- In the ISDOTDOT case we have to unlock the dvp before locking the child,
if this fails we must relock dvp before returning an error. This was
missing before.
Sponsored by: Isilon Systems, Inc.
- Network filesystems are written with a special idiom that checks the
cache first, and may even unlock dvp before discovering that a network
round-trip is required to resolve the name. I believe dvp is prevented
from being recycled even in the forced unmount case by the shared lock
on the mount point. If not, this code should grow checks for VI_DOOMED
after it relocks dvp or it will access NULL v_data fields.
Sponsored by: Isilon Systems, Inc.
these filesystems will support shared locks until they are explicitly
modified to do so. Careful review must be done to ensure that this
is safe for each individual filesystem.
Sponsored by: Isilon Systems, Inc.
these filesystems will support shared locks until they are explicitly
modified to do so. Careful review must be done to ensure that this
is safe for each individual filesystem.
Sponsored by: Isilon Systems, Inc.
that they set v->v_vnlock. This is true for all filesystems in the
tree.
- Remove all uses of LK_THISLAYER. If the lower layer is locked, the
null layer is locked. We only use vget() to get a reference now.
null essentially does no locking. This fixes LOOKUP_SHARED with
nullfs.
- Remove the special LK_DRAIN considerations, I do not believe this is
needed now as LK_DRAIN doesn't destroy the lower vnode's lock, and
it's hardly used anymore.
- Add one well commented hack to prevent the lowervp from going away
while we're in it's VOP_LOCK routine. This can only happen if we're
forcibly unmounted while some callers are waiting in the lock. In
this case the lowervp could be recycled after we drop our last ref
in null_reclaim(). Prevent this with a vhold().
as suggested by Matt's comment. Also fix some style and paranoia issues.
The entire function could benefit from review by a VFS guru.
MFC after: 6 weeks
Since we used an sbuf of size resid to accumulate dirents, we would end
up returning one byte short when we had enough dirents to fill or exceed
the size of the sbuf (the last byte being lost to bogus NUL termination)
causing the next call to return EINVAL due to an unaligned offset. This
went undetected for a long time because I did most of my testing in
single-user mode, where there are rarely enough processes to fill the
4096-byte buffer ls(1) uses. The most common symptom of this bug is that
tab completion of /proc or /compat/linux/proc does not work properly when
many processes are running.
Also, a check near the top would return EINVAL if resid was smaller than
PFS_DELEN, even if it was 0, which is frequently the case and perfectly
allowable. Change the test so that it returns 0 if resid is 0.
MFC after: 2 weeks
the filesystem. Check that rather than VI_XLOCK.
- VOP_INACTIVE should no longer drop the vnode lock.
- The vnode lock is required around calls to vrecycle() and vgone().
Sponsored by: Isilon Systems, Inc.
vnode lock. Remove the c_lock and use the vn lock in its place.
- Keep the coda lock functions so that the debugging information is
preserved, but call directly to the vop_std*lock routines for the real
functionality.
Sponsored by: Isilon Systems, Inc.
long filename. Each substring is indexed by the windows ID, a
sequential one-based value. The previous code was extremely slow,
doing a malloc/strcpy/free for each substring.
This code optimizes these routines with this in mind, using the ID
to index into a single array and concatenating each WIN_CHARS chunk
at once. (The last chunk is variable-length.)
This code has been tested as working on an FS with difficult filename
sizes (255, 13, 26, etc.) It gives a 77.1% decrease in profiled
time (total across all functions) and a 73.7% decrease in wall time.
Test was "ls -laR > /dev/null".
Per-function time savings:
mbnambuf_init: -90.7%
mbnambuf_write: -18.7%
mbnambuf_flush: -67.1%
MFC after: 1 month
List devfs_dirents rather than vnodes off their shared struct cdev, this
saves a pointer field in the vnode at the expense of a field in the
devfs_dirent. There are often 100 times more vnodes so this is bargain.
In addition it makes it harder for people to try to do stypid things like
"finding the vnode from cdev".
Since DEVFS handles all VCHR nodes now, we can do the vnode related
cleanup in devfs_reclaim() instead of in dev_rel() and vgonel().
Similarly, we can do the struct cdev related cleanup in dev_rel()
instead of devfs_reclaim().
rename idestroy_dev() to destroy_devl() for consistency.
Add LIST_ENTRY de_alias to struct devfs_dirent.
Remove v_specnext from struct vnode.
Change si_hlist to si_alist in struct cdev.
String new devfs vnodes' devfs_dirent on si_alist when
we create them and take them off in devfs_reclaim().
Fix devfs_revoke() accordingly. Also don't clear fields
devfs_reclaim() will clear when called from vgone();
Let devfs_reclaim() call dev_rel() instead of vgonel().
Move the usecount tracking from dev_rel() to devfs_reclaim(),
and let dev_rel() take a struct cdev argument instead of vnode.
Destroy SI_CHEAPCLONE devices in dev_rel() (instead of
devfs_reclaim()) when they are no longer used. (This
should maybe happen in devfs_close() instead.)
mounted (is it Joliet, RockRidge, High Sierra) based on bootverbose.
Most file systems don't generate log messages based on details of the
file system superblock, and these log messages disrupt sysinstall output
during a new install from CD. We may want to explore exposing this
status information using nmount() at some point.
MFC after: 3 days
on my P3, microbenchmarks show the unrolled version is 78x faster. In
actual use (recursive ls), this gives an average of 9% improvement in
system time and 2% improvement in wall time.
called in "open", causing mmap() to fail.
Where possible, pass size of file to vnode_create_vobject() rather
than having it find it out the hard way via VOP_LOOKUP
Reviewed by: phk
mmap() on NTFS files was hosed, returning pages offset from the
start of the disk rather than the start of the file. (ie, "cp" of
a 1-block file would get you a copy of the boot sector, not the
data in the file.) The solution isn't ideal, but gives a functioning
filesystem.
Cached vnode lookup was also broken, resulting in vnode haemorrhage.
A lookup on the same file twice would give you two vnodes, and the
resulting cached pages.
Just recently, mmap() was broken due to a lack of a call to
vnode_create_vobject() in ntfs_open().
Discussed with: phk@
with NFS.
We are moving responsibility for creating the vnode_pager object into
the filesystems which own the vnode, and this is one of the places
we have to cover.
We call vnode_create_vobject() directly because we own the vnode.
If we can get the size easily, pass it as an argument to save the
call to VOP_GETATTR() in vnode_create_vobject()
and KASSERT coverage.
After this check there is only one "nasty" cast in this code but there
is a KASSERT to protect against the wrong argument structure behind
that cast.
Un-inlining the meat of VOP_FOO() saves 35kB of text segment on a typical
kernel with no change in performance.
We also now run the checking and tracing on VOP's which have been layered
by nullfs, umapfs, deadfs or unionfs.
Add new (non-inline) VOP_FOO_AP() functions which take a "struct
foo_args" argument and does everything the VOP_FOO() macros
used to do with checks and debugging code.
Add KASSERT to VOP_FOO_AP() check for argument type being
correct.
Slim down VOP_FOO() inline functions to just stuff arguments
into the struct foo_args and call VOP_FOO_AP().
Put function pointer to VOP_FOO_AP() into vop_foo_desc structure
and make VCALL() use it instead of the current offsetoff() hack.
Retire vcall() which implemented the offsetoff()
Make deadfs and unionfs use VOP_FOO_AP() calls instead of
VCALL(), we know which specific call we want already.
Remove unneeded arguments to VCALL() in nullfs and umapfs bypass
functions.
Remove unused vdesc_offset and VOFFSET().
Generally improve style/readability of the generated code.
I'm not sure why a credential was added to these in the first place, it is
not used anywhere and it doesn't make much sense:
The credentials for syncing a file (ability to write to the
file) should be checked at the system call level.
Credentials for syncing one or more filesystems ("none")
should be checked at the system call level as well.
If the filesystem implementation needs a particular credential
to carry out the syncing it would logically have to the
cached mount credential, or a credential cached along with
any delayed write data.
Discussed with: rwatson
After disscussing things I have decided to take the easy and
consistent 90% solution instead of aiming for the very involved 99%
solution.
If we allow forceful unmounts of DEVFS we need to decide how to handle
the devices which are in use through this filesystem at the time.
We cannot just readopt the open devices in the main /dev instance since
that would open us to security issues.
For the majority of the devices, this is relatively straightforward
as we can just pretend they got revoke(2)'ed.
Some devices get tricky: /dev/console and /dev/tty for instance
does a sort of recursive open of the real console device. Other devices
may be mmap'ed (kill the processes ?).
And then there are disk devices which are mounted.
The correct thing here would be to recursively unmount the filesystems
mounte from devices from our DEVFS instance (forcefully) and if
this succeeds, complete the forcefully unmount of DEVFS. But if
one of the forceful unmounts fail we cannot complete the forceful
unmount of DEVFS, but we are likely to already have severed a lot
of stuff in the process of trying.
Event attempting this would be a lot of code for a very far out
corner-case which most people would never see or get in touch with.
It's just not worth it.
methods:
Read can see O_NONBLOCK and O_DIRECT.
Write can see O_NONBLOCK, O_DIRECT and O_FSYNC.
In addition O_DIRECT is shadowed as IO_DIRECT for now for backwards
compatibility.
fcntl.h.
This is in preparation for making the flags passed to device drivers be
consistently from fcntl.h for all entrypoints.
Today open, close and ioctl uses fcntl.h flags, while read and write
uses vnode.h flags.
split the conversion of the remaining three filesystems out from the root
mounting changes, so in one go:
cd9660:
Convert to nmount.
Add omount compat shims.
Remove dedicated rootfs mounting code.
Use vfs_mountedfrom()
Rely on vfs_mount.c calling VFS_STATFS()
nfs(client):
Convert to nmount (the simple way, mount_nfs(8) is still necessary).
Add omount compat shims.
Drop COMPAT_PRELITE2 mount arg compatibility.
ffs:
Convert to nmount.
Add omount compat shims.
Remove dedicated rootfs mounting code.
Use vfs_mountedfrom()
Rely on vfs_mount.c calling VFS_STATFS()
Remove vfs_omount() method, all filesystems are now converted.
Remove MNTK_WANTRDWR, handling RO/RW conversions is a filesystem
task, and they all do it now.
Change rootmounting to use DEVFS trampoline:
vfs_mount.c:
Mount devfs on /. Devfs needs no 'from' so this is clean.
symlink /dev to /. This makes it possible to lookup /dev/foo.
Mount "real" root filesystem on /.
Surgically move the devfs mountpoint from under the real root
filesystem onto /dev in the real root filesystem.
Remove now unnecessary getdiskbyname().
kern_init.c:
Don't do devfs mounting and rootvnode assignment here, it was
already handled by vfs_mount.c.
Remove now unused bdevvp(), addaliasu() and addalias(). Put the
few necessary lines in devfs where they belong. This eliminates the
second-last source of bogo vnodes, leaving only the lemming-syncer.
Remove rootdev variable, it doesn't give meaning in a global context and
was not trustworth anyway. Correct information is provided by
statfs(/).
Same comment as msdosfs applies: It would be nice if we had generic option
names for charset conversions.
Use vfs_mountefrom(). Rely on vfs_mount.c calling VFS_STATFS().
Add a vfs_cmount() function which converts omount argument stucture
to nmount arguments.
Convert vfs_omount() to vfs_mount() and parse nmount arguments.
This is 100% compatible with existing userland.
Later on, but before userland gets converted to nmount we may want
to revisit the names of the mountoptions, for instance it may make
sense to use consistent options for charset conversion etc.
doesn't. Most of the implementations have grown weeds for this so they
copy some fields from mnt_stat if the passed argument isn't that.
Fix this the cleaner way: Always call the implementation on mnt_stat
and copy that in toto to the VFS_STATFS argument if different.
initializations but we did have lofty goals and big ideals.
Adjust to more contemporary circumstances and gain type checking.
Replace the entire vop_t frobbing thing with properly typed
structures. The only casualty is that we can not add a new
VOP_ method with a loadable module. History has not given
us reason to belive this would ever be feasible in the the
first place.
Eliminate in toto VOCALL(), vop_t, VNODEOP_SET() etc.
Give coda correct prototypes and function definitions for
all vop_()s.
Generate a bit more data from the vnode_if.src file: a
struct vop_vector and protype typedefs for all vop methods.
Add a new vop_bypass() and make vop_default be a pointer
to another struct vop_vector.
Remove a lot of vfs_init since vop_vector is ready to use
from the compiler.
Cast various vop_mumble() to void * with uppercase name,
for instance VOP_PANIC, VOP_NULL etc.
Implement VCALL() by making vdesc_offset the offsetof() the
relevant function pointer in vop_vector. This is disgusting
but since the code is generated by a script comparatively
safe. The alternative for nullfs etc. would be much worse.
Fix up all vnode method vectors to remove casts so they
become typesafe. (The bulk of this is generated by scripts)
Null_open() was only here to handle MNT_NODEV, but since that does
not affect any filesystems anymore, it could only have any effect
if you nullfs mounted a devfs but didn't want devices to show up.
If you need that, there are easier ways.
instead of a vnode for it.
The vnode_pager does not and should not have any interest in what
the filesystem uses for backend.
(vfs_cluster doesn't use the backing store argument.)
to a cdev and a devsw, doing all the relevant checks along the way.
Add the check to see if fp->f_vnode->v_rdev differs from our cached
fp->f_data copy of our cdev. If it does the device was revoked and
we return ENXIO.
Use this in all the places where sleeping with the lock held is not
an issue.
The distinction will become significant once we finalize the exact
lock-type to use for this kind of case.
The tunable vfs.devfs.fops controls this feature and defaults to off.
When enabled (vfs.devfs.fops=1 in loader), device vnodes opened
through a filedescriptor gets a special fops vector which instead
of the detour through the vnode layer goes directly to DEVFS.
Amongst other things this allows us to run Giant free read/write to
device drivers which have been weaned off D_NEEDGIANT.
Currently this means /dev/null, /dev/zero, disks, (and maybe the
random stuff ?)
On a 700MHz K7 machine this doubles the speed of
dd if=/dev/zero of=/dev/null bs=1 count=1000000
This roughly translates to shaving 2usec of each read/write syscall.
The poll/kqfilter paths need more work before they are giant free,
this work is ongoing in p4::phk_bufwork
Please test this and report any problems, LORs etc.
buf->b-dev.
Put a bio between the buf passed to dev_strategy() and the device driver
strategy routine in order to not clobber fields in the buf.
Assert copyright on vfs_bio.c and update copyright message to canonical
text. There is no legal difference between John Dysons two-clause
abbreviated BSD license and the canonical text.
We keep si_bsize_phys around for now as that is the simplest way to pull
the number out of disk device drivers in devfs_open(). The correct solution
would be to do an ioctl(DIOCGSECTORSIZE), but the point is probably mooth
when filesystems sit on GEOM, so don't bother for now.
jest, of most excellent fancy: he hath taught me lessons a thousand
times; and now, how abhorred in my imagination it is! my gorge rises
at it. Here were those hacks that I have curs'd I know not how
oft. Where be your kludges now? your workarounds? your layering
violations, that were wont to set the table on a roar?
Move the skeleton of specfs into devfs where it now belongs and
bury the rest.
the raw values including for child process statistics and only compute the
system and user timevals on demand.
- Fix the various kern_wait() syscall wrappers to only pass in a rusage
pointer if they are going to use the result.
- Add a kern_getrusage() function for the ABI syscalls to use so that they
don't have to play stackgap games to call getrusage().
- Fix the svr4_sys_times() syscall to just call calcru() to calculate the
times it needs rather than calling getrusage() twice with associated
stackgap, etc.
- Add a new rusage_ext structure to store raw time stats such as tick counts
for user, system, and interrupt time as well as a bintime of the total
runtime. A new p_rux field in struct proc replaces the same inline fields
from struct proc (i.e. p_[isu]ticks, p_[isu]u, and p_runtime). A new p_crux
field in struct proc contains the "raw" child time usage statistics.
ruadd() has been changed to handle adding the associated rusage_ext
structures as well as the values in rusage. Effectively, the values in
rusage_ext replace the ru_utime and ru_stime values in struct rusage. These
two fields in struct rusage are no longer used in the kernel.
- calcru() has been split into a static worker function calcru1() that
calculates appropriate timevals for user and system time as well as updating
the rux_[isu]u fields of a passed in rusage_ext structure. calcru() uses a
copy of the process' p_rux structure to compute the timevals after updating
the runtime appropriately if any of the threads in that process are
currently executing. It also now only locks sched_lock internally while
doing the rux_runtime fixup. calcru() now only requires the caller to
hold the proc lock and calcru1() only requires the proc lock internally.
calcru() also no longer allows callers to ask for an interrupt timeval
since none of them actually did.
- calcru() now correctly handles threads executing on other CPUs.
- A new calccru() function computes the child system and user timevals by
calling calcru1() on p_crux. Note that this means that any code that wants
child times must now call this function rather than reading from p_cru
directly. This function also requires the proc lock.
- This finishes the locking for rusage and friends so some of the Giant locks
in exit1() and kern_wait() are now gone.
- The locking in ttyinfo() has been tweaked so that a shared lock of the
proctree lock is used to protect the process group rather than the process
group lock. By holding this lock until the end of the function we now
ensure that the process/thread that we pick to dump info about will no
longer vanish while we are trying to output its info to the console.
Submitted by: bde (mostly)
MFC after: 1 month
with different file systems. This may cause ill things
with my previous fix. Now it translate fsid of direct child of
mount point directory only.
Pointed out by: Uwe Doering
frobbing the cdevsw.
In both cases we examine only the cdevsw and it is a good question if we
weren't better off copying those properties into the cdev in the first
place. This question will be revisited.
of the number of threads which are inside whatever is behind the
cdevsw for this particular cdev.
Make the device mutex visible through dev_lock() and dev_unlock().
We may want finer granularity later.
Replace spechash_mtx use with dev_lock()/dev_unlock().
too much kernel copying, but it is not the right way to do it, and it is
in the way for straightening out the buffer cache.
The right way is to pass the VM page array down through the struct
bio to the disk device driver and DMA directly in to/out off the
physical memory. Once the VM/buf thing is sorted out it is next on
the list.
Retire most of vnode method. ffs_getpages(). It is not clear if what is
left shouldn't be in the default implementation which we now fall back to.
Retire specfs_getpages() as well, as it has no users now.
preparation for integration of p4::phk_bufwork. In the future,
local filesystems will talk to GEOM directly and they will consequently
be able to issue BIO_DELETE directly. Since the removal of the fla
driver, BIO_DELETE has effectively been a no-op anyway.
