summaryrefslogtreecommitdiffstats
path: root/fs
AgeCommit message (Collapse)Author
2010-08-24ceph: initialize fields on new dentry_infosSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-24ceph: maintain i_head_snapc when any caps are dirty, not just for dataSage Weil
We used to use i_head_snapc to keep track of which snapc the current epoch of dirty data was dirtied under. It is used by queue_cap_snap to set up the cap_snap. However, since we queue cap snaps for any dirty caps, not just for dirty file data, we need to keep a valid i_head_snapc anytime we have dirty|flushing caps. This fixes a NULL pointer deref in queue_cap_snap when writing back dirty caps without data (e.g., snaptest-authwb.sh). Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: fix osd request lru adjustment when sending requestHenry C Chang
Fix argument order. We want to move the item to the end of the list, not change the position of the head. Signed-off-by: Henry C Chang <henry_c_chang@tcloudcomputing.com> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: don't improperly set dir complete when holding EXCL capSage Weil
If we hold the EXCL cap, we cannot trust the dir stats from the MDS (num files, subdirs) and must not incorrectly conclude that the directory is empty. If we do, we get can bad results from lookup (bad ENOENT) and bad readdir results. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22mm: exporting account_page_dirtyMichael Rubin
This allows code outside of the mm core to safely manipulate page state and not worry about the other accounting. Not using these routines means that some code will lose track of the accounting and we get bugs. This has happened once already. Signed-off-by: Michael Rubin <mrubin@google.com> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: direct requests in snapped namespace based on nonsnap parentSage Weil
When making a request in the virtual snapdir or a snapped portion of the namespace, we should choose the MDS based on the first nonsnap parent (and its caps). If that is not the best place, we will get forward hints to find the right MDS in the cluster. This fixes ESTALE errors when using the .snap directory and namespace with multiple MDSs. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: queue cap snap writeback for realm children on snap updateSage Weil
When a realm is updated, we need to queue writeback on inodes in that realm _and_ its children. Otherwise, if the inode gets cowed on the server, we can get a hang later due to out-of-sync cap/snap state. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: include dirty xattrs state in snapped capsSage Weil
When we snapshot dirty metadata that needs to be written back to the MDS, include dirty xattr metadata. Make the capsnap reference the encoded xattr blob so that it will be written back in the FLUSHSNAP op. Also fix the capsnap creation guard to include dirty auth or file bits, not just tests specific to dirty file data or file writes in progress (this fixes auth metadata writeback). Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: fix xattr cap writebackSage Weil
We should include the xattr metadata blob in the cap update message any time we are flushing dirty state, NOT just when we are also dropping the cap. This fixes async xattr writeback. Also, clean up the code slightly to avoid duplicating the bit test. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-22ceph: fix multiple mds session shutdownSage Weil
The use of a completion when waiting for session shutdown during umount is inappropriate, given the complexity of the condition. For multiple MDS's, this resulted in the umount thread spinning, often preventing the session close message from being processed in some cases. Switch to a waitqueue and defined a condition helper. This cleans things up nicely. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-10ceph: generalize mon requests, add pool op supportYehuda Sadeh
Generalize the current statfs synchronous requests, and support pool_ops. Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-05ceph: only queue async writeback on cap revocation if there is dirty dataSage Weil
Normally, if the Fb cap bit is being revoked, we queue an async writeback. If there is no dirty data but we still hold the cap, this leaves the client sitting around doing nothing until the cap timeouts expire and the cap is released on its own (as it would have been without the revocation). Instead, only queue writeback if the bit is actually used (i.e., we have dirty data). If not, we can reply to the revocation immediately. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-03ceph: do not ignore osd_idle_ttl mount optionSage Weil
Actually apply the mount option to the mount_args struct. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-03ceph: constify dentry_operationsSage Weil
This makes checkpatch happy. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-03ceph: whitespace cleanupSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: add flock/fcntl lock supportGreg Farnum
Implement flock inode operation to support advisory file locking. All lock/unlock operations are synchronous with the MDS. Lock state is sent when reconnecting to a recovering MDS to restore the shared lock state. Signed-off-by: Greg Farnum <gregf@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: define on-wire types, constants for file locking supportGreg Farnum
Define the MDS operations and data types for doing file advisory locking with the MDS. Signed-off-by: Greg Farnum <gregf@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: add CEPH_FEATURE_FLOCK to the supported feature bitsGreg Farnum
This informs the server that we will accept v2 client_caps format and v2 client_reconnect format messages. Signed-off-by: Greg Farnum <gregf@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: support v2 reconnect encodingSage Weil
Encode either old or v2 encoding of client_reconnect message, depending on whether the peer has the FLOCK feature bit. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: support v2 client_caps encodingSage Weil
Add support for v2 encoding of MClientCaps, which includes a flock blob. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: move AES iv definition to shared headerSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-02ceph: fix decoding of pool snap infoSage Weil
The pool info contains a vector for snap_info_t, not snap ids. This fixes the broken decoding, which would declare teh update corrupt when a pool snapshot was created. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: make ->sync_fs not wait if wait==0Sage Weil
The ->sync_fs() super op only needs to wait if wait is true. Otherwise, just get some dirty cap writeback started. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: warn on missing snap realmSage Weil
Well, this Shouldn't Happen, so it would be helpful to know the caller when it does. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: print useful error message when crush rule not foundSage Weil
Include the crush_ruleset in the error message. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: use %pU to print uuid (fsid)Sage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: sync header defs with server codeSage Weil
Define ROLLBACK op, IFLOCK inode lock (for advisory file locking). Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: clean up header guardsSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: strip misleading/obsolete version, feature infoSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: specify supported features in super.hSage Weil
Specify the supported/required feature bits in super.h client code instead of using the definitions from the shared kernel/userspace headers (which will go away shortly). Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: clean up fsid mount optionSage Weil
Specify the fsid mount option in hex, not via the major/minor u64 hackery we had before. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: remove unused 'monport' mount optionSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: handle ESTALE properly; on receipt send to authority if it wasn'tGreg Farnum
Signed-off-by: Greg Farnum <gregf@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: add ceph_get_cap_for_mds function.Greg Farnum
Signed-off-by: Greg Farnum <gregf@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: connect to export targets on cap exportSage Weil
When we get a cap EXPORT message, make sure we are connected to all export targets to ensure we can handle the matching IMPORT. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: connect to export targets if mds is laggySage Weil
If an MDS we are talking to may have failed, we need to open sessions to its potential export targets to ensure that any in-progress migration that may have involved some of our caps is properly handled. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: introduce helper to connect to mds export targetsSage Weil
There are a few cases where we need to open sessions with a given mds's potential export targets. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: only set num_pages in calc_layoutSage Weil
Setting it elsewhere is unnecessary and more fragile. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: do caps accounting per mds_clientYehuda Sadeh
Caps related accounting is now being done per mds client instead of just being global. This prepares ground work for a later revision of the caps preallocated reservation list. Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: track laggy state of mds from mdsmapSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: code cleanupYehuda Sadeh
Mainly fixing minor issues reported by sparse. Signed-off-by: Yehuda Sadeh <yehuda@hq.newdream.net> Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: skip if no auth cap in flush_snapsSage Weil
If we have a capsnap but no auth cap (e.g. because it is migrating to another mds), bail out and do nothing for now. Do NOT remove the capsnap from the flush list. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: simplify caps revocation, fix for multimdsSage Weil
The caps revocation should either initiate writeback, invalidateion, or call check_caps to ack or do the dirty work. The primary question is whether we can get away with only checking the auth cap or whether all caps need to be checked. The old code was doing...something else. At the very least, revocations from non-auth MDSs could break by triggering the "check auth cap only" case. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: simplify add_cap_releasesSage Weil
No functional change, aside from more useful debug output. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: drop unused argumentSage Weil
Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: perform lazy reads when file mode and caps permitSage Weil
If the file mode is marked as "lazy," perform cached/buffered reads when the caps permit it. Adjust the rdcache_gen and invalidation logic accordingly so that we manage our cache based on the FILE_CACHE -or- FILE_LAZYIO cap bits. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: perform lazy writes when file mode and caps permitSage Weil
If we have marked a file as "lazy" (using the ceph ioctl), perform buffered writes when the MDS caps allow it. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: add LAZYIO ioctl to mark a file description for lazy consistencySage Weil
Allow an application to mark a file descriptor for lazy file consistency semantics, allowing buffered reads and writes when multiple clients are accessing the same file. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01ceph: request FILE_LAZYIO cap when LAZY file mode is setSage Weil
Also clean up the file flags -> file mode -> wanted caps functions while we're at it. This resyncs this file with userspace. Signed-off-by: Sage Weil <sage@newdream.net>
2010-08-01NFS: Fix a typo in include/linux/nfs_fs.hTrond Myklebust
nfs_commit_inode() needs to be defined irrespectively of whether or not we are supporting NFSv3 and NFSv4. Allow the compiler to optimise away code in the NFSv2-only case by converting it into an inlined stub function. Reported-and-tested-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>