staging/lustre: pass fsync() range through RPC/IO stack
authorAndreas Dilger <andreas.dilger@intel.com>
Sun, 27 Apr 2014 17:07:09 +0000 (13:07 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sun, 27 Apr 2014 17:31:01 +0000 (10:31 -0700)
commit05289927d029ffed093424acac0b12f8ee06bfc2
tree0c726dae4e637ae33121a7030651e79b3fbffce6
parentb7d0254ce83019aebdc6bc3a7dcd4e720895ef1f
staging/lustre: pass fsync() range through RPC/IO stack

The Linux VFS and Lustre OST_SYNC RPC are both capable of specifying
fsync() on a sub-extent of the file {start, end} instead of the full
file.  This allows less than the full amount of data to be flushed,
reducing or possibly eliminating the work needed before the syscall
can return.

However, the handling of sub-extent of the file for fsync was lost
with the move to CLIO on the client and OSD API on the server.  They
were ignoring the passed {start, end} and using {0, OBD_OBJECT_EOF}
instead.

Return the ability to pass a sub-extent for fsync() from the client,
to the specific stripes/OSTs that need the sync operation, and pass
it down to the OSD.  The ZFS OSD doesn't handle this yet, but there
is room for improvement in a separate patch.

Signed-off-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-on: http://review.whamcloud.com/8626
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4388
Reviewed-by: Bobi Jam <bobijam@gmail.com>
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/staging/lustre/lustre/include/dt_object.h
drivers/staging/lustre/lustre/llite/file.c