__getblk() spinning when size != bdev->...i_blkbits

[Zach Brown <zach.brown@oracle.com>]

Recently I debugged something that was spinning in the for(;;) in
__getblk_slow().  It turned out it was calling __getblk() with a size
argument that didn't match the bdev's inode's i_blkbits.

grow_buffers() would add a page at the index calculated from the 4k size
argument but when __getblk_slow() got back around to looking for the
instantiated page __find_get_block_slow() would look at the index
calculated from the 1k i_blkbits and not find the new page.  Round and
round she goes.

If the different size inputs are intentional, would a simple BUG_ON() be
acceptible so that we can tell straight away if the caller messed up?

- z