pagebuf can now (theoretically) take a configurable sector size.
This is effectively no visible change to users though - there is
still no way to change the sector size in XFS. This'll also let
me go and do a bit of work on the tools (several header macros I
need now exist), and then this can be revisited later.
Date: Tue Nov 12 23:49:10 PST 2002
The following file(s) were checked into:
linux/fs/xfs/xfs_sb.h - 1.54
- Add macros for doing sect to BB conversions, FSB to sect conversions,
linux/fs/xfs/xfs_vfsops.c - 1.399
- Set the device sector size based on the (existing) superblock field
value (currently this is always 512), rather than hardcoding the
value to 512.
linux/fs/xfs/xfs_mount.h - 1.163
- Squeeze in an 8 byte sectbb field (no change to xfs_mount_t size),
similar semantics to blkbb log field (except for sector size, not
linux/fs/xfs/xfs_mount.c - 1.312
- Read a sector from the end of the device when checking log/data
device sizes, not a BB. Currently only 512 byte sectors exist,
so no real change here. Move mount type initialisation earlier
on in the piece.
linux/fs/xfs/xfs_alloc_btree.h - 1.22
- Add some macros relating to sector sizes, update some comments.
linux/fs/xfs/linux/xfs_super.h - 1.32
linux/fs/xfs/linux/xfs_super.c - 1.229
- When initialising a new pagebuf target, allow a sector size to be
linux/fs/xfs/pagebuf/page_buf.c - 1.76
linux/fs/xfs/pagebuf/page_buf.h - 1.45
- Remove the hardcoded sector size of 512 bytes, make it variable.