From: James Simmons <jsimmons@infradead.org>
To: Andreas Dilger <adilger@whamcloud.com>,
Oleg Drokin <green@whamcloud.com>, NeilBrown <neilb@suse.de>
Cc: Andriy Skulysh <c17819@cray.com>,
Lustre Development List <lustre-devel@lists.lustre.org>
Subject: [lustre-devel] [PATCH 14/19] lustre: llite: skip request slot for lmv_revalidate_slaves()
Date: Sun, 28 Nov 2021 18:27:49 -0500 [thread overview]
Message-ID: <1638142074-5945-15-git-send-email-jsimmons@infradead.org> (raw)
In-Reply-To: <1638142074-5945-1-git-send-email-jsimmons@infradead.org>
From: Andriy Skulysh <c17819@cray.com>
Some syscalls need lmv_revalidate_slaves(). It requires
second lock enqueue and the it can be blocked by
lack of RPC slots.
Don't acquire rpc slot for second lock enqueue.
HPE-bug-id: LUS-8416
WC-bug-id: https://jira.whamcloud.com/browse/LU-15121
Lustre-commit: 7e781c605c4189ea1 ("LU-15121 llite: skip request slot for lmv_revalidate_slaves()")
Signed-off-by: Andriy Skulysh <c17819@cray.com>
Reviewed-on: https://review.whamcloud.com/45275
Reviewed-by: Vitaly Fertman <c17818@cray.com>
Reviewed-by: Alexander Zarochentsev <c17826@cray.com>
Reviewed-by: Vitaly Fertman <vitaly.fertman@hpe.com>
Reviewed-by: Alexander Zarochentsev <alexander.zarochentsev@hpe.com>
Reviewed-by: Lai Siyao <lai.siyao@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
---
fs/lustre/include/lustre_dlm.h | 7 +++++--
fs/lustre/include/obd.h | 1 +
fs/lustre/ldlm/ldlm_request.c | 18 +++++++++++-------
fs/lustre/llite/statahead.c | 1 +
fs/lustre/lmv/lmv_intent.c | 2 ++
fs/lustre/mdc/mdc_dev.c | 3 ++-
fs/lustre/mdc/mdc_locks.c | 5 +++--
fs/lustre/osc/osc_request.c | 2 +-
8 files changed, 26 insertions(+), 13 deletions(-)
diff --git a/fs/lustre/include/lustre_dlm.h b/fs/lustre/include/lustre_dlm.h
index 1fc199b..a2fe9676 100644
--- a/fs/lustre/include/lustre_dlm.h
+++ b/fs/lustre/include/lustre_dlm.h
@@ -1018,7 +1018,9 @@ struct ldlm_enqueue_info {
/* whether enqueue slave stripes */
unsigned int ei_enq_slave:1;
/* whether acquire rpc slot */
- unsigned int ei_enq_slot:1;
+ unsigned int ei_req_slot:1;
+ /** whether acquire mod rpc slot */
+ unsigned int ei_mod_slot:1;
};
extern struct obd_ops ldlm_obd_ops;
@@ -1343,7 +1345,8 @@ int ldlm_prep_elc_req(struct obd_export *exp,
int ldlm_cli_enqueue_fini(struct obd_export *exp, struct ptlrpc_request *req,
struct ldlm_enqueue_info *einfo, u8 with_policy,
u64 *flags, void *lvb, u32 lvb_len,
- const struct lustre_handle *lockh, int rc);
+ const struct lustre_handle *lockh, int rc,
+ bool request_slot);
int ldlm_cli_convert_req(struct ldlm_lock *lock, u32 *flags, u64 new_bits);
int ldlm_cli_convert(struct ldlm_lock *lock,
enum ldlm_cancel_flags cancel_flags);
diff --git a/fs/lustre/include/obd.h b/fs/lustre/include/obd.h
index 27acd33..58a5803 100644
--- a/fs/lustre/include/obd.h
+++ b/fs/lustre/include/obd.h
@@ -722,6 +722,7 @@ enum md_cli_flags {
CLI_API32 = BIT(3),
CLI_MIGRATE = BIT(4),
CLI_DIRTY_DATA = BIT(5),
+ CLI_NO_SLOT = BIT(6),
};
enum md_op_code {
diff --git a/fs/lustre/ldlm/ldlm_request.c b/fs/lustre/ldlm/ldlm_request.c
index 746c45b..44e1ec2 100644
--- a/fs/lustre/ldlm/ldlm_request.c
+++ b/fs/lustre/ldlm/ldlm_request.c
@@ -359,7 +359,9 @@ static bool ldlm_request_slot_needed(struct ldlm_enqueue_info *einfo)
/* exclude EXTENT locks and DOM-only IBITS locks because they
* are asynchronous and don't wait on server being blocked.
*/
- return einfo->ei_type == LDLM_FLOCK || einfo->ei_type == LDLM_IBITS;
+ return einfo->ei_req_slot &&
+ (einfo->ei_type == LDLM_FLOCK ||
+ einfo->ei_type == LDLM_IBITS);
}
/**
@@ -371,7 +373,7 @@ int ldlm_cli_enqueue_fini(struct obd_export *exp, struct ptlrpc_request *req,
struct ldlm_enqueue_info *einfo,
u8 with_policy, u64 *ldlm_flags, void *lvb,
u32 lvb_len, const struct lustre_handle *lockh,
- int rc)
+ int rc, bool request_slot)
{
struct ldlm_namespace *ns = exp->exp_obd->obd_namespace;
const struct lu_env *env = NULL;
@@ -380,7 +382,7 @@ int ldlm_cli_enqueue_fini(struct obd_export *exp, struct ptlrpc_request *req,
struct ldlm_reply *reply;
int cleanup_phase = 1;
- if (ldlm_request_slot_needed(einfo))
+ if (request_slot)
obd_put_request_slot(&req->rq_import->imp_obd->u.cli);
ptlrpc_put_mod_rpc_slot(req);
@@ -726,6 +728,7 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp,
int is_replay = *flags & LDLM_FL_REPLAY;
int req_passed_in = 1;
int rc, err;
+ bool need_req_slot;
struct ptlrpc_request *req;
ns = exp->exp_obd->obd_namespace;
@@ -829,13 +832,14 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp,
* that threads that are waiting for a modify RPC slot are not polluting
* our rpcs in flight counter.
*/
- if (einfo->ei_enq_slot)
+ if (einfo->ei_mod_slot)
ptlrpc_get_mod_rpc_slot(req);
- if (ldlm_request_slot_needed(einfo)) {
+ need_req_slot = ldlm_request_slot_needed(einfo);
+ if (need_req_slot) {
rc = obd_get_request_slot(&req->rq_import->imp_obd->u.cli);
if (rc) {
- if (einfo->ei_enq_slot)
+ if (einfo->ei_mod_slot)
ptlrpc_put_mod_rpc_slot(req);
failed_lock_cleanup(ns, lock, einfo->ei_mode);
LDLM_LOCK_RELEASE(lock);
@@ -855,7 +859,7 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp,
rc = ptlrpc_queue_wait(req);
err = ldlm_cli_enqueue_fini(exp, req, einfo, policy ? 1 : 0, flags,
- lvb, lvb_len, lockh, rc);
+ lvb, lvb_len, lockh, rc, need_req_slot);
/*
* If ldlm_cli_enqueue_fini did not find the lock, we need to free
diff --git a/fs/lustre/llite/statahead.c b/fs/lustre/llite/statahead.c
index 4806e99..39ffb9d 100644
--- a/fs/lustre/llite/statahead.c
+++ b/fs/lustre/llite/statahead.c
@@ -380,6 +380,7 @@ static int ll_statahead_interpret(struct ptlrpc_request *req,
einfo->ei_cb_cp = ldlm_completion_ast;
einfo->ei_cb_gl = NULL;
einfo->ei_cbdata = NULL;
+ einfo->ei_req_slot = 1;
return minfo;
}
diff --git a/fs/lustre/lmv/lmv_intent.c b/fs/lustre/lmv/lmv_intent.c
index 93da2b3..906ca16 100644
--- a/fs/lustre/lmv/lmv_intent.c
+++ b/fs/lustre/lmv/lmv_intent.c
@@ -106,6 +106,7 @@ static int lmv_intent_remote(struct obd_export *exp, struct lookup_intent *it,
}
op_data->op_bias = MDS_CROSS_REF;
+ op_data->op_cli_flags = CLI_NO_SLOT;
CDEBUG(D_INODE, "REMOTE_INTENT with fid=" DFID " -> mds #%u\n",
PFID(&body->mbo_fid1), tgt->ltd_index);
@@ -203,6 +204,7 @@ int lmv_revalidate_slaves(struct obd_export *exp,
* it's remote object.
*/
op_data->op_bias = MDS_CROSS_REF;
+ op_data->op_cli_flags = CLI_NO_SLOT;
tgt = lmv_tgt(lmv, lsm->lsm_md_oinfo[i].lmo_mds);
if (!tgt) {
diff --git a/fs/lustre/mdc/mdc_dev.c b/fs/lustre/mdc/mdc_dev.c
index b2f60ea..0b1d257 100644
--- a/fs/lustre/mdc/mdc_dev.c
+++ b/fs/lustre/mdc/mdc_dev.c
@@ -66,6 +66,7 @@ static void mdc_lock_build_einfo(const struct lu_env *env,
einfo->ei_cb_cp = ldlm_completion_ast;
einfo->ei_cb_gl = mdc_ldlm_glimpse_ast;
einfo->ei_cbdata = osc; /* value to be put into ->l_ast_data */
+ einfo->ei_req_slot = 1;
}
static void mdc_lock_lvb_update(const struct lu_env *env,
@@ -664,7 +665,7 @@ int mdc_enqueue_interpret(const struct lu_env *env, struct ptlrpc_request *req,
/* Complete obtaining the lock procedure. */
rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, &einfo, 1, aa->oa_flags,
aa->oa_lvb, aa->oa_lvb ?
- sizeof(*aa->oa_lvb) : 0, lockh, rc);
+ sizeof(*aa->oa_lvb) : 0, lockh, rc, true);
/* Complete mdc stuff. */
rc = mdc_enqueue_fini(aa->oa_exp, req, aa->oa_upcall, aa->oa_cookie,
lockh, mode, aa->oa_flags, rc);
diff --git a/fs/lustre/mdc/mdc_locks.c b/fs/lustre/mdc/mdc_locks.c
index 4135c3a..66f0039 100644
--- a/fs/lustre/mdc/mdc_locks.c
+++ b/fs/lustre/mdc/mdc_locks.c
@@ -984,7 +984,8 @@ int mdc_enqueue_base(struct obd_export *exp, struct ldlm_enqueue_info *einfo,
req->rq_sent = ktime_get_real_seconds() + resends;
}
- einfo->ei_enq_slot = !mdc_skip_mod_rpc_slot(it);
+ einfo->ei_req_slot = !(op_data->op_cli_flags & CLI_NO_SLOT);
+ einfo->ei_mod_slot = !mdc_skip_mod_rpc_slot(it);
/* With Data-on-MDT the glimpse callback is needed too.
* It is set here in advance but not in mdc_finish_enqueue()
@@ -1371,7 +1372,7 @@ static int mdc_intent_getattr_async_interpret(const struct lu_env *env,
rc = -ETIMEDOUT;
rc = ldlm_cli_enqueue_fini(exp, req, einfo, 1, &flags, NULL, 0,
- lockh, rc);
+ lockh, rc, true);
if (rc < 0) {
CERROR("%s: ldlm_cli_enqueue_fini() failed: rc = %d\n",
exp->exp_obd->obd_name, rc);
diff --git a/fs/lustre/osc/osc_request.c b/fs/lustre/osc/osc_request.c
index cf79808..e065eab 100644
--- a/fs/lustre/osc/osc_request.c
+++ b/fs/lustre/osc/osc_request.c
@@ -2824,7 +2824,7 @@ int osc_enqueue_interpret(const struct lu_env *env, struct ptlrpc_request *req,
/* Complete obtaining the lock procedure. */
rc = ldlm_cli_enqueue_fini(aa->oa_exp, req, &einfo, 1, aa->oa_flags,
- lvb, lvb_len, lockh, rc);
+ lvb, lvb_len, lockh, rc, false);
/* Complete osc stuff. */
rc = osc_enqueue_fini(req, aa->oa_upcall, aa->oa_cookie, lockh, mode,
aa->oa_flags, aa->oa_speculative, rc);
--
1.8.3.1
_______________________________________________
lustre-devel mailing list
lustre-devel@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-devel-lustre.org
next prev parent reply other threads:[~2021-11-28 23:28 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-28 23:27 [lustre-devel] [PATCH 00/19] lustre: update to OpenSFS tree Nov 28, 2021 James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 01/19] lnet: fix delay rule crash James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 02/19] lnet: change tp_nid to 16byte in lnet_test_peer James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 03/19] lnet: extend preferred nids in struct lnet_peer_ni James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 04/19] lnet: switch to large lnet_processid for matching James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 05/19] lnet: libcfs: add timeout to cfs_race() to fix race James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 06/19] lustre: llite: tighten condition for fault not drop mmap_sem James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 07/19] lnet: o2iblnd: map_on_demand not needed for frag interop James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 08/19] lnet: o2iblnd: Fix logic for unaligned transfer James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 09/19] lnet: Reset ni_ping_count only on receive James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 10/19] lustre: ptlrpc: fix timeout after spurious wakeup James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 11/19] lnet: Fail peer add for existing gw peer James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 12/19] lustre: ptlrpc: remove bogus LASSERT James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 13/19] lustre: quota: optimize capability check for root squash James Simmons
2021-11-28 23:27 ` James Simmons [this message]
2021-11-28 23:27 ` [lustre-devel] [PATCH 15/19] lnet: set eth routes needed for multi rail James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 16/19] lustre: llite: Do not count tiny write twice James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 17/19] lustre: llite: mend the trunc_sem_up_write() James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 18/19] lnet: Netlink improvements James Simmons
2021-11-28 23:27 ` [lustre-devel] [PATCH 19/19] lnet: libcfs: separate daemon_list from cfs_trace_data James Simmons
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1638142074-5945-15-git-send-email-jsimmons@infradead.org \
--to=jsimmons@infradead.org \
--cc=adilger@whamcloud.com \
--cc=c17819@cray.com \
--cc=green@whamcloud.com \
--cc=lustre-devel@lists.lustre.org \
--cc=neilb@suse.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).