reference to the parent directory's vnode instead of its smbnode to
avoid a use-after-free bug causing a panic when a smbfs mount is
forcefully unmounted.
Keep trying to flush the vnode list for the mount while some are still
busy and we are making progress towards making them not busy. This
stops attempts to unmount idle smbfs mounts failing with EBUSY.
The easiest way to reproduce the above problem, from what I have seen is:
1) Assume /s is a smbfs mount point.
2) mount /s
3) stat /s/foo/1
4) umount /s
Returns error because the file system is busy.
5) Shutdown the machine: panic in smbfs_reclaim because vrele
accesses already-released memory.
array's contents and returning all the requested pages. Otherwise there
are problems (accessing invalid memory) when the a_m vector is passed
uninitialized as the NFS server code does. Fixes PR kern/34959.
Note that this is not a "real" fix. While this makes tmpfs's getpages
operation consistent with the behavior of other file systems, it does
not resolve the different semantics between uvn_get and uao_get as
described in PR kern/32166. I'm adding a comment in the code mentioning
exactly this so that it can be reviewed when this last problem is
addressed.
it, not the mtime of the file itself. This fixes the problems exposed when
unpacking software under a tmpfs and trying to build it because dependencies
were not calculated properly (e.g. autoconf 2.60 as reported by tls@).
loops where vnodes can get removed or added during the loops. This could
lead to panic's on unmount since nodes are skipped or otherwise
TAILQ_NEXT(0xdeadbeef, ...) was dereferenced.
all waiters *before* trying to get the syncer lock necessary for
dounmount(). This prevents a deadlock if the userspace server dies
while the syncer is running.
getnewvnode() while holding on to any vnode lock deadlocks the
system if the file system is being forcibly unmounted.
Normal file systems don't trigger this problem because of two reaons:
1) they don't hold on to vnode locks while idling who-knows-where, so
the race doesn't trigger
2) they aren't usually unmounted with FORCE; puffs is, in case "someone"
manages to make a crashy userspace server
Nevertheless, a real solution is slowly being braised.
It contains the VFS attachment and userspace message-passing interface.
This work was initially started and completed for Google SoC 2005
and tweaked to work a bit better in the past few weeks. While
being far from complete, it is functional enough to be able and
stable to host a fairly general-purpose in-memory file system in
userspace. Even so, puffs should be considered experimental and
no binary compatibility for interfaces or crash-freedom or zero
security implications should be relied upon just yet.
The GSoC project was mentored by William Studenmund and the final
review for the code was done by Christos.
vnodes were synced and processed backwards. This meant that the last
accessed node was processed first and the earlierst last.
An extra benefit is the removal of the ugly hack from the Berkly days on
LFS.
In the proces, i've also replaced the various variations hand written loops
by the TAILQ_FOREACH() macro's.
the udf_verbose variable. So when something goes wrong, it can be examined
on the spot without needing to reboot a new kernel and possibly loosing
state.
. get rid of struct adirent which didn't match struct dirent anymore
. fix cookies, move all the code handling them to the end of the function
Includes many minor changes to the code of this function.
"Add a 3rd entry in the cache, which keeps the end position
from just before extending a file.
This has the desired effect of keeping the write speed constant."
And yes, that helps a lot copying large files... always at full speed
now. This closes my PR kern/30868 "Poor performance copying large files
on msdosfs".
Also remove a 2 if-statements testing the same condition, combine them.
All that from Rhialto, thank you very much.
If you perform this request on a directory with exactly 50 files
(plus '.' and '..' which brings the total to 52 objects), the first
reply for the SMB server completely satisfies the query (server
side is Windows 2000 Professional).
The smbfs client then performs a TRANS2_FIND_NEXT2 using the last
file name as the resume key. The response returns a SearchCount
of zero (ctx->f_ecnt == 0) and an EndOfSearch code of zero.
Any attempt to get more entries with calls to TRANS2_FIND_NEXT2
result in Badfid (bad file descriptor). I suspect the return of
SearchCount of zero means that end-of-search has been reached and
the Sid is now closed.
The solution is to set "SMB_RDD_EOF | SMB_RDD_NOCLOSE" after getting
back a zero SearchCount, I've tested this in the field on a quite
a few systems, aggressively accessing Windows shares over smbfs
and it appears flawless.
I was initially concerned about the possibility of resource exhaustion
on the Windows server. I was afraid by not officially closing the
search, it would leave a resource hung-up and over time, exhaust
some sort of "open search table" limit. I've since convinced myself
this is NOT the case.
Windows needs to be able to handle clients that come and go over
time. If the search is not closed, Windows will close it if it
finds it needs more resources. I've testing this on directory
searches descending into 10's of thousands of folders, with 100's
of thousands of files.
it was initialised quite late due to its reliance on disc data the mount
process could have stopped before initialising and thus could panic again
only now for uninitialising an not initialised pool! *sigh*
the same memory block allocated as before and it bombs out on its
descriptor pool allready being initialised. It turns out that the pool was
not allways destroyed. This fix ought to clean it up whatever the cause of
the mishap that results in a reject.
253 of the superblock be zero. Searching the net failed to find any
justification for checking these bytes; all available references say
that they are part of the boot code and not BOOTSIG2 and BOOTSIG3.
Modify the MSDOS 7.1 bootsector definition to have 420 bytes of boot
code and no BOOTSIG[23], rather than 418 bytes of boot code, to follow
available references and apparent Windows practice. A test build
showed that these defines are not used other than in the check removed
by this commit.
Patch tested on netbsd-3, and enabled mounting of a 4 GB CF formatted
under Windows XP and then in a digital camera. The CF was previously
unmountable.
Concept approved on tech-kern by christos@ and martin@.
While touching all vptofh/fhtovp functions, get rid of VFS_MAXFIDSIZ,
version the getfh(2) syscall and explicitly pass the size available in
the filehandle from userland.
Discussed on tech-kern, with lots of help from yamt (thanks!).
allocated space was 2048 bytes, but when adding 1024 to the variable
`unix_name' to split the allocated space in half it effectively starts just
OUTSIDE the allocated space. This ought to fix memory corruption bugs when
using UDF.
This is a routine to revisit one day.
- struct timeval time is gone
time.tv_sec -> time_second
- struct timeval mono_time is gone
mono_time.tv_sec -> time_uptime
- access to time via
{get,}{micro,nano,bin}time()
get* versions are fast but less precise
- support NTP nanokernel implementation (NTP API 4)
- further reading:
Timecounter Paper: http://phk.freebsd.dk/pubs/timecounter.pdf
NTP Nanokernel: http://www.eecis.udel.edu/~mills/ntp/html/kern.html
- use vmspace rather than proc or lwp where appropriate.
the latter is more natural to specify an address space.
(and less likely to be abused for random purposes.)
- fix a swdmover race.
The code supports read access to all media types that CD/DVD type drives
can recognize including DVD-RAM and BD- drives as well as harddisc partions
and vnd devices. UDF versions upto the latest 2.60 are to be supported
though due to lack of test media version 2.50 and 2.60 are not implemented
yet though easy to add. Both open and closed media are supported.
Write access is planned and in preparation. To facilitate this some hooks
are present in the code that are not strictly needed in a read-only
implementation but which allow writing to be added more easily.
Implemented and tested media types are CD-ROM, CD-R, CD-RW, CD-MRW,
DVD-ROM, DVD*R, DVD*RW, DVD+MRW but the same code can also read DVD-RAM,
HD-DVD and BluRay discs. Also vnd devices have been tested with several
sector sizes.
Discs created and written by UDFclient, Nero's InCD and Roxio's
DirectCD/Drag2Disc read fine.
Let it get recycled along with all of the others. This is important
as if the root vnode has already been reclaimed, then we get a panic
when we try to vget it.
This addresses PR: kern/31382
- for structure fields that are conditionally present,
make those fields always present.
- for functions which are conditionally inline, make them never inline.
- remove some other functions which are conditionally defined but
don't actually do anything anymore.
- make a lock-debugging function conditional on only LOCKDEBUG.
as discussed on tech-kern some time back.
- make sure that kernel only files don't compile in userland using #error
- XXX: some kernel only files still get installed.
- XXX: some files used in userland, don't get installed.
because VOP_UPDATE() usually succeeded, spec_close() was not usually
called. Only skip the spec_close() step if VOP_UPDATE() returns
an error result. Now /dev/watchdog works as expected when /dev/
is a tmpfs; previously, it was impossible to disarm a user-tickled
watchdog.
per yamt's suggestion. Previously, if /dev/ was mounted on a tmpfs,
block device buffers were never flushed to disk. Trying to unmount
a dirty filesystem (umount /dev/wd0e, say) caused an endless stream
of vflushbuf warnings, because tmpfs_bwrite was not flushing buffers.
The fix told to me by yamt solves the problem.
ptyfs_write() rather than setting a flag and updating these times
through ptyfs_itimes() at some indeterminate time in the future.
However, just use the "time" variable to set the times instead of
using a potentially expensive call to nanotime(). A HZ resolution
on these timestamps is more than enough.
(Possibly incomplete) fix for PR kern/31430.
OK'd be christos@.
implementation of getpages and putpages and the use of UBC in the read and
write operations), the worst problem has gone away which was a panic when
a file's contents were modified in the original file system and then read
through the NFS mount point.
Also remove the entry about optimization. While tmpfs still has room for
improvement, it has become a lot better lately, thanks to the string pools
and the changes yamt@ did in the anonymous objects handling.
tmpfs' "API" and was already rotting.
Instead, merge all the relevant comments into the code. This includes
acknowledgements to Google's Summer of Code 2005 program (they were in the
AUTHORS section of tmpfs(9) before), so all the files need to be changed
to include this sentence alongside the title. (Note that this was not a
requirement of the program.)
to modify it (I hope this is the correct way to go). Avoids triggering an
assertion in the tmpfs_dir_detach function, shown by the t_rename
regression test.
- Remove all NFS related stuff from file system specific code.
- Drop the vfs_checkexp hook and generalize it in the new nfs_check_export
function, thus removing redundancy from all file systems.
- Move all NFS export-related stuff from kern/vfs_subr.c to the new
file sys/nfs/nfs_export.c. The former was becoming large and its code
is always compiled, regardless of the build options. Using the latter,
the code is only compiled in when NFSSERVER is enabled. While doing this,
also make some functions in nfs_subs.c conditional to NFSSERVER.
- Add a new command in nfssvc(2), called NFSSVC_SETEXPORTSLIST, that takes a
path and a set of export entries. At the moment it can only clear the
exports list or append entries, one by one, but it is done in a way that
allows setting the whole set of entries atomically in the future (see the
comment in mountd_set_exports_list or in doc/TODO).
- Change mountd(8) to use the nfssvc(2) system call instead of mount(2) so
that it becomes file system agnostic. In fact, all this whole thing was
done to remove a 'XXX' block from this utility!
- Change the mount*, newfs and fsck* userland utilities to not deal with NFS
exports initialization; done internally by the kernel when initializing
the NFS support for each file system.
- Implement an interface for VFS (called VFS hooks) so that several kernel
subsystems can run arbitrary code upon receipt of specific VFS events.
At the moment, this only provides support for unmount and is used to
destroy NFS exports lists from the file systems being unmounted, though it
has room for extension.
Thanks go to yamt@, chs@, thorpej@, wrstuden@ and others for their comments
and advice in the development of this patch.
from macros to real functions. Original patch and review from chuq.
Note: ext2fs only keeps seconds in the on-disk inode, and msdosfs does not
have enough precision for all fields, so this is not very useful for those
two.
was developed as part of Google's Summer of Code 2005 program. This
change adds the kernel code, the mount_tmpfs utility, a regression test
suite and does all other related changes to integrate these.
The file-system is still *experimental*. Therefore, it is disabled by
default in all kernels. However, as typically done, a commented-out
entry is added in them to ease its setup.
Note that I haven't commited the required mountd(8) changes to be able
to export tmpfs file-systems because NFS support is still very unstable
and because, before enabling it, I'd like to do some other changes.
OK'ed by my project mentor, William Studenmund (wrstuden@).
don't bother trying to write files bigger than this. Just return
EFBIG to caller, rather than panic()ing later.
From OpenBSD.
This closes my PR kern/30864: "panic when copying files of >4GB on msdosfs"
The __UNCONST macro is now used only where necessary and the RW macros
are gone. Most of the changes here are consumers of the
sysctl_createv(9) interface that now takes a pair of const pointers
which used not to be.
into the "vfsops" link set.
- Use VFS_ATTACH() where vfsops are declared for individual file systems.
- In vfsinit(), traverse the "vfsops" link set, rather than vfs_list_initial[].
cache; this appears to fix the random file content corruption which
happens when more than one cluster is read at the same time, i.e. for
files > 3*cluster_size
Fixes PR kern/23835
change obtained from FreeBSD ntfs_subr.c rev. 1.31
this means we can no longer look at the vnode size to determine how many
pages to request in a fault, which is good since for NFS the size can change
out from under us on the server anyway. there's also a new flag UBC_UNMAP
for ubc_release(), so that the file system code can make the decision about
whether to cache mappings for files being used as executables.
foo_mountfs() to foo_mount(), to match the new mountroot API.
Also, for ext2fs and lfs, copy some restructuring from ffs to allow changing
file system parameters without specifying the device name.
(ntfs could use some more work.)
and just passes it on to the file system functions. This avoids opening and
closing the device several times.
Mentioned on tech-kern some time ago, IIRC. I've been running this for a
long time.
and adjust callers appropriately. cn_nameptr isn't NUL-terminated
for non-leaf components, so it's incorrect to assume it is always
NUL-terminated.
This fixes previous utf8-ization changes to not panic for cases
like 'echo */*/*.ps' in case the intermediate path components
are not in cache (yet).
the root of the file system, and slaves and masters alternate starting
at 3. This means the inode numbers won't be really large until you
really allocate a really lot of ptys.
some strange type names with their C99 equivalent.
This avoids sign extension of 32bit fields (like the file size) on 64bit
archs.
Based on a patch supplied by Olaf 'Rhialto' Seibert in PR kern/28471;
should fix that PR.
zero in ptyfs_getattr(). There's no reason to make them something
else and no other devices (except for disk devices) report anything
other than zero here, so why should we be different?
into UTF-8, rather than filtering them to ISO-8859-1 subset
provide vfs.cd9660.utf8_joliet sysctl to switch to the former
iso-8859-1-only handling, default is to UTF-8 encode
by permissions. Ie denial is up to server with clients/openers needing to use
advisory locks for further control. From FreeBSD.
Fixes file access problem when the file is already open on the server.
Problem reported by George Abdelmalik <gabdelmalik at avdat dot com dot au>
Reviewed by jdolecek.
* Rather than using mnt_maxsymlinklen to indicate that a file systems returns
d_type fields(!), add a new internal flag, IMNT_DTYPE.
Add 3 new elements to ufsmount:
* um_maxsymlinklen, replaces mnt_maxsymlinklen (which never should have existed
in the first place).
* um_dirblksiz, which tracks the current directory block size, eliminating the
FS-specific checks littered throughout the code. This may be used later to
make the block size variable.
* um_maxfilesize, which is the maximum file size, possibly adjusted lower due
to implementation issues.
Sync some bug fixes from FFS into ext2fs, particularly:
* ffs_lookup.c 1.21, 1.28, 1.33, 1.48
* ffs_inode.c 1.43, 1.44, 1.45, 1.66, 1.67
* ffs_vnops.c 1.84, 1.85, 1.86
Clean up some crappy pointer frobnication.
