Josef fixed btrfs_page_mkwrite to properly release reserved
extents if there was an error. But if we fail to get a reservation
and we fail to dirty the inode (for ENOSPC reasons), we'll end up
trying to release a reservation we never had.
This makes sure we only release if we were able to reserve.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
If we span a long area in a bitmap we could end up taking a lot of time
searching to the next free area if we're searching from the original
window_start, so advance window_start in order to make sure we don't do any
superficial searching. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
btree_releasepage is a callback and can be passed unknown gfp flags and then
they may end up in kmem_cache_alloc called from alloc_extent_state, slab
allocator will BUG_ON when there is HIGHMEM or DMA32 flag set.
This may happen when btrfs is mounted from a loop device, which masks out
__GFP_IO flag. The check in try_release_extent_state
3399 if ((mask & GFP_NOFS) == GFP_NOFS)
3400 mask = GFP_NOFS;
will not work and passes unfiltered flags further resulting in crash at
mm/slab.c:2963
[<000000000024ae4c>] cache_alloc_refill+0x3b4/0x5c8
[<000000000024c810>] kmem_cache_alloc+0x204/0x294
[<00000000001fd3c2>] mempool_alloc+0x52/0x170
[<000003c000ced0b0>] alloc_extent_state+0x40/0xd4 [btrfs]
[<000003c000cee5ae>] __clear_extent_bit+0x38a/0x4cc [btrfs]
[<000003c000cee78c>] try_release_extent_state+0x9c/0xd4 [btrfs]
[<000003c000cc4c66>] btree_releasepage+0x7e/0xd0 [btrfs]
[<0000000000210d84>] shrink_page_list+0x6a0/0x724
[<0000000000211394>] shrink_inactive_list+0x230/0x578
[<0000000000211bb8>] shrink_list+0x6c/0x120
[<0000000000211e4e>] shrink_zone+0x1e2/0x228
[<0000000000211f24>] shrink_zones+0x90/0x254
[<0000000000213410>] do_try_to_free_pages+0xac/0x420
[<0000000000213ae0>] try_to_free_pages+0x13c/0x1b0
[<0000000000204e6c>] __alloc_pages_nodemask+0x5b4/0x9a8
[<00000000001fb04a>] grab_cache_page_write_begin+0x7e/0xe8
Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
When we did sysbench test for inline files, enospc error happened easily though
there was lots of free disk space which could be allocated for new chunks.
Reproduce steps:
# mkfs.btrfs -b $((2 * 1024 * 1024 * 1024)) <test partition>
# mount <test partition> /mnt
# ulimit -n 102400
# cd /mnt
# sysbench --num-threads=1 --test=fileio --file-num=81920 \
> --file-total-size=80M --file-block-size=1K --file-io-mode=sync \
> --file-test-mode=seqwr prepare
# sysbench --num-threads=1 --test=fileio --file-num=81920 \
> --file-total-size=80M --file-block-size=1K --file-io-mode=sync \
> --file-test-mode=seqwr run
<soon later, BUG_ON() was triggered by enospc error>
The reason of this bug is:
Now, we can reserve space which is larger than the free space in the chunks if
we have enough free disk space which can be used for new chunks. By this way,
the space allocator should allocate a new chunk by force if there is no free
space in the free space cache. But there are two wrong checks which break this
operation.
One is
if (ret == -ENOSPC && num_bytes > min_alloc_size)
in btrfs_reserve_extent(), it is wrong, we should try to allocate a new chunk
even we fail to allocate free space by minimum allocable size.
The other is
if (space_info->force_alloc)
force = space_info->force_alloc;
in do_chunk_alloc(). It makes the allocator ignore CHUNK_ALLOC_FORCE If someone
sets ->force_alloc to CHUNK_ALLOC_LIMITED, and makes the enospc error happen.
Fix these two wrong checks. Especially the second one, we fix it by changing
the value of CHUNK_ALLOC_LIMITED and CHUNK_ALLOC_FORCE, and make
CHUNK_ALLOC_FORCE greater than CHUNK_ALLOC_LIMITED since CHUNK_ALLOC_FORCE has
higher priority. And if the value which is passed in by the caller is greater
than ->force_alloc, use the passed value.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
xfstests 218 complains that btrfs defrags a file partially:
After: 1
Write backwards sync, but contiguous - should defrag to 1 extent
Before: 10
-After: 1
+After: 2
To fix this, we need to set max_to_defrag count properly.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
There have been 4 warnings on 32-bit build, they are herewith fixed.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We specifically set window_start in the cluster struct to indicate where the
cluster starts in a bitmap, but we've been using min_start to indicate where
we're searching from. This is usually the start of the blockgroup, so
essentially means we're constantly searching from the start of any bitmap we
find, which completely negates all the trouble we go to in order to setup a
cluster. So start using window_start to make sure we actually use the area we
found. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
A user has encountered a NULL pointer kernel oops in btrfs when
encountering media errors. The problem has been identified
as an unhandled NULL pointer returned from find_get_page().
This modification simply checks for a NULL page, and returns
with an error if found (the extent_range_uptodate() function
returns 1 on errors).
After testing this patch, the user reported that the error with
the NULL pointer oops was solved. However, there is still a
remaining problem with a thread becoming stuck in
wait_on_page_locked(page) in the read_extent_buffer_pages(...)
function in extent_io.c
for (i = start_i; i < num_pages; i++) {
page = extent_buffer_page(eb, i);
wait_on_page_locked(page);
if (!PageUptodate(page))
ret = -EIO;
}
This patch leaves the issue with the locked page yet to be resolved.
Signed-off-by: Mitch Harder <mitch.harder@sabayonlinux.org>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
wait_log_commit() and wait_for_writer() were using slightly different
conditions for deciding whether they should call schedule() and whether they
should continue in the wait loop. Thus it could happen that we busylooped when
the first condition was not true while the second one was. That is burning CPU
cycles needlessly and is deadly on UP machines...
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
We have only been checking for min_bytes available in bitmap entries, but we
won't successfully setup a bitmap cluster unless it has at least bytes in the
bitmap, so in the common case min_bytes is 4k and we want something like 2MB, so
if there are a bunch of bitmap entries with less than 2mb's in them, we'll
search all them anyway, which is suboptimal. Fix this check. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Added initialization with the declaration of ret. It isn't set later on the
switch-default branch (which should never be taken).
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Quoth Ben Myers:
"Please pull in the following bugfix for xfs. We forgot to drop a lock on
error in xfs_readlink. It hasn't been through -next yet, but there is no
-next tree tomorrow. The fix is clear so I'm sending this request today."
* 'for-linus' of git://oss.sgi.com/xfs/xfs:
xfs: Fix missing xfs_iunlock() on error recovery path in xfs_readlink()
The data encryption was moved from ecryptfs_write_end into
ecryptfs_writepage, this patch moves the corresponding function
comments to be consistent with the modification.
Signed-off-by: Li Wang <liwang@nudt.edu.cn>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Says Tyler:
"Tim's logging message update will be really helpful to users when
they're trying to locate a problematic file in the lower filesystem
with filename encryption enabled.
You'll recognize the fix from Li, as you commented on that.
You should also be familiar with my setattr/truncate improvements,
since you were the one that pointed them out to us (thanks again!).
Andrew noted the /dev/ecryptfs write count sanitization needed to be
improved, so I've got a fix in there for that along with some other
less important cleanups of the /dev/ecryptfs read/write code."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tyhicks/ecryptfs:
eCryptfs: Fix oops when printing debug info in extent crypto functions
eCryptfs: Remove unused ecryptfs_read()
eCryptfs: Check inode changes in setattr
eCryptfs: Make truncate path killable
eCryptfs: Infinite loop due to overflow in ecryptfs_write()
eCryptfs: Replace miscdev read/write magic numbers
eCryptfs: Report errors in writes to /dev/ecryptfs
eCryptfs: Sanitize write counts of /dev/ecryptfs
ecryptfs: Remove unnecessary variable initialization
ecryptfs: Improve metadata read failure logging
MAINTAINERS: Update eCryptfs maintainer address
If pages passed to the eCryptfs extent-based crypto functions are not
mapped and the module parameter ecryptfs_verbosity=1 was specified at
loading time, a NULL pointer dereference will occur.
Note that this wouldn't happen on a production system, as you wouldn't
pass ecryptfs_verbosity=1 on a production system. It leaks private
information to the system logs and is for debugging only.
The debugging info printed in these messages is no longer very useful
and rather than doing a kmap() in these debugging paths, it will be
better to simply remove the debugging paths completely.
https://launchpad.net/bugs/913651
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Reported-by: Daniel DeFreez
Cc: <stable@vger.kernel.org>
ecryptfs_read() has been ifdef'ed out for years now and it was
apparently unused before then. It is time to get rid of it for good.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Most filesystems call inode_change_ok() very early in ->setattr(), but
eCryptfs didn't call it at all. It allowed the lower filesystem to make
the call in its ->setattr() function. Then, eCryptfs would copy the
appropriate inode attributes from the lower inode to the eCryptfs inode.
This patch changes that and actually calls inode_change_ok() on the
eCryptfs inode, fairly early in ecryptfs_setattr(). Ideally, the call
would happen earlier in ecryptfs_setattr(), but there are some possible
inode initialization steps that must happen first.
Since the call was already being made on the lower inode, the change in
functionality should be minimal, except for the case of a file extending
truncate call. In that case, inode_newsize_ok() was never being
called on the eCryptfs inode. Rather than inode_newsize_ok() catching
maximum file size errors early on, eCryptfs would encrypt zeroed pages
and write them to the lower filesystem until the lower filesystem's
write path caught the error in generic_write_checks(). This patch
introduces a new function, called ecryptfs_inode_newsize_ok(), which
checks if the new lower file size is within the appropriate limits when
the truncate operation will be growing the lower file.
In summary this change prevents eCryptfs truncate operations (and the
resulting page encryptions), which would exceed the lower filesystem
limits or FSIZE rlimits, from ever starting.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Reviewed-by: Li Wang <liwang@nudt.edu.cn>
Cc: <stable@vger.kernel.org>
ecryptfs_write() handles the truncation of eCryptfs inodes. It grabs a
page, zeroes out the appropriate portions, and then encrypts the page
before writing it to the lower filesystem. It was unkillable and due to
the lack of sparse file support could result in tying up a large portion
of system resources, while encrypting pages of zeros, with no way for
the truncate operation to be stopped from userspace.
This patch adds the ability for ecryptfs_write() to detect a pending
fatal signal and return as gracefully as possible. The intent is to
leave the lower file in a useable state, while still allowing a user to
break out of the encryption loop. If a pending fatal signal is detected,
the eCryptfs inode size is updated to reflect the modified inode size
and then -EINTR is returned.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Cc: <stable@vger.kernel.org>
ecryptfs_write() can enter an infinite loop when truncating a file to a
size larger than 4G. This only happens on architectures where size_t is
represented by 32 bits.
This was caused by a size_t overflow due to it incorrectly being used to
store the result of a calculation which uses potentially large values of
type loff_t.
[tyhicks@canonical.com: rewrite subject and commit message]
Signed-off-by: Li Wang <liwang@nudt.edu.cn>
Signed-off-by: Yunchuan Wen <wenyunchuan@kylinos.com.cn>
Reviewed-by: Cong Wang <xiyou.wangcong@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
ecryptfs_miscdev_read() and ecryptfs_miscdev_write() contained many
magic numbers for specifying packet header field sizes and offsets. This
patch defines those values and replaces the magic values.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Errors in writes to /dev/ecryptfs were being incorrectly reported by
returning 0 or the value of the original write count.
This patch clears up the return code assignment in error paths.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
A malicious count value specified when writing to /dev/ecryptfs may
result in a a very large kernel memory allocation.
This patch peeks at the specified packet payload size, adds that to the
size of the packet headers and compares the result with the write count
value. The resulting maximum memory allocation size is approximately 532
bytes.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Reported-by: Sasha Levin <levinsasha928@gmail.com>
Cc: <stable@vger.kernel.org>
Removes unneeded variable initialization in ecryptfs_read_metadata(). Also adds
a small comment to help explain metadata reading logic.
[tyhicks@canonical.com: Pulled out of for-stable patch and wrote commit msg]
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Print inode on metadata read failure. The only real
way of dealing with metadata read failures is to delete
the underlying file system file. Having the inode
allows one to 'find . -inum INODE`.
[tyhicks@canonical.com: Removed some minor not-for-stable parts]
Signed-off-by: Tim Gardner <tim.gardner@canonical.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: stable@vger.kernel.org
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Commit b52a360b forgot to call xfs_iunlock() when it detected corrupted
symplink and bailed out. Fix it by jumping to 'out' instead of doing return.
CC: stable@kernel.org
CC: Carlos Maiolino <cmaiolino@redhat.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: Alex Elder <elder@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
quota: Pass information that quota is stored in system file to userspace
ext2: protect inode changes in the SETVERSION and SETFLAGS ioctls
jbd: Issue cache flush after checkpointing
Recently an OOPS was observed from the usb serial io_ti driver when it tried to remove
sysfs directories. Upon investigation it turns out this driver was always buggy
and that a recent sysfs change had stopped guarding itself against removing attributes
from sysfs directories that had already been removed. :(
Historically we have been silent about attempting to files from nonexistent sysfs
directories and have politely returned error codes. That has resulted in people writing
broken code that ignores the error codes.
Issue a kernel WARNING and a stack backtrace to make it clear in no uncertain
terms that abusing sysfs is not ok, and the callers need to fix their code.
This change transforms the io_ti OOPS into a more comprehensible error message
and stack backtrace.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Reported-by: Wolfgang Frisch <wfpub@roembden.net>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Fix new kernel-doc warnings:
Warning(fs/debugfs/file.c:556): No description found for parameter 'nregs'
Warning(fs/debugfs/file.c:556): Excess function parameter 'mregs' description in 'debugfs_print_regs32'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The usual kernel-doc fixups from Randy. Some of them David acked as
merged in his tree, this is the random left-overs.
* kernel-doc:
docbook: fix sched source file names in device-drivers book
docbook: change iomap source filename in deviceiobook
docbook: don't use serial_core.h in device-drivers book
kernel-doc: fix kernel-doc warnings in sched
kernel-doc: fix new warnings in cfg80211.h
kernel-doc: fix new warning in usb.h
kernel-doc: fix new warnings in device.h
kernel-doc: fix new warnings in debugfs
kernel-doc: fix new warning in regulator core
kernel-doc: fix new warnings in pci
kernel-doc: fix new warnings in driver-core
kernel-doc: fix new warnings in auditsc.c
scripts/kernel-doc: fix fatal error caused by cfg80211.h
Quoth Andrew:
"Random fixes. And a simple new LED driver which I'm trying to sneak
in while you're not looking."
Sneaking successful.
* akpm:
score: fix off-by-one index into syscall table
mm: fix rss count leakage during migration
SHM_UNLOCK: fix Unevictable pages stranded after swap
SHM_UNLOCK: fix long unpreemptible section
kdump: define KEXEC_NOTE_BYTES arch specific for s390x
mm/hugetlb.c: undo change to page mapcount in fault handler
mm: memcg: update the correct soft limit tree during migration
proc: clear_refs: do not clear reserved pages
drivers/video/backlight/l4f00242t03.c: return proper error in l4f00242t03_probe if regulator_get() fails
drivers/video/backlight/adp88x0_bl.c: fix bit testing logic
kprobes: initialize before using a hlist
ipc/mqueue: simplify reading msgqueue limit
leds: add led driver for Bachmann's ot200
mm: __count_immobile_pages(): make sure the node is online
mm: fix NULL ptr dereference in __count_immobile_pages
mm: fix warnings regarding enum migrate_mode
* git://git.samba.org/sfrench/cifs-2.6:
CIFS: Rename *UCS* functions to *UTF16*
[CIFS] ACL and FSCACHE support no longer EXPERIMENTAL
[CIFS] Fix build break with multiuser patch when LANMAN disabled
cifs: warn about impending deprecation of legacy MultiuserMount code
cifs: fetch credentials out of keyring for non-krb5 auth multiuser mounts
cifs: sanitize username handling
keys: add a "logon" key type
cifs: lower default wsize when unix extensions are not used
cifs: better instrumentation for coalesce_t2
cifs: integer overflow in parse_dacl()
cifs: Fix sparse warning when calling cifs_strtoUCS
CIFS: Add descriptions to the brlock cache functions
Fix new kernel-doc warnings:
Warning(fs/debugfs/file.c:556): No description found for parameter 'nregs'
Warning(fs/debugfs/file.c:556): Excess function parameter 'mregs' description in 'debugfs_print_regs32'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
/proc/pid/clear_refs is used to clear the Referenced and YOUNG bits for
pages and corresponding page table entries of the task with PID pid, which
includes any special mappings inserted into the page tables in order to
provide things like vDSOs and user helper functions.
On ARM this causes a problem because the vectors page is mapped as a
global mapping and since ec706dab ("ARM: add a vma entry for the user
accessible vector page"), a VMA is also inserted into each task for this
page to aid unwinding through signals and syscall restarts. Since the
vectors page is required for handling faults, clearing the YOUNG bit (and
subsequently writing a faulting pte) means that we lose the vectors page
*globally* and cannot fault it back in. This results in a system deadlock
on the next exception.
To see this problem in action, just run:
$ echo 1 > /proc/self/clear_refs
on an ARM platform (as any user) and watch your system hang. I think this
has been the case since 2.6.37
This patch avoids clearing the aforementioned bits for reserved pages,
therefore leaving the vectors page intact on ARM. Since reserved pages
are not candidates for swap, this change should not have any impact on the
usefulness of clear_refs.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Reported-by: Moussa Ba <moussaba@micron.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Acked-by: Nicolas Pitre <nico@linaro.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: <stable@vger.kernel.org> [2.6.37+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/accounting, proc: Fix /proc/stat interrupts sum
* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
tracepoints/module: Fix disabling tracepoints with taint CRAP or OOT
x86/kprobes: Add arch/x86/tools/insn_sanity to .gitignore
x86/kprobes: Fix typo transferred from Intel manual
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86, syscall: Need __ARCH_WANT_SYS_IPC for 32 bits
x86, tsc: Fix SMI induced variation in quick_pit_calibrate()
x86, opcode: ANDN and Group 17 in x86-opcode-map.txt
x86/kconfig: Move the ZONE_DMA entry under a menu
x86/UV2: Add accounting for BAU strong nacks
x86/UV2: Ack BAU interrupt earlier
x86/UV2: Remove stale no-resources test for UV2 BAU
x86/UV2: Work around BAU bug
x86/UV2: Fix BAU destination timeout initialization
x86/UV2: Fix new UV2 hardware by using native UV2 broadcast mode
x86: Get rid of dubious one-bit signed bitfield
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
qnx4: don't leak ->BitMap on late failure exits
qnx4: reduce the insane nesting in qnx4_checkroot()
qnx4: di_fname is an array, for crying out loud...
vfs: remove printk from set_nlink()
wake up s_wait_unfrozen when ->freeze_fs fails
to reflect the unicode encoding used by CIFS protocol.
Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Acked-by: Jeff Layton <jlayton@samba.org>
Reviewed-by: Shirish Pargaonkar <shirishpargaonkar@gmail.com>
CIFS ACL support and FSCACHE support have been in long enough
to be no longer considered experimental. Remove obsolete Kconfig
dependency.
Signed-off-by: Steve French <sfrench@us.ibm.com>
Acked-by: Jeff Layton <jlayton@redhat.com>
We'll allow a grace period of 2 releases (3.3 and 3.4) and then remove
the legacy code in 3.5.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
Fix up multiuser mounts to set the secType and set the username and
password from the key payload in the vol info for non-krb5 auth types.
Look for a key of type "secret" with a description of
"cifs🅰️<server address>" or "cifs:d:<domainname>". If that's found,
then scrape the username and password out of the key payload and use
that to create a new user session.
Finally, don't have the code enforce krb5 auth on multiuser mounts,
but do require a kernel with keys support.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
Currently, it's not very clear whether you're allowed to have a NULL
vol->username or ses->user_name. Some places check for it and some don't.
Make it clear that a NULL pointer is OK in these fields, and ensure that
all the callers check for that.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
We've had some reports of servers (namely, the Solaris in-kernel CIFS
server) that don't deal properly with writes that are "too large" even
though they set CAP_LARGE_WRITE_ANDX. Change the default to better
mirror what windows clients do.
Cc: stable@vger.kernel.org
Cc: Pavel Shilovsky <piastry@etersoft.ru>
Reported-by: Nick Davis <phireph0x@yahoo.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
When coalesce_t2 returns an error, have it throw a cFYI message that
explains the reason. Also rename some variables to clarify what they
represent.
Reported-and-Tested-by: Konstantinos Skarlatos <k.skarlatos@gmail.com>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/audit: (29 commits)
audit: no leading space in audit_log_d_path prefix
audit: treat s_id as an untrusted string
audit: fix signedness bug in audit_log_execve_info()
audit: comparison on interprocess fields
audit: implement all object interfield comparisons
audit: allow interfield comparison between gid and ogid
audit: complex interfield comparison helper
audit: allow interfield comparison in audit rules
Kernel: Audit Support For The ARM Platform
audit: do not call audit_getname on error
audit: only allow tasks to set their loginuid if it is -1
audit: remove task argument to audit_set_loginuid
audit: allow audit matching on inode gid
audit: allow matching on obj_uid
audit: remove audit_finish_fork as it can't be called
audit: reject entry,always rules
audit: inline audit_free to simplify the look of generic code
audit: drop audit_set_macxattr as it doesn't do anything
audit: inline checks for not needing to collect aux records
audit: drop some potentially inadvisable likely notations
...
Use evil merge to fix up grammar mistakes in Kconfig file.
Bad speling and horrible grammar (and copious swearing) is to be
expected, but let's keep it to commit messages and comments, rather than
expose it to users in config help texts or printouts.
* 'for-linus' of git://oss.sgi.com/xfs/xfs:
xfs: cleanup xfs_file_aio_write
xfs: always return with the iolock held from xfs_file_aio_write_checks
xfs: remove the i_new_size field in struct xfs_inode
xfs: remove the i_size field in struct xfs_inode
xfs: replace i_pin_wait with a bit waitqueue
xfs: replace i_flock with a sleeping bitlock
xfs: make i_flags an unsigned long
xfs: remove the if_ext_max field in struct xfs_ifork
xfs: remove the unused dm_attrs structure
xfs: cleanup xfs_iomap_eof_align_last_fsb
xfs: remove xfs_itruncate_data
* 'btrfs' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
btrfs: take allocation of ->tree_root into open_ctree()
btrfs: let ->s_fs_info point to fs_info, not root...
btrfs: consolidate failure exits in btrfs_mount() a bit
btrfs: make free_fs_info() call ->kill_sb() unconditional
btrfs: merge free_fs_info() calls on fill_super failures
btrfs: kill pointless reassignment of ->s_fs_info in btrfs_fill_super()
btrfs: make open_ctree() return int
btrfs: sanitizing ->fs_info, part 5
btrfs: sanitizing ->fs_info, part 4
btrfs: sanitizing ->fs_info, part 3
btrfs: sanitizing ->fs_info, part 2
btrfs: sanitizing ->fs_info, part 1
btrfs: fix a deadlock in btrfs_scan_one_device()
btrfs: fix mount/umount race
btrfs: get ->kill_sb() of its own
btrfs: preparation to fixing mount/umount race
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (62 commits)
Btrfs: use larger system chunks
Btrfs: add a delalloc mutex to inodes for delalloc reservations
Btrfs: space leak tracepoints
Btrfs: protect orphan block rsv with spin_lock
Btrfs: add allocator tracepoints
Btrfs: don't call btrfs_throttle in file write
Btrfs: release space on error in page_mkwrite
Btrfs: fix btrfsck error 400 when truncating a compressed
Btrfs: do not use btrfs_end_transaction_throttle everywhere
Btrfs: add balance progress reporting
Btrfs: allow for resuming restriper after it was paused
Btrfs: allow for canceling restriper
Btrfs: allow for pausing restriper
Btrfs: add skip_balance mount option
Btrfs: recover balance on mount
Btrfs: save balance parameters to disk
Btrfs: soft profile changing mode (aka soft convert)
Btrfs: implement online profile changing
Btrfs: do not reduce profile in do_chunk_alloc()
Btrfs: virtual address space subset filter
...
Fix up trivial conflict in fs/btrfs/ioctl.c due to the use of the new
mnt_drop_write_file() helper.
Jüri Aedla reported that the /proc/<pid>/mem handling really isn't very
robust, and it also doesn't match the permission checking of any of the
other related files.
This changes it to do the permission checks at open time, and instead of
tracking the process, it tracks the VM at the time of the open. That
simplifies the code a lot, but does mean that if you hold the file
descriptor open over an execve(), you'll continue to read from the _old_
VM.
That is different from our previous behavior, but much simpler. If
somebody actually finds a load where this matters, we'll need to revert
this commit.
I suspect that nobody will ever notice - because the process mapping
addresses will also have changed as part of the execve. So you cannot
actually usefully access the fd across a VM change simply because all
the offsets for IO would have changed too.
Reported-by: Jüri Aedla <asd@ut.ee>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
dd slept infinitely when fsfeeze failed because of EIO.
To fix this problem, if ->freeze_fs fails, freeze_super() wakes up
the tasks waiting for the filesystem to become unfrozen.
When s_frozen isn't SB_UNFROZEN in __generic_file_aio_write(),
the function sleeps until FITHAW ioctl wakes up s_wait_unfrozen.
However, if ->freeze_fs fails, s_frozen is set to SB_UNFROZEN and then
freeze_super() returns an error number. In this case, FITHAW ioctl returns
EINVAL because s_frozen is already SB_UNFROZEN. There is no way to wake up
s_wait_unfrozen, so __generic_file_aio_write() sleeps infinitely.
Signed-off-by: Kazuya Mio <k-mio@sx.jp.nec.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Just a code cleanup really. We don't need to make a function call just for
it to return on error. This also makes the VFS function even easier to follow
and removes a conditional on a hot path.
Signed-off-by: Eric Paris <eparis@redhat.com>
At the moment we allow tasks to set their loginuid if they have
CAP_AUDIT_CONTROL. In reality we want tasks to set the loginuid when they
log in and it be impossible to ever reset. We had to make it mutable even
after it was once set (with the CAP) because on update and admin might have
to restart sshd. Now sshd would get his loginuid and the next user which
logged in using ssh would not be able to set his loginuid.
Systemd has changed how userspace works and allowed us to make the kernel
work the way it should. With systemd users (even admins) are not supposed
to restart services directly. The system will restart the service for
them. Thus since systemd is going to loginuid==-1, sshd would get -1, and
sshd would be allowed to set a new loginuid without special permissions.
If an admin in this system were to manually start an sshd he is inserting
himself into the system chain of trust and thus, logically, it's his
loginuid that should be used! Since we have old systems I make this a
Kconfig option.
Signed-off-by: Eric Paris <eparis@redhat.com>
The function always deals with current. Don't expose an option
pretending one can use it for something. You can't.
Signed-off-by: Eric Paris <eparis@redhat.com>
With all the size field updates out of the way xfs_file_aio_write can
be further simplified by pushing all iolock handling into
xfs_file_dio_aio_write and xfs_file_buffered_aio_write and using
the generic generic_write_sync helper for synchronous writes.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ben Myers <bpm@sgi.com>
While xfs_iunlock is fine with 0 lockflags the calling conventions are much
cleaner if xfs_file_aio_write_checks never returns without the iolock held.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ben Myers <bpm@sgi.com>
Now that we use the VFS i_size field throughout XFS there is no need for the
i_new_size field any more given that the VFS i_size field gets updated
in ->write_end before unlocking the page, and thus is always uptodate when
writeback could see a page. Removing i_new_size also has the advantage that
we will never have to trim back di_size during a failed buffered write,
given that it never gets updated past i_size.
Note that currently the generic direct I/O code only updates i_size after
calling our end_io handler, which requires a small workaround to make
sure di_size actually makes it to disk. I hope to fix this properly in
the generic code.
A downside is that we lose the support for parallel non-overlapping O_DIRECT
appending writes that recently was added. I don't think keeping the complex
and fragile i_new_size infrastructure for this is a good tradeoff - if we
really care about parallel appending writers we should investigate turning
the iolock into a range lock, which would also allow for parallel
non-overlapping buffered writers.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
There is no fundamental need to keep an in-memory inode size copy in the XFS
inode. We already have the on-disk value in the dinode, and the separate
in-memory copy that we need for regular files only in the XFS inode.
Remove the xfs_inode i_size field and change the XFS_ISIZE macro to use the
VFS inode i_size field for regular files. Switch code that was directly
accessing the i_size field in the xfs_inode to XFS_ISIZE, or in cases where
we are limited to regular files direct access of the VFS inode i_size field.
This also allows dropping some fairly complicated code in the write path
which dealt with keeping the xfs_inode i_size uptodate with the VFS i_size
that is getting updated inside ->write_end.
Note that we do not bother resetting the VFS i_size when truncating a file
that gets freed to zero as there is no point in doing so because the VFS inode
is no longer in use at this point. Just relax the assert in xfs_ifree to
only check the on-disk size instead.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ben Myers <bpm@sgi.com>
Replace i_pin_wait, which is only used during synchronous inode flushing
with a bit waitqueue. This trades off a much smaller inode against
slightly slower wakeup performance, and saves 12 (32-bit) or 20 (64-bit)
bytes in the XFS inode.
Reviewed-by: Alex Elder <aelder@sgi.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Ben Myers <bpm@sgi.com>
We almost never block on i_flock, the exception is synchronous inode
flushing. Instead of bloating the inode with a 16/24-byte completion
that we abuse as a semaphore just implement it as a bitlock that uses
a bit waitqueue for the rare sleeping path. This primarily is a
tradeoff between a much smaller inode and a faster non-blocking
path vs faster wakeups, and we are much better off with the former.
A small downside is that we will lose lockdep checking for i_flock, but
given that it's always taken inside the ilock that should be acceptable.
Note that for example the inode writeback locking is implemented in a
very similar way.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
To be used for bit wakeup i_flags needs to be an unsigned long or we'll
run into trouble on big endian systems. Because of the 1-byte i_update
field right after it this actually causes a fairly large size increase
on its own (4 or 8 bytes), but that increase will be more than offset
by the next two patches.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Alex Elder <aelder@sgi.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
We spent a lot of effort to maintain this field, but it always equals to the
fork size divided by the constant size of an extent. The prime use of it is
to assert that the two stay in sync. Just divide the fork size by the extent
size in the few places that we actually use it and remove the overhead
of maintaining it. Also introduce a few helpers to consolidate the places
where we actually care about the value.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=beF6
-----END PGP SIGNATURE-----
Merge tag 'nfs-for-3.3-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
NFS client bugfixes and cleanups for Linux 3.3 (pull 2)
* tag 'nfs-for-3.3-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
pnfsblock: alloc short extent before submit bio
pnfsblock: remove rpc_call_ops from struct parallel_io
pnfsblock: move find lock page logic out of bl_write_pagelist
pnfsblock: cleanup bl_mark_sectors_init
pnfsblock: limit bio page count
pnfsblock: don't spinlock when freeing block_dev
pnfsblock: clean up _add_entry
pnfsblock: set read/write tk_status to pnfs_error
pnfsblock: acquire im_lock in _preload_range
NFS4: fix compile warnings in nfs4proc.c
nfs: check for integer overflow in decode_devicenotify_args()
NFS: cleanup endian type in decode_ds_addr()
NFS: add an endian notation
system chunks by default are very small. This makes them slightly
larger and also fixes the conditional checks to make sure we don't
allocate a billion of them at once.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
I was using i_mutex for this, but we're getting bogus lockdep warnings by doing
that and theres no real way to get rid of those, so just stop using i_mutex to
protect delalloc metadata reservations and use a delalloc mutex instead. This
shouldn't be contended often at all, only if you are writing and mmap writing to
the file at the same time. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
This in addition to a script in my btrfs-tracing tree will help track down space
leaks when we're getting space left over in block groups on umount. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
We've been seeing warnings coming out of the orphan commit stuff forever from
ceph. Turns out it's because we're racing with checking if the orphan block
reserve is set, because we clear it outside of the spin_lock. So leave the
normal fastpath checks where they are, but take the spin_lock and _recheck_ to
make sure we haven't had an orphan block rsv added in the meantime. Then clear
the root's orphan block rsv and release the lock. With this patch a user said
the warnings went away and they usually showed up pretty soon after he started
ceph. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
I used these tracepoints when figuring out what the cluster stuff was doing, so
add them to mainline in case we need to profile this stuff again. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Btrfs_throttle will make us wait if there is a currently committing transaction
until we can open new transactions, which is ridiculous since we don't actually
start any transactions within the file write path anyway, so all this does is
introduce big latencies if we have a sync/fsync heavy workload going on while
somebody else is trying to do work. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
If updating the inode gave us an ENOSPC we were just returning in page_mkwrite,
which is a problem since we make our reservation right before trying to update
the inode, so fix the out label so that we actually free our reservation.
Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Reproduce steps:
# mkfs.btrfs /dev/sdb5
# mount /dev/sdb5 -o compress=lzo /mnt
# dd if=/dev/zero of=/mnt/tmpfile bs=128K count=1
# sync
# truncate -s 64K /mnt/tmpfile
root 5 inode 257 errors 400
This is because of the wrong if condition, which is used to check if we should
subtract the bytes of the dropped range from i_blocks/i_bytes of i-node or not.
When we truncate a compressed extent, btrfs substracts the bytes of the whole
extent, it's wrong. We should substract the real size that we truncate, no
matter it is a compressed extent or not. Fix it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
A user reported a problem where things like open with O_CREAT would take up to
30 seconds when he had nfs activity on the same mount. This is because all of
our quick metadata operations, like create, symlink etc all do
btrfs_end_transaction_throttle, which if the transaction is blocked will wait
for the commit to complete before it returns. This adds a ridiculous amount of
latency and isn't really needed. The normal btrfs_end_transaction will mark the
transaction as blocked and wake the transaction kthread up if it thinks the
transaction needs to end (this being in the running out of global reserve space
scenario), and this is all that is really needed since we've already done
everything we're going to do, we just need to return. This should help people
with the latency they were seeing when using synchronous heavy workloads.
Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Recognize BTRFS_BALANCE_RESUME flag passed from userspace. We use the
same heuristics used when recovering balance after a crash to try to
start where we left off last time.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Implement an ioctl for canceling restriper. Currently we wait until
relocation of the current block group is finished, in future this can be
done by triggering a commit. Balance item is deleted and no memory
about the interrupted balance is kept.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Implement an ioctl for pausing restriper. This pauses the relocation,
but balance is still considered to be "in progress": balance item is
not deleted, other volume operations cannot be started, etc. If paused
in the middle of profile changing operation we will continue making
allocations with the target profile.
Add a hook to close_ctree() to pause restriper and free its data
structures on unmount. (It's safe to unmount when restriper is in
"paused" state, we will resume with the same parameters on the next
mount)
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Since restriper kthread starts involuntarily on mount and can suck cpu
and memory bandwidth add a mount option to forcefully skip it. The
restriper in that case hangs around in paused state and can be resumed
from userspace when it's convenient.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
On mount, if balance item is found, resume balance in a separate
kernel thread.
Try to be smart to continue roughly where previous balance (or convert)
was interrupted. For chunk types that were being converted to some
profile we turn on soft convert, in case of a simple balance we turn on
usage filter and relocate only less-than-90%-full chunks of that type.
These are just heuristics but they help quite a bit, and can be improved
in future.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Introduce a new btree objectid for storing balance item. The reason is
to be able to resume restriper after a crash with the same parameters.
Balance item has a very high objectid and goes into tree of tree roots.
The key for the new item is as follows:
[ BTRFS_BALANCE_OBJECTID ; BTRFS_BALANCE_ITEM_KEY ; 0 ]
Older kernels simply ignore it so it's safe to mount with an older
kernel and then go back to the newer one.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
When doing convert from one profile to another if soft mode is on
restriper won't touch chunks that already have the profile we are
converting to. This is useful if e.g. half of the FS was converted
earlier.
The soft mode switch is (like every other filter) per-type. This means
that we can convert for example meta chunks the "hard" way while
converting data chunks selectively with soft switch.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Profile changing is done by launching a balance with
BTRFS_BALANCE_CONVERT bits set and target fields of respective
btrfs_balance_args structs initialized. Profile reducing code in this
case will pick restriper's target profile if it's available instead of
doing a blind reduce. If target profile is not yet available it goes
back to a plain reduce.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Every caller of do_chunk_alloc() feeds it the reduced allocation
profile, so stop trying to reduce it one more time. Instead check the
validity of the passed profile.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Select chunks which have at least one byte located inside a given
[vstart, vend) virtual address space range.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Select chunks which have at least one byte of at least one stripe
located on a device with devid X in a given [pstart,pend) physical
address range.
This filter only works when devid filter is turned on.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
This allows to have a separate set of filters for each chunk type
(data,meta,sys). The code however is generic and switch on chunk type
is only done once.
This commit also adds a type filter: it allows to balance for example
meta and system chunks w/o touching data ones.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Add basic restriper infrastructure: extended balancing ioctl and all
related ioctl data structures, add data structure for tracking
restriper's state to fs_info, etc. The semantics of the old balancing
ioctl are fully preserved.
Explicitly disallow any volume operations when balance is in progress.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Currently when new chunks are created respective avail_alloc_bits field
is updated to reflect profiles of all chunks present in the system.
However when chunks are removed profile bits are never cleared.
This patch clears profile bit of respective avail_alloc_bits field when
the last chunk with that profile is removed. Restriper needs this to
properly operate when "downgrading".
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Right now on-disk BTRFS_BLOCK_GROUP_* profile bits are used for
avail_{data,metadata,system}_alloc_bits fields, which gather info about
available allocation profiles in the FS. When chunk is created or read
from disk, its profile is OR'ed with the corresponding avail_alloc_bits
field. Since SINGLE is denoted by 0 in the on-disk format, currently
there is no way to tell when such chunks become avaialble. Restriper
needs that information, so add a separate bit for SINGLE profile.
This bit is going to be in-memory only, it should never be written out
to disk, so it's not a disk format change. However to avoid remappings
in future, reserve corresponding on-disk bit.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Chunk's type and profile are encoded in u64 flags field. Introduce
masks to easily access them. Also fix the type of BTRFS_BLOCK_GROUP_*
constants, it should be ULL.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
