mlelstv@ points out FreeBSD fixed the same thing a couple of years
ago - here's the commit message they used on rev 1.127:
Fixes a bug that caused UFS2 filesystems bigger than 2TB to
prematurely report that they were full and/or to panic the kernel
with the message ``ffs_clusteralloc: allocated out of group''.
Submitted by: Henry Whincup <henry@jot.to>
its not on a free list.
Also change buf_init() to not automatically mark buffers `busy' since this
only makes sense for bufcache buffers.
Mark all buf_init'd buffers 'busy' on the places where they ought to be
flagged as such to not confuse the buffer cache.
Fixes PR 38923.
If the number of deletes in progress is getting too high, newdirrem()
requests the syncer to flush faster, and in some cases will block to
prevent deletes accumulating faster than the disk can service them.
The syncer will try to lock vnodes that the remover holds locked, leading
to the syncer and remover proceeding in lockstep and making very little
overall forward progress.
Put a hook into ufs_rmdir() and ufs_remove() so that the softdep code
can pace itself without holding vnode locks if the number of deletes is
running out of control.
Use ufs_getlbns()/bread() instead.
Saves some reads and removes deep recursion with possible deadlock
when ffs_balloc() runs copy-on-write on the buffer returned.
run through copy-on-write. Call fscow_run() with valid data where possible.
The LP_UFSCOW hack is no longer needed to protect ffs_copyonwrite() against
endless recursion.
- Add a flag B_MODIFY to bread(), breada() and breadn(). If set the caller
intends to modify the buffer returned.
- Always run copy-on-write on buffers returned from ffs_balloc().
- Add new function ffs_getblk() that gets a buffer, assigns a new blkno,
may clear the buffer and runs copy-on-write. Process possible errors
from getblk() or fscow_run(). Part of PR kern/38664.
Welcome to 4.99.63
Reviewed by: YAMAMOTO Takashi <yamt@netbsd.org>
Make VFS hooks dynamic while we're here and say farewell to VFS_ATTACH and
VFS_HOOKS_ATTACH linksets.
As a consequence, most of the file systems can now be loaded as new style
modules.
Quick sanity check by ad@.
Simplify the mount locking. Remove all the crud to deal with recursion on
the mount lock, and crud to deal with unmount as another weirdo lock.
Hopefully this will once and for all fix the deadlocks with this. With this
commit there are two locks on each mount:
- krwlock_t mnt_unmounting. This is used to prevent unmount across critical
sections like getnewvnode(). It's only ever read locked with rw_tryenter(),
and is only ever write locked in dounmount(). A write hold can't be taken
on this lock if the current LWP could hold a vnode lock.
- kmutex_t mnt_updating. This is taken by threads updating the mount, for
example when going r/o -> r/w, and is only present to serialize updates.
In order to take this lock, a read hold must first be taken on
mnt_unmounting, and the two need to be held across the operation.
One effect of this change: previously if an unmount failed, we would make a
half hearted attempt to back out of it gracefully, but that was unlikely to
work in a lot of cases. Now while an unmount that will be aborted is in
progress, new file operations within the mount will fail instead of being
delayed. That is unlikely to be a problem though, because if the admin
requests unmount of a file system then s(he) has made a decision to deny
access to the resource.
The previous fix worked, but it opened a window where mounts could have
disappeared from mountlist while the caller was traversing it using
vfs_trybusy(). Fix that.
The symptom was that sometimes file systems would occasionally not appear
in output from 'df' or 'mount' if the system was busy. Resolution:
- Make mount locks work somewhat like vm_map locks.
- vfs_trybusy() now only fails if the mount is gone, or if someone is
unmounting the file system. Simple contention on mnt_lock doesn't
cause it to fail.
- vfs_busy() will wait even if the file system is being unmounted.
we no longer need to guard against access from hardware interrupt handlers.
Additionally, if cloning a process with CLONE_SIGHAND, arrange to have the
child process share the parent's lock so that signal state may be kept in
sync. Partially addresses PR kern/37437.
proclist_mutex and proclist_lock into a single adaptive mutex (proc_lock).
Implications:
- Inspecting process state requires thread context, so signals can no longer
be sent from a hardware interrupt handler. Signal activity must be
deferred to a soft interrupt or kthread.
- As the proc state locking is simplified, it's now safe to take exit()
and wait() out from under kernel_lock.
- The system spends less time at IPL_SCHED, and there is less lock activity.
initialization/finalization of snapshot private data on creation/deletion
of struct ufsmount.
Snapshot mounts no longer may fail silently because kmem_alloc() fails.
Welcome to 4.99.60
Ok: Andrew Doran <ad@netbsd.org>
pushing the syncer before considering rate limiting the deletes. We hold
vnodes locked and it's likely that the syncer will try to lock them while
flushing, leading to the syncer and remover proceeding in lockstep and
making very little forward progress. XXX this is not a solution.
- Reference count the mfsnode to fix an aincent bug. Only destroy when
reference count drops to zero. In mfs_start(), busy the mount and get
a reference to the mfsnode to prevent it disappearing while the server
is running. If the file system is gone already, vfs_busy() will fail.
- Always destroy the bufq.
- Use a global mfs_lock for simplicity.
- Replace use of malloc/free. Fixes broken MALLOC_TYPE change.
- Initialize si_vnlock in si_mount_init().
- Also initialize vl_recursecnt to zero.
- Destroy it only in si_mount_dtor().
- Simplify the v_lock <-> si_vnlock exchange.
- Don't abuse the overall error variable for LK_NOWAIT errors.
- ffs_snapremove: release the vnode one instead of three times.
- Do reference counting for 'struct mount'. Each vnode associated with a
mount takes a reference, and in turn the mount takes a reference to the
vfsops.
- Now that mounts are reference counted, replace the overcomplicated mount
locking inherited from 4.4BSD with a recursable rwlock.
