At the bottom of pagebuf_delalloc_convert there's a test to make
sure we really get something out of bmap, sort of a safety
valve so that if we were asking for direct allocation, but
got back a delayed allocation, we try again. We were testing
"flags == PBF_DIRECT" when it should be "flags & PBF_DIRECT"
since PBF_WRITE would also be set.
If we hit this, it probably means something else has not gone
as expected, but it's a safety valve that was "stuck" until
this change.
Date: Sun Feb 10 19:34:28 PST 2002
Workarea: stout.americas.sgi.com:/localhome/eric/2.4.x-xfs/workarea-reallyclean
The following file(s) were checked into:
bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs
Modid: 2.4.x-xfs:slinx:111476a
linux/fs/xfs/pagebuf/page_buf_io.c - 1.7
- fix test at bottom of pagebuf_delalloc_convert
|