staging/lustre/ptlrpc: false alarm in AT network latency measuring
authorLiang Zhen <liang.zhen@intel.com>
Thu, 26 Mar 2015 01:53:19 +0000 (21:53 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 26 Mar 2015 10:10:24 +0000 (11:10 +0100)
commit06fbc01a870eea4431ca340642da7bfd4a5a262e
treeee32e0402d1398450b0374d0f726537db99ceb41
parentf42894004727ffdfb3b0826b0047eaf44cbdece6
staging/lustre/ptlrpc: false alarm in AT network latency measuring

If early reply of client RPC is lost and client RPC is expired and
resent, server will drop the resent RPC because it's already in
processing, server may also send reply or early reply to client,
which can still match reply buffer of the original request.
In this case, client is measuring time from resent time, but server
is reporting service time of original RPC, which is longer than
the time measured by client.

Signed-off-by: Liang Zhen <liang.zhen@intel.com>
Reviewed-on: http://review.whamcloud.com/12855
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5545
Reviewed-by: Li Wei <wei.g.li@intel.com>
Reviewed-by: Johann Lombardi <johann.lombardi@intel.com>
Signed-off-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/staging/lustre/lustre/ptlrpc/client.c