Pull nowait read support from Al Viro:
"Support IOCB_NOWAIT for buffered reads and block devices"
* 'work.read_write' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
block_dev: support RFW_NOWAIT on block device nodes
fs: support RWF_NOWAIT for buffered reads
fs: support IOCB_NOWAIT in generic_file_buffered_read
fs: pass iocb to do_generic_file_read
Pull mount flag updates from Al Viro:
"Another chunk of fmount preparations from dhowells; only trivial
conflicts for that part. It separates MS_... bits (very grotty
mount(2) ABI) from the struct super_block ->s_flags (kernel-internal,
only a small subset of MS_... stuff).
This does *not* convert the filesystems to new constants; only the
infrastructure is done here. The next step in that series is where the
conflicts would be; that's the conversion of filesystems. It's purely
mechanical and it's better done after the merge, so if you could run
something like
list=$(for i in MS_RDONLY MS_NOSUID MS_NODEV MS_NOEXEC MS_SYNCHRONOUS MS_MANDLOCK MS_DIRSYNC MS_NOATIME MS_NODIRATIME MS_SILENT MS_POSIXACL MS_KERNMOUNT MS_I_VERSION MS_LAZYTIME; do git grep -l $i fs drivers/staging/lustre drivers/mtd ipc mm include/linux; done|sort|uniq|grep -v '^fs/namespace.c$')
sed -i -e 's/\<MS_RDONLY\>/SB_RDONLY/g' \
-e 's/\<MS_NOSUID\>/SB_NOSUID/g' \
-e 's/\<MS_NODEV\>/SB_NODEV/g' \
-e 's/\<MS_NOEXEC\>/SB_NOEXEC/g' \
-e 's/\<MS_SYNCHRONOUS\>/SB_SYNCHRONOUS/g' \
-e 's/\<MS_MANDLOCK\>/SB_MANDLOCK/g' \
-e 's/\<MS_DIRSYNC\>/SB_DIRSYNC/g' \
-e 's/\<MS_NOATIME\>/SB_NOATIME/g' \
-e 's/\<MS_NODIRATIME\>/SB_NODIRATIME/g' \
-e 's/\<MS_SILENT\>/SB_SILENT/g' \
-e 's/\<MS_POSIXACL\>/SB_POSIXACL/g' \
-e 's/\<MS_KERNMOUNT\>/SB_KERNMOUNT/g' \
-e 's/\<MS_I_VERSION\>/SB_I_VERSION/g' \
-e 's/\<MS_LAZYTIME\>/SB_LAZYTIME/g' \
$list
and commit it with something along the lines of 'convert filesystems
away from use of MS_... constants' as commit message, it would save a
quite a bit of headache next cycle"
* 'work.mount' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
VFS: Differentiate mount flags (MS_*) from internal superblock flags
VFS: Convert sb->s_flags & MS_RDONLY to sb_rdonly(sb)
vfs: Add sb_rdonly(sb) to query the MS_RDONLY flag on s_flags
* Media error handling support in the Block Translation Table (BTT)
driver is reworked to address sleeping-while-atomic locking and
memory-allocation-context conflicts.
* The dax_device lookup overhead for xfs and ext4 is moved out of the
iomap hot-path to a mount-time lookup.
* A new 'ecc_unit_size' sysfs attribute is added to advertise the
read-modify-write boundary property of a persistent memory range.
* Preparatory fix-ups for arm and powerpc pmem support are included
along with other miscellaneous fixes.
-----BEGIN PGP SIGNATURE-----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=5iGY
-----END PGP SIGNATURE-----
Merge tag 'libnvdimm-for-4.14' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm
Pull libnvdimm from Dan Williams:
"A rework of media error handling in the BTT driver and other updates.
It has appeared in a few -next releases and collected some late-
breaking build-error and warning fixups as a result.
Summary:
- Media error handling support in the Block Translation Table (BTT)
driver is reworked to address sleeping-while-atomic locking and
memory-allocation-context conflicts.
- The dax_device lookup overhead for xfs and ext4 is moved out of the
iomap hot-path to a mount-time lookup.
- A new 'ecc_unit_size' sysfs attribute is added to advertise the
read-modify-write boundary property of a persistent memory range.
- Preparatory fix-ups for arm and powerpc pmem support are included
along with other miscellaneous fixes"
* tag 'libnvdimm-for-4.14' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (26 commits)
libnvdimm, btt: fix format string warnings
libnvdimm, btt: clean up warning and error messages
ext4: fix null pointer dereference on sbi
libnvdimm, nfit: move the check on nd_reserved2 to the endpoint
dax: fix FS_DAX=n BLOCK=y compilation
libnvdimm: fix integer overflow static analysis warning
libnvdimm, nd_blk: remove mmio_flush_range()
libnvdimm, btt: rework error clearing
libnvdimm: fix potential deadlock while clearing errors
libnvdimm, btt: cache sector_size in arena_info
libnvdimm, btt: ensure that flags were also unchanged during a map_read
libnvdimm, btt: refactor map entry operations with macros
libnvdimm, btt: fix a missed NVDIMM_IO_ATOMIC case in the write path
libnvdimm, nfit: export an 'ecc_unit_size' sysfs attribute
ext4: perform dax_device lookup at mount
ext2: perform dax_device lookup at mount
xfs: perform dax_device lookup at mount
dax: introduce a fs_dax_get_by_bdev() helper
libnvdimm, btt: check memory allocation failure
libnvdimm, label: fix index block size calculation
...
Pull quota scaling updates from Jan Kara:
"This contains changes to make the quota subsystem more scalable.
Reportedly it improves number of files created per second on ext4
filesystem on fast storage by about a factor of 2x"
* 'quota_scaling' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs: (28 commits)
quota: Add lock annotations to struct members
quota: Reduce contention on dq_data_lock
fs: Provide __inode_get_bytes()
quota: Inline dquot_[re]claim_reserved_space() into callsite
quota: Inline inode_{incr,decr}_space() into callsites
quota: Inline functions into their callsites
ext4: Disable dirty list tracking of dquots when journalling quotas
quota: Allow disabling tracking of dirty dquots in a list
quota: Remove dq_wait_unused from dquot
quota: Move locking into clear_dquot_dirty()
quota: Do not dirty bad dquots
quota: Fix possible corruption of dqi_flags
quota: Propagate ->quota_read errors from v2_read_file_info()
quota: Fix error codes in v2_read_file_info()
quota: Push dqio_sem down to ->read_file_info()
quota: Push dqio_sem down to ->write_file_info()
quota: Push dqio_sem down to ->get_next_id()
quota: Push dqio_sem down to ->release_dqblk()
quota: Remove locking for writing to the old quota format
quota: Do not acquire dqio_sem for dquot overwrites in v2 format
...
Pull block layer updates from Jens Axboe:
"This is the first pull request for 4.14, containing most of the code
changes. It's a quiet series this round, which I think we needed after
the churn of the last few series. This contains:
- Fix for a registration race in loop, from Anton Volkov.
- Overflow complaint fix from Arnd for DAC960.
- Series of drbd changes from the usual suspects.
- Conversion of the stec/skd driver to blk-mq. From Bart.
- A few BFQ improvements/fixes from Paolo.
- CFQ improvement from Ritesh, allowing idling for group idle.
- A few fixes found by Dan's smatch, courtesy of Dan.
- A warning fixup for a race between changing the IO scheduler and
device remova. From David Jeffery.
- A few nbd fixes from Josef.
- Support for cgroup info in blktrace, from Shaohua.
- Also from Shaohua, new features in the null_blk driver to allow it
to actually hold data, among other things.
- Various corner cases and error handling fixes from Weiping Zhang.
- Improvements to the IO stats tracking for blk-mq from me. Can
drastically improve performance for fast devices and/or big
machines.
- Series from Christoph removing bi_bdev as being needed for IO
submission, in preparation for nvme multipathing code.
- Series from Bart, including various cleanups and fixes for switch
fall through case complaints"
* 'for-4.14/block' of git://git.kernel.dk/linux-block: (162 commits)
kernfs: checking for IS_ERR() instead of NULL
drbd: remove BIOSET_NEED_RESCUER flag from drbd_{md_,}io_bio_set
drbd: Fix allyesconfig build, fix recent commit
drbd: switch from kmalloc() to kmalloc_array()
drbd: abort drbd_start_resync if there is no connection
drbd: move global variables to drbd namespace and make some static
drbd: rename "usermode_helper" to "drbd_usermode_helper"
drbd: fix race between handshake and admin disconnect/down
drbd: fix potential deadlock when trying to detach during handshake
drbd: A single dot should be put into a sequence.
drbd: fix rmmod cleanup, remove _all_ debugfs entries
drbd: Use setup_timer() instead of init_timer() to simplify the code.
drbd: fix potential get_ldev/put_ldev refcount imbalance during attach
drbd: new disk-option disable-write-same
drbd: Fix resource role for newly created resources in events2
drbd: mark symbols static where possible
drbd: Send P_NEG_ACK upon write error in protocol != C
drbd: add explicit plugging when submitting batches
drbd: change list_for_each_safe to while(list_first_entry_or_null)
drbd: introduce drbd_recv_header_maybe_unplug
...
Merge updates from Andrew Morton:
- various misc bits
- DAX updates
- OCFS2
- most of MM
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (119 commits)
mm,fork: introduce MADV_WIPEONFORK
x86,mpx: make mpx depend on x86-64 to free up VMA flag
mm: add /proc/pid/smaps_rollup
mm: hugetlb: clear target sub-page last when clearing huge page
mm: oom: let oom_reap_task and exit_mmap run concurrently
swap: choose swap device according to numa node
mm: replace TIF_MEMDIE checks by tsk_is_oom_victim
mm, oom: do not rely on TIF_MEMDIE for memory reserves access
z3fold: use per-cpu unbuddied lists
mm, swap: don't use VMA based swap readahead if HDD is used as swap
mm, swap: add sysfs interface for VMA based swap readahead
mm, swap: VMA based swap readahead
mm, swap: fix swap readahead marking
mm, swap: add swap readahead hit statistics
mm/vmalloc.c: don't reinvent the wheel but use existing llist API
mm/vmstat.c: fix wrong comment
selftests/memfd: add memfd_create hugetlbfs selftest
mm/shmem: add hugetlbfs support to memfd_create()
mm, devm_memremap_pages: use multi-order radix for ZONE_DEVICE lookups
mm/vmalloc.c: halve the number of comparisons performed in pcpu_get_vm_areas()
...
All users of pagevec_lookup() and pagevec_lookup_range() now pass
PAGEVEC_SIZE as a desired number of pages.
Just drop the argument.
Link: http://lkml.kernel.org/r/20170726114704.7626-11-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Both occurences of pagevec_lookup() actually want only pages from a
given range. Use pagevec_lookup_range() for the lookup.
Link: http://lkml.kernel.org/r/20170726114704.7626-7-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use pagevec_lookup_range() in ext4_find_unwritten_pgoff() since we are
interested only in pages in the given range. Simplify the logic as a
result of not getting pages out of range and index getting automatically
advanced.
Link: http://lkml.kernel.org/r/20170726114704.7626-6-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make pagevec_lookup() (and underlying find_get_pages()) update index to
the next page where iteration should continue. Most callers want this
and also pagevec_lookup_tag() already does this.
Link: http://lkml.kernel.org/r/20170726114704.7626-3-jack@suse.cz
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When servicing mmap() reads from file holes the current DAX code
allocates a page cache page of all zeroes and places the struct page
pointer in the mapping->page_tree radix tree.
This has three major drawbacks:
1) It consumes memory unnecessarily. For every 4k page that is read via
a DAX mmap() over a hole, we allocate a new page cache page. This
means that if you read 1GiB worth of pages, you end up using 1GiB of
zeroed memory. This is easily visible by looking at the overall
memory consumption of the system or by looking at /proc/[pid]/smaps:
7f62e72b3000-7f63272b3000 rw-s 00000000 103:00 12 /root/dax/data
Size: 1048576 kB
Rss: 1048576 kB
Pss: 1048576 kB
Shared_Clean: 0 kB
Shared_Dirty: 0 kB
Private_Clean: 1048576 kB
Private_Dirty: 0 kB
Referenced: 1048576 kB
Anonymous: 0 kB
LazyFree: 0 kB
AnonHugePages: 0 kB
ShmemPmdMapped: 0 kB
Shared_Hugetlb: 0 kB
Private_Hugetlb: 0 kB
Swap: 0 kB
SwapPss: 0 kB
KernelPageSize: 4 kB
MMUPageSize: 4 kB
Locked: 0 kB
2) It is slower than using a common zero page because each page fault
has more work to do. Instead of just inserting a common zero page we
have to allocate a page cache page, zero it, and then insert it. Here
are the average latencies of dax_load_hole() as measured by ftrace on
a random test box:
Old method, using zeroed page cache pages: 3.4 us
New method, using the common 4k zero page: 0.8 us
This was the average latency over 1 GiB of sequential reads done by
this simple fio script:
[global]
size=1G
filename=/root/dax/data
fallocate=none
[io]
rw=read
ioengine=mmap
3) The fact that we had to check for both DAX exceptional entries and
for page cache pages in the radix tree made the DAX code more
complex.
Solve these issues by following the lead of the DAX PMD code and using a
common 4k zero page instead. As with the PMD code we will now insert a
DAX exceptional entry into the radix tree instead of a struct page
pointer which allows us to remove all the special casing in the DAX
code.
Note that we do still pretty aggressively check for regular pages in the
DAX radix tree, especially where we take action based on the bits set in
the page. If we ever find a regular page in our radix tree now that
most likely means that someone besides DAX is inserting pages (which has
happened lots of times in the past), and we want to find that out early
and fail loudly.
This solution also removes the extra memory consumption. Here is that
same /proc/[pid]/smaps after 1GiB of reading from a hole with the new
code:
7f2054a74000-7f2094a74000 rw-s 00000000 103:00 12 /root/dax/data
Size: 1048576 kB
Rss: 0 kB
Pss: 0 kB
Shared_Clean: 0 kB
Shared_Dirty: 0 kB
Private_Clean: 0 kB
Private_Dirty: 0 kB
Referenced: 0 kB
Anonymous: 0 kB
LazyFree: 0 kB
AnonHugePages: 0 kB
ShmemPmdMapped: 0 kB
Shared_Hugetlb: 0 kB
Private_Hugetlb: 0 kB
Swap: 0 kB
SwapPss: 0 kB
KernelPageSize: 4 kB
MMUPageSize: 4 kB
Locked: 0 kB
Overall system memory consumption is similarly improved.
Another major change is that we remove dax_pfn_mkwrite() from our fault
flow, and instead rely on the page fault itself to make the PTE dirty
and writeable. The following description from the patch adding the
vm_insert_mixed_mkwrite() call explains this a little more:
"To be able to use the common 4k zero page in DAX we need to have our
PTE fault path look more like our PMD fault path where a PTE entry
can be marked as dirty and writeable as it is first inserted rather
than waiting for a follow-up dax_pfn_mkwrite() =>
finish_mkwrite_fault() call.
Right now we can rely on having a dax_pfn_mkwrite() call because we
can distinguish between these two cases in do_wp_page():
case 1: 4k zero page => writable DAX storage
case 2: read-only DAX storage => writeable DAX storage
This distinction is made by via vm_normal_page(). vm_normal_page()
returns false for the common 4k zero page, though, just as it does
for DAX ptes. Instead of special casing the DAX + 4k zero page case
we will simplify our DAX PTE page fault sequence so that it matches
our DAX PMD sequence, and get rid of the dax_pfn_mkwrite() helper.
We will instead use dax_iomap_fault() to handle write-protection
faults.
This means that insert_pfn() needs to follow the lead of
insert_pfn_pmd() and allow us to pass in a 'mkwrite' flag. If
'mkwrite' is set insert_pfn() will do the work that was previously
done by wp_page_reuse() as part of the dax_pfn_mkwrite() call path"
Link: http://lkml.kernel.org/r/20170724170616.25810-4-ross.zwisler@linux.intel.com
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In the case of a kzalloc failure when allocating sbi we end up
with a null pointer dereference on sbi when assigning sbi->s_daxdev.
Fix this by moving the assignment of sbi->s_daxdev to after the
null pointer check of sbi.
Detected by CoverityScan CID#1455379 ("Dereference before null check")
Fixes: 5e405595e5 ("ext4: perform dax_device lookup at mount")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This is based on the old idea and code from Milosz Tanski. With the aio
nowait code it becomes mostly trivial now. Buffered writes continue to
return -EOPNOTSUPP if RWF_NOWAIT is passed.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The ->iomap_begin() operation is a hot path, so cache the
fs_dax_get_by_host() result at mount time to avoid the incurring the
hash lookup overhead on a per-i/o basis.
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Reviewed-by: Jan Kara <jack@suse.cz>
Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Avoid a 32-bit time overflow in recently_deleted() since i_dtime
(inode deletion time) is stored only as a 32-bit value on disk.
Since i_dtime isn't used for much beyond a boolean value in e2fsck
and is otherwise only used in this function in the kernel, there is
no benefit to use more space in the inode for this field on disk.
Instead, compare only the relative deletion time with the low
32 bits of the time using the newly-added time_before32() helper,
which is similar to time_before() and time_after() for jiffies.
Increase RECENTCY_DIRTY to 300s based on Ted's comments about
usage experience at Google.
Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
If an ext4 filesystem is mounted with both the DAX and read-only
options, executables on that filesystem will fail to start (claiming
'Segmentation fault') due to the fault handler returning
VM_FAULT_SIGBUS.
This is due to the DAX fault handler (see ext4_dax_huge_fault)
attempting to write to the journal when FAULT_FLAG_WRITE is set. This is
the wrong behavior for write faults which will lead to a COW page; in
particular, this fails for readonly mounts.
This change avoids journal writes for faults that are expected to COW.
It might be the case that this could be better handled in
ext4_iomap_begin / ext4_iomap_end (called via iomap_ops inside
dax_iomap_fault). These is some overlap already (e.g. grabbing journal
handles).
Signed-off-by: Randy Dodgen <dodgen@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Quota does not get enabled for read-only mounts if filesystem
has quota feature, so that quotas cannot updated during orphan
cleanup, which will lead to quota inconsistency.
This patch turn on quotas during orphan cleanup for this case,
make sure quotas can be updated correctly.
Reported-by: Jan Kara <jack@suse.cz>
Signed-off-by: zhangyi (F) <yi.zhang@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: stable@vger.kernel.org # 3.18+
Current ext4 quota should always "usage enabled" if the
quota feautre is enabled. But in ext4_orphan_cleanup(), it
turn quotas off directly (used for the older journaled
quota), so we cannot turn it on again via "quotaon" unless
umount and remount ext4.
Simple reproduce:
mkfs.ext4 -O project,quota /dev/vdb1
mount -o prjquota /dev/vdb1 /mnt
chattr -p 123 /mnt
chattr +P /mnt
touch /mnt/aa /mnt/bb
exec 100<>/mnt/aa
rm -f /mnt/aa
sync
echo c > /proc/sysrq-trigger
#reboot and mount
mount -o prjquota /dev/vdb1 /mnt
#query status
quotaon -Ppv /dev/vdb1
#output
quotaon: Cannot find mountpoint for device /dev/vdb1
quotaon: No correct mountpoint specified.
This patch add check for journaled quotas to avoid incorrect
quotaoff when ext4 has quota feautre.
Signed-off-by: zhangyi (F) <yi.zhang@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: stable@vger.kernel.org # 3.18
On transformation of str to hash, computed value is initialised before
first byte modulo 4. But it is already initialised before entering loop
and after processing last byte modulo 4. So the corresponding test and
initialisation could be removed.
Signed-off-by: Damien Guibouret <damien.guibouret@partition-saving.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Original Lustre ea_inode feature did not have ref counts on xattr inodes
because there was always one parent that referenced it. New
implementation expects ref count to be initialized which is not true for
Lustre case. Handle this by detecting Lustre created xattr inode and set
its ref count to 1.
The quota handling of xattr inodes have also changed with deduplication
support. New implementation manually manages quotas to support sharing
across multiple users. A consequence is that, a referencing inode
incorporates the blocks of xattr inode into its own i_block field.
We need to know how a xattr inode was created so that we can reverse the
block charges during reference removal. This is handled by introducing a
EXT4_STATE_LUSTRE_EA_INODE flag. The flag is set on a xattr inode if
inode appears to have been created by Lustre. During xattr inode reference
removal, the manual quota uncharge is skipped if the flag is set.
Signed-off-by: Tahsin Erdogan <tahsin@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Changing behavior based on the version code is a timebomb waiting to
happen, and not easily bisectable. Drop it and leave any removal
to explicit developer action. (And I don't think file system
should _ever_ remove backwards compatibility that has no explicit
flag, but I'll leave that to the ext4 folks).
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Eric Biggers <ebiggers@google.com>
Replace the specification of data structures by pointer dereferences
as the parameter for the operator "sizeof" to make the corresponding size
determination a bit safer according to the Linux coding style convention.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
In the ext4 implementations of SEEK_HOLE and SEEK_DATA, make sure we
return -ENXIO for negative offsets instead of banging around inside
the extent code and returning -EFSCORRUPTED.
Reported-by: Mateusz S <muttdini@gmail.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org # 4.6
While running number of creating file threads concurrently,
we found heavy lock contention on group spinlock:
FUNC TOTAL_TIME(us) COUNT AVG(us)
ext4_create 1707443399 1440000 1185.72
_raw_spin_lock 1317641501 180899929 7.28
jbd2__journal_start 287821030 1453950 197.96
jbd2_journal_get_write_access 33441470 73077185 0.46
ext4_add_nondir 29435963 1440000 20.44
ext4_add_entry 26015166 1440049 18.07
ext4_dx_add_entry 25729337 1432814 17.96
ext4_mark_inode_dirty 12302433 5774407 2.13
most of cpu time blames to _raw_spin_lock, here is some testing
numbers with/without patch.
Test environment:
Server : SuperMicro Sever (2 x E5-2690 v3@2.60GHz, 128GB 2133MHz
DDR4 Memory, 8GbFC)
Storage : 2 x RAID1 (DDN SFA7700X, 4 x Toshiba PX02SMU020 200GB
Read Intensive SSD)
format command:
mkfs.ext4 -J size=4096
test command:
mpirun -np 48 mdtest -n 30000 -d /ext4/mdtest.out -F -C \
-r -i 1 -v -p 10 -u #first run to load inode
mpirun -np 48 mdtest -n 30000 -d /ext4/mdtest.out -F -C \
-r -i 3 -v -p 10 -u
Kernel version: 4.13.0-rc3
Test 1,440,000 files with 48 directories by 48 processes:
Without patch:
File Creation File removal
79,033 289,569 ops/per second
81,463 285,359
79,875 288,475
With patch:
File Creation File removal
810669 301694
812805 302711
813965 297670
Creation performance is improved more than 10X with large
journal size. The main problem here is we test bitmap
and do some check and journal operations which could be
slept, then we test and set with lock hold, this could
be racy, and make 'inode' steal by other process.
However, after first try, we could confirm handle has
been started and inode bitmap journaled too, then
we could find and set bit with lock hold directly, this
will mostly gurateee success with second try.
Tested-by: Shuichi Ihara <sihara@ddn.com>
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
avoid duplicated codes, also we need goto
next group in case we found reserved inode.
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
In recently_deleted() function we want to check whether inode is still
cached in buffer cache. Use sb_find_get_block() for that instead of
sb_getblk() to avoid unnecessary allocation of bdev page and buffer
heads.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
This way we don't need a block_device structure to submit I/O. The
block_device has different life time rules from the gendisk and
request_queue and is usually only available when the block device node
is open. Other callers need to explicitly create one (e.g. the lightnvm
passthrough code, or the new nvme multipathing code).
For the actual I/O path all that we need is the gendisk, which exists
once per block device. But given that the block layer also does
partition remapping we additionally need a partition index, which is
used for said remapping in generic_make_request.
Note that all the block drivers generally want request_queue or
sometimes the gendisk, so this removes a layer of indirection all
over the stack.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
dq_data_lock is currently used to protect all modifications of quota
accounting information, consistency of quota accounting on the inode,
and dquot pointers from inode. As a result contention on the lock can be
pretty heavy.
Reduce the contention on the lock by protecting quota accounting
information by a new dquot->dq_dqb_lock and consistency of quota
accounting with inode usage by inode->i_lock.
This change reduces time to create 500000 files on ext4 on ramdisk by 50
different processes in separate directories by 6% when user quota is
turned on. When those 50 processes belong to 50 different users, the
improvement is about 9%.
Signed-off-by: Jan Kara <jack@suse.cz>
When journalling quotas, we writeback all dquots immediately after
changing them as part of current transation. Thus there's no need to
write anything in dquot_writeback_dquots() and so we can avoid updating
list of dirty dquots to reduce dq_list_lock contention.
This change reduces time to create 500000 files on ext4 on ramdisk by 50
different processes in separate directories by 15% when user quota is
turned on.
Signed-off-by: Jan Kara <jack@suse.cz>
When updating an extended attribute, if the padded value sizes are the
same, a shortcut is taken to avoid the bulk of the work. This was fine
until the xattr hash update was moved inside ext4_xattr_set_entry().
With that change, the hash update got missed in the shortcut case.
Thanks to ZhangYi (yizhang089@gmail.com) for root causing the problem.
Fixes: daf8328172 ("ext4: eliminate xattr entry e_hash recalculation for removes")
Reported-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Tahsin Erdogan <tahsin@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Arnd Bergmann <arnd@arndb.de>
As Stefan pointed out, I misremembered what clang can do specifically,
and it turns out that the variable-length array at the end of the
structure did not work (a flexible array would have worked here
but not solved the problem):
fs/ext4/mballoc.c:2303:17: error: fields must have a constant size:
'variable length array in structure' extension will never be supported
ext4_grpblk_t counters[blocksize_bits + 2];
This reverts part of my previous patch, using a fixed-size array
again, but keeping the check for the array overflow.
Fixes: 2df2c3402f ("ext4: fix warning about stack corruption")
Reported-by: Stefan Agner <stefan@agner.ch>
Tested-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
This bug was found by a static code checker tool for copy paste
problems.
Signed-off-by: Maninder Singh <maninder1.s@samsung.com>
Signed-off-by: Vaneet Narang <v.narang@samsung.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
On a 32-bit platform, the value of n_blcoks_count may be wrong during
the file system is resized to size larger than 2^32 blocks. This may
caused the superblock being corrupted with zero blocks count.
Fixes: 1c6bd7173d
Signed-off-by: Jerry Lee <jerrylee@qnap.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@vger.kernel.org # 3.7+
When upgrading from old format, try to set project id
to old file first time, it will return EOVERFLOW, but if
that file is dirtied(touch etc), changing project id will
be allowed, this might be confusing for users, we could
try to expand @i_extra_isize here too.
Reported-by: Zhang Yi <yi.zhang@huawei.com>
Signed-off-by: Miao Xie <miaoxie@huawei.com>
Signed-off-by: Wang Shilong <wshilong@ddn.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Clean up some goto statement, make ext4_expand_extra_isize_ea() clearer.
Signed-off-by: Miao Xie <miaoxie@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Wang Shilong <wshilong@ddn.com>
Current ext4_expand_extra_isize just tries to expand extra isize, if
someone is holding xattr lock or some check fails, it will give up.
So rename its name to ext4_try_to_expand_extra_isize.
Besides that, we clean up unnecessary check and move some relative checks
into it.
Signed-off-by: Miao Xie <miaoxie@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Wang Shilong <wshilong@ddn.com>
We should avoid the contention between the i_extra_isize update and
the inline data insertion, so move the xattr trylock in front of
i_extra_isize update.
Signed-off-by: Miao Xie <miaoxie@huawei.com>
Reviewed-by: Wang Shilong <wshilong@ddn.com>
ext4_xattr_inode_read() currently reads each block sequentially while
waiting for io operation to complete before moving on to the next
block. This prevents request merging in block layer.
Add a ext4_bread_batch() function that starts reads for all blocks
then optionally waits for them to complete. A similar logic is used
in ext4_find_entry(), so update that code to use the new function.
Signed-off-by: Tahsin Erdogan <tahsin@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When an xattr block has a single reference, block is updated inplace
and it is reinserted to the cache. Later, a cache lookup is performed
to see whether an existing block has the same contents. This cache
lookup will most of the time return the just inserted entry so
deduplication is not achieved.
Running the following test script will produce two xattr blocks which
can be observed in "File ACL: " line of debugfs output:
mke2fs -b 1024 -I 128 -F -O extent /dev/sdb 1G
mount /dev/sdb /mnt/sdb
touch /mnt/sdb/{x,y}
setfattr -n user.1 -v aaa /mnt/sdb/x
setfattr -n user.2 -v bbb /mnt/sdb/x
setfattr -n user.1 -v aaa /mnt/sdb/y
setfattr -n user.2 -v bbb /mnt/sdb/y
debugfs -R 'stat x' /dev/sdb | cat
debugfs -R 'stat y' /dev/sdb | cat
This patch defers the reinsertion to the cache so that we can locate
other blocks with the same contents.
Signed-off-by: Tahsin Erdogan <tahsin@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>
ext4_alloc_file_blocks() does not use its mode parameter. Remove it.
Signed-off-by: Tahsin Erdogan <tahsin@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
After commit 62d1034f53e3 ("fortify: use WARN instead of BUG for now"),
we get a warning about possible stack overflow from a memcpy that
was not strictly bounded to the size of the local variable:
inlined from 'ext4_mb_seq_groups_show' at fs/ext4/mballoc.c:2322:2:
include/linux/string.h:309:9: error: '__builtin_memcpy': writing between 161 and 1116 bytes into a region of size 160 overflows the destination [-Werror=stringop-overflow=]
We actually had a bug here that would have been found by the warning,
but it was already fixed last year in commit 30a9d7afe7 ("ext4: fix
stack memory corruption with 64k block size").
This replaces the fixed-length structure on the stack with a variable-length
structure, using the correct upper bound that tells the compiler that
everything is really fine here. I also change the loop count to check
for the same upper bound for consistency, but the existing code is
already correct here.
Note that while clang won't allow certain kinds of variable-length arrays
in structures, this particular instance is fine, as the array is at the
end of the structure, and the size is strictly bounded.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The dir_nlink feature has been enabled by default for new ext4
filesystems since e2fsprogs-1.41 in 2008, and was automatically
enabled by the kernel for older ext4 filesystems since the
dir_nlink feature was added with ext4 in kernel 2.6.28+ when
the subdirectory count exceeded EXT4_LINK_MAX-1.
Automatically adding the file system features such as dir_nlink is
generally frowned upon, since it could cause the file system to not be
mountable on older kernel, thus preventing the administrator from
rolling back to an older kernel if necessary.
In this case, the administrator might also want to disable the feature
because glibc's fts_read() function does not correctly optimize
directory traversal for directories that use st_nlinks field of 1 to
indicate that the number of links in the directory are not tracked by
the file system, and could fail to traverse the full directory
hierarchy. Fortunately, in the past ten years very few users have
complained about incomplete file system traversal by glibc's
fts_read().
This commit also changes ext4_inc_count() to allow i_nlinks to reach
the full EXT4_LINK_MAX links on the parent directory (including "."
and "..") before changing i_links_count to be 1.
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=196405
Signed-off-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
I get a static checker warning:
fs/ext4/ext4.h:3091 ext4_set_de_type()
error: buffer overflow 'ext4_type_by_mode' 15 <= 15
It seems unlikely that we would hit this read overflow in real life, but
it's also simple enough to make the array 16 bytes instead of 15.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
ext4_find_unwritten_pgoff() does not properly handle a situation when
starting index is in the middle of a page and blocksize < pagesize. The
following command shows the bug on filesystem with 1k blocksize:
xfs_io -f -c "falloc 0 4k" \
-c "pwrite 1k 1k" \
-c "pwrite 3k 1k" \
-c "seek -a -r 0" foo
In this example, neither lseek(fd, 1024, SEEK_HOLE) nor lseek(fd, 2048,
SEEK_DATA) will return the correct result.
Fix the problem by neglecting buffers in a page before starting offset.
Reported-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jan Kara <jack@suse.cz>
CC: stable@vger.kernel.org # 3.8+
We've changed the discard command handling into parallel manner.
But, in this change, I forgot decreasing the usage count of the bio
which was used to send discard request. I'm sorry about that.
Fixes: a015434480 ("ext4: send parallel discards on commit completions")
Signed-off-by: Daeho Jeong <daeho.jeong@samsung.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
For some odd reason, it forces a byte-by-byte copy of each field. A
plain old swap() on most of these fields would be more efficient. We
do need to retain the memswap of i_data however as that field is an array.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Jan Kara <jack@suse.cz>
For Lustre, if ea_inode fails in hash validation but passes parent
inode and generation checks, it won't be added to the cache as well
as the error "-EFSCORRUPTED" should be cleared, otherwise it will
cause "Structure needs cleaning" when running getfattr command.
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-9723
Cc: stable@vger.kernel.org
Fixes: dec214d00e
Signed-off-by: Emoly Liu <emoly.liu@intel.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>
Reviewed-by: tahsin@google.com
When new directory 'DIR1' is created in a directory 'DIR0' with SGID bit
set, DIR1 is expected to have SGID bit set (and owning group equal to
the owning group of 'DIR0'). However when 'DIR0' also has some default
ACLs that 'DIR1' inherits, setting these ACLs will result in SGID bit on
'DIR1' to get cleared if user is not member of the owning group.
Fix the problem by moving posix_acl_update_mode() out of
__ext4_set_acl() into ext4_set_acl(). That way the function will not be
called when inheriting ACLs which is what we want as it prevents SGID
bit clearing and the mode has been properly set by posix_acl_create()
anyway.
Fixes: 073931017b
CC: stable@vger.kernel.org
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Andreas Gruenbacher <agruenba@redhat.com>
When changing a file's acl mask, __ext4_set_acl() will first set the group
bits of i_mode to the value of the mask, and only then set the actual
extended attribute representing the new acl.
If the second part fails (due to lack of space, for example) and the file
had no acl attribute to begin with, the system will from now on assume
that the mask permission bits are actual group permission bits, potentially
granting access to the wrong users.
Prevent this by only changing the inode mode after the acl has been set.
Signed-off-by: Ernesto A. Fernández <ernesto.mnd.fernandez@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Linus Torvalds