- 19 May, 2011 28 commits
-
-
Sean Finney authored
A relatively minor nit, but also clarified the "consensus" from the preceding comments that it is in fact better to try for the kstrdup early and cleanup while cleaning up is still a simple thing to do. Reviewed-By:
Steve French <smfrench@gmail.com> Signed-off-by:
Sean Finney <seanius@seanius.net> Signed-off-by:
Pavel Shilovsky <piastry@etersoft.ru> Signed-off-by:
Steve French <sfrench@us.ibm.com>
-
Sean Finney authored
With CONFIG_DFS_UPCALL enabled, maintain the submount options in cifs_sb->mountdata, simplifying the code just a bit as well as making corner-case allocation problems less likely. Reviewed-by:
Jeff Layton <jlayton@redhat.com> Signed-off-by:
-
Sean Finney authored
To keep strings passed to cifs_parse_mount_options re-usable (which is needed to clean up the DFS referral handling), tokenize a copy of the mount options instead. If values are needed from this tokenized string, they too must be duplicated (previously, some options were copied and others duplicated). Since we are not on the critical path and any cleanup is relatively easy, the extra memory usage shouldn't be a problem (and it is a bit simpler than trying to implement something smarter). Reviewed-by:
-
Sean Finney authored
Windows 2008 CIFS servers do not always return PATH_NOT_COVERED when attempting to access a DFS share. Therefore, when checking for remote shares, unconditionally ask for a DFS referral for the UNC (w/out prepath) before continuing with previous behavior of attempting to access the UNC + prepath and checking for PATH_NOT_COVERED. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=31092Reviewed-by:
-
Sean Finney authored
The logic behind the expansion of DFS referrals is now extracted from cifs_mount into a new static function, expand_dfs_referral. This will reduce duplicate code in upcoming commits. Reviewed-by:
-
Jeff Layton authored
It's a bad idea to have macro functions that reference variables more than once, as the arguments could have side effects. Turn BCC() into a static inlined function instead. While we're at it, make it return a void * to discourage anyone from dereferencing it as-is. Reported-and-acked-by:
David Howells <dhowells@redhat.com> Signed-off-by:
-
Jeff Layton authored
This is the same patch as originally posted, just with some merge conflicts fixed up... Currently, the ByteCount is usually converted to host-endian on receive. This is confusing however, as we need to keep two sets of routines for accessing it, and keep track of when to use each routine. Munging received packets like this also limits when the signature can be calulated. Simplify the code by keeping the received ByteCount in little-endian format. This allows us to eliminate a set of routines for accessing it and we can now drop the *_le suffixes from the accessor functions since that's now implied. While we're at it, switch all of the places that read the ByteCount directly to use the get_bcc inline which should also clean up some unaligned accesses. Signed-off-by:
-
Jeff Layton authored
fs/cifs/cifsacl.c: In function ‘id_rb_search’: fs/cifs/cifsacl.c:215:19: warning: variable ‘linkto’ set but not used [-Wunused-but-set-variable] fs/cifs/cifsacl.c:214:18: warning: variable ‘parent’ set but not used [-Wunused-but-set-variable] Reviewed-by:
Shirish Pargaonkar <shirishpargaonkar@gmail.com> Signed-off-by:
-
Pavel Shilovsky authored
Simplify many places when we call cifs_revalidate/invalidate to make it do what it exactly needs. Reviewed-by:
Jeff Layton <jlayton@samba.org> Signed-off-by:
-
Pavel Shilovsky authored
Recently introduced strictcache mode brought a new code that can be efficiently used by directio part. That's let us add vectored operations and break unnecessary cifs_user_read and cifs_user_write. Signed-off-by:
-
Steve French authored
There is one big endian field in the cifs protocol, the RFC1001 length, which cifs code (unlike in the smb2 code) had been handling as u32 until the last possible moment, when it was converted to be32 (its native form) before sending on the wire. To remove the last sparse endian warning, and to make this consistent with the smb2 implementation (which always treats the fields in their native size and endianness), convert all uses of smb_buf_length to be32. This version incorporates Christoph's comment about using be32_add_cpu, and fixes a typo in the second version of the patch. Signed-off-by: -
Shirish Pargaonkar authored
rb tree search and insertion routines. A SID which needs to be mapped, is looked up in one of the rb trees depending on whether SID is either owner or group SID. If found in the tree, a (mapped) id from that node is assigned to uid or gid as appropriate. If unmapped, an upcall is attempted to map the SID to an id. If upcall is successful, node is marked as mapped. If upcall fails, node stays marked as unmapped and a mapping is attempted again only after an arbitrary time period has passed. To map a SID, which can be either a Owner SID or a Group SID, key description starts with the string "os" or "gs" followed by SID converted to a string. Without "os" or "gs", cifs.upcall does not know whether SID needs to be mapped to either an uid or a gid. Nodes in rb tree have fields to prevent multiple upcalls for a SID. Searching, adding, and removing nodes is done within global locks. Whenever a node is either found or inserted in a tree, a reference is taken on that node. Shrinker routine prunes a node if it has expired but does not prune an expired node if its refcount is not zero (i.e. sid/id of that node is_being/will_be accessed). Thus a node, if its SID needs to be mapped by making an upcall, can safely stay and its fields accessed without shrinker pruning it. A reference (refcount) is put on the node without holding the spinlock but a reference is get on the node by holding the spinlock. Every time an existing mapped node is accessed or mapping is attempted, its timestamp is updated to prevent it from getting erased or a to prevent multiple unnecessary repeat mapping retries respectively. For now, cifs.upcall is only used to map a SID to an id (uid or gid) but it would be used to obtain an SID for an id. Signed-off-by:
-
Shirish Pargaonkar authored
Define (global) data structures to store ids, uids and gids, to which a SID maps. There are two separate trees, one for SID/uid and another one for SID/gid. A new type of key, cifs_idmap_key_type, is used. Keys are instantiated and searched using credential of the root by overriding and restoring the credentials of the caller requesting the key. Id mapping functions are invoked under config option of cifs acl. Signed-off-by:
-
Pavel Shilovsky authored
Add this let us drop filemap_write_and_wait from cifs_invalidate_mapping and simplify the code to properly process invalidate logic. Reviewed-by:
-
Steve French authored
As with Linux nfs client, which uses "nfsvers=" or "vers=" to indicate which protocol to use for mount, specifying "vers=smb2" or "vers=2" will force an smb2 mount. When vers is not specified cifs is used ie "vers=cifs" or "vers=1" We can eventually autonegotiate down from smb2 to cifs when smb2 is stable enough to make it the default, but this is for the future. At that time we could also implement a "maxprotocol" mount option as smbclient and Samba have today, but that would be premature until smb2 is stable. Intially the smb2 Kconfig option will depend on "BROKEN" until the merge is complete, and then be "EXPERIMENTAL" When it is no longer experimental we can consider changing the default protocol to attempt first. Reviewed-by:
-
Pavel Shilovsky authored
Use invalidate_inode_pages2 that don't leave pages even if shrink_page_list() has a temp ref on them. It prevents a data coherency problem when cifs_invalidate_mapping didn't invalidate pages but the client thinks that a data from the cache is uptodate according to an oplock level (exclusive or II). Signed-off-by:
-
Jeff Layton authored
The comment about checking the bcc is in the wrong place. Also make it match kernel coding style. Reported-and-acked-by:
-
Jeff Layton authored
I originally intended to remove this warning in 2.6.34, but it's not in a high performance codepath and might help us to catch bugs later. Let's keep it, but fix the comment to allay confusion about its removal. Signed-off-by:
-
Steve French authored
Allow setting cifs_acl on the server. Pass on to the server the ACL blob generated by an application. cifs is just a pass-through, it does not monitor or inspect the contents of the blob, server decides whether to enforce/apply the ACL blob composed by an application. If setting of ACL is succeessful, mark the inode for revalidation. Signed-off-by:
-
Steve French authored
local cifs functions (repost) Using kernel crypto APIs for DES encryption during LM and NT hash generation instead of local functions within cifs. Source file smbdes.c is deleted sans four functions, one of which uses ecb des functionality provided by kernel crypto APIs. Remove function SMBOWFencrypt. Add return codes to various functions such as calc_lanman_hash, SMBencrypt, and SMBNTencrypt. Includes fix noticed by Dan Carpenter. Signed-off-by:
-
Shirish Pargaonkar authored
Remove config flag CIFS_EXPERIMENTAL. Do export operations under new config flag CIFS_NFSD_EXPORT Signed-off-by:
-
Steve French authored
SMB2 is the followon to the CIFS (and SMB) protocols and the default for Windows since Windows Vista, and also now implemented by various non-Windows servers. SMB2 is more secure, has various performance advantages, including larger i/o sizes, flow control, better caching model and more. SMB2 also resolves some scalability limits in the cifs protocol and adds many new features while being much simpler (only a few dozen commands instead of hundreds) and since the protocol is clearer it is also more consistently implemented across servers and thus easier to optimize. After much discussion with Jeff Layton, Jeremy Allison and others at Connectathon, we decided to move the smb2 code from a distinct .ko and fstype into distinct C files that optionally build in cifs.ko. As a result the Kconfig gets simpler. To avoid destabilizing cifs, the smb2 code is going to be moved into its own experimental CONFIG_CIFS_SMB2 ifdef as it is merged and rereviewed. The changes to stable cifs (builds with the smb2 ifdef off) are expected to be fairly small. Reviewed-by:
-
Steve French authored
We were reserving MAX_USERNAME (now 256) on stack for something which only needs to fit about 24 bytes ie string krb50x + printf version of uid Signed-off-by: -
Justin P. Mattock authored
The patch below removes an extra "l" in the word. Signed-off-by:
Justin P. Mattock <justinmattock@gmail.com> Signed-off-by:
-
Steve French authored
Recent Windows versions now create symlinks more frequently and they do use this "reparse point" symlink mechanism. We can of course do symlinks nicely to Samba and other servers which support the CIFS Unix Extensions and we can also do SFU symlinks and "client only" "MF" symlinks optionally, but for recent Windows we currently can not handle the common "reparse point" symlinks fully, removing the caller for this. We will need to extend and reenable this "reparse point" worker code in cifs and fix cifs_symlink to call this. In the interim this code has been moved to its own config option so it is not compiled in by default until cifs_symlink fixed up (and tested) to use this. CC: Jeff Layton <jlayton@redhat.com> Signed-off-by: -
Steve French authored
The CIFSSMBNotify worker is unused, pending changes to allow it to be called via inotify, so move it into its own experimental config option so it does not get built in, until the necessary VFS support is fixed. It used to be used in dnotify, but according to Jeff, inotify needs minor changes before we can reenable this. CC: Jeff Layton <jlayton@redhat.com> Signed-off-by: -
Shirish Pargaonkar authored
ino is unused in function cifs_root_iget(). Signed-off-by:
-
Linus Torvalds authored
-
- 18 May, 2011 12 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2Linus Torvalds authored
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jlbec/ocfs2: configfs: Fix race between configfs_readdir() and configfs_d_iput() configfs: Don't try to d_delete() negative dentries. ocfs2/dlm: Target node death during resource migration leads to thread spin ocfs2: Skip mount recovery for hard-ro mounts ocfs2/cluster: Heartbeat mismatch message improved ocfs2/cluster: Increase the live threshold for global heartbeat ocfs2/dlm: Use negotiated o2dlm protocol version ocfs2: skip existing hole when removing the last extent_rec in punching-hole codes. ocfs2: Initialize data_ac (might be used uninitialized)
-
git://git.secretlab.ca/git/linux-2.6Linus Torvalds authored
* 'devicetree/merge' of git://git.secretlab.ca/git/linux-2.6: drivercore: revert addition of of_match to struct device of: fix race when matching drivers
-
git://git.linux-mips.org/pub/scm/upstream-linusLinus Torvalds authored
* 'upstream' of git://git.linux-mips.org/pub/scm/upstream-linus: MIPS: Kludge IP27 build for 2.6.39. MIPS: AR7: Fix GPIO register size for Titan variant. MIPS: Fix duplicate invocation of notify_die. MIPS: RB532: Fix iomap resource size miscalculation.
-
Grant Likely authored
Commit b826291c, "drivercore/dt: add a match table pointer to struct device" added an of_match pointer to struct device to cache the of_match_table entry discovered at driver match time. This was unsafe because matching is not an atomic operation with probing a driver. If two or more drivers are attempted to be matched to a driver at the same time, then the cached matching entry pointer could get overwritten. This patch reverts the of_match cache pointer and reworks all users to call of_match_device() directly instead. Signed-off-by:
Grant Likely <grant.likely@secretlab.ca>
-
Milton Miller authored
If two drivers are probing devices at the same time, both will write their match table result to the dev->of_match cache at the same time. Only write the result if the device matches. In a thread titled "SBus devices sometimes detected, sometimes not", Meelis reported his SBus hme was not detected about 50% of the time. From the debug suggested by Grant it was obvious another driver matched some devices between the call to match the hme and the hme discovery failling. Reported-by:
Meelis Roos <mroos@linux.ee> Signed-off-by:
Milton Miller <miltonm@bga.com> [grant.likely: modified to only call of_match_device() once] Signed-off-by:
-
git://git.kernel.dk/linux-2.6-blockLinus Torvalds authored
* 'for-linus' of git://git.kernel.dk/linux-2.6-block: block: don't delay blk_run_queue_async scsi: remove performance regression due to async queue run blk-throttle: Use task_subsys_state() to determine a task's blkio_cgroup block: rescan partitions on invalidated devices on -ENOMEDIA too cdrom: always check_disk_change() on open block: unexport DISK_EVENT_MEDIA_CHANGE for legacy/fringe drivers
-
Ralf Baechle authored
Signed-off-by:Ralf Baechle <ralf@linux-mips.org>
-
Florian Fainelli authored
The 'size' variable contains the correct register size for both AR7 and Titan, but we never used it to ioremap the correct register size. This problem only shows up on Titan. [ralf@linux-mips.org: Fixed the fix. The original patch as in patchwork recognizes the problem correctly then fails to fix it ...] Reported-by:
Alexander Clouter <alex@digriz.org.uk> Signed-off-by:
Florian Fainelli <florian@openwrt.org> Patchwork: https://patchwork.linux-mips.org/patch/2380/Signed-off-by:
-
Ralf Baechle authored
Initial patch by Yury Polyanskiy <ypolyans@princeton.edu>. Signed-off-by:
-
Ralf Baechle authored
This is the MIPS portion of Joe Perches <joe@perches.com>'s https://patchwork.linux-mips.org/patch/2172/ which seems to have been lost in time and space. Signed-off-by:
-
Joel Becker authored
configfs_readdir() will use the existing inode numbers of inodes in the dcache, but it makes them up for attribute files that aren't currently instantiated. There is a race where a closing attribute file can be tearing down at the same time as configfs_readdir() is trying to get its inode number. We want to get the inode number of open attribute files, because they should match while instantiated. We can't lock down the transition where dentry->d_inode is set to NULL, so we just check for NULL there. We can, however, ensure that an inode we find isn't iput() in configfs_d_iput() until after we've accessed it. Signed-off-by:Joel Becker <jlbec@evilplan.org>
-
Joel Becker authored
When configfs is faking mkdir() on its subsystem or default group objects, it starts by adding a negative dentry. It then tries to instantiate the group. If that should fail, it must clean up after itself. I was using d_delete() here, but configfs_attach_group() promises to return an empty dentry on error. d_delete() explodes with the entry dentry. Let's try d_drop() instead. The unhashing is what we want for our dentry. Signed-off-by:
-
