049525126a
remove the XXX comment about holes again - it appears to work fine
2003-02-25 10:14:29 +00:00
f9d42ade36
add __KERNEL_RCSID()
2003-02-25 09:09:31 +00:00
4f6180461f
switch to hash32_strn() from <sys/hash.h> as the hash function -
...
former smbfs_hash() may have accessed memory past buffer (ouch!), and accessed
memory in alignment-unsafe way
as an added bonus, hash32_strn() appears to have better distribution
2003-02-24 21:52:53 +00:00
b0c6be8be5
add XXX comment regarding holes - they are not handled properly
...
in write, this needs to be fixed eventually (unless it would work
by luck)
2003-02-24 21:29:48 +00:00
968351dd0d
g/c smbfs_hashprint()
2003-02-24 21:27:55 +00:00
0c3d39c642
enable mmap and seek vnode ops - this makes mmap(2), lseek(2),
...
p{read,write}{,v}(2) work
2003-02-24 19:09:02 +00:00
26d1332b7c
since smbfs_fctx.f_fname is always used as 'char', may as well define it
...
that way rather than u_char
2003-02-24 18:45:35 +00:00
b5a66cd65b
use pool to allocate memory for smbfs nodes
...
rename smbfs_uninit() to smbfs_done()
some whitespace fixes + use FREE() (not free()) for MALLOC()ed memory
2003-02-24 18:41:04 +00:00
6d0edf1d1b
smbfs_lookup():
...
* make sure islastcn and lockparent is set before used when the entry is cached
* add VWRITE check for nameiop == RENAME, add necessary unlock/relock for
ISDOTDOT case
adjust smbfs_print() to have slightly nicer output
enable vnode locking
2003-02-24 16:19:05 +00:00
ba728ab921
smbfs_node_alloc():
...
* use LK_RETRY for the vget()
* turn the check for dead parent vnode to #ifdef DIAGNOSTIC panic
* reformat the hash walk code a little, and only assign vp if entries match
* add some comments
2003-02-24 15:55:08 +00:00
a7a5a493b5
SMBERROR() purge - turn into #ifdef DIAGNOSTIC panics
2003-02-24 10:01:02 +00:00
38b6d81ba3
update for changed locking requirements for smb_share_put()
2003-02-24 09:57:31 +00:00
da9a088d46
SMBERROR() purge - replace with KASSERT()s, panic or remove altogether
...
as appropriate
2003-02-24 09:30:42 +00:00
d8ab9af5e5
need <sys/mount.h> now
2003-02-24 09:18:56 +00:00
f31ba31b00
change smbfs_args:
...
* g/c unused members 'mount_point', 'root_path', which only take space
* rename 'dev' to 'dev_fd' to more closely match what it is
* add 'export' member to avoid changing this again if ever smbfs would
be made exportable
bump SMBFS_VERMIN - ABI/API change; mount_smbfs needs to be recompiled
to work again
2003-02-23 22:31:17 +00:00
999273657a
put the printf about version mismatch inside #ifdef SMB_VNODE_DEBUG
...
do not touch sm_args.mount_point[]
2003-02-23 22:20:05 +00:00
8bf6b72335
style; no space after cast, and add cast to the assignment case too.
2003-02-23 22:03:11 +00:00
e211d06862
add genfsops necessary for genfs_*{get|put}pages() to actually work
...
use 'compat' getpages and gop_write - smbfs doesn't have bmap vnode op (yet)
2003-02-23 21:55:20 +00:00
ee219745fe
remove bogus (u_long) cast for uvm_vnp_setsize() arg in smbfs_setattr()
2003-02-23 21:54:05 +00:00
a7bf0f25db
g/c some dead code
...
make smbfs_name_{alloc|free} inline
turn some checks for smbfs code errors to KASSERT()s
KNF some
also initialize n_size in smbfs_node_alloc()
g/c write-only variable in smbfs_nget()
2003-02-23 21:33:06 +00:00
ccf56de797
KASSERT() smp->sm_root is nonnull in unmount
...
g/c unneeded variable in smbfs_root()
2003-02-23 21:27:33 +00:00
9a7b95383e
fix KASSERT() in smbfs_findnextLM1() to actually compile
2003-02-23 21:26:26 +00:00
5c7688a6f3
g/c compatibility defines PROC_LOCK()/PROC_UNLOCK()
2003-02-23 19:35:14 +00:00
24a773c6ef
g/c smbfs_pbuf_freecnt, it's not used (it was not even declared on NetBSD)
2003-02-23 19:30:13 +00:00
21a1fae4a8
g/c smbfs_debuglevel, it's not used anywhere
...
pointed out by Krister Walfridsson
2003-02-23 18:47:10 +00:00
eb5d35152e
remove non-NetBSD smbfs_vnodeop_entries[], smbfs_vnodeop_opv_desc
2003-02-23 17:28:48 +00:00
c3fb3ac103
g/c smbfs_getpages()/smbfs_putpages() we don't use on NetBSD
2003-02-23 17:28:01 +00:00
164de9f034
make DIFF1970TO1601 static
...
problem reported in FreeBSD patch request #43613 by Matt Emmerton
2003-02-21 21:46:45 +00:00
5a426e7c8f
g/c smbfs_fastlookup
...
make smbfs_readvdir() slightly more readable, and don't adjust uio_offset
by hand
g/c the lock code around smbfs_readvdir() - a) it's not needed here b) it
didn't do anything anyway
add some KASSERTs
simplify smbfs_vinvalbuf() a bit, use plain interruptible sleep instead
of smb_proc_intr()
2003-02-21 20:19:00 +00:00
dfc181b14f
some KNF, add a KASSERT() for smbfs_findnextLM1() case
...
replace some 'never should happen' printfs with KASSERTs
2003-02-21 20:15:01 +00:00
50fa881e7b
g/c unneeded local variables in smbfs_read()/smbfs_readdir()
2003-02-21 20:13:21 +00:00
b4cbac6f51
remove SMBFS_NAME_DEBUG code, mainly because it didn't work on alignment-picky
...
archs anyway
don't bother to zero terminate smbnode name (penny saved ...)
2003-02-20 15:39:58 +00:00
7be3184fb7
don't assume struct smbnode -> n_name is zero terminated,
...
use .*s to print it
2003-02-20 15:35:55 +00:00
02f8f49e38
g/c commented out code
2003-02-20 15:34:42 +00:00
5c10202b37
do previous differently - pass ap->a_id down to smbfs_smb_lock()
...
and smbfs_smb_lockandx() as caddr_t, and mask it to 16bit value
in smbfs_smb_lockandx()
okayed by Martin
2003-02-19 13:51:24 +00:00
681bcd85a7
SMB lock ids apparently are 16bit items. And the code uses id 1 always.
...
No need to cast it back and forth to a caddr_t and u_int32_t - which does
not work well on 64bit archs.
2003-02-19 12:44:47 +00:00
bb1e147039
add nfs-style n_ctime, n_nctime
...
#ifdef smbfs_getpages(), smbfs_putpages() - it's not used on NetBSD
2003-02-18 20:09:01 +00:00
d3d721833b
make smbfs_lookup() look more like nfs_lookup(), particularily account
...
for different cache_lookup() semantics
fix smbfs_pathcheck() to not assume zero-ended component name
remove the bogus code in smbfs_close(), it's not needed on NetBSD
fix couple other vnop routines to do vput()/vrele()/PNBUF_PUT() as appropriate
KNF
2003-02-18 20:07:50 +00:00
e737730e68
g/c VI_(UN)LOCK macros and just use simple_(un)lock(vp->v_interlock)
...
#if 0 smbfs_hashprint(), and mark it static
do NOT reinit vp->v_lock - getnewvnode() did it for us
smbfs_reclaim(): unconditionaly do LIST_REMOVE(), and de-obfuscate parent
directory vrele() call
2003-02-18 20:00:35 +00:00
e75d9c776e
vrele() the reference to root vnode before calling vflush() in smbfs_unmount(),
...
so that the filesystem wouldn't appear as busy when it's not, and vref()
it back if the vflush() fails
LK_DRAIN sm_hashlock before freeing the smbmount struct
g/c one meaningless check in smbfs_root()
2003-02-18 19:40:36 +00:00
309faeafd5
smbfs_smb_create(): use SMB_COM_CREATE_NEW rather than SMB_COM_CREATE; the
...
latter truncates the file if it exists, which we don't want to do
change the code a bit to be easier to read
2003-02-18 12:52:34 +00:00
0550f349b8
couple cosmetic changes:
...
* g/c superfluous printfs
* use vcp on one more place
* don't need to check result of allocation with M_WAITOK
* just exlicitly return (0) on the success path, rather than return (error)
2003-02-18 10:27:17 +00:00
399550a191
do age attrcache on NetBSD - use mono_time.tv_sec instead of FreeBSD time_second
...
don't need to erase *vpp on top smbfs_nget()
s/bzero/memset/ for struct smbnode allocation
2003-02-18 10:24:58 +00:00
baec08a11d
constify a little
2003-02-17 09:55:49 +00:00
f01bbafeaa
terminate FB_CURRENT specific struct smbnode member
2003-02-16 19:42:21 +00:00
245fa74e31
g/c SMBFS_USEZONE and some other redundant code
...
terminate the check if options SMB was included
remove the #ifndef FB_CURRENT around f_mntonname; FB_CURRENT used to be defined
in netsmb headers, and this caused f_mntonname to not be initialized
terminate the messy code setting f_mntfromname, and use snprintf() instead
2003-02-16 19:35:16 +00:00
16a0b9753c
Wrap the smbfs_pbuf_freecnt var in #ifndef NetBSD since it's entirely unused.
2003-02-01 21:02:03 +00:00
b193480908
Add extensible malloc types, adapted from FreeBSD. This turns
...
malloc types into a structure, a pointer to which is passed around,
instead of an int constant. Allow the limit to be adjusted when the
malloc type is defined, or with a function call, as suggested by
Jonathan Stone.
2003-02-01 06:23:35 +00:00
14757120f4
Fix compiler warnings after 64bit daddr_t change.
2003-01-27 04:08:45 +00:00
a3ff3a3038
Bump daddr_t to 64 bits. Replace it with int32_t in all places where
...
it was used on-disk, so that on-disk formats remain the same.
Remove ufs_daddr_t and ufs_lbn_t for the time being.
2003-01-24 21:55:02 +00:00
c821903287
Remove variable that is only assigned too but not referenced.
2003-01-20 02:25:49 +00:00
b223c57195
Don't scale file block numbers. Fixes mmap() and UBCified read() problems.
...
Reviewed by Ignatios.
2003-01-09 18:34:41 +00:00
5a91827c81
through, not throught.
2003-01-06 12:29:48 +00:00
6621e1674f
Typo and grammar fixes in comments and a printf string.
2003-01-06 12:27:42 +00:00
98d58548c7
move msdosfs code from sys/msdosfs to sys/fs/msdosfs
2002-12-26 12:31:23 +00:00
3a8872dee7
move cd9660 filesystem code from sys/isofs/cd9660 to sys/fs/cd9660
2002-12-23 17:52:02 +00:00
9accf4df15
move NTFS code from sys/ntfs to sys/fs/ntfs
2002-12-23 17:38:19 +00:00
5356de155a
move filecorefs code from sys/filecorefs to sys/fs/filecorefs
2002-12-23 17:30:36 +00:00
8974c3cf90
move adosfs files from sys/adosfs to sys/fs/adosfs
2002-12-23 17:15:20 +00:00
0635de35a3
Remove KDIR=, since SYS_INCLUDE=symlinks and KDIR are not supported any more.
2002-11-26 23:30:07 +00:00
6f3945a88d
MNT_GETARGS support
2002-09-21 18:10:34 +00:00
178d83d503
Die, qaddr_t, die! - mnt_data in struct mount is already effectively
...
a void *, so stop pretending otherwise.
2002-07-30 07:40:07 +00:00
37dc008ca3
Cleanup how file system configuration information is declared, grouping
...
related information together, with the file system code itself.
This is just low-hanging fruit -- more to come.
2002-04-16 23:14:05 +00:00
4687da33a2
descend into smbfs
2002-01-09 21:07:54 +00:00
408ba63df0
smbfs makefile, INCLUDES commented out for now
2002-01-09 21:06:53 +00:00
4738ff87d6
netbsd changes
2002-01-09 17:43:28 +00:00
f5cc2966a0
import freebsd smbfs code
2002-01-09 17:25:32 +00:00
jdolecek