Rework XFS read/write path so that there is one common read and one common
write path for all the different I/O variants. This means we can now support
true async I/O.
Date: Tue Jul 29 09:33:46 PDT 2003
Workarea: jen.americas.sgi.com:/src/lord/xfs-linux.2.5
The following file(s) were checked into:
bonnie.engr.sgi.com:/isms/slinx/2.5.x-xfs
Modid: 2.5.x-xfs:slinx:154475a
linux/mm/filemap.c - 1.160
- Export __generic_file_aio_read so xfs can use it
linux/include/linux/fs.h - 1.220
- Prototype for __generic_file_aio_read
linux/fs/xfs/linux/xfs_lrw.h - 1.38
- change prototypes for xfs_read and xfs_write
linux/fs/xfs/linux/xfs_lrw.c - 1.190
- Pass an iocb down through xfs_read and xfs_write to allow them to be
used
for true async I/O.
linux/fs/xfs/linux/xfs_file.c - 1.92
- lip over to using the aio interface for read write. Change the next
layer down to take an iocb instead of a file pointer.
linux/fs/xfs/linux/xfs_vnode.h - 1.83
- pass an iocb down into the read and write path
|