diff options
| author |   Gregoire Pichon <gregoire.pichon@bull.net> | 2016-11-10 10:51:13 -0500 |
|---|---|---|
| committer |   Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2016-11-15 11:01:17 +0100 |
| commit | 7cb15d044867c83b27ed434fe7002797846002c0 (patch) | |
| tree | 0c929206e606c24ecf14362d44219be4fb6676b6 /drivers/staging/lustre/lustre/mdc/mdc_locks.c | |
| parent | staging: lustre: hsm: Use file lease to implement migration (diff) | |
| download | linux-dev-7cb15d044867c83b27ed434fe7002797846002c0.tar.xz linux-dev-7cb15d044867c83b27ed434fe7002797846002c0.zip | |
staging: lustre: mdc: manage number of modify RPCs in flight
This patch is the main client part of a new feature that supports
multiple modify metadata RPCs in parallel. Its goal is to improve
metadata operations performance of a single client, while maintening
the consistency of MDT reply reconstruction and MDT recovery
mechanisms.
It allows to manage the number of modify RPCs in flight within
the client obd structure and to assign a virtual index (the tag) to
each modify RPC to help server side cleaning of reply data.
The mdc component uses this feature to send multiple modify RPCs
in parallel.
Signed-off-by: Gregoire Pichon <gregoire.pichon@bull.net>
Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-5319
Reviewed-on: http://review.whamcloud.com/14374
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Alex Zhuravlev <alexey.zhuravlev@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'drivers/staging/lustre/lustre/mdc/mdc_locks.c')
| -rw-r--r-- | drivers/staging/lustre/lustre/mdc/mdc_locks.c | 13 |
1 files changed, 7 insertions, 6 deletions
diff --git a/drivers/staging/lustre/lustre/mdc/mdc_locks.c b/drivers/staging/lustre/lustre/mdc/mdc_locks.c index b9ca14005800..42a128faf03c 100644 --- a/drivers/staging/lustre/lustre/mdc/mdc_locks.c +++ b/drivers/staging/lustre/lustre/mdc/mdc_locks.c @@ -766,15 +766,16 @@ resend: req->rq_sent = ktime_get_real_seconds() + resends; } - /* It is important to obtain rpc_lock first (if applicable), so that - * threads that are serialised with rpc_lock are not polluting our - * rpcs in flight counter. We do not do flock request limiting, though + /* It is important to obtain modify RPC slot first (if applicable), so + * that threads that are waiting for a modify RPC slot are not polluting + * our rpcs in flight counter. + * We do not do flock request limiting, though */ if (it) { - mdc_get_rpc_lock(obddev->u.cli.cl_rpc_lock, it); + mdc_get_mod_rpc_slot(req, it); rc = obd_get_request_slot(&obddev->u.cli); if (rc != 0) { - mdc_put_rpc_lock(obddev->u.cli.cl_rpc_lock, it); + mdc_put_mod_rpc_slot(req, it); mdc_clear_replay_flag(req, 0); ptlrpc_req_finished(req); return rc; @@ -801,7 +802,7 @@ resend: } obd_put_request_slot(&obddev->u.cli); - mdc_put_rpc_lock(obddev->u.cli.cl_rpc_lock, it); + mdc_put_mod_rpc_slot(req, it); if (rc < 0) { CDEBUG(D_INFO, "%s: ldlm_cli_enqueue failed: rc = %d\n", |