Seigo Tanimura (tanimura) posted the initial delta.
I've polished it quite a bit reducing the need for locking and
adapting it for KSE.
Locks:
1 mutex in each filedesc
protects all the fields.
protects "struct file" initialization, while a struct file
is being changed from &badfileops -> &pipeops or something
the filedesc should be locked.
1 mutex in each struct file
protects the refcount fields.
doesn't protect anything else.
the flags used for garbage collection have been moved to
f_gcflag which was the FILLER short, this doesn't need
locking because the garbage collection is a single threaded
container.
could likely be made to use a pool mutex.
1 sx lock for the global filelist.
struct file * fhold(struct file *fp);
/* increments reference count on a file */
struct file * fhold_locked(struct file *fp);
/* like fhold but expects file to locked */
struct file * ffind_hold(struct thread *, int fd);
/* finds the struct file in thread, adds one reference and
returns it unlocked */
struct file * ffind_lock(struct thread *, int fd);
/* ffind_hold, but returns file locked */
I still have to smp-safe the fget cruft, I'll get to that asap.
against VM_WAIT in the pageout code. Both fixes involve adjusting
the lockmgr's timeout capability so locks obtained with timeouts do not
interfere with locks obtained without a timeout.
Hopefully MFC: before the 4.5 release
- crhold() returns a reference to the ucred whose refcount it bumps.
- crcopy() now simply copies the credentials from one credential to
another and has no return value.
- a new crshared() primitive is added which returns true if a ucred's
refcount is > 1 and false (0) otherwise.
Note ALL MODULES MUST BE RECOMPILED
make the kernel aware that there are smaller units of scheduling than the
process. (but only allow one thread per process at this time).
This is functionally equivalent to teh previousl -current except
that there is a thread associated with each process.
Sorry john! (your next MFC will be a doosie!)
Reviewed by: peter@freebsd.org, dillon@freebsd.org
X-MFC after: ha ha ha ha
systems were repo-copied from sys/miscfs to sys/fs.
- Renamed the following file systems and their modules:
fdesc -> fdescfs, portal -> portalfs, union -> unionfs.
- Renamed corresponding kernel options:
FDESC -> FDESCFS, PORTAL -> PORTALFS, UNION -> UNIONFS.
- Install header files for the above file systems.
- Removed bogus -I${.CURDIR}/../../sys CFLAGS from userland
Makefiles.
vm_mtx does not recurse and is required for most low level
vm operations.
faults can not be taken without holding Giant.
Memory subsystems can now call the base page allocators safely.
Almost all atomic ops were removed as they are covered under the
vm mutex.
Alpha and ia64 now need to catch up to i386's trap handlers.
FFS and NFS have been tested, other filesystems will need minor
changes (grabbing the vm lock when twiddling page properties).
Reviewed (partially) by: jake, jhb
vn_start_write() on the given vnode will be successful. VOP_LEASE() may
help to solve this problem, but its return value ignored nearly everywhere.
For now just assume that the missing upper layer on write means insufficient
access rights (which is correct for most cases).
the number of references on the filesystem root vnode to be both
expected and released. Many filesystems hold an extra reference on
the filesystem root vnode, which must be accounted for when
determining if the filesystem is busy and then released if it isn't
busy. The old `skipvp' approach required individual filesystem
xxx_unmount functions to re-implement much of vflush()'s logic to
deal with the root vnode.
All 9 filesystems that hold an extra reference on the root vnode
got the logic wrong in the case of forced unmounts, so `umount -f'
would always fail if there were any extra root vnode references.
Fix this issue centrally in vflush(), now that we can.
This commit also fixes a vnode reference leak in devfs, which could
result in idle devfs filesystems that refuse to unmount.
Reviewed by: phk, bp
other "system" header files.
Also help the deprecation of lockmgr.h by making it a sub-include of
sys/lock.h and removing sys/lockmgr.h form kernel .c files.
Sort sys/*.h includes where possible in affected files.
OK'ed by: bde (with reservations)
Make 7 filesystems which don't really know about VOP_BMAP rely
on the default vector, rather than more or less complete local
vop_nopbmap() implementations.
An initial tidyup of the mount() syscall and VFS mount code.
This code replaces the earlier work done by jlemon in an attempt to
make linux_mount() work.
* the guts of the mount work has been moved into vfs_mount().
* move `type', `path' and `flags' from being userland variables into being
kernel variables in vfs_mount(). `data' remains a pointer into
userspace.
* Attempt to verify the `type' and `path' strings passed to vfs_mount()
aren't too long.
* rework mount() and linux_mount() to take the userland parameters
(besides data, as mentioned) and pass kernel variables to vfs_mount().
(linux_mount() already did this, I've just tidied it up a little more.)
* remove the copyin*() stuff for `path'. `data' still requires copyin*()
since its a pointer into userland.
* set `mount->mnt_statf_mntonname' in vfs_mount() rather than in each
filesystem. This variable is generally initialised with `path', and
each filesystem can override it if they want to.
* NOTE: f_mntonname is intiailised with "/" in the case of a root mount.
<sys/proc.h> to <sys/systm.h>.
Correctly document the #includes needed in the manpage.
Add one now needed #include of <sys/systm.h>.
Remove the consequent 48 unused #includes of <sys/proc.h>.
Add lockdestroy() and appropriate invocations, which corresponds to
lockinit() and must be called to clean up after a lockmgr lock is no
longer needed.
the gating of system calls that cause modifications to the underlying
filesystem. The gating can be enabled by any filesystem that needs
to consistently suspend operations by adding the vop_stdgetwritemount
to their set of vnops. Once gating is enabled, the function
vfs_write_suspend stops all new write operations to a filesystem,
allows any filesystem modifying system calls already in progress
to complete, then sync's the filesystem to disk and returns. The
function vfs_write_resume allows the suspended write operations to
begin again. Gating is not added by default for all filesystems as
for SMP systems it adds two extra locks to such critical kernel
paths as the write system call. Thus, gating should only be added
as needed.
Details on the use and current status of snapshots in FFS can be
found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness
is not included here. Unless and until you create a snapshot file,
these changes should have no effect on your system (famous last words).
<sys/bio.h>.
<sys/bio.h> is now a prerequisite for <sys/buf.h> but it shall
not be made a nested include according to bdes teachings on the
subject of nested includes.
Diskdrivers and similar stuff below specfs::strategy() should no
longer need to include <sys/buf.> unless they need caching of data.
Still a few bogus uses of struct buf to track down.
Repocopy by: peter
field in struct buf: b_iocmd. The b_iocmd is enforced to have
exactly one bit set.
B_WRITE was bogusly defined as zero giving rise to obvious coding
mistakes.
Also eliminate the redundant struct buf flag B_CALL, it can just
as efficiently be done by comparing b_iodone to NULL.
Should you get a panic or drop into the debugger, complaining about
"b_iocmd", don't continue. It is likely to write on your disk
where it should have been reading.
This change is a step in the direction towards a stackable BIO capability.
A lot of this patch were machine generated (Thanks to style(9) compliance!)
Vinum users: Greg has not had time to test this yet, be careful.
is an application space macro and the applications are supposed to be free
to use it as they please (but cannot). This is consistant with the other
BSD's who made this change quite some time ago. More commits to come.
* lockstatus() and VOP_ISLOCKED() gets a new process argument and a new
return value: LK_EXCLOTHER, when the lock is held exclusively by another
process.
* The ASSERT_VOP_(UN)LOCKED family is extended to use what this gives them
* Extend the vnode_if.src format to allow more exact specification than
locked/unlocked.
This commit should not do any semantic changes unless you are using
DEBUG_VFS_LOCKS.
Discussed with: grog, mch, peter, phk
Reviewed by: peter
Note: Previous commit to these files (except coda_vnops and devfs_vnops)
that claimed to remove WILLRELE from VOP_RENAME actually removed it from
VOP_MKNOD.
Merge the contents (less some trivial bordering the silly comments)
of <vm/vm_prot.h> and <vm/vm_inherit.h> into <vm/vm.h>. This puts
the #defines for the vm_inherit_t and vm_prot_t types next to their
typedefs.
This paves the road for the commit to follow shortly: change
useracc() to use VM_PROT_{READ|WRITE} rather than B_{READ|WRITE}
as argument.
a lower layer to an upper layer. I'm not sure how necessary this is
for reading.
Fix bug in union_lookup() (note: there are probably still several bugs
in union_lookup()). This one set lerror as a side effect without
setting lowervp, causing copyup code further on down to crash on a null
lowervp pointer. Changed the side effect to use a temporary variable
instead.
fixed (many due to changing semantics in other parts of the kernel and not
the original author's fault), including one critical one: unionfs could
cause UFS corruption in the fronting store due to calling VOP_OPEN for
writing without turning on vmio for the UFS vnode.
Most of the bugs were related to semantics changes in VOP calls, lock
ordering problems (causing deadlocks), improper handling of a read-only
backing store (such as an NFS mount), improper referencing and locking
of vnodes, not using real struct locks for vnode locking, not using
recursive locks when accessing the fronting store, and things like that.
New functionality has been added: unionfs now has mmap() support, but
only partially tested, and rename has been enhanced considerably.
There are still some things that unionfs cannot do. You cannot
rename a directory without confusing unionfs, and there are issues
with softlinks, hardlinks, and special files. unionfs mostly doesn't
understand them (and never did).
There are probably still panic situations, but hopefully no where near
as many as before this commit.
The unionfs in this commit has been tested overlayed on /usr/src
(backing /usr/src being a read-only NFS mount, fronting /usr/src being
a local filesystem). kernel builds have been tested, buildworld is
undergoing testing. More testing is necessary.
reasonable defaults.
This avoids confusing and ugly casting to eopnotsupp or making dummy functions.
Bogus casting of filesystem sysctls to eopnotsupp() have been removed.
This should make *_vfsops.c more readable and reduce bloat.
Reviewed by: msmith, eivind
Approved by: phk
Tested by: Jeroen Ruigrok/Asmodai <asmodai@wxs.nl>
format errors exposed by this. It has nothing to do with diagnostics
since it does little more than control tracing of normal operation.
Actual diagnostics for the union file system are still controlled by
the DIAGNOSTIC option.
references to them.
The change a couple of days ago to ignore these numbers in statically
configured vfsconf structs was slightly premature because the cd9660,
cfs, devfs, ext2fs, nfs vfs's still used MOUNT_* instead of the number
in their vfsconf struct.
as the value in b_vp is often not really what you want.
(and needs to be frobbed). more cleanups will follow this.
Reviewed by: Bruce Evans <bde@freebsd.org>
---------
Make callers of namei() responsible for releasing references or locks
instead of having the underlying filesystems do it. This eliminates
redundancy in all terminal filesystems and makes it possible for stacked
transport layers such as umapfs or nullfs to operate correctly.
Quality testing was done with testvn, and lat_fs from the lmbench suite.
Some NFS client testing courtesy of Patrik Kudo.
vop_mknod and vop_symlink still release the returned vpp. vop_rename
still releases 4 vnode arguments before it returns. These remaining cases
will be corrected in the next set of patches.
---------
Submitted by: Michael Hancock <michaelh@cet.co.jp>
Reverse the VFS_VRELE patch. Reference counting of vnodes does not need
to be done per-fs. I noticed this while fixing vfs layering violations.
Doing reference counting in generic code is also the preference cited by
John Heidemann in recent discussions with him.
The implementation of alternative vnode management per-fs is still a valid
requirement for some filesystems but will be revisited sometime later,
most likely using a different framework.
Submitted by: Michael Hancock <michaelh@cet.co.jp>
a complement to all ops that return a vpp, VFS_VRELE. This is
initially only for file systems that implement the following ops
that do a WILLRELE:
vop_create, vop_whiteout, vop_mknod, vop_remove, vop_link,
vop_rename, vop_mkdir, vop_rmdir, vop_symlink
This is initial DNA that doesn't do anything yet. VFS_VRELE is
implemented but not called.
A default vfs_vrele was created for fs implementations that use the
standard vnode management routines.
VFS_VRELE implementations were made for the following file systems:
Standard (vfs_vrele)
ffs mfs nfs msdosfs devfs ext2fs
Custom
union umapfs
Just EOPNOTSUPP
fdesc procfs kernfs portal cd9660
These implementations may change as VOP changes are implemented.
In the next phase, in the vop implementations calls to vrele and the vrele
part of vput will be moved to the top layer vfs_vnops and made visible
to all layers. vput will be replaced by unlock in these cases. Unlocking
will still be done in the per fs layer but the refcount decrement will be
triggered at the top because it doesn't hurt to hold a vnode reference a
little longer. This will have minimal impact on the structure of the
existing code.
This will only be done for vnode arguments that are released by the various
fs vop implementations.
Wider use of VFS_VRELE will likely require restructuring of the code.
Reviewed by: phk, dyson, terry et. al.
Submitted by: Michael Hancock <michaelh@cet.co.jp>
- Set UN_ULOCK in union_lock() when UN_KLOCK is set. Caller expects
that vnode is locked correctly, and may call another function which
expects locked vnode and may unlock the vnode.
- Do not assume the behavior of inside functions in FreeBSD's
vfs_suber.c is same as 4.4BSD-Lite2. Vnode may be locked in
vget() even though flag is zero. (Locked vnode is, of course,
unlocked before returning from vget.)
Rename vn_default_error to vop_defaultop all over the place.
Move vn_bwrite from vfs_bio.c to vfs_default.c and call it vop_stdbwrite.
Use vop_null instead of nullop.
Move vop_nopoll from vfs_subr.c to vfs_default.c
Move vop_sharedlock from vfs_subr.c to vfs_default.c
Move vop_nolock from vfs_subr.c to vfs_default.c
Move vop_nounlock from vfs_subr.c to vfs_default.c
Move vop_noislocked from vfs_subr.c to vfs_default.c
Use vop_ebadf instead of *_ebadf.
Add vop_defaultop for getpages on master vnode in MFS.
1. Remove VOP_UPDATE, it is (also) an UFS/{FFS,LFS,EXT2FS,MFS}
intereface function, and now lives in the ufsmount structure.
2. Remove VOP_SEEK, it was unused.
3. Add mode default vops:
VOP_ADVLOCK vop_einval
VOP_CLOSE vop_null
VOP_FSYNC vop_null
VOP_IOCTL vop_enotty
VOP_MMAP vop_einval
VOP_OPEN vop_null
VOP_PATHCONF vop_einval
VOP_READLINK vop_einval
VOP_REALLOCBLKS vop_eopnotsupp
And remove identical functionality from filesystems
4. Add vop_stdpathconf, which returns the canonical stuff. Use
it in the filesystems. (XXX: It's probably wrong that specfs
and fifofs sets this vop, shouldn't it come from the "host"
filesystem, for instance ufs or cd9660 ?)
5. Try to make system wide VOP functions have vop_* names.
6. Initialize the um_* vectors in LFS.
(Recompile your LKMS!!!)
1. Remove comment stating the blatantly obvious.
2. Align in two columns.
3. Sort all but the default element alphabetically.
4. Remove XXX comments pointing out entries not needed.
Distribute all but the most fundamental malloc types. This time I also
remembered the trick to making things static: Put "static" in front of
them.
A couple of finer points by: bde
1. Clustered I/O is switched by the MNT_NOCLUSTERR and MNT_NOCLUSTERW
bits of the mnt_flag. The sysctl variables, vfs.foo.doclusterread
and vfs.foo.doclusterwrite are deleted. Only mount option can
control clustered I/O from userland.
2. When foofs_mount mounts block device, foofs_mount checks D_CLUSTERR
and D_CLUSTERW bits of the d_flags member in the block device switch
table. If D_NOCLUSTERR / D_NOCLUSTERW are set, MNT_NOCLUSTERR /
MNT_NOCLUSTERW bits will be set. In this case, MNT_NOCLUSTERR and
MNT_NOCLUSTERW cannot be cleared from userland.
3. Vnode driver disables both clustered read and write.
4. Union filesystem disables clutered write.
Reviewed by: bde
plus the previous changes to use the zone allocator decrease the useage
of malloc by half. The Zone allocator will be upgradeable to be able
to use per CPU-pools, and has more intelligent usage of SPLs. Additionally,
it has reasonable stats gathering capabilities, while making most calls
inline.
socket addresses in mbufs. (Socket buffers are the one exception.) A number
of kernel APIs needed to get fixed in order to make this happen. Also,
fix three protocol families which kept PCBs in mbufs to not malloc them
instead. Delete some old compatibility cruft while we're at it, and add
some new routines in the in_cksum family.
uerror == 0 && lerror == EACCES, lowervp == NULLVP and union_allocvp
doesn't find existing union node and new union node is created.
Sicne it is dificult to cover all the case, union_lookup always
returns when union_lookup1() returns EACCES.
Submitted by: Naofumi Honda <honda@Kururu.math.sci.hokudai.ac.jp>
Obtained from: NetBSD/pc98
in savedvp variable and it is used for the argument of
MOUNTTOUNIONMOUNT(). I didn't realize ap->a_vp is modified before
MOUNTTOUNIONMOUNT(), so the change by revision 1.22 is incorrect.
UN_KLOCK flag.
When UN_KLOCK is set, VOP_UNLOCK should keep uppervp locked and clear
UN_ULOCK flag. To do this, when UN_KLOCK is set, (1) union_unlock
clears UN_ULOCK and does not clear UN_KLOCK, (2) union_lock() does not
access uppervp and does not clear UN_KLOCK, and (3) callers of
vput/VOP_UNLOCK should clear UN_KLOCK. For example, vput becomes:
SETKLOCK(union_node);
vput(vnode);
CLEARKLOCK(union_node);
where SETKLOCK macro sets UN_KLOCK and CLEARKLOCK macro clears
UN_KLOCK.
Our vput calls vm_object_deallocate() --> vm_object_terminate(). The
vm_object_terminate() calls vn_lock(), since UN_LOCKED has been
already cleared in union_unlock(). Then, union_lock locks upper vnode
when UN_ULOCK is not set. The upper vnode is not unlocked when
UN_KLOCK is set in union_unlock(), thus, union_lock tries to lock
locked vnode and we get panic.
UN_ULOCK flag. This shows a locking violation but I couldn't find the
reason UN_ULOCK is not set or upper vnode is not unlocked. I added
the code that detect this case and adjust un_flags. DIAGNOSTIC kernel
doesn't adjust un_flags, but just panic here to help debug by kernel
hackers.
# mount -t union (or null) dir1 dir2
# mount -t union (or null) dir2 dir1
The function namei in union_mount calls union_root. The upper vnode
has been already locked and vn_lock in union_root causes above panic.
Add printf's included in `#ifdef DIAGNOSTIC' for EDEADLK cases.
is NULLVP, union node will have neither uppervp nor lowervp. This
causes page fault trap.
The union_removed_upper just remove union node from cache and it
doesn't set uppervp to NULLVP. Since union node is removed from
cache, it will not be referenced.
The code that remove union node from cache was copied from
union_inactive.
VOP_LINK(). The reason of strange behavior was wrong order of the
argument, that is, the operation
# ln foo bar
in a union fs tried to do
# ln bar foo
in ufs layer.
Now we can make a link in a union fs.
fix!
The ufs_link() assumes that vnode is not unlocked and tries to lock it
in certain case. Because union_link calls VOP_LINK after locking vnode,
vn_lock in ufs_link causes above panic.
Currently, I don't know the real fix for a locking violation in
union_link, but I think it is important to avoid panic.
A vnode is unlocked before calling VOP_LINK and is locked after it if
the vnode is not union fs. Even though panic went away, the process
that access the union fs in which link was made will hang-up.
Hang-up can be easily reproduced by following operation:
mount -t union a b
cd b
ln foo bar
ls
same directory pair.
If we do:
mount -t union a b
mount -t union a b
then, (1) namei tries to lock fs which has been already locked by
first union mount and (2) union_root() tries to lock locked fs. To
avoid first deadlock condition, unlock vnode if lowerrootvp is union
node, and to avoid second case, union_mount returns EDEADLK when multi
union mount is detected.
dolock is not set (that is, targetvp == overlaying vnode object).
Current code use FIXUP macro to do this, and never unlocks overlaying
vnode object in union_fsync. So, the vnode object will be locked
twice and never unlocked.
PR: 3271
Submitted by: kato
relookup() in union_relookup() is succeeded. However, if relookup()
returns non-zero value, that is relookup fails, VOP_MKDIR is never
called (c.f. union_mkshadow). Thus, pathname buffer is never FREEed.
Reviewed by: phk
Submitted by: kato
PR: 3262
changes, so don't expect to be able to run the kernel as-is (very well)
without the appropriate Lite/2 userland changes.
The system boots and can mount UFS filesystems.
Untested: ext2fs, msdosfs, NFS
Known problems: Incorrect Berkeley ID strings in some files.
Mount_std mounts will not work until the getfsent
library routine is changed.
Reviewed by: various people
Submitted by: Jeffery Hsu <hsu@freebsd.org>
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
it 1138 times (:-() in casts and a few more times in declarations.
This change is null for the i386.
The type has to be `typedef int vop_t(void *)' and not `typedef
int vop_t()' because `gcc -Wstrict-prototypes' warns about the
latter. Since vnode op functions are called with args of different
(struct pointer) types, neither of these function types is any use
for type checking of the arg, so it would be preferable not to use
the complete function type, especially since using the complete
type requires adding 1138 casts to avoid compiler warnings and
another 40+ casts to reverse the function pointer conversions before
calling the functions.
calls.
Found by: gcc -Wstrict-prototypes after I supplied some of the 5000+
missing prototypes. Now I have 9000+ lines of warnings and errors
about bogus conversions of function pointers.
wrong vp's ops vector being used by changing the VOP_LINK's argument order.
The special-case hack doesn't go far enough and breaks the generic
bypass routine used in some non-leaf filesystems. Pointed out by Kirk
McKusick.
VFCF_NETWORK (this FS goes over the net)
VFCF_READONLY (read-write mounts do not make any sense)
VFCF_SYNTHETIC (data in this FS is not real)
VFCF_LOOPBACK (this FS aliases something else)
cd9660 is readonly; nullfs, umapfs, and union are loopback; NFS is netowkr;
procfs, kernfs, and fdesc are synthetic.
Find enclosed a short bugfix to get the union filesystem up and running
in FreeBSD-current. We don't think we've got all the problems yet but
these fixes sort out the major ones (which mostly concert bad locking
of vnodes), no doubt we'll post others as necessary. Known problems
include the inability of the umount command (not the system call) to unmount
unions in certain circumstances (this is due the way "realpath" works),
and the failure of direntries to always get all available files in
unioned subdirectories. We are, as they say, working on it.
Submitted by: tim@cs.city.ac.uk (Tim Wilkinson)
- Make a number of filesystems work again when they are statically compiled
(blush)
- FIFOs are no longer optional; ``options FIFO'' removed from distributed
config files.
- set args->lkm_offset correctly so that VFS modules can be unloaded
- initialize _fs_vfsops.vfc_refcount correctly so that VFS modules can
be unloaded
- include kernel.h in a few placves to get the correct definition of DATA_SET