Commit Graph

1007 Commits

Author SHA1 Message Date
John Baldwin
3634d5b241 Add dedicated routines to toggle lockmgr flags such as LK_NOSHARE and
LK_CANRECURSE after a lock is created.  Use them to implement macros that
otherwise manipulated the flags directly.  Assert that the associated
lockmgr lock is exclusively locked by the current thread when manipulating
these flags to ensure the flag updates are safe.  This last change required
some minor shuffling in a few filesystems to exclusively lock a brand new
vnode slightly earlier.

Reviewed by:	kib
MFC after:	3 days
2010-08-20 19:46:50 +00:00
Konstantin Belousov
691401eef8 Softdep_process_worklist() should unsuspend not only before processing
the worklist (in softdep_process_journal), but also after flushing the
workitems. Might be, we should even do this before bwillwrite() too, but
this seems to be not needed for now.

Fs might be suspended during processing the queue, and then there is
nobody around to unsuspend.

In collaboration with:	pho
Tested by:	bz
Reviewed by:	jeff
2010-08-12 08:35:24 +00:00
John Baldwin
61e1c19319 Revert the previous commit. The race is not applicable to the lockmgr
implementation in 8.0 and later as its flags field does not hold dynamic
state such as waiters flags, but is only modified in lockinit() aside
from VN_LOCK_*().

Discussed with:	attilio
2010-07-16 19:52:03 +00:00
John Baldwin
dbfcf8cfea When the MNTK_EXTENDED_SHARED mount option was added, some filesystems were
changed to defer the setting of VN_LOCK_ASHARE() (which clears LK_NOSHARE
in the vnode lock's flags) until after they had determined if the vnode was
a FIFO.  This occurs after the vnode has been inserted a VFS hash or some
similar table, so it is possible for another thread to find this vnode via
vget() on an i-node number and block on the vnode lock.  If the lockmgr
interlock (vnode interlock for vnode locks) is not held when clearing the
LK_NOSHARE flag, then the lk_flags field can be clobbered.  As a result
the thread blocked on the vnode lock may never get woken up.  Fix this by
holding the vnode interlock while modifying the lock flags in this case.

MFC after:	3 days
2010-07-16 19:20:20 +00:00
Jeff Roberson
9f9c8c59ae - Handle the truncation of an inode with an effective link count of 0 in
the context of the process that reduced the effective count.  Previously
   all truncation as a result of unlink happened in the softdep flush
   thread.  This had the effect of being impossible to rate limit properly
   with the journal code.  Now the process issuing unlinks is suspended
   when the journal files.  This has a side-effect of improving rm
   performance by allowing more concurrent work.
 - Handle two cases in inactive, one for effnlink == 0 and another when
   nlink finally reaches 0.
 - Eliminate the SPACECOUNTED related code since the truncation is no
   longer delayed.

Discussed with:	mckusick
2010-07-06 07:11:04 +00:00
Andriy Gapon
d89c217f30 ffs_softdep: change K&R in function defintions to ANSI prototypes
Apparently it's bad when we first have an ANSI prototype in function
declaration, but then use K&R in its defintion.

Complaint from:	clang
MFC after:	2 weeks
2010-06-11 18:26:53 +00:00
Andriy Gapon
0b9626482b ffs_mount: accept and drop userland-only options that can be passed from
loader(8)

In r193192 loader(8) has grown an ability to pass root mount options
from fstab via vfs.root.mountfrom.options.  Unfortunately, some options
that can be present in fstab are for userland only and lead to root
mounting failure when seen by kernel.
Rather than teaching loader about FFS-specific options that should be
filtered out, ffs_mount recognizes those options as valid, but ignores
and deletes[1] them.

[1] is suggested by jh.

PR:		kern/141050
Reported by:	many
Reviewed by:	jh, bde
MFC after:	4 days
2010-05-19 09:32:11 +00:00
Jeff Roberson
f0268739c7 - Don't immediately re-run softdepflush if we didn't make any progress
on the last iteration.  This can lead to a deadlock when we have
   worklist items that cannot be immediately satisfied.

Reported by:	uqs, Dimitry Andric <dimitry@andric.com>

 - Remove some unnecessary debugging code and place some other under
   SUJ_DEBUG.
 - Examine the journal state in softdep_slowdown().
 - Re-format some comments so I may more easily add flag descriptions.
2010-05-19 06:18:01 +00:00
Jeff Roberson
8ef48de888 - Call softdep_prealloc() before any of the balloc routines in the
snapshot code.
 - Don't fsync() vnodes in prealloc if copy on write is in progress.  It
   is not safe to recurse back into the write path here.

Reported by:	Vladimir Grebenschikov <vova@fbsd.ru>
2010-05-07 08:45:21 +00:00
Jeff Roberson
2c3ae115b6 - Use the correct flag mask when determining whether an inode has
successfully made it to the free list yet or not.  This fixes
   a deadlock that can occur with unlinked but referenced files.
   Journal space and inodedeps were not correctly reclaimed because
   the inode block was not left dirty.

Tested/Reported by:	lwindschuh@googlemail.com
2010-05-07 08:20:56 +00:00
Alan Cox
eb00b276ab Eliminate page queues locking around most calls to vm_page_free(). 2010-05-06 18:58:32 +00:00
Alan Cox
5ac59343be Acquire the page lock around all remaining calls to vm_page_free() on
managed pages that didn't already have that lock held.  (Freeing an
unmanaged page, such as the various pmaps use, doesn't require the page
lock.)

This allows a change in vm_page_remove()'s locking requirements.  It now
expects the page lock to be held instead of the page queues lock.
Consequently, the page queues lock is no longer required at all by callers
to vm_page_rename().

Discussed with: kib
2010-05-05 18:16:06 +00:00
Edward Tomasz Napierala
b5f770bd86 Move checking against RLIMIT_FSIZE into one place, vn_rlimit_fsize().
Reviewed by:	kib
2010-05-05 16:44:25 +00:00
Andriy Gapon
deb3b115e2 ffs_vfsops: restore alphabetic order of options in ffs_opts
The order was not correct only for nfsv4acls.
("no" prefix is ignored)

MFC after:	1 week
2010-04-29 10:04:00 +00:00
Jeff Roberson
2bd20091e4 - When canceling jaddrefs they may not yet be in the journal if this is via
a revert call.  In this case don't attempt to remove something that
   has not yet been added.  Otherwise this jaddref must hang around
   to prevent the bitmap write as normal.
2010-04-28 07:57:37 +00:00
Jeff Roberson
3b32573a9f - Fix builds without SOFTUPDATES defined in the kernel config. 2010-04-28 07:26:41 +00:00
Pawel Jakub Dawidek
a8750f2dca Fix build for UFS without SOFTUPDATES. 2010-04-24 07:36:33 +00:00
Jeff Roberson
113db2dddb - Merge soft-updates journaling from projects/suj/head into head. This
brings in support for an optional intent log which eliminates the need
   for background fsck on unclean shutdown.

Sponsored by:   iXsystems, Yahoo!, and Juniper.
With help from: McKusick and Peter Holm
2010-04-24 07:05:35 +00:00
Andriy Gapon
ecaf3257be ffs_mount: remove redundant assignment of geom consumer to devvp.v_bufobj
The assignment is already done in g_vfs_open.
Redundant assignment is harmless, but can become a problem if g_vfs_open
logic is changed.

MFC after:	1 week
2010-04-03 08:25:04 +00:00
Konstantin Belousov
2950ff259c When ffs_realloccg() failed to allocate bigger fragment and, because
pending blocks are scheduled for removal, goes to retry the (re)allocation,
clear the bp pointer. It might happen that meantime free space is really
exhausted and we are entering nospace: label without bread()ing buffer,
causing stale bp value to be brelse()d again.

Tested by:	pho
    (Producing a scenario to reliably reproduce the
     race appeared to be much harder then fixing the bug)
MFC after:	1 week
2010-02-13 10:34:50 +00:00
Kirk McKusick
81479e688b One last pass to get all the unsigned comparisons correct. 2010-02-11 18:14:53 +00:00
Kirk McKusick
e870d1e6f9 This fix corrects a problem in the file system that treats large
inode numbers as negative rather than unsigned. For a default
(16K block) file system, this bug began to show up at a file system
size above about 16Tb.

To fully handle this problem, newfs must be updated to ensure that
it will never create a filesystem with more than 2^32 inodes. That
patch will be forthcoming soon.

Reported by: Scott Burns, John Kilburg, Bruce Evans
Followup by: Jeff Roberson
PR:          133980
MFC after:   2 weeks
2010-02-10 20:10:35 +00:00
Edward Tomasz Napierala
619961810c Remove unused variable. 2010-02-10 18:56:49 +00:00
Kirk McKusick
53298164b8 Cast 64-bit quantity to intptr_t rather than int so as to work properly
with 64-bit architectures (such as amd64).

Reported by:	bz
2010-01-11 22:42:06 +00:00
Kirk McKusick
e268f54cb4 Background:
When renaming a directory it passes through several intermediate
states. First its new name will be created causing it to have two
names (from possibly different parents). Next, if it has different
parents, its value of ".." will be changed from pointing to the old
parent to pointing to the new parent. Concurrently, its old name
will be removed bringing it back into a consistent state. When fsck
encounters an extra name for a directory, it offers to remove the
"extraneous hard link"; when it finds that the names have been
changed but the update to ".." has not happened, it offers to rewrite
".." to point at the correct parent. Both of these changes were
considered unexpected so would cause fsck in preen mode or fsck in
background mode to fail with the need to run fsck manually to fix
these problems. Fsck running in preen mode or background mode now
corrects these expected inconsistencies that arise during directory
rename. The functionality added with this update is used by fsck
running in background mode to make these fixes.

Solution:

This update adds three new fsck sysctl commands to support background
fsck in correcting expected inconsistencies that arise from incomplete
directory rename operations. They are:

setcwd(dirinode) - set the current directory to dirinode in the
    filesystem associated with the snapshot.
setdotdot(oldvalue, newvalue) - Verify that the inode number for ".."
    in the current directory is oldvalue then change it to newvalue.
unlink(nameptr, oldvalue) - Verify that the inode number associated
    with nameptr in the current directory is oldvalue then unlink it.

As with all other fsck sysctls, these new ones may only be used by
processes with appropriate priviledge.

Reported by:    	jeff
Security issues:	rwatson
2010-01-11 20:44:05 +00:00
Martin Blapp
c2ede4b379 Remove extraneous semicolons, no functional changes.
Submitted by:	Marc Balmer <marc@msys.ch>
MFC after:	1 week
2010-01-07 21:01:37 +00:00
Edward Tomasz Napierala
9340fc72e6 Implement NFSv4 ACL support for UFS.
Reviewed by:	rwatson
2009-12-21 19:39:10 +00:00
Konstantin Belousov
49e3050e6c VI_OBJDIRTY vnode flag mirrors the state of OBJ_MIGHTBEDIRTY vm object
flag. Besides providing the redundand information, need to update both
vnode and object flags causes more acquisition of vnode interlock.
OBJ_MIGHTBEDIRTY is only checked for vnode-backed vm objects.

Remove VI_OBJDIRTY and make sure that OBJ_MIGHTBEDIRTY is set only for
vnode-backed vm objects.

Suggested and reviewed by:	alc
Tested by:	pho
MFC after:	3 weeks
2009-12-21 12:29:38 +00:00
Konstantin Belousov
6cc745d2d7 insmntque_stddtr() clears vp->v_data and resets vp->v_op to
dead_vnodeops before calling vgone(). Revert r189706 and corresponding
part of the r186560.

Noted and reviewed by:	tegge
Approved by:	des (pseudofs part)
MFC after:	3 days
2009-09-07 11:55:34 +00:00
Konstantin Belousov
5c61c646a3 The clear_remove() and clear_inodedeps() call vn_start_write(NULL, &mp,
V_NOWAIT) on the non-busied mount point. Unmount might free ufs-specific
mp data, causing ffs_vgetf() to access freed memory.

Busy mountpoint before dropping softdep lk.

Noted and reviewed by:	tegge
Tested by:	pho
MFC after:	1 week
2009-09-06 11:46:51 +00:00
Konstantin Belousov
165a3b418f When a UFS node is truncated to the zero length, e.g. by explicit
truncate(2) call, or by being removed or truncated on open, either
new softupdate freeblks structure is allocated to track the freed
blocks of the node, or truncation is done syncronously when too many SU
dependencies are accumulated. The decision does not take into account
the allocated freeblks dependencies, allowing workloads that do huge
amount of truncations to exhaust the kernel memory.

Take the number of allocated freeblks into consideration for
softdep_slowdown().

Reported by:	pluknet gmail com
Diagnosed and tested by:	pho
Approved by:	re (rwatson)
MFC after:	1 month
2009-08-14 11:00:38 +00:00
Konstantin Belousov
f1eccd05ec In vn_vget_ino() and their inline equivalents, mnt_ref() the mount point
around the sequence that drop vnode lock and then busies the mount point.
Not having vlocked node or direct reference to the mp allows for the
forced unmount to proceed, making mp unmounted or reused.

Tested by:	pho
Reviewed by:	jeff
Approved by:	re (kensmith)
MFC after:	2 weeks
2009-07-02 18:02:55 +00:00
Edward Tomasz Napierala
4bc61fd4ec Don't panic on attempt to set ACL on a block device file.
This is just a part of kern/125613.

PR:		kern/125613
Submitted by:	Jaakko Heinonen <jh at saunalahti dot fi>
Reviewed by:	rwatson
Approved by:	re (kib)
2009-07-01 22:30:36 +00:00
Konstantin Belousov
cfba50c070 For SU mounts, softdep_fsync() might drop vnode lock, allowing other
threads to put dirty buffers on the vnode bufobj list. For regular files
and synchronous fsync requests, check for the condition and restart the
fsync vop if a new dirty buffer arrived.

Tested by:	pho
Approved by:	re (kensmith)
MFC after:	1 month
2009-06-30 10:07:33 +00:00
Konstantin Belousov
a50d1b2a66 Softdep_fsync() may need to lock parent directory of the synced vnode.
Use inlined (due to FFSV_FORCEINSMQ) version of vn_vget_ino() to prevent
mountpoint from being unmounted and freed while no vnodes are locked.

Tested by:	pho
Approved by:	re (kensmith)
MFC after:	1 month
2009-06-30 10:07:00 +00:00
Robert Watson
bcf11e8d00 Move "options MAC" from opt_mac.h to opt_global.h, as it's now in GENERIC
and used in a large number of files, but also because an increasing number
of incorrect uses of MAC calls were sneaking in due to copy-and-paste of
MAC-aware code without the associated opt_mac.h include.

Discussed with:	pjd
2009-06-05 14:55:22 +00:00
Attilio Rao
f083018223 Handle lock recursion differenty by always checking against LO_RECURSABLE
instead the lock own flag itself.

Tested by:	pho
2009-06-02 13:03:35 +00:00
Alan Cox
6e5982caf7 Introduce vfs_bio_set_valid() and use it from ffs_realloccg(). This
eliminates the misuse of vfs_bio_clrbuf() by ffs_realloccg().

In collaboration with:	tegge
2009-05-17 20:26:00 +00:00
Attilio Rao
dfd233edd5 Remove the thread argument from the FSD (File-System Dependent) parts of
the VFS.  Now all the VFS_* functions and relating parts don't want the
context as long as it always refers to curthread.

In some points, in particular when dealing with VOPs and functions living
in the same namespace (eg. vflush) which still need to be converted,
pass curthread explicitly in order to retain the old behaviour.
Such loose ends will be fixed ASAP.

While here fix a bug: now, UFS_EXTATTR can be compiled alone without the
UFS_EXTATTR_AUTOSTART option.

VFS KPI is heavilly changed by this commit so thirdy parts modules needs
to be recompiled.  Bump __FreeBSD_version in order to signal such
situation.
2009-05-11 15:33:26 +00:00
Robert Watson
885868cd8f Remove VOP_LEASE and supporting functions. This hasn't been used since
the removal of NQNFS, but was left in in case it was required for NFSv4.
Since our new NFSv4 client and server can't use it for their
requirements, GC the old mechanism, as well as other unused lease-
related code and interfaces.

Due to its impact on kernel programming and binary interfaces, this
change should not be MFC'd.

Proposed by:    jeff
Reviewed by:    jeff
Discussed with: rmacklem, zach loafman @ isilon
2009-04-10 10:52:19 +00:00
Konstantin Belousov
bc364c4e99 When removing or renaming snaphost, do not delve into request_cleanup().
The later may need blocks from the underlying device that belongs
to normal files, that should not be locked while snap lock is held.

Reported and tested by:	pho
MFC after:	1 month
2009-04-04 12:19:52 +00:00
Konstantin Belousov
02e06d99e6 Correct typo.
Noted by:	kensmith
2009-03-27 15:46:02 +00:00
Konstantin Belousov
c1d8b5e82c Fix two issues with bufdaemon, often causing the processes to hang in
the "nbufkv" sleep.

First, ffs background cg group block write requests a new buffer for
the shadow copy. When ffs_bufwrite() is called from the bufdaemon due
to buffers shortage, requesting the buffer deadlock bufdaemon.
Introduce a new flag for getnewbuf(), GB_NOWAIT_BD, to request getblk
to not block while allocating the buffer, and return failure
instead. Add a flag argument to the geteblk to allow to pass the flags
to getblk(). Do not repeat the getnewbuf() call from geteblk if buffer
allocation failed and either GB_NOWAIT_BD is specified, or geteblk()
is called from bufdaemon (or its helper, see below). In
ffs_bufwrite(), fall back to synchronous cg block write if shadow
block allocation failed.

Since r107847, buffer write assumes that vnode owning the buffer is
locked. The second problem is that buffer cache may accumulate many
buffers belonging to limited number of vnodes. With such workload,
quite often threads that own the mentioned vnodes locks are trying to
read another block from the vnodes, and, due to buffer cache
exhaustion, are asking bufdaemon for help. Bufdaemon is unable to make
any substantial progress because the vnodes are locked.

Allow the threads owning vnode locks to help the bufdaemon by doing
the flush pass over the buffer cache before getnewbuf() is going to
uninterruptible sleep. Move the flushing code from buf_daemon() to new
helper function buf_do_flush(), that is called from getnewbuf().  The
number of buffers flushed by single call to buf_do_flush() from
getnewbuf() is limited by new sysctl vfs.flushbufqtarget.  Prevent
recursive calls to buf_do_flush() by marking the bufdaemon and threads
that temporarily help bufdaemon by TDP_BUFNEED flag.

In collaboration with:	pho
Reviewed by:	 tegge (previous version)
Tested by:	 glebius, yandex ...
MFC after:	 3 weeks
2009-03-16 15:39:46 +00:00
Konstantin Belousov
e65f5a4ead The non-modifying EA VOPs are executed with only shared vnode lock taken.
Provide a custom lock around initializing and tearing down EA area,
to prevent both memory leaks and double-free of it. Count the number
of EA area accessors.

Lock protocol requires either holding exclusive vnode lock to modify
i_ea_area, or shared vnode lock and owning IN_EA_LOCKED flag in i_flag.

Noted by:	YAMAMOTO, Taku <taku tackymt homeip net>
Tested by:	pho (previous version)
MFC after:	2 weeks
2009-03-12 12:43:56 +00:00
Konstantin Belousov
a9d9537110 Do not double-free the struct inode when insmntque failed. Default
insmntque destructor reclaims the vnode, and ufs_reclaim frees the memory.

Reviewed by:	tegge
MFC after:	3 days
2009-03-11 19:45:52 +00:00
John Baldwin
33fc362512 Add a new internal mount flag (MNTK_EXTENDED_SHARED) to indicate that a
filesystem supports additional operations using shared vnode locks.
Currently this is used to enable shared locks for open() and close() of
read-only file descriptors.
- When an ISOPEN namei() request is performed with LOCKSHARED, use a
  shared vnode lock for the leaf vnode only if the mount point has the
  extended shared flag set.
- Set LOCKSHARED in vn_open_cred() for requests that specify O_RDONLY but
  not O_CREAT.
- Use a shared vnode lock around VOP_CLOSE() if the file was opened with
  O_RDONLY and the mountpoint has the extended shared flag set.
- Adjust md(4) to upgrade the vnode lock on the vnode it gets back from
  vn_open() since it now may only have a shared vnode lock.
- Don't enable shared vnode locks on FIFO vnodes in ZFS and UFS since
  FIFO's require exclusive vnode locks for their open() and close()
  routines.  (My recent MPSAFE patches for UDF and cd9660 already included
  this change.)
- Enable extended shared operations on UFS, cd9660, and UDF.

Submitted by:	ups
Reviewed by:	pjd (ZFS bits)
MFC after:	1 month
2009-03-11 14:13:47 +00:00
John Baldwin
5bd65606f4 Adjust some variables (mostly related to the buffer cache) that hold
address space sizes to be longs instead of ints.  Specifically, the follow
values are now longs: runningbufspace, bufspace, maxbufspace,
bufmallocspace, maxbufmallocspace, lobufspace, hibufspace, lorunningspace,
hirunningspace, maxswzone, maxbcache, and maxpipekva.  Previously, a
relatively small number (~ 44000) of buffers set in kern.nbuf would result
in integer overflows resulting either in hangs or bogus values of
hidirtybuffers and lodirtybuffers.  Now one has to overflow a long to see
such problems.  There was a check for a nbuf setting that would cause
overflows in the auto-tuning of nbuf.  I've changed it to always check and
cap nbuf but warn if a user-supplied tunable would cause overflow.

Note that this changes the ABI of several sysctls that are used by things
like top(1), etc., so any MFC would probably require a some gross shims
to allow for that.

MFC after:	1 month
2009-03-09 19:35:20 +00:00
Edward Tomasz Napierala
4f560d7595 Right now, when trying to unmount a device that's already gone,
msdosfs_unmount() and ffs_unmount() exit early after getting ENXIO.
However, dounmount() treats ENXIO as a success and proceeds with
unmounting.  In effect, the filesystem gets unmounted without closing
GEOM provider etc.

Reviewed by:	kib
Approved by:	rwatson (mentor)
Tested by:	dho
Sponsored by:	FreeBSD Foundation
2009-02-23 21:09:28 +00:00
Edward Tomasz Napierala
3c140b2df4 Refactor, moving error checking outside of the
'if (mp->mnt_flag & MNT_SOFTDEP)' conditional.  No functional
changes.

Reviewed by:	kib
Approved by:	rwatson (mentor)
Tested by:	pho
Sponsored by:	FreeBSD Foundation
2009-02-23 20:56:27 +00:00
John Baldwin
ee445a69c5 - If the g_access() call for the initial root mount fails, then fully
cleanup.  Before the GEOM consumer would not have been closed.
- Bump the reference on the character device being mounted while the
  associated devfs vnode is locked.

Reviewed by:	kib
2009-02-11 22:19:54 +00:00