When iputting the inode, We may leave the delayed nodes if they have some
delayed items that have not been dealt with. So when the inode is read again,
we must look up the relative delayed node, and use the information in it to
initialize the inode. Or we will get inconsonant inode information, it may
cause that the same directory index number is allocated again, and hit the
following oops:
[ 5447.554187] err add delayed dir index item(name: pglog_0.965_0) into the
insertion tree of the delayed node(root id: 262, inode id: 258, errno: -17)
[ 5447.569766] ------------[ cut here ]------------
[ 5447.575361] kernel BUG at fs/btrfs/delayed-inode.c:1301!
[SNIP]
[ 5447.790721] Call Trace:
[ 5447.793191] [<ffffffffa0641c4e>] btrfs_insert_dir_item+0x189/0x1bb [btrfs]
[ 5447.800156] [<ffffffffa0651a45>] btrfs_add_link+0x12b/0x191 [btrfs]
[ 5447.806517] [<ffffffffa0651adc>] btrfs_add_nondir+0x31/0x58 [btrfs]
[ 5447.812876] [<ffffffffa0651d6a>] btrfs_create+0xf9/0x197 [btrfs]
[ 5447.818961] [<ffffffff8111f840>] vfs_create+0x72/0x92
[ 5447.824090] [<ffffffff8111fa8c>] do_last+0x22c/0x40b
[ 5447.829133] [<ffffffff8112076a>] path_openat+0xc0/0x2ef
[ 5447.834438] [<ffffffff810c58e2>] ? __perf_event_task_sched_out+0x24/0x44
[ 5447.841216] [<ffffffff8103ecdd>] ? perf_event_task_sched_out+0x59/0x67
[ 5447.847846] [<ffffffff81121a79>] do_filp_open+0x3d/0x87
[ 5447.853156] [<ffffffff811e126c>] ? strncpy_from_user+0x43/0x4d
[ 5447.859072] [<ffffffff8111f1f5>] ? getname_flags+0x2e/0x80
[ 5447.864636] [<ffffffff8111f179>] ? do_getname+0x14b/0x173
[ 5447.870112] [<ffffffff8111f1b7>] ? audit_getname+0x16/0x26
[ 5447.875682] [<ffffffff8112b1ab>] ? spin_lock+0xe/0x10
[ 5447.880882] [<ffffffff81112d39>] do_sys_open+0x69/0xae
[ 5447.886153] [<ffffffff81112db1>] sys_open+0x20/0x22
[ 5447.891114] [<ffffffff813b9aab>] system_call_fastpath+0x16/0x1b
Fix it by reusing the old delayed node.
Reported-by: Jim Schutt <[email protected]>
Signed-off-by: Miao Xie <[email protected]>
Tested-by: Jim Schutt <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Changelog V5 -> V6:
- Fix oom when the memory load is high, by storing the delayed nodes into the
root's radix tree, and letting btrfs inodes go.
Changelog V4 -> V5:
- Fix the race on adding the delayed node to the inode, which is spotted by
Chris Mason.
- Merge Chris Mason's incremental patch into this patch.
- Fix deadlock between readdir() and memory fault, which is reported by
Itaru Kitayama.
Changelog V3 -> V4:
- Fix nested lock, which is reported by Itaru Kitayama, by updating space cache
inode in time.
Changelog V2 -> V3:
- Fix the race between the delayed worker and the task which does delayed items
balance, which is reported by Tsutomu Itoh.
- Modify the patch address David Sterba's comment.
- Fix the bug of the cpu recursion spinlock, reported by Chris Mason
Changelog V1 -> V2:
- break up the global rb-tree, use a list to manage the delayed nodes,
which is created for every directory and file, and used to manage the
delayed directory name index items and the delayed inode item.
- introduce a worker to deal with the delayed nodes.
Compare with Ext3/4, the performance of file creation and deletion on btrfs
is very poor. the reason is that btrfs must do a lot of b+ tree insertions,
such as inode item, directory name item, directory name index and so on.
If we can do some delayed b+ tree insertion or deletion, we can improve the
performance, so we made this patch which implemented delayed directory name
index insertion/deletion and delayed inode update.
Implementation:
- introduce a delayed root object into the filesystem, that use two lists to
manage the delayed nodes which are created for every file/directory.
One is used to manage all the delayed nodes that have delayed items. And the
other is used to manage the delayed nodes which is waiting to be dealt with
by the work thread.
- Every delayed node has two rb-tree, one is used to manage the directory name
index which is going to be inserted into b+ tree, and the other is used to
manage the directory name index which is going to be deleted from b+ tree.
- introduce a worker to deal with the delayed operation. This worker is used
to deal with the works of the delayed directory name index items insertion
and deletion and the delayed inode update.
When the delayed items is beyond the lower limit, we create works for some
delayed nodes and insert them into the work queue of the worker, and then
go back.
When the delayed items is beyond the upper bound, we create works for all
the delayed nodes that haven't been dealt with, and insert them into the work
queue of the worker, and then wait for that the untreated items is below some
threshold value.
- When we want to insert a directory name index into b+ tree, we just add the
information into the delayed inserting rb-tree.
And then we check the number of the delayed items and do delayed items
balance. (The balance policy is above.)
- When we want to delete a directory name index from the b+ tree, we search it
in the inserting rb-tree at first. If we look it up, just drop it. If not,
add the key of it into the delayed deleting rb-tree.
Similar to the delayed inserting rb-tree, we also check the number of the
delayed items and do delayed items balance.
(The same to inserting manipulation)
- When we want to update the metadata of some inode, we cached the data of the
inode into the delayed node. the worker will flush it into the b+ tree after
dealing with the delayed insertion and deletion.
- We will move the delayed node to the tail of the list after we access the
delayed node, By this way, we can cache more delayed items and merge more
inode updates.
- If we want to commit transaction, we will deal with all the delayed node.
- the delayed node will be freed when we free the btrfs inode.
- Before we log the inode items, we commit all the directory name index items
and the delayed inode update.
I did a quick test by the benchmark tool[1] and found we can improve the
performance of file creation by ~15%, and file deletion by ~20%.
Before applying this patch:
Create files:
Total files: 50000
Total time: 1.096108
Average time: 0.000022
Delete files:
Total files: 50000
Total time: 1.510403
Average time: 0.000030
After applying this patch:
Create files:
Total files: 50000
Total time: 0.932899
Average time: 0.000019
Delete files:
Total files: 50000
Total time: 1.215732
Average time: 0.000024
[1] http://marc.info/?l=linux-btrfs&m=128212635122920&q=p3
Many thanks for Kitayama-san's help!
Signed-off-by: Miao Xie <[email protected]>
Reviewed-by: David Sterba <[email protected]>
Tested-by: Tsutomu Itoh <[email protected]>
Tested-by: Itaru Kitayama <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Link count of the inode is not decreased if btrfs_set_inode_index()
fails.
Signed-off-by: Miao Xie <[email protected]>
Singed-off-by: Li Zefan <[email protected]>
the object id of the space cache inode's key is allocated from the relative
root, just like the regular file. So we can't identify space cache inode by
checking the object id of the inode's key, and we have to clear __GFP_FS flag
at the time we look up the space cache inode.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Liu Bo <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Using the GFP_HIGHUSER_MOVABLE flag to allocate the metadata's page may cause
deadlock.
Task1
open()
...
btrfs_search_slot()
...
btrfs_cow_block()
...
alloc_page()
wait for reclaiming
shrink_slab()
...
shrink_icache_memory()
...
btrfs_evict_inode()
...
btrfs_search_slot()
If the path is locked by task1, the deadlock happens.
So the btree's page cache is different with the file's page cache, it can not
allocate pages by GFP_HIGHUSER_MOVABLE flag, we must clear __GFP_FS flag in
GFP_HIGHUSER_MOVABLE flag.
Reported-by: Itaru Kitayama <[email protected]>
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
In the filesystem context, we must allocate memory by GFP_NOFS,
or we may start another filesystem operation and make kswap thread hang up.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
btrfs_link() will insert 3 items(inode ref, dir name item and dir index item)
into the b+ tree and update 2 items(its inode, and parent's inode) in the b+
tree. So we should reserve space for these 5 items, not 3 items.
Reported-by: Tsutomu Itoh <[email protected]>
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
fixup, which is allocated when starting page write to fix up the
extent without ORDERED bit set, should be freed after this work
is done.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Li Zefan <[email protected]>
When we store data by raid profile in btrfs with two or more different size
disks, df command shows there is some free space in the filesystem, but the
user can not write any data in fact, df command shows the wrong free space
information of btrfs.
# mkfs.btrfs -d raid1 /dev/sda9 /dev/sda10
# btrfs-show
Label: none uuid: a95cd49e-6e33-45b8-8741-a36153ce4b64
Total devices 2 FS bytes used 28.00KB
devid 1 size 5.01GB used 2.03GB path /dev/sda9
devid 2 size 10.00GB used 2.01GB path /dev/sda10
# btrfs device scan /dev/sda9 /dev/sda10
# mount /dev/sda9 /mnt
# dd if=/dev/zero of=tmpfile0 bs=4K count=9999999999
(fill the filesystem)
# sync
# df -TH
Filesystem Type Size Used Avail Use% Mounted on
/dev/sda9 btrfs 17G 8.6G 5.4G 62% /mnt
# btrfs-show
Label: none uuid: a95cd49e-6e33-45b8-8741-a36153ce4b64
Total devices 2 FS bytes used 3.99GB
devid 1 size 5.01GB used 5.01GB path /dev/sda9
devid 2 size 10.00GB used 4.99GB path /dev/sda10
It is because btrfs cannot allocate chunks when one of the pairing disks has
no space, the free space on the other disks can not be used for ever, and should
be subtracted from the total space, but btrfs doesn't subtract this space from
the total. It is strange to the user.
This patch fixes it by calcing the free space that can be used to allocate
chunks.
Implementation:
1. get all the devices free space, and align them by stripe length.
2. sort the devices by the free space.
3. check the free space of the devices,
3.1. if it is not zero, and then check the number of the devices that has
more free space than this device,
if the number of the devices is beyond the min stripe number, the free
space can be used, and add into total free space.
if the number of the devices is below the min stripe number, we can not
use the free space, the check ends.
3.2. if the free space is zero, check the next devices, goto 3.1
This implementation is just likely fake chunk allocation.
After appling this patch, df can show correct space information:
# df -TH
Filesystem Type Size Used Avail Use% Mounted on
/dev/sda9 btrfs 17G 8.6G 0 100% /mnt
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
With this patch, we change the handling method when we can not get enough free
extents with default size.
Implementation:
1. Look up the suitable free extent on each device and keep the search result.
If not find a suitable free extent, keep the max free extent
2. If we get enough suitable free extents with default size, chunk allocation
succeeds.
3. If we can not get enough free extents, but the number of the extent with
default size is >= min_stripes, we just change the mapping information
(reduce the number of stripes in the extent map), and chunk allocation
succeeds.
4. If the number of the extent with default size is < min_stripes, sort the
devices by its max free extent's size descending
5. Use the size of the max free extent on the (num_stripes - 1)th device as the
stripe size to allocate the device space
By this way, the chunk allocator can allocate chunks as large as possible when
the devices' space is not enough and make full use of the devices.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
- make it return the start position and length of the max free space when it can
not find a suitable free space.
- make it more readability
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
There are two tiny problem:
- One is When we check the chunk size is greater than the max chunk size or not,
we should take mirrors into account, but the original code didn't.
- The other is btrfs shouldn't use the size of the residual free space as the
length of of a dup chunk when doing chunk allocation. It is because the device
space that a dup chunk needs is twice as large as the chunk size, if we use
the size of the residual free space as the length of a dup chunk, we can not
get enough free space. Fix it.
Signed-off-by: Miao Xie <[email protected]>
Reviewed-by: Josef Bacik <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
We cannot write data into files when when there is tiny space in the filesystem.
Reproduce steps:
# mkfs.btrfs /dev/sda1
# mount /dev/sda1 /mnt
# dd if=/dev/zero of=/mnt/tmpfile0 bs=4K count=1
# dd if=/dev/zero of=/mnt/tmpfile1 bs=4K count=99999999999999
(fill the filesystem)
# umount /mnt
# mount /dev/sda1 /mnt
# rm -f /mnt/tmpfile0
# dd if=/dev/zero of=/mnt/tmpfile0 bs=4K count=1
(failed with nospec)
But if we do the last step again, we can write data successfully. The reason of
the problem is that btrfs didn't try to commit the current transaction and
reclaim some space when chunk allocation failed.
This patch fixes it by committing the current transaction to reclaim some
space when chunk allocation fails.
Signed-off-by: Miao Xie <[email protected]>
Reviewed-by: Josef Bacik <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Josef has implemented mixed data/metadata chunks, we must add those chunks'
space just like data chunks.
Signed-off-by: Miao Xie <[email protected]>
Reviewed-by: Josef Bacik <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
extent_bio_alloc() and compressed_bio_alloc() are similar, cleanup
similar source code.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
bio_endio() will free dip and dip->csums, so dip and dip->csums twice will
be freed twice. Fix it.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
This patch reduces the CPU time spent in the extent buffer search by using the
radix tree instead of the rbtree and using the rcu lock instead of the spin
lock.
I did a quick test by the benchmark tool[1] and found the patch improve the
file creation/deletion performance problem that I have reported[2].
Before applying this patch:
Create files:
Total files: 50000
Total time: 0.971531
Average time: 0.000019
Delete files:
Total files: 50000
Total time: 1.366761
Average time: 0.000027
After applying this patch:
Create files:
Total files: 50000
Total time: 0.927455
Average time: 0.000019
Delete files:
Total files: 50000
Total time: 1.292280
Average time: 0.000026
[1] http://marc.info/?l=linux-btrfs&m=128212635122920&q=p3
[2] http://marc.info/?l=linux-btrfs&m=128212635122920&w=2
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
restructure try_release_extent_buffer() and write a function to release the
extent buffer. It will be used later.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
ext3 didn't update the ctime of the file when its permission was changed.
Steps to reproduce:
# touch aaa
# stat -c %Z aaa
1275289822
# setfacl -m 'u::x,g::x,o::x' aaa
# stat -c %Z aaa
1275289822 <- unchanged
But, according to the spec of the ctime, ext3 must update it.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Jan Kara <[email protected]>
when we use remap_file_pages() to remap a file, remap_file_pages always return
error. It is because btrfs didn't set VM_CAN_NONLINEAR for vma.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
We cannot use the loop device which has been connected to a file in the btrf
The reproduce steps is following:
# dd if=/dev/zero of=vdev0 bs=1M count=1024
# losetup /dev/loop0 vdev0
# mkfs.btrfs /dev/loop0
...
failed to zero device start -5
The reason is that the btrfs don't implement either ->write_begin or ->write
the VFS API, so we fix it by setting ->write to do_sync_write().
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Before applying this patch, cpuset updates task->mems_allowed and
mempolicy by setting all new bits in the nodemask first, and clearing all
old unallowed bits later. But in the way, the allocator may find that
there is no node to alloc memory.
The reason is that cpuset rebinds the task's mempolicy, it cleans the
nodes which the allocater can alloc pages on, for example:
(mpol: mempolicy)
task1 task1's mpol task2
alloc page 1
alloc on node0? NO 1
1 change mems from 1 to 0
1 rebind task1's mpol
0-1 set new bits
0 clear disallowed bits
alloc on node1? NO 0
...
can't alloc page
goto oom
This patch fixes this problem by expanding the nodes range first(set newly
allowed bits) and shrink it lazily(clear newly disallowed bits). So we
use a variable to tell the write-side task that read-side task is reading
nodemask, and the write-side task clears newly disallowed nodes after
read-side task ends the current memory allocation.
[[email protected]: fix spello]
Signed-off-by: Miao Xie <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Paul Menage <[email protected]>
Cc: Lee Schermerhorn <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Ravikiran Thirumalai <[email protected]>
Cc: KOSAKI Motohiro <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Andi Kleen <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Nick Piggin reported that the allocator may see an empty nodemask when
changing cpuset's mems[1]. It happens only on the kernel that do not do
atomic nodemask_t stores. (MAX_NUMNODES > BITS_PER_LONG)
But I found that there is also a problem on the kernel that can do atomic
nodemask_t stores. The problem is that the allocator can't find a node to
alloc page when changing cpuset's mems though there is a lot of free
memory. The reason is like this:
(mpol: mempolicy)
task1 task1's mpol task2
alloc page 1
alloc on node0? NO 1
1 change mems from 1 to 0
1 rebind task1's mpol
0-1 set new bits
0 clear disallowed bits
alloc on node1? NO 0
...
can't alloc page
goto oom
I can use the attached program reproduce it by the following step:
# mkdir /dev/cpuset
# mount -t cpuset cpuset /dev/cpuset
# mkdir /dev/cpuset/1
# echo `cat /dev/cpuset/cpus` > /dev/cpuset/1/cpus
# echo `cat /dev/cpuset/mems` > /dev/cpuset/1/mems
# echo $$ > /dev/cpuset/1/tasks
# numactl --membind=`cat /dev/cpuset/mems` ./cpuset_mem_hog <nr_tasks> &
<nr_tasks> = max(nr_cpus - 1, 1)
# killall -s SIGUSR1 cpuset_mem_hog
# ./change_mems.sh
several hours later, oom will happen though there is a lot of free memory.
This patchset fixes this problem by expanding the nodes range first(set
newly allowed bits) and shrink it lazily(clear newly disallowed bits). So
we use a variable to tell the write-side task that read-side task is
reading nodemask, and the write-side task clears newly disallowed nodes
after read-side task ends the current memory allocation.
This patch:
In order to fix no node to alloc memory, when we want to update mempolicy
and mems_allowed, we expand the set of nodes first (set all the newly
nodes) and shrink the set of nodes lazily(clean disallowed nodes), But the
mempolicy's rebind functions may breaks the expanding.
So we restructure the mempolicy's rebind functions and split the rebind
work to two steps, just like the update of cpuset's mems: The 1st step:
expand the set of the mempolicy's nodes. The 2nd step: shrink the set of
the mempolicy's nodes. It is used when there is no real lock to protect
the mempolicy in the read-side. Otherwise we can do rebind work at once.
In order to implement it, we define
enum mpol_rebind_step {
MPOL_REBIND_ONCE,
MPOL_REBIND_STEP1,
MPOL_REBIND_STEP2,
MPOL_REBIND_NSTEP,
};
If the mempolicy needn't be updated by two steps, we can pass
MPOL_REBIND_ONCE to the rebind functions. Or we can pass
MPOL_REBIND_STEP1 to do the first step of the rebind work and pass
MPOL_REBIND_STEP2 to do the second step work.
Besides that, it maybe long time between these two step and we have to
release the lock that protects mempolicy and mems_allowed. If we hold the
lock once again, we must check whether the current mempolicy is under the
rebinding (the first step has been done) or not, because the task may
alloc a new mempolicy when we don't hold the lock. So we defined the
following flag to identify it:
#define MPOL_F_REBINDING (1 << 2)
The new functions will be used in the next patch.
Signed-off-by: Miao Xie <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Paul Menage <[email protected]>
Cc: Lee Schermerhorn <[email protected]>
Cc: Hugh Dickins <[email protected]>
Cc: Ravikiran Thirumalai <[email protected]>
Cc: KOSAKI Motohiro <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Andi Kleen <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
btrfs_find_create_tree_block() may return NULL, so we must check the returned
value, or we will access a NULL pointer.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
btrfs_mkdir() must jump to the place of ending transaction after
btrfs_find_free_objectid() failed. Or this transaction can't end.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Mounting a bad filesystem caused a BUG_ON(). The following is steps to
reproduce it.
# mkfs.btrfs /dev/sda2
# mount /dev/sda2 /mnt
# mkfs.btrfs /dev/sda1 /dev/sda2
(the program says that /dev/sda2 was mounted, and then exits. )
# umount /mnt
# mount /dev/sda1 /mnt
At the third step, mkfs.btrfs exited in the way of make filesystem. So the
initialization of the filesystem didn't finish. So the filesystem was bad, and
it caused BUG_ON() when mounting it. But BUG_ON() should be called by the wrong
code, not user's operation, so I think it is a bug of btrfs.
This patch fixes it.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
This patch removes tree_search() in extent_map.c because it is not called by
anything.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Chris Mason <[email protected]>
Fix allocating page cache/slab object on the unallowed node when memory
spread is set by updating tasks' mems_allowed after its cpuset's mems is
changed.
In order to update tasks' mems_allowed in time, we must modify the code of
memory policy. Because the memory policy is applied in the process's
context originally. After applying this patch, one task directly
manipulates anothers mems_allowed, and we use alloc_lock in the
task_struct to protect mems_allowed and memory policy of the task.
But in the fast path, we didn't use lock to protect them, because adding a
lock may lead to performance regression. But if we don't add a lock,the
task might see no nodes when changing cpuset's mems_allowed to some
non-overlapping set. In order to avoid it, we set all new allowed nodes,
then clear newly disallowed ones.
[[email protected]:
The rework of mpol_new() to extract the adjusting of the node mask to
apply cpuset and mpol flags "context" breaks set_mempolicy() and mbind()
with MPOL_PREFERRED and a NULL nodemask--i.e., explicit local
allocation. Fix this by adding the check for MPOL_PREFERRED and empty
node mask to mpol_new_mpolicy().
Remove the now unneeded 'nodes = NULL' from mpol_new().
Note that mpol_new_mempolicy() is always called with a non-NULL
'nodes' parameter now that it has been removed from mpol_new().
Therefore, we don't need to test nodes for NULL before testing it for
'empty'. However, just to be extra paranoid, add a VM_BUG_ON() to
verify this assumption.]
[[email protected]:
I don't think the function name 'mpol_new_mempolicy' is descriptive
enough to differentiate it from mpol_new().
This function applies cpuset set context, usually constraining nodes
to those allowed by the cpuset. However, when the 'RELATIVE_NODES flag
is set, it also translates the nodes. So I settled on
'mpol_set_nodemask()', because the comment block for mpol_new() mentions
that we need to call this function to "set nodes".
Some additional minor line length, whitespace and typo cleanup.]
Signed-off-by: Miao Xie <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Paul Menage <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Yasunori Goto <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: David Rientjes <[email protected]>
Signed-off-by: Lee Schermerhorn <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The kernel still allocates the page caches on old node after modifying its
cpuset's mems when 'memory_spread_page' was set, or it didn't spread the
page cache evenly over all the nodes that faulting task is allowed to usr
after memory_spread_page was set. it is caused by the old mem_allowed and
flags of the task, the current kernel doesn't updates them unless some
function invokes cpuset_update_task_memory_state(), it is too late
sometimes.We must update the mem_allowed and the flags of the tasks in
time.
Slab has the same problem.
The following patches fix this bug by updating tasks' mem_allowed and
spread flag after its cpuset's mems or spread flag is changed.
This patch:
Extract a function from cpuset_update_task_memory_state(). It will be
used later for update tasks' page/slab spread flags after its cpuset's
flag is set
Signed-off-by: Miao Xie <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Christoph Lameter <[email protected]>
Cc: Paul Menage <[email protected]>
Cc: Nick Piggin <[email protected]>
Cc: Yasunori Goto <[email protected]>
Cc: Pekka Enberg <[email protected]>
Cc: David Rientjes <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Lockdep reported some possible circular locking info when we tested cpuset on
NUMA/fake NUMA box.
=======================================================
[ INFO: possible circular locking dependency detected ]
2.6.29-rc1-00224-ga652504 #111
-------------------------------------------------------
bash/2968 is trying to acquire lock:
(events){--..}, at: [<ffffffff8024c8cd>] flush_work+0x24/0xd8
but task is already holding lock:
(cgroup_mutex){--..}, at: [<ffffffff8026ad1e>] cgroup_lock_live_group+0x12/0x29
which lock already depends on the new lock.
......
-------------------------------------------------------
Steps to reproduce:
# mkdir /dev/cpuset
# mount -t cpuset xxx /dev/cpuset
# mkdir /dev/cpuset/0
# echo 0 > /dev/cpuset/0/cpus
# echo 0 > /dev/cpuset/0/mems
# echo 1 > /dev/cpuset/0/memory_migrate
# cat /dev/zero > /dev/null &
# echo $! > /dev/cpuset/0/tasks
This is because async_rebuild_sched_domains has the following lock sequence:
run_workqueue(async_rebuild_sched_domains)
-> do_rebuild_sched_domains -> cgroup_lock
But, attaching tasks when memory_migrate is set has following:
cgroup_lock_live_group(cgroup_tasks_write)
-> do_migrate_pages -> flush_work
This patch fixes it by using a separate workqueue thread.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Lai Jiangshan <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
I found a bug on my dual-cpu box. I created a sub cpuset in top cpuset
and assign 1 to its cpus. And then we attach some tasks into this sub
cpuset. After this, we offline CPU1. Now, the tasks in this new cpuset
are moved into top cpuset automatically because there is no cpu in sub
cpuset. Then we online CPU1, we find all the tasks which doesn't belong
to top cpuset originally just run on CPU0.
We fix this bug by setting task's cpu_allowed to cpu_possible_map when
attaching it into top cpuset. This method needn't modify the current
behavior of cpusets on CPU hotplug, and all of tasks in top cpuset use
cpu_possible_map to initialize their cpu_allowed.
Signed-off-by: Miao Xie <[email protected]>
Cc: Paul Menage <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
After adding a node into the machine, top cpuset's mems isn't updated.
By reviewing the code, we found that the update function
cpuset_track_online_nodes()
was invoked after node_states[N_ONLINE] changes. It is wrong because
N_ONLINE just means node has pgdat, and if node has/added memory, we use
N_HIGH_MEMORY. So, We should invoke the update function after
node_states[N_HIGH_MEMORY] changes, just like its commit says.
This patch fixes it. And we use notifier of memory hotplug instead of
direct calling of cpuset_track_online_nodes().
Signed-off-by: Miao Xie <[email protected]>
Acked-by: Yasunori Goto <[email protected]>
Cc: David Rientjes <[email protected]>
Cc: Paul Menage <[email protected]
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
When trying to build 2.6.28-rc1 on ia64, make aborts with:
CC drivers/acpi/processor_perflib.o
drivers/acpi/processor_perflib.c:41:28: error: asm/cpufeature.h: No such file or directory
drivers/acpi/processor_perflib.c: In function ‘acpi_processor_get_performance_info’:
drivers/acpi/processor_perflib.c:364: error: implicit declaration of function ‘boot_cpu_has’
drivers/acpi/processor_perflib.c:364: error: ‘X86_FEATURE_EST’ undeclared (first use in this function)
drivers/acpi/processor_perflib.c:364: error: (Each undeclared identifier is reported only once
drivers/acpi/processor_perflib.c:364: error: for each function it appears in.)
make[2]: *** [drivers/acpi/processor_perflib.o] Error 1
make[1]: *** [drivers/acpi] Error 2
make: *** [drivers] Error 2
this patch fix it.
Signed-off-by: Miao Xie <[email protected]>
Acked-by: Thomas Renninger <[email protected]>
Signed-off-by: Len Brown <[email protected]>
If NR_CPUS isn't a multiple of 32, we get a truncated string of sched
domains by catting /proc/schedstat. This is caused by the wrong mask_len.
This patch fixes it.
Signed-off-by: Miao Xie <[email protected]>
Cc: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
On the tickless system(CONFIG_NO_HZ=y and CONFIG_HIGH_RES_TIMERS=n), after
I made an offlined cpu online, I found this cpu's event handler was
tick_handle_periodic, not tick_nohz_handler.
After debuging, I found this bug was caused by the wrong tick mode. the
tick mode is not changed to NOHZ_MODE_INACTIVE when the cpu is offline.
This patch fixes this bug.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
When an error occured, datagram_send_ctl() should exit immediately rather than
continue to run the for loop. Otherwise, the variable err might be changed and
the error might be hidden.
Fix this bug by using "goto" instead of "break".
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
The bug is that a task may run on the cpu/node which is not in its
cpuset.cpus/ cpuset.mems.
It can be reproduced by the following commands:
-----------------------------------
# mkdir /dev/cpuset
# mount -t cpuset xxx /dev/cpuset
# mkdir /dev/cpuset/0
# echo 0-1 > /dev/cpuset/0/cpus
# echo 0 > /dev/cpuset/0/mems
# echo $$ > /dev/cpuset/0/tasks
# echo 0 > /sys/devices/system/cpu/cpu1/online
# echo 1 > /sys/devices/system/cpu/cpu1/online
-----------------------------------
There is only CPU0 in cpuset.cpus, but the task in this cpuset runs on
both CPU0 and CPU1.
It is because the task's cpu_allowed didn't get updated after we did CPU
offline/online manipulation. Similar for mem_allowed.
This patch fixes this bug expect for root cpuset. Because there is a
problem about root cpuset, in that whether it is necessary to update all
the tasks in root cpuset or not after cpu/node offline/online.
If updating, some kernel threads which is bound into a specified cpu will
be unbound.
If not updating, there is a bug in root cpuset. This bug is also caused
by offline/online manipulation. For example, there is a dual-cpu machine.
we create a sub cpuset in root cpuset and assign 1 to its cpus. And then
we attach some tasks into this sub cpuset. After this, we offline CPU1.
Now, the tasks in this new cpuset are moved into root cpuset automatically
because there is no cpu in sub cpuset. Then we online CPU1, we find all
the tasks which doesn't belong to root cpuset originally just run on CPU0.
Maybe we need to add a flag in the task_struct to mark which task can't be
unbound?
Signed-off-by: Miao Xie <[email protected]>
Acked-by: Paul Jackson <[email protected]>
Cc: Li Zefan <[email protected]>
Cc: Paul Jackson <[email protected]>
Cc: Paul Menage <[email protected]>
Cc: David Rientjes <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
When I echoed 0 into the "cpu.shares" file, a Div0 error occured.
We found it is caused by the following calling.
sched_group_set_shares(tg, shares)
set_se_shares(tg->se[i], shares/nr_cpu_ids)
__set_se_shares(se, shares)
div64_64((1ULL<<32), shares)
When the echoed value was less than the number of processores, the result of the
sentence "shares/nr_cpu_ids" was 0, and then the system called div64() to divide
the result, the Div0 error occured.
It is unnecessary that the shares value is divided by nr_cpu_ids, I think.
Because in the function __update_group_shares_cpu() and init_tg_cfs_entry(),
the shares value isn't divided by nr_cpu_ids when setting shares of the sched
entity.
This patch fixes this bug. And echoing ULONG_MAX value into cpu.shares also
causes Div0 error, so we set a macro MAX_SHARES to limit the max value of
shares.
Signed-off-by: Miao Xie <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Function sys_sched_rr_get_interval returns wrong time slice value for
SCHED_FIFO tasks. The time slice for SCHED_FIFO tasks should be 0.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
I found that there is a buffer overflow problem in the following code.
Version: 2.6.24-rc2,
File: kernel/time/clocksource.c:417-432
--------------------------------------------------------------------
static ssize_t
sysfs_show_available_clocksources(struct sys_device *dev, char *buf)
{
struct clocksource *src;
char *curr = buf;
spin_lock_irq(&clocksource_lock);
list_for_each_entry(src, &clocksource_list, list) {
curr += sprintf(curr, "%s ", src->name);
}
spin_unlock_irq(&clocksource_lock);
curr += sprintf(curr, "\n");
return curr - buf;
}
-----------------------------------------------------------------------
sysfs_show_current_clocksources() also has the same problem though in practice
the size of current clocksource's name won't exceed PAGE_SIZE.
I fix the bug by using snprintf according to the specification of the kernel
(Version:2.6.24-rc2,File:Documentation/filesystems/sysfs.txt)
Fix sysfs_show_available_clocksources() and sysfs_show_current_clocksources()
buffer overflow problem with snprintf().
Signed-off-by: Miao Xie <[email protected]>
Cc: WANG Cong <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: john stultz <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
I found that there is a off-by-one problem in the following code.
Version: 2.6.24-rc2
File: fs/sysfs/file.c:118-122
Function: fill_read_buffer
--------------------------------------------------------------------
count = ops->show(kobj, attr_sd->s_attr.attr, buffer->page);
sysfs_put_active_two(attr_sd);
BUG_ON(count > (ssize_t)PAGE_SIZE);
--------------------------------------------------------------------
Because according to the specification of the sysfs and the implement of
the show methods, the show methods return the number of bytes which would
be generated for the given input, excluding the trailing null.So if the
return value of the show methods equals PAGE_SIZE - 1, the buffer is full
in fact. And if the return value equals PAGE_SIZE, the resulting string
was already truncated,or buffer overflow occurred.
This patch fixes an off-by-one error in fill_read_buffer.
Signed-off-by: Miao Xie <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Acked-by: Tejun Heo <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>
I find a function(clockevents_unregister_notifier) which is not called by
anything in tree.
Signed-off-by: Miao Xie <[email protected]>
Acked-by: Thomas Gleixner <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Miao Xie