3 Fix memory leak in statfs when mounted to very old servers (e.g.
4 Windows 9x). Add new feature "POSIX open" which allows servers
5 which support the current POSIX Extensions to provide better semantics
6 (e.g. delete for open files opened with posix open). Take into
7 account umask on posix mkdir not just older style mkdir. Add
8 ability to mount to IPC$ share (which allows CIFS named pipes to be
9 opened, read and written as if they were files). When 1st tree
10 connect fails (e.g. due to signing negotiation failure) fix
11 leak that causes cifsd not to stop and rmmod to fail to cleanup
12 cifs_request_buffers pool. Fix problem with POSIX Open/Mkdir on
13 bigendian architectures. Fix possible memory corruption when
14 EAGAIN returned on kern_recvmsg. Return better error if server
15 requires packet signing but client has disabled it. When mounted
16 with cifsacl mount option - mode bits are approximated based
17 on the contents of the files ACL.
21 Fix NTLMv2 signing. NFS server mounted over cifs works (if cifs mount is
22 done with "serverino" mount option). Add support for POSIX Unlink
23 (helps with certain sharing violation cases when server such as
24 Samba supports newer POSIX CIFS Protocol Extensions). Add "nounix"
25 mount option to allow disabling the CIFS Unix Extensions for just
26 that mount. Fix hang on spinlock in find_writable_file (race when
27 reopening file after session crash). Byte range unlock request to
28 windows server could unlock more bytes (on server copy of file)
29 than intended if start of unlock request is well before start of
30 a previous byte range lock that we issued.
34 IPv6 support. Enable ipv6 addresses to be passed on mount (put the ipv6
35 address after the "ip=" mount option, at least until mount.cifs is fixed to
36 handle DNS host to ipv6 name translation). Accept override of uid or gid
37 on mount even when Unix Extensions are negotiated (it used to be ignored
38 when Unix Extensions were ignored). This allows users to override the
39 default uid and gid for files when they are certain that the uids or
40 gids on the server do not match those of the client. Make "sec=none"
41 mount override username (so that null user connection is attempted)
42 to match what documentation said. Support for very large reads, over 127K,
43 available to some newer servers (such as Samba 3.0.26 and later but
44 note that it also requires setting CIFSMaxBufSize at module install
45 time to a larger value which may hurt performance in some cases).
46 Make sign option force signing (or fail if server does not support it).
50 Fix mtime bouncing around from local idea of last write times to remote time.
51 Fix hang (in i_size_read) when simultaneous size update of same remote file
52 on smp system corrupts sequence number. Do not reread unnecessarily partial page
53 (which we are about to overwrite anyway) when writing out file opened rw.
54 When DOS attribute of file on non-Unix server's file changes on the server side
55 from read-only back to read-write, reflect this change in default file mode
56 (we had been leaving a file's mode read-only until the inode were reloaded).
57 Allow setting of attribute back to ATTR_NORMAL (removing readonly dos attribute
58 when archive dos attribute not set and we are changing mode back to writeable
59 on server which does not support the Unix Extensions). Remove read only dos
60 attribute on chmod when adding any write permission (ie on any of
61 user/group/other (not all of user/group/other ie 0222) when
62 mounted to windows. Add support for POSIX MkDir (slight performance
63 enhancement and eliminates the network race between the mkdir and set
64 path info of the mode).
69 Fix oops in list_del during mount caused by unaligned string.
70 Fix file corruption which could occur on some large file
71 copies caused by writepages page i/o completion bug.
72 Seek to SEEK_END forces check for update of file size for non-cached
73 files. Allow file size to be updated on remote extend of locally open,
74 non-cached file. Fix reconnect to newer Samba servers (or other servers
75 which support the CIFS Unix/POSIX extensions) so that we again tell the
76 server the Unix/POSIX cifs capabilities which we support (SetFSInfo).
77 Add experimental support for new POSIX Open/Mkdir (which returns
78 stat information on the open, and allows setting the mode).
82 Support deep tree mounts. Better support OS/2, Win9x (DOS) time stamps.
83 Allow null user to be specified on mount ("username="). Do not return
84 EINVAL on readdir when filldir fails due to overwritten blocksize
85 (fixes FC problem). Return error in rename 2nd attempt retry (ie report
86 if rename by handle also fails, after rename by path fails, we were
87 not reporting whether the retry worked or not). Fix NTLMv2 to
88 work to Windows servers (mount with option "sec=ntlmv2").
92 Do not time out lockw calls when using posix extensions. Do not
93 time out requests if server still responding reasonably fast
94 on requests on other threads. Improve POSIX locking emulation,
95 (lock cancel now works, and unlock of merged range works even
96 to Windows servers now). Fix oops on mount to lanman servers
97 (win9x, os/2 etc.) when null password. Do not send listxattr
98 (SMB to query all EAs) if nouser_xattr specified. Fix SE Linux
99 problem (instantiate inodes/dentries in right order for readdir).
103 Rewritten sessionsetup support, including support for legacy SMB
104 session setup needed for OS/2 and older servers such as Windows 95 and 98.
105 Fix oops on ls to OS/2 servers. Add support for level 1 FindFirst
106 so we can do search (ls etc.) to OS/2. Do not send NTCreateX
107 or recent levels of FindFirst unless server says it supports NT SMBs
108 (instead use legacy equivalents from LANMAN dialect). Fix to allow
109 NTLMv2 authentication support (now can use stronger password hashing
110 on mount if corresponding /proc/fs/cifs/SecurityFlags is set (0x4004).
111 Allow override of global cifs security flags on mount via "sec=" option(s).
115 POSIX locking to servers which support CIFS POSIX Extensions
116 (disabled by default controlled by proc/fs/cifs/Experimental).
117 Handle conversion of long share names (especially Asian languages)
118 to Unicode during mount. Fix memory leak in sess struct on reconnect.
119 Fix rare oops after acpi suspend. Fix O_TRUNC opens to overwrite on
120 cifs open which helps rare case when setpathinfo fails or server does
125 Fix slow oplock break when mounted to different servers at the same time and
126 the tids match and we try to find matching fid on wrong server. Fix read
127 looping when signing required by server (2.6.16 kernel only). Fix readdir
128 vs. rename race which could cause each to hang. Return . and .. even
129 if server does not. Allow searches to skip first three entries and
130 begin at any location. Fix oops in find_writeable_file.
134 Fix NTLMv2 security (can be enabled in /proc/fs/cifs) so customers can
135 configure stronger authentication. Fix sfu symlinks so they can
136 be followed (not just recognized). Fix wraparound of bcc on
137 read responses when buffer size over 64K and also fix wrap of
138 max smb buffer size when CIFSMaxBufSize over 64K. Fix oops in
139 cifs_user_read and cifs_readpages (when EAGAIN on send of smb
140 on socket is returned over and over). Add POSIX (advisory) byte range
141 locking support (requires server with newest CIFS UNIX Extensions
142 to the protocol implemented). Slow down negprot slightly in port 139
143 RFC1001 case to give session_init time on buggy servers.
147 Use fsuid (fsgid) more consistently instead of uid (gid). Improve performance
148 of readpages by eliminating one extra memcpy. Allow update of file size
149 from remote server even if file is open for write as long as mount is
150 directio. Recognize share mode security and send NTLM encrypted password
151 on tree connect if share mode negotiated.
155 Defer close of a file handle slightly if pending writes depend on that handle
156 (this reduces the EBADF bad file handle errors that can be logged under heavy
157 stress on writes). Modify cifs Kconfig options to expose CONFIG_CIFS_STATS2
158 Fix SFU style symlinks and mknod needed for servers which do not support the
159 CIFS Unix Extensions. Fix setfacl/getfacl on bigendian. Timeout negative
160 dentries so files that the client sees as deleted but that later get created
161 on the server will be recognized. Add client side permission check on setattr.
162 Timeout stuck requests better (where server has never responded or sent corrupt
167 Fix tcp socket retransmission timeouts (e.g. on ENOSPACE from the socket)
168 to be smaller at first (but increasing) so large write performance performance
169 over GigE is better. Do not hang thread on illegal byte range lock response
170 from Windows (Windows can send an RFC1001 size which does not match smb size) by
171 allowing an SMBs TCP length to be up to a few bytes longer than it should be.
172 wsize and rsize can now be larger than negotiated buffer size if server
173 supports large readx/writex, even when directio mount flag not specified.
174 Write size will in many cases now be 16K instead of 4K which greatly helps
175 file copy performance on lightly loaded networks. Fix oops in dnotify
176 when experimental config flag enabled. Make cifsFYI more granular.
180 Fix readdir caching when unlink removes file in current search buffer,
181 and this is followed by a rewind search to just before the deleted entry.
182 Do not attempt to set ctime unless atime and/or mtime change requested
183 (most servers throw it away anyway). Fix length check of received smbs
184 to be more accurate. Fix big endian problem with mapchars mount option,
185 and with a field returned by statfs.
189 Add support for mounting to older pre-CIFS servers such as Windows9x and ME.
190 For these older servers, add option for passing netbios name of server in
191 on mount (servernetbiosname). Add suspend support for power management, to
192 avoid cifsd thread preventing software suspend from working.
193 Add mount option for disabling the default behavior of sending byte range lock
194 requests to the server (necessary for certain applications which break with
195 mandatory lock behavior such as Evolution), and also mount option for
196 requesting case insensitive matching for path based requests (requesting
197 case sensitive is the default).
201 Add writepage performance improvements. Fix path name conversions
202 for long filenames on mounts which were done with "mapchars" mount option
203 specified. Ensure multiplex ids do not collide. Fix case in which
204 rmmod can oops if done soon after last unmount. Fix truncated
205 search (readdir) output when resume filename was a long filename.
206 Fix filename conversion when mapchars mount option was specified and
207 filename was a long filename.
211 Fix error mapping of the TOO_MANY_LINKS (hardlinks) case.
212 Do not oops if root user kills cifs oplock kernel thread or
213 kills the cifsd thread (NB: killing the cifs kernel threads is not
214 recommended, unmount and rmmod cifs will kill them when they are
215 no longer needed). Fix readdir to ASCII servers (ie older servers
216 which do not support Unicode) and also require asterisk.
217 Fix out of memory case in which data could be written one page
218 off in the page cache.
222 Fix caching problem, in which readdir of directory containing a file
223 which was cached could cause the file's time stamp to be updated
224 without invalidating the readahead data (so we could get stale
225 file data on the client for that file even as the server copy changed).
226 Cleanup response processing so cifsd can not loop when abnormally
232 Fix oops in ls when Transact2 FindFirst (or FindNext) returns more than one
233 transact response for an SMB request and search entry split across two frames.
234 Add support for lsattr (getting ext2/ext3/reiserfs attr flags from the server)
235 as new protocol extensions. Do not send Get/Set calls for POSIX ACLs
236 unless server explicitly claims to support them in CIFS Unix extensions
237 POSIX ACL capability bit. Fix packet signing when multiuser mounting with
238 different users from the same client to the same server. Fix oops in
239 cifs_close. Add mount option for remapping reserved characters in
240 filenames (also allow recognizing files with created by SFU which have any
241 of these seven reserved characters, except backslash, to be recognized).
242 Fix invalid transact2 message (we were sometimes trying to interpret
243 oplock breaks as SMB responses). Add ioctl for checking that the
244 current uid matches the uid of the mounter (needed by umount.cifs).
245 Reduce the number of large buffer allocations in cifs response processing
246 (significantly reduces memory pressure under heavy stress with multiple
247 processes accessing the same server at the same time).
251 Fix updates of DOS attributes and time fields so that files on NT4 servers
252 do not get marked delete on close. Display sizes of cifs buffer pools in
253 cifs stats. Fix oops in unmount when cifsd thread being killed by
254 shutdown. Add generic readv/writev and aio support. Report inode numbers
255 consistently in readdir and lookup (when serverino mount option is
256 specified use the inode number that the server reports - for both lookup
257 and readdir, otherwise by default the locally generated inode number is used
258 for inodes created in either path since servers are not always able to
259 provide unique inode numbers when exporting multiple volumes from under one
264 Allow new nouser_xattr mount parm to disable xattr support for user namespace.
265 Do not flag user_xattr mount parm in dmesg. Retry failures setting file time
266 (mostly affects NT4 servers) by retry with handle based network operation.
267 Add new POSIX Query FS Info for returning statfs info more accurately.
268 Handle passwords with multiple commas in them.
272 Fix default mode in sysfs of cifs module parms. Remove old readdir routine.
273 Fix capabilities flags for large readx so as to allow reads larger than 64K.
277 Add module init parm for large SMB buffer size (to allow it to be changed
278 from its default of 16K) which is especially useful for large file copy
279 when mounting with the directio mount option. Fix oops after
280 returning from mount when experimental ExtendedSecurity enabled and
281 SpnegoNegotiated returning invalid error. Fix case to retry better when
282 peek returns from 1 to 3 bytes on socket which should have more data.
283 Fixed path based calls (such as cifs lookup) to handle path names
284 longer than 530 (now can handle PATH_MAX). Fix pass through authentication
285 from Samba server to DC (Samba required dummy LM password).
289 Turn off DNOTIFY (directory change notification support) by default
290 (unless built with the experimental flag) to fix hang with KDE
291 file browser. Fix DNOTIFY flag mappings. Fix hang (in wait_event
292 waiting on an SMB response) in SendReceive when session dies but
293 reconnects quickly from another task. Add module init parms for
294 minimum number of large and small network buffers in the buffer pools,
295 and for the maximum number of simultaneous requests.
299 Add setfacl support to allow setting of ACLs remotely to Samba 3.10 and later
300 and other POSIX CIFS compliant servers. Fix error mapping for getfacl
301 to EOPNOTSUPP when server does not support posix acls on the wire. Fix
302 improperly zeroed buffer in CIFS Unix extensions set times call.
306 Fix internationalization problem in cifs readdir with filenames that map to
307 longer UTF-8 strings than the string on the wire was in Unicode. Add workaround
308 for readdir to netapp servers. Fix search rewind (seek into readdir to return
309 non-consecutive entries). Do not do readdir when server negotiates
310 buffer size to small to fit filename. Add support for reading POSIX ACLs from
311 the server (add also acl and noacl mount options).
315 Optionally allow using server side inode numbers, rather than client generated
316 ones by specifying mount option "serverino" - this is required for some apps
317 to work which double check hardlinked files and have persistent inode numbers.
321 Multiple bigendian fixes. On little endian systems (for reconnect after
322 network failure) fix tcp session reconnect code so we do not try first
323 to reconnect on reverse of port 445. Treat reparse points (NTFS junctions)
324 as directories rather than symlinks because we can do follow link on them.
328 Add config option to enable XATTR (extended attribute) support, mapping
329 xattr names in the "user." namespace space to SMB/CIFS EAs. Lots of
330 minor fixes pointed out by the Stanford SWAT checker (mostly missing
331 or out of order NULL pointer checks in little used error paths).
335 Add new mount parm to control whether mode check (generic_permission) is done
336 on the client. If Unix extensions are enabled and the uids on the client
337 and server do not match, client permission checks are meaningless on
338 server uids that do not exist on the client (this does not affect the
339 normal ACL check which occurs on the server). Fix default uid
340 on mknod to match create and mkdir. Add optional mount parm to allow
341 override of the default uid behavior (in which the server sets the uid
342 and gid of newly created files). Normally for network filesystem mounts
343 user want the server to set the uid/gid on newly created files (rather than
344 using uid of the client processes you would in a local filesystem).
348 Make transaction counts more consistent. Merge /proc/fs/cifs/SimultaneousOps
349 info into /proc/fs/cifs/DebugData. Fix oops in rare oops in readdir
350 (in build_wildcard_path_from_dentry). Fix mknod to pass type field
351 (block/char/fifo) properly. Remove spurious mount warning log entry when
352 credentials passed as mount argument. Set major/minor device number in
353 inode for block and char devices when unix extensions enabled.
357 Fix /proc/fs/cifs/Stats and DebugData display to handle larger
358 amounts of return data. Properly limit requests to MAX_REQ (50
359 is the usual maximum active multiplex SMB/CIFS requests per server).
360 Do not kill cifsd (and thus hurt the other SMB session) when more than one
361 session to the same server (but with different userids) exists and one
362 of the two user's smb sessions is being removed while leaving the other.
363 Do not loop reconnecting in cifsd demultiplex thread when admin
364 kills the thread without going through unmount.
368 Do not rename hardlinked files (since that should be a noop). Flush
369 cached write behind data when reopening a file after session abend,
370 except when already in write. Grab per socket sem during reconnect
371 to avoid oops in sendmsg if overlapping with reconnect. Do not
372 reset cached inode file size on readdir for files open for write on
378 Update number of blocks in file so du command is happier (in Linux a fake
379 blocksize of 512 is required for calculating number of blocks in inode).
380 Fix prepare write of partial pages to read in data from server if possible.
381 Fix race on tcpStatus field between unmount and reconnection code, causing
382 cifsd process sometimes to hang around forever. Improve out of memory
383 checks in cifs_filldir
387 Fix incorrect file size in file handle based setattr on big endian hardware.
388 Fix oops in build_path_from_dentry when out of memory. Add checks for invalid
389 and closing file structs in writepage/partialpagewrite. Add statistics
390 for each mounted share (new menuconfig option). Fix endianness problem in
391 volume information displayed in /proc/fs/cifs/DebugData (only affects
392 affects big endian architectures). Prevent renames while constructing
393 path names for open, mkdir and rmdir.
397 Change to mempools for alloc smb request buffers and multiplex structs
398 to better handle low memory problems (and potential deadlocks).
402 Fix incomplete listings of large directories on Samba servers when Unix
403 extensions enabled. Fix oops when smb_buffer can not be allocated. Fix
404 rename deadlock when writing out dirty pages at same time.
408 Fix open of files in which O_CREATE can cause the mode to change in
409 some cases. Fix case in which retry of write overlaps file close.
410 Fix PPC64 build error. Reduce excessive stack usage in smb password
411 hashing. Fix overwrite of Linux user's view of file mode to Windows servers.
415 Fixes for large file copy, signal handling, socket retry, buffer
416 allocation and low memory situations.
420 Better port 139 support to Windows servers (RFC1001/RFC1002 Session_Initialize)
421 also now allowing support for specifying client netbiosname. NT4 support added.
425 Fix reconnection (and certain failed mounts) to properly wake up the
426 blocked users thread so it does not seem hung (in some cases was blocked
427 until the cifs receive timeout expired). Fix spurious error logging
428 to kernel log when application with open network files killed.
432 Fix /proc/fs module unload warning message (that could be logged
433 to the kernel log). Fix intermittent failure in connectathon
434 test7 (hardlink count not immediately refreshed in case in which
435 inode metadata can be incorrectly kept cached when time near zero)
439 Allow file_mode and dir_mode (specified at mount time) to be enforced
440 locally (the server already enforced its own ACLs too) for servers
441 that do not report the correct mode (do not support the
442 CIFS Unix Extensions).
446 Fix some small memory leaks in some unmount error paths. Fix major leak
447 of cache pages in readpages causing multiple read oriented stress
448 testcases (including fsx, and even large file copy) to fail over time.
452 Send NTCreateX with ATTR_POSIX if Linux/Unix extensions negotiated with server.
453 This allows files that differ only in case and improves performance of file
454 creation and file open to such servers. Fix semaphore conflict which causes
455 slow delete of open file to Samba (which unfortunately can cause an oplock
456 break to self while vfs_unlink held i_sem) which can hang for 20 seconds.
460 fixes to cifs_readpages for fsx test case
464 Fix caching data integrity bug when extending file size especially when no
465 oplock on file. Fix spurious logging of valid already parsed mount options
466 that are parsed outside of the cifs vfs such as nosuid.
471 Connect to server when port number override not specified, and tcp port
472 unitialized. Reset search to restart at correct file when kernel routine
473 filldir returns error during large directory searches (readdir).
477 Fix caching problem when files opened by multiple clients in which
478 page cache could contain stale data, and write through did
479 not occur often enough while file was still open when read ahead
480 (read oplock) not allowed. Treat "sep=" when first mount option
481 as an override of comma as the default separator between mount
486 Allow passwords longer than 16 bytes. Allow null password string.
490 Gracefully clean up failed mounts when attempting to mount to servers such as
491 Windows 98 that terminate tcp sessions during protocol negotiation. Handle
492 embedded commas in mount parsing of passwords.
496 Invalidate local inode cached pages on oplock break and when last file
497 instance is closed so that the client does not continue using stale local
498 copy rather than later modified server copy of file. Do not reconnect
499 when server drops the tcp session prematurely before negotiate
500 protocol response. Fix oops in reopen_file when dentry freed. Allow
501 the support for CIFS Unix Extensions to be disabled via proc interface.
505 Fix hang in commit_write during reconnection of open files under heavy load.
506 Fix unload_nls oops in a mount failure path. Serialize writes to same socket
507 which also fixes any possible races when cifs signatures are enabled in SMBs
508 being sent out of signature sequence number order.
512 Fix byte range locking bug (endian problem) causing bad offset and
517 Fix oops (in send_sig) caused by CIFS unmount code trying to
518 wake up the demultiplex thread after it had exited. Do not log
519 error on harmless oplock release of closed handle.
523 Fix unsafe global variable usage and password hash failure on gcc 3.3.1
524 Fix problem reconnecting secondary mounts to same server after session
525 failure. Fix invalid dentry - race in mkdir when directory gets created
526 by another client between the lookup and mkdir.
530 Fix to list processing in reopen_files. Fix reconnection when server hung
531 but tcpip session still alive. Set proper timeout on socket read.
535 Add missing mount options including iocharset. SMP fixes in write and open.
536 Fix errors in reconnecting after TCP session failure. Fix module unloading
537 of default nls codepage
541 Active smb transactions should never go negative (fix double FreeXid). Fix
542 list processing in file routines. Check return code on kmalloc in open.
543 Fix spinlock usage for SMP.
547 Fix oops in reopen_files when invalid dentry. drop dentry on server rename
548 and on revalidate errors. Fix cases where pid is now tgid. Fix return code
549 on create hard link when server does not support them.
553 Fix scheduling while atomic error in getting inode info on newly created file.
554 Fix truncate of existing files opened with O_CREAT but not O_TRUNC set.
558 Fix oops on write to dead tcp session. Remove error log write for case when file open
559 O_CREAT but not O_EXCL
563 Fix non-POSIX behavior on rename of open file and delete of open file by taking
564 advantage of trans2 SetFileInfo rename facility if available on target server.
565 Retry on ENOSPC and EAGAIN socket errors.
569 Fix oops on big endian readdir. Set blksize to be even power of two (2**blkbits) to fix
570 allocation size miscalculation. After oplock token lost do not read through
575 Fix oops on empty file readahead. Fix for file size handling for locally cached files.
579 Fix oops in mkdir when server fails to return inode info. Fix oops in reopen_files
580 during auto reconnection to server after server recovered from failure.
584 Finish support for Linux 2.5 open/create changes, which removes the
585 redundant NTCreate/QPathInfo/close that was sent during file create.
586 Enable oplock by default. Enable packet signing by default (needed to
587 access many recent Windows servers)
591 Fix oops when mounting to long server names caused by inverted parms to kmalloc.
592 Fix MultiuserMount (/proc/fs/cifs configuration setting) so that when enabled
593 we will choose a cifs user session (smb uid) that better matches the local
594 uid if a) the mount uid does not match the current uid and b) we have another
595 session to the same server (ip address) for a different mount which
596 matches the current local uid.
600 Add support for mknod of block or character devices. Fix oplock
601 code (distributed caching) to properly send response to oplock
606 Finish up CIFS packet digital signing for the default
607 NTLM security case. This should help Windows 2003
608 network interoperability since it is common for
609 packet signing to be required now. Fix statfs (stat -f)
610 which recently started returning errors due to
611 invalid value (-1 instead of 0) being set in the
612 struct kstatfs f_ffiles field.
616 Fix oops on stopping oplock thread when removing cifs when
621 Fix mount options for ro (readonly), uid, gid and file and directory mode.
625 Fix errors displayed on failed mounts to be more understandable.
626 Fixed various incorrect or misleading smb to posix error code mappings.
630 Fix display of NTFS DFS junctions to display as symlinks.
631 They are the network equivalent. Fix oops in
632 cifs_partialpagewrite caused by missing spinlock protection
633 of openfile linked list. Allow writebehind caching errors to
634 be returned to the application at file close.
638 Clean up options displayed in /proc/mounts by show_options to
639 be more consistent with other filesystems.
643 Fix delete of readonly file to Windows servers. Reflect
644 presence or absence of read only dos attribute in mode
645 bits for servers that do not support CIFS Unix extensions.
646 Fix shortened results on readdir of large directories to
647 servers supporting CIFS Unix extensions (caused by
648 incorrect resume key).
652 Fix truncate bug (set file size) that could cause hangs e.g. running fsx
656 unload nls if mount fails.
660 Add resume key support to search (readdir) code to workaround
661 Windows bug. Add /proc/fs/cifs/LookupCacheEnable which
662 allows disabling caching of attribute information for
667 Add more oplock handling (distributed caching code). Remove
668 dead code. Remove excessive stack space utilization from
673 Fix oops in get dfs referral (triggered when null path sent in to
674 mount). Add support for overriding rsize at mount time.
678 Fix buffer overrun in readdir which caused intermittent kernel oopses.
679 Fix writepage code to release kmap on write data. Allow "-ip=" new
680 mount option to be passed in on parameter distinct from the first part
681 (server name portion of) the UNC name. Allow override of the
682 tcp port of the target server via new mount option "-port="
686 Fix search handle leak on rewind. Fix setuid and gid so that they are
687 reflected in the local inode immediately. Cleanup of whitespace
688 to make 2.4 and 2.5 versions more consistent.
693 Fix signal sending so that captive thread (cifsd) exits on umount
694 (which was causing the warning in kmem_cache_free of the request buffers
695 at rmmod time). This had broken as a sideeffect of the recent global
696 kernel change to daemonize. Fix memory leak in readdir code which
697 showed up in "ls -R" (and applications that did search rewinding).
701 Reconnect tids and fids after session reconnection (still do not
702 reconnect byte range locks though). Fix problem caching
703 lookup information for directory inodes, improving performance,
704 especially in deep directory trees. Fix various build warnings.
708 Finish fixes to commit write for caching/readahead consistency. fsx
709 now works to Samba servers. Fix oops caused when readahead
710 was interrupted by a signal.
714 Fix data corruption (in partial page after truncate) that caused fsx to
715 fail to Windows servers. Cleaned up some extraneous error logging in
716 common error paths. Add generic sendfile support.
720 Fix memory leak in AllocMidQEntry.
721 Finish reconnection logic, so connection with server can be dropped
722 (or server rebooted) and the cifs client will reconnect.
726 Fix temporary socket leak when bad userid or password specified
727 (or other SMBSessSetup failure). Increase maximum buffer size to slightly
728 over 16K to allow negotiation of up to Samba and Windows server default read
729 sizes. Add support for readpages
733 Fix oops when username not passed in on mount. Extensive fixes and improvements
734 to error logging (strip redundant newlines, change debug macros to ensure newline
735 passed in and to be more consistent). Fix writepage wrong file handle problem,
736 a readonly file handle could be incorrectly used to attempt to write out
737 file updates through the page cache to multiply open files. This could cause
738 the iozone benchmark to fail on the fwrite test. Fix bug mounting two different
739 shares to the same Windows server when using different usernames
740 (doing this to Samba servers worked but Windows was rejecting it) - now it is
741 possible to use different userids when connecting to the same server from a
742 Linux client. Fix oops when treeDisconnect called during unmount on
743 previously freed socket.
747 Fix oops in readpages caused by not setting address space operations in inode in
752 Includes support for deleting of open files and renaming over existing files (per POSIX
753 requirement). Add readlink support for Windows junction points (directory symlinks).
757 Changed read and write to go through pagecache. Added additional address space operations.
758 Memory mapped operations now working.
762 Added writepage code for additional memory mapping support. Fixed leak in xids causing
763 the simultaneous operations counter (/proc/fs/cifs/SimultaneousOps) to increase on
764 every stat call. Additional formatting cleanup.
768 Fix bigendian bug in order of time conversion. Merge 2.5 to 2.4 version. Formatting cleanup.
772 Fixes from Zwane Mwaikambo for adding missing return code checking in a few places.
773 Also included a modified version of his fix to protect global list manipulation of
774 the smb session and tree connection and mid related global variables.
778 Fix problem with captive thread hanging around at unmount time. Adjust to 2.5.42-pre
779 changes to superblock layout. Remove wasteful allocation of smb buffers (now the send
780 buffer is reused for responses). Add more oplock handling. Additional minor cleanup.
784 More stylistic updates to better match kernel style. Add additional statistics
785 for filesystem which can be viewed via /proc/fs/cifs. Add more pieces of NTLMv2
786 and CIFS Packet Signing enablement.
790 Replace call to sleep_on with safer wait_on_event.
791 Make stylistic changes to better match kernel style recommendations.
792 Remove most typedef usage (except for the PDUs themselves).
796 Update mount so the -unc mount option is no longer required (the ip address can be specified
797 in a UNC style device name. Implementation of readpage/writepage started.
801 Fix intermittent problem with incorrect smb header checking on badly
802 fragmented tcp responses
806 Fixes to setting of allocation size and file size.
810 Various 2.5.38 fixes. Now works on 2.5.38
814 Prepare for 2.5 kernel merge. Remove ifdefs.
818 Socket buffer management fixes. Fix dual free.
822 Various big endian fixes for hardlinks and symlinks and also for dfs.
826 Various big endian fixes for servers with Unix extensions such as Samba
830 Various FindNext fixes for incorrect filenames on large directory searches on big endian
831 clients. basic posix file i/o tests now work on big endian machines, not just le
835 SessionSetup and NegotiateProtocol now work from Big Endian machines.
836 Various Big Endian fixes found during testing on the Linux on 390. Various fixes for compatibility with older
837 versions of 2.4 kernel (now builds and works again on kernels at least as early as 2.4.7).
841 Various minor fixes for Connectathon Posix "basic" file i/o test suite. Directory caching fixed so hardlinked
842 files now return the correct number of links on fstat as they are repeatedly linked and unlinked.
846 Implemented "Raw" (i.e. not encapsulated in SPNEGO) NTLMSSP (i.e. the Security Provider Interface used to negotiate
847 session advanced session authentication). Raw NTLMSSP is preferred by Windows 2000 Professional and Windows XP.
848 Began implementing support for SPNEGO encapsulation of NTLMSSP based session authentication blobs
849 (which is the mechanism preferred by Windows 2000 server in the absence of Kerberos).
853 Introduced optional mount helper utility mount.cifs and made coreq changes to cifs vfs to enable
854 it. Fixed a few bugs in the DFS code (e.g. bcc two bytes too short and incorrect uid in PDU).
858 Rewrote much of connection and mount/unmount logic to handle bugs with
859 multiple uses to same share, multiple users to same server etc.
863 Fixed major problem with dentry corruption (missing call to dput)
867 Rewrite of readdir code to fix bug. Various fixes for bigendian machines.
868 Begin adding oplock support. Multiusermount and oplockEnabled flags added to /proc/fs/cifs
869 although corresponding function not fully implemented in the vfs yet
873 Fixed dentry caching bug, misc. cleanup
877 Fixed 2.5 support to handle build and configure changes as well as misc. 2.5 changes. Now can build
878 on current 2.5 beta version (2.5.24) of the Linux kernel as well as on 2.4 Linux kernels.
879 Support for STATUS codes (newer 32 bit NT error codes) added. DFS support begun to be added.
883 Unix extensions (symlink, readlink, hardlink, chmod and some chgrp and chown) implemented
884 and tested against Samba 2.2.5
889 1) Fixed lockrange to be correct (it was one byte too short)
891 2) Fixed GETLK (i.e. the fcntl call to test a range of bytes in a file to see if locked) to correctly
892 show range as locked when there is a conflict with an existing lock.
894 3) default file perms are now 2767 (indicating support for mandatory locks) instead of 777 for directories
895 in most cases. Eventually will offer optional ability to query server for the correct perms.
897 3) Fixed eventual trap when mounting twice to different shares on the same server when the first succeeded
898 but the second one was invalid and failed (the second one was incorrectly disconnecting the tcp and smb
901 4) Fixed error logging of valid mount options
903 5) Removed logging of password field.
905 6) Moved negotiate, treeDisconnect and uloggoffX (only tConx and SessSetup remain in connect.c) to cifssmb.c
906 and cleaned them up and made them more consistent with other cifs functions.
908 7) Server support for Unix extensions is now fully detected and FindFirst is implemented both ways
909 (with or without Unix extensions) but FindNext and QueryPathInfo with the Unix extensions are not completed,
910 nor is the symlink support using the Unix extensions
912 8) Started adding the readlink and follow_link code
git://ftp.safe.ca / safe / jmp / linux-2.6 / blob