This should fix the assert seen from debug kernels, and the NFS oops
reported by Matteo Centonza. Assorted other I/O path problems from the
last couple of days should probably settle down as well.
Steve
Date: Thu Apr 5 15:25:46 PDT 2001
Workarea: jen.americas.sgi.com:/src/lord/xfs-linux.2.4
The following file(s) were checked into:
bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs
Modid: 2.4.x-xfs:slinx:91786a
linux/fs/xfs/linux/xfs_lrw.c - 1.88
- cleanup the eof hole zeroing code to be more straightforward
linux/include/linux/page_buf.h - 1.85
- changed proto for pagebuf_iozero
linux/fs/pagebuf/page_buf_io.c - 1.75
- Fix pagebuf_iozero to call the higher level commit_write function and
bump the inode size as required. Remove the cases were prepare_write
was
not being called - as this was broken (assert trip and sometimes
panic).
|