
echo "File content before:"
od -t x1 $SCRATCH_MNT/foo
# Simulate a crash/power loss.
_load_flakey_table $FLAKEY_DROP_WRITES
_unmount_flakey
# After mounting the fs again, the fsync log was replayed.
# The btrfs fsync log replay code didn't update the i_size of the persisted
# inode because the inode item in the log had a special generation with a
# value of 0 (and it couldn't know the correct i_size, since that inode item
# had a 0 i_size too). This made the last 4Kb of file data inaccessible and
# effectively lost.
echo "File content after:"
od -t x1 $SCRATCH_MNT/foo
This isn't a new issue/regression. This problem has been around since the
log tree code was added in 2008:
If btrfs_find_item is called with NULL path it allocates one locally but
does not free it. Affected paths are inserting an orphan item for a file
and for a subvol root.
Move the path allocation to the callers.
Fixes: 3f870c289900 ("btrfs: expand btrfs_find_item() to include find_orphan_item functionality") Signed-off-by: David Sterba <dsterba@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
It turns out it's possible to get __remove_osd() called twice on the
same OSD. That doesn't sit well with rb_erase() - depending on the
shape of the tree we can get a NULL dereference, a soft lockup or
a random crash at some point in the future as we end up touching freed
memory. One scenario that I was able to reproduce is as follows:
A case can be made that osd refcounting is imperfect and reworking it
would be a proper resolution, but for now Sage and I decided to fix
this by adding a safe guard around __remove_osd().
Fixes: http://tracker.ceph.com/issues/8087 Cc: Sage Weil <sage@redhat.com> Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Reviewed-by: Sage Weil <sage@redhat.com> Reviewed-by: Alex Elder <elder@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since kernel 3.14 the backlight control has been broken on various Samsung
Atom based netbooks. This has been bisected and this problem happens since
commit b35684b8fa94 ("drm/i915: do full backlight setup at enable time")
This has been reported and discussed in detail here:
http://lists.freedesktop.org/archives/intel-gfx/2014-July/049395.html
Unfortunately no-one has been able to fix this. This only affects Samsung
Atom netbooks, and the Linux kernel and the BIOS of those laptops have never
worked well together. All affected laptops already have a quirk to avoid using
the standard acpi-video interface and instead use the samsung specific SABI
interface which samsung-laptop uses. It seems that recent fixes to the i915
driver have also broken backlight control through the SABI interface.
The intel_backlight driver OTOH works fine, and also allows for finer grained
backlight control. So add a new use_native_backlight quirk, and replace the
broken_acpi_video quirk with this quirk for affected models. This new quirk
disables acpi-video as before and also stops samsung-laptop from registering
the SABI based samsung_laptop backlight interface, leaving only the working
intel_backlight interface.
This commit enables this new quirk for 3 models which are known to be affected,
chances are that it needs to be used on other models too.
When DRAM errors occur on memory controllers after EDAC_MAX_MCS (16),
the kernel fatally dereferences unallocated structures, see splat below;
this occurs on at least NumaConnect systems.
Fix by checking if a memory controller info structure was found.
Causes an RCW cycle to be forced even when the array is degraded.
A degraded array cannot support RCW as that requires reading all data
blocks, and one may be missing.
Forcing an RCW when it is not possible causes a live-lock and the code
spins, repeatedly deciding to do something that cannot succeed.
So change the condition to only force RCW on non-degraded arrays.
Commit f6edb53c4993ffe92ce521fb449d1c146cea6ec2 converted the probe to
a CPU wide event first (pid == -1). For kernels that do not support
the PERF_FLAG_FD_CLOEXEC flag the probe fails with EINVAL. Since this
errno is not handled pid is not reset to 0 and the subsequent use of
pid = -1 as an argument brings in an additional failure path if
perf_event_paranoid > 0:
$ perf record -- sleep 1
perf_event_open(..., 0) failed unexpectedly with error 13 (Permission denied)
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.007 MB /tmp/perf.data (11 samples) ]
Also, ensure the fd of the confirmation check is closed and comment why
pid = -1 is used.
Needs to go to 3.18 stable tree as well.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Based-on-patch-by: David Ahern <david.ahern@oracle.com> Acked-by: David Ahern <david.ahern@oracle.com> Cc: David Ahern <dsahern@gmail.com> Link: http://lkml.kernel.org/r/54EC610C.8000403@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We have two race conditions in the probe code which could lead to a null
pointer dereference in the interrupt handler.
The interrupt handler accesses the clockevent device, which may not yet be
registered.
First race condition happens when the interrupt handler gets registered before
the interrupts get disabled. The second race condition happens when the
interrupts get enabled, but the clockevent device is not yet registered.
Fix that by disabling the interrupts before we register the interrupt and enable
the interrupts after the clockevent device got registered.
Reported-by: Gongbae Park <yongbae2@gmail.com> Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The KSTK_EIP() and KSTK_ESP() macros should return the user program
counter (PC) and stack pointer (A0StP) of the given task. These are used
to determine which VMA corresponds to the user stack in
/proc/<pid>/maps, and for the user PC & A0StP in /proc/<pid>/stat.
However for Meta the PC & A0StP from the task's kernel context are used,
resulting in broken output. For example in following /proc/<pid>/maps
output, the 3afff000-3b021000 VMA should be described as the stack:
And in the following /proc/<pid>/stat output, the PC is in kernel code
(1074234964 = 0x40078654) and the A0StP is in the kernel heap
(1335981392 = 0x4fa17550):
Fix the definitions of KSTK_EIP() and KSTK_ESP() to use
task_pt_regs(tsk)->ctx rather than (tsk)->thread.kernel_context. This
gets the registers from the user context stored after the thread info at
the base of the kernel stack, which is from the last entry into the
kernel from userland, regardless of where in the kernel the task may
have been interrupted, which results in the following more correct
/proc/<pid>/maps output:
For filesystems without separate project quota inode field in the
superblock we just reuse project quota file for group quotas (and vice
versa) if project quota file is allocated and we need group quota file.
When we reuse the file, quota structures on disk suddenly have wrong
type stored in d_flags though. Nobody really cares about this (although
structure type reported to userspace was wrong as well) except
that after commit 14bf61ffe6ac (quota: Switch ->get_dqblk() and
->set_dqblk() to use bytes as space units) assertion in
xfs_qm_scall_getquota() started to trigger on xfs/106 test (apparently I
was testing without XFS_DEBUG so I didn't notice when submitting the
above commit).
Fix the problem by properly resetting ddq->d_flags when running quotacheck
for a quota file.
Reported-by: Al Viro <viro@ZenIV.linux.org.uk> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The gpio_chip operations receive a pointer the gpio_chip struct which is
contained in the driver's private struct, yet the container_of call in those
functions point to the mfd struct defined in include/linux/mfd/tps65912.h.
assumed that only one gpio-chip is registred per of-node.
Some drivers register more than one chip per of-node, so
adjust the matching function of_gpiochip_find_and_xlate to
not stop looking for chips if a node-match is found and
the translation fails.
Fixes: 7b8792bbdffd ("gpiolib: of: Correct error handling in of_get_named_gpiod_flags") Signed-off-by: Hans Holmberg <hans.holmberg@intel.com> Acked-by: Alexandre Courbot <acourbot@nvidia.com> Tested-by: Robert Jarzmik <robert.jarzmik@free.fr> Tested-by: Tyler Hall <tylerwhall@gmail.com> Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The native (64-bit) sigval_t union contains sival_int (32-bit) and
sival_ptr (64-bit). When a compat application invokes a syscall that
takes a sigval_t value (as part of a larger structure, e.g.
compat_sys_mq_notify, compat_sys_timer_create), the compat_sigval_t
union is converted to the native sigval_t with sival_int overlapping
with either the least or the most significant half of sival_ptr,
depending on endianness. When the corresponding signal is delivered to a
compat application, on big endian the current (compat_uptr_t)sival_ptr
cast always returns 0 since sival_int corresponds to the top part of
sival_ptr. This patch fixes copy_siginfo_to_user32() so that sival_int
is copied to the compat_siginfo_t structure.
Since the removal of CONFIG_REGULATOR_DUMMY option, the touchscreen stopped
working. This patch enables the "replacement" for REGULATOR_DUMMY and
allows the touchscreen to work even though there is no regulator for "vcc".
Signed-off-by: Martin Vajnar <martin.vajnar@gmail.com> Signed-off-by: Robert Jarzmik <robert.jarzmik@free.fr> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
projects / linux / log