kernel client
- setattr should check lease before believing something is a no-op
- trim expired leases so we don't indefinitely hold dcache refs...
-- carry wrbuffer/rdcache caps until data is flushed
- - this should make the utimes bit kick in
- procfs/debugfs
- adjust granular debug levels too
- should we be using debugfs?
- convert rest of ceph_fs.h to le32/le64 notation, cleaning up userspace appropriately
- vfs
- can we use dentry_path(), if it gets merged into mainline?
-- osd client
- - readpages (readahead)
- - async (caching) mode
- - sync mode (write-through)
+- io / osd client
+ - carry wrbuffer/rdcache caps until data is flushed
+ - this should make the utimes bit kick in
+ - invalidate cache pages?
+ - kick requests when new map arrives
+ - sync?
+ - sync on unmount?
- capability changes (flush.. need way to initiate/block on writeback initiated by mds)
- osd ack vs commit handling. hmm!
- - handle map changes (resubmit ops)
- mon client
- work out message resend logic..?
- unmount
}
+static int ceph_syncfs(struct super_block *sb, int wait)
+{
+ dout(10, "sync_fs %d\n", wait);
+ return 0;
+}
+
+
/**
* ceph_show_options - Show mount options in /proc/mounts
* @m: seq_file to write to
.alloc_inode = ceph_alloc_inode,
.destroy_inode = ceph_destroy_inode,
.write_inode = ceph_write_inode,
-/* .sync_fs = ceph_syncfs, */
+ .sync_fs = ceph_syncfs,
.put_super = ceph_put_super,
.show_options = ceph_show_options,
.statfs = ceph_statfs,