TAKE - Fix lockup on highmem system

[Eric Sandeen <sandeen@xxxxxxx>]

The addition of __pb_block_prepare_write to pagebuf_iozero
caused us to kmap a page twice... this will break on highmem
systems.  Change the second kmap() to page_address().


Date:  Mon Mar 11 14:17:53 PST 2002
Workarea:  stout.americas.sgi.com:/localhome/eric/2.4.x-xfs/workarea-alwaysclean

The following file(s) were checked into:
  bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs


Modid:  2.4.x-xfs:slinx:113806a
linux/fs/xfs/pagebuf/page_buf_io.c - 1.16
        - Don't try to kmap() a page twice in a row...