kernel.git/fs/hfsplus/super.c, branch linux-2.6.37.y

new helper: mount_bdev()

2010-10-29T08:16:13Z

... and switch of the obvious get_sb_bdev() users to ->mount() Signed-off-by: Al Viro

hfsplus: use atomic bitops for the superblock flags

2010-10-01T03:45:20Z

The flags in the HFS+-specific superlock do get modified during runtime, use atomic bitops to make the modifications SMP safe. Signed-off-by: Christoph Hellwig

hfsplus: add per-superblock lock for volume header updates

2010-10-01T03:45:08Z

Lock updates to the mutal fields in the volume header, and document the locing in the hfsplus_sb_info structure. Signed-off-by: Christoph Hellwig

hfsplus: remove the rsrc_inodes list

2010-10-01T03:44:02Z

We never walk the list - the only reason for it is to make the resource fork inodes appear hashed to the writeback code. Borrow a trick from JFS to do that without needing a list head. Signed-off-by: Christoph Hellwig

hfsplus: do not cache and write next_alloc

2010-10-01T03:43:58Z

We never look at it, nor change the next_alloc field in the superblock. So don't bother caching it or writing it out in hfsplus_sync_fs. Signed-off-by: Christoph Hellwig

hfsplus: clean up hfsplus_write_inode

2010-10-01T03:43:43Z

Add a new hfsplus_system_write_inode for writing the special system inodes and streamline the fastpath write_inode code. Signed-off-by: Christoph Hellwig

hfsplus: clean up hfsplus_iget

2010-10-01T03:43:41Z

Add a new hfsplus_system_read_inode for reading the special system inodes and streamline the fastpath iget code. Signed-off-by: Christoph Hellwig

hfsplus: fix HFSPLUS_I calling convention

2010-10-01T03:43:31Z

HFSPLUS_I doesn't return a pointer to the hfsplus-specific inode information like all other FOO_I macros, but dereference the pointer in a way that made it look like a direct struct derefence. This only works as long as the HFSPLUS_I macro is used directly and prevents us from keepig a local hfsplus_inode_info pointer. Fix the calling convention and introduce a local hip variable in all functions that use it constantly. Signed-off-by: Christoph Hellwig

hfsplus: fix HFSPLUS_SB calling convention

2010-10-01T03:42:59Z

HFSPLUS_SB doesn't return a pointer to the hfsplus-specific superblock information like all other FOO_SB macros, but dereference the pointer in a way that made it look like a direct struct derefence. This only works as long as the HFSPLUS_SB macro is used directly and prevents us from keepig a local hfsplus_sb_info pointer. Fix the calling convention and introduce a local sbi variable in all functions that use it constantly. Signed-off-by: Christoph Hellwig

hfsplus: remove BKL from hfsplus_put_super

2010-10-01T03:41:53Z

Except for ->put_super the BKL is now gone from HFS, which means it's superflous there too as ->put_super is serialized by the VFS. Signed-off-by: Christoph Hellwig