* [PATCH 00/14] Lustre fixes
@ 2016-11-03 1:24 Oleg Drokin
2016-11-03 1:24 ` [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field Oleg Drokin
` (13 more replies)
0 siblings, 14 replies; 29+ messages in thread
From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw)
To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons
Cc: Linux Kernel Mailing List, Lustre Development List, Oleg Drokin
This batch of patches represents mostly recent fixes,
also a couple of cleanups and a couple of changes that fixes depend on.
Amir Shehata (1):
staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs
Andrew Perepechko (1):
staging/lustre/llite: drop_caches hangs in cl_inode_fini()
Andriy Skulysh (1):
staging/lustre: conflicting PW & PR extent locks on a client
Bobi Jam (2):
staging/lustre/llite: protect from accessing NULL lli_clob
staging/lustre/llite: update ras window correctly
Hongchao Zhang (2):
staging/lustre/ldlm: fix export reference problem
staging/lustre/ldlm: engage ELC for all ldlm enqueue req
Jinshan Xiong (2):
staging/lustre: Get rid of cl_env hash table
staging/lustre/llite: do not clear uptodate bit in page delete
Niu Yawei (1):
staging/lustre/llite: clear inode timestamps after losing UPDATE lock
Oleg Drokin (3):
staging/lustre/ldlm: Drop unused blocking_refs flock field
staging/lustre/ldlm: Reinstate ldlm_enqueue_pack()
staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users
Patrick Farrell (1):
staging/lustre/ptlrpc: Suppress error for flock requests
drivers/staging/lustre/lustre/include/cl_object.h | 22 --
drivers/staging/lustre/lustre/include/lustre_dlm.h | 2 -
drivers/staging/lustre/lustre/include/lustre_lib.h | 2 -
.../staging/lustre/lustre/include/obd_support.h | 3 +
drivers/staging/lustre/lustre/ldlm/ldlm_extent.c | 20 ++
drivers/staging/lustre/lustre/ldlm/ldlm_lib.c | 3 +
drivers/staging/lustre/lustre/ldlm/ldlm_pool.c | 9 -
drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 61 ++--
drivers/staging/lustre/lustre/llite/file.c | 26 +-
drivers/staging/lustre/lustre/llite/lcommon_cl.c | 11 +-
drivers/staging/lustre/lustre/llite/lcommon_misc.c | 8 +-
drivers/staging/lustre/lustre/llite/llite_mmap.c | 59 ++--
drivers/staging/lustre/lustre/llite/namei.c | 10 +
drivers/staging/lustre/lustre/llite/rw.c | 21 +-
drivers/staging/lustre/lustre/llite/rw26.c | 52 +--
drivers/staging/lustre/lustre/llite/vvp_page.c | 1 -
drivers/staging/lustre/lustre/lov/lov_io.c | 5 -
drivers/staging/lustre/lustre/lov/lov_object.c | 7 +-
.../staging/lustre/lustre/obdclass/cl_internal.h | 23 --
drivers/staging/lustre/lustre/obdclass/cl_io.c | 1 -
drivers/staging/lustre/lustre/obdclass/cl_object.c | 389 ++++-----------------
drivers/staging/lustre/lustre/obdclass/lu_object.c | 4 -
.../staging/lustre/lustre/obdecho/echo_client.c | 14 +-
drivers/staging/lustre/lustre/osc/osc_cache.c | 6 +-
drivers/staging/lustre/lustre/osc/osc_lock.c | 116 +++---
drivers/staging/lustre/lustre/osc/osc_page.c | 6 +-
drivers/staging/lustre/lustre/osc/osc_request.c | 1 +
drivers/staging/lustre/lustre/ptlrpc/client.c | 4 +-
drivers/staging/lustre/lustre/ptlrpc/import.c | 19 +-
drivers/staging/lustre/lustre/ptlrpc/service.c | 3 +-
30 files changed, 298 insertions(+), 610 deletions(-)
--
2.7.4
^ permalink raw reply [flat|nested] 29+ messages in thread* [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:38 ` James Simmons 2016-11-03 1:24 ` [PATCH 02/14] staging/lustre/ldlm: fix export reference problem Oleg Drokin ` (12 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Oleg Drokin blocking_refs is only used on the server, so drop it on the client. Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/include/lustre_dlm.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/staging/lustre/lustre/include/lustre_dlm.h b/drivers/staging/lustre/lustre/include/lustre_dlm.h index 1c6b7b8..f770b86 100644 --- a/drivers/staging/lustre/lustre/include/lustre_dlm.h +++ b/drivers/staging/lustre/lustre/include/lustre_dlm.h @@ -550,8 +550,6 @@ struct ldlm_flock { __u64 owner; __u64 blocking_owner; struct obd_export *blocking_export; - /* Protected by the hash lock */ - __u32 blocking_refs; __u32 pid; }; -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field 2016-11-03 1:24 ` [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field Oleg Drokin @ 2016-11-07 1:38 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:38 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List > blocking_refs is only used on the server, so drop it on the client. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/include/lustre_dlm.h | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/include/lustre_dlm.h b/drivers/staging/lustre/lustre/include/lustre_dlm.h > index 1c6b7b8..f770b86 100644 > --- a/drivers/staging/lustre/lustre/include/lustre_dlm.h > +++ b/drivers/staging/lustre/lustre/include/lustre_dlm.h > @@ -550,8 +550,6 @@ struct ldlm_flock { > __u64 owner; > __u64 blocking_owner; > struct obd_export *blocking_export; > - /* Protected by the hash lock */ > - __u32 blocking_refs; > __u32 pid; > }; > > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 02/14] staging/lustre/ldlm: fix export reference problem 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin 2016-11-03 1:24 ` [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:37 ` James Simmons 2016-11-03 1:24 ` [PATCH 03/14] staging/lustre: conflicting PW & PR extent locks on a client Oleg Drokin ` (11 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Hongchao Zhang, Oleg Drokin From: Hongchao Zhang <hongchao.zhang@intel.com> 1, in client_import_del_conn, the export returned from class_conn2export is not released after using it. 2, in ptlrpc_connect_interpret, the export is not released if the connect_flags isn't compatible. Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com> Reviewed-on: http://review.whamcloud.com/22031 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8500 Reviewed-by: James Simmons <uja.ornl@yahoo.com> Reviewed-by: Bobi Jam <bobijam@hotmail.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/ldlm/ldlm_lib.c | 3 +++ drivers/staging/lustre/lustre/ptlrpc/import.c | 19 ++++++++++--------- 2 files changed, 13 insertions(+), 9 deletions(-) diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c b/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c index 4f9480e..06d3cc6 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c @@ -170,6 +170,9 @@ int client_import_del_conn(struct obd_import *imp, struct obd_uuid *uuid) ptlrpc_connection_put(dlmexp->exp_connection); dlmexp->exp_connection = NULL; } + + if (dlmexp) + class_export_put(dlmexp); } list_del(&imp_conn->oic_item); diff --git a/drivers/staging/lustre/lustre/ptlrpc/import.c b/drivers/staging/lustre/lustre/ptlrpc/import.c index 46ba5a4..05fd92d 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/import.c +++ b/drivers/staging/lustre/lustre/ptlrpc/import.c @@ -972,6 +972,16 @@ static int ptlrpc_connect_interpret(const struct lu_env *env, spin_unlock(&imp->imp_lock); + if (!exp) { + /* This could happen if export is cleaned during the + * connect attempt + */ + CERROR("%s: missing export after connect\n", + imp->imp_obd->obd_name); + rc = -ENODEV; + goto out; + } + /* check that server granted subset of flags we asked for. */ if ((ocd->ocd_connect_flags & imp->imp_connect_flags_orig) != ocd->ocd_connect_flags) { @@ -982,15 +992,6 @@ static int ptlrpc_connect_interpret(const struct lu_env *env, goto out; } - if (!exp) { - /* This could happen if export is cleaned during the - * connect attempt - */ - CERROR("%s: missing export after connect\n", - imp->imp_obd->obd_name); - rc = -ENODEV; - goto out; - } old_connect_flags = exp_connect_flags(exp); exp->exp_connect_data = *ocd; imp->imp_obd->obd_self_export->exp_connect_data = *ocd; -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 02/14] staging/lustre/ldlm: fix export reference problem 2016-11-03 1:24 ` [PATCH 02/14] staging/lustre/ldlm: fix export reference problem Oleg Drokin @ 2016-11-07 1:37 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:37 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Hongchao Zhang > From: Hongchao Zhang <hongchao.zhang@intel.com> > > 1, in client_import_del_conn, the export returned from > class_conn2export is not released after using it. > > 2, in ptlrpc_connect_interpret, the export is not released > if the connect_flags isn't compatible. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com> > Reviewed-on: http://review.whamcloud.com/22031 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8500 > Reviewed-by: James Simmons <uja.ornl@yahoo.com> > Reviewed-by: Bobi Jam <bobijam@hotmail.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/ldlm/ldlm_lib.c | 3 +++ > drivers/staging/lustre/lustre/ptlrpc/import.c | 19 ++++++++++--------- > 2 files changed, 13 insertions(+), 9 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c b/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c > index 4f9480e..06d3cc6 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_lib.c > @@ -170,6 +170,9 @@ int client_import_del_conn(struct obd_import *imp, struct obd_uuid *uuid) > ptlrpc_connection_put(dlmexp->exp_connection); > dlmexp->exp_connection = NULL; > } > + > + if (dlmexp) > + class_export_put(dlmexp); > } > > list_del(&imp_conn->oic_item); > diff --git a/drivers/staging/lustre/lustre/ptlrpc/import.c b/drivers/staging/lustre/lustre/ptlrpc/import.c > index 46ba5a4..05fd92d 100644 > --- a/drivers/staging/lustre/lustre/ptlrpc/import.c > +++ b/drivers/staging/lustre/lustre/ptlrpc/import.c > @@ -972,6 +972,16 @@ static int ptlrpc_connect_interpret(const struct lu_env *env, > > spin_unlock(&imp->imp_lock); > > + if (!exp) { > + /* This could happen if export is cleaned during the > + * connect attempt > + */ > + CERROR("%s: missing export after connect\n", > + imp->imp_obd->obd_name); > + rc = -ENODEV; > + goto out; > + } > + > /* check that server granted subset of flags we asked for. */ > if ((ocd->ocd_connect_flags & imp->imp_connect_flags_orig) != > ocd->ocd_connect_flags) { > @@ -982,15 +992,6 @@ static int ptlrpc_connect_interpret(const struct lu_env *env, > goto out; > } > > - if (!exp) { > - /* This could happen if export is cleaned during the > - * connect attempt > - */ > - CERROR("%s: missing export after connect\n", > - imp->imp_obd->obd_name); > - rc = -ENODEV; > - goto out; > - } > old_connect_flags = exp_connect_flags(exp); > exp->exp_connect_data = *ocd; > imp->imp_obd->obd_self_export->exp_connect_data = *ocd; > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 03/14] staging/lustre: conflicting PW & PR extent locks on a client 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin 2016-11-03 1:24 ` [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field Oleg Drokin 2016-11-03 1:24 ` [PATCH 02/14] staging/lustre/ldlm: fix export reference problem Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:52 ` James Simmons 2016-11-03 1:24 ` [PATCH 04/14] staging/lustre/llite: clear inode timestamps after losing UPDATE lock Oleg Drokin ` (10 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Andriy Skulysh, Oleg Drokin From: Andriy Skulysh <andriy.skulysh@seagate.com> PW lock isn't replayed once a lock is marked LDLM_FL_CANCELING and glimpse lock doesn't wait for conflicting locks on the client. So the server will grant a PR lock in response to the glimpse lock request, which conflicts with the PW lock in LDLM_FL_CANCELING state on the client. Lock in LDLM_FL_CANCELING state may still have pending IO, so it should be replayed until LDLM_FL_BL_DONE is set to avoid granted conflicting lock by a server. Seagate-bug-id: MRP-3311 Signed-off-by: Andriy Skulysh <andriy.skulysh@seagate.com> Reviewed-on: http://review.whamcloud.com/20345 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8175 Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/include/obd_support.h | 3 +++ drivers/staging/lustre/lustre/ldlm/ldlm_extent.c | 20 ++++++++++++++++++++ drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 4 ++-- drivers/staging/lustre/lustre/osc/osc_request.c | 1 + 4 files changed, 26 insertions(+), 2 deletions(-) diff --git a/drivers/staging/lustre/lustre/include/obd_support.h b/drivers/staging/lustre/lustre/include/obd_support.h index 7f3f8cd..aaedec7 100644 --- a/drivers/staging/lustre/lustre/include/obd_support.h +++ b/drivers/staging/lustre/lustre/include/obd_support.h @@ -321,6 +321,8 @@ extern char obd_jobid_var[]; #define OBD_FAIL_LDLM_CP_CB_WAIT4 0x322 #define OBD_FAIL_LDLM_CP_CB_WAIT5 0x323 +#define OBD_FAIL_LDLM_GRANT_CHECK 0x32a + /* LOCKLESS IO */ #define OBD_FAIL_LDLM_SET_CONTENTION 0x385 @@ -343,6 +345,7 @@ extern char obd_jobid_var[]; #define OBD_FAIL_OSC_CP_ENQ_RACE 0x410 #define OBD_FAIL_OSC_NO_GRANT 0x411 #define OBD_FAIL_OSC_DELAY_SETTIME 0x412 +#define OBD_FAIL_OSC_DELAY_IO 0x414 #define OBD_FAIL_PTLRPC 0x500 #define OBD_FAIL_PTLRPC_ACK 0x501 diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c b/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c index ecf472e..a7b34e4 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c @@ -193,6 +193,26 @@ void ldlm_extent_add_lock(struct ldlm_resource *res, * add the locks into grant list, for debug purpose, .. */ ldlm_resource_add_lock(res, &res->lr_granted, lock); + + if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_GRANT_CHECK)) { + struct ldlm_lock *lck; + + list_for_each_entry_reverse(lck, &res->lr_granted, + l_res_link) { + if (lck == lock) + continue; + if (lockmode_compat(lck->l_granted_mode, + lock->l_granted_mode)) + continue; + if (ldlm_extent_overlap(&lck->l_req_extent, + &lock->l_req_extent)) { + CDEBUG(D_ERROR, "granting conflicting lock %p %p\n", + lck, lock); + ldlm_resource_dump(D_ERROR, res); + LBUG(); + } + } + } } /** Remove cancelled lock from resource interval tree. */ diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c index 43856ff..6e704c7 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c @@ -1846,7 +1846,7 @@ static int ldlm_chain_lock_for_replay(struct ldlm_lock *lock, void *closure) * bug 17614: locks being actively cancelled. Get a reference * on a lock so that it does not disappear under us (e.g. due to cancel) */ - if (!(lock->l_flags & (LDLM_FL_FAILED | LDLM_FL_CANCELING))) { + if (!(lock->l_flags & (LDLM_FL_FAILED | LDLM_FL_BL_DONE))) { list_add(&lock->l_pending_chain, list); LDLM_LOCK_GET(lock); } @@ -1915,7 +1915,7 @@ static int replay_one_lock(struct obd_import *imp, struct ldlm_lock *lock) int flags; /* Bug 11974: Do not replay a lock which is actively being canceled */ - if (ldlm_is_canceling(lock)) { + if (ldlm_is_bl_done(lock)) { LDLM_DEBUG(lock, "Not replaying canceled lock:"); return 0; } diff --git a/drivers/staging/lustre/lustre/osc/osc_request.c b/drivers/staging/lustre/lustre/osc/osc_request.c index 091558e..8023561 100644 --- a/drivers/staging/lustre/lustre/osc/osc_request.c +++ b/drivers/staging/lustre/lustre/osc/osc_request.c @@ -1823,6 +1823,7 @@ int osc_build_rpc(const struct lu_env *env, struct client_obd *cli, DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %ur/%dw in flight", page_count, aa, cli->cl_r_in_flight, cli->cl_w_in_flight); + OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_DELAY_IO, cfs_fail_val); ptlrpcd_add_req(req); rc = 0; -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 03/14] staging/lustre: conflicting PW & PR extent locks on a client 2016-11-03 1:24 ` [PATCH 03/14] staging/lustre: conflicting PW & PR extent locks on a client Oleg Drokin @ 2016-11-07 1:52 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:52 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Andriy Skulysh > From: Andriy Skulysh <andriy.skulysh@seagate.com> > > PW lock isn't replayed once a lock is marked > LDLM_FL_CANCELING and glimpse lock doesn't wait for > conflicting locks on the client. So the server will > grant a PR lock in response to the glimpse lock request, > which conflicts with the PW lock in LDLM_FL_CANCELING > state on the client. > > Lock in LDLM_FL_CANCELING state may still have pending IO, > so it should be replayed until LDLM_FL_BL_DONE is set to > avoid granted conflicting lock by a server. Reviewed-by: James Simmons <jsimmons@infradead.org> > Seagate-bug-id: MRP-3311 > Signed-off-by: Andriy Skulysh <andriy.skulysh@seagate.com> > Reviewed-on: http://review.whamcloud.com/20345 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8175 > Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/include/obd_support.h | 3 +++ > drivers/staging/lustre/lustre/ldlm/ldlm_extent.c | 20 ++++++++++++++++++++ > drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 4 ++-- > drivers/staging/lustre/lustre/osc/osc_request.c | 1 + > 4 files changed, 26 insertions(+), 2 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/include/obd_support.h b/drivers/staging/lustre/lustre/include/obd_support.h > index 7f3f8cd..aaedec7 100644 > --- a/drivers/staging/lustre/lustre/include/obd_support.h > +++ b/drivers/staging/lustre/lustre/include/obd_support.h > @@ -321,6 +321,8 @@ extern char obd_jobid_var[]; > #define OBD_FAIL_LDLM_CP_CB_WAIT4 0x322 > #define OBD_FAIL_LDLM_CP_CB_WAIT5 0x323 > > +#define OBD_FAIL_LDLM_GRANT_CHECK 0x32a > + > /* LOCKLESS IO */ > #define OBD_FAIL_LDLM_SET_CONTENTION 0x385 > > @@ -343,6 +345,7 @@ extern char obd_jobid_var[]; > #define OBD_FAIL_OSC_CP_ENQ_RACE 0x410 > #define OBD_FAIL_OSC_NO_GRANT 0x411 > #define OBD_FAIL_OSC_DELAY_SETTIME 0x412 > +#define OBD_FAIL_OSC_DELAY_IO 0x414 > > #define OBD_FAIL_PTLRPC 0x500 > #define OBD_FAIL_PTLRPC_ACK 0x501 > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c b/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c > index ecf472e..a7b34e4 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_extent.c > @@ -193,6 +193,26 @@ void ldlm_extent_add_lock(struct ldlm_resource *res, > * add the locks into grant list, for debug purpose, .. > */ > ldlm_resource_add_lock(res, &res->lr_granted, lock); > + > + if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_GRANT_CHECK)) { > + struct ldlm_lock *lck; > + > + list_for_each_entry_reverse(lck, &res->lr_granted, > + l_res_link) { > + if (lck == lock) > + continue; > + if (lockmode_compat(lck->l_granted_mode, > + lock->l_granted_mode)) > + continue; > + if (ldlm_extent_overlap(&lck->l_req_extent, > + &lock->l_req_extent)) { > + CDEBUG(D_ERROR, "granting conflicting lock %p %p\n", > + lck, lock); > + ldlm_resource_dump(D_ERROR, res); > + LBUG(); > + } > + } > + } > } > > /** Remove cancelled lock from resource interval tree. */ > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > index 43856ff..6e704c7 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > @@ -1846,7 +1846,7 @@ static int ldlm_chain_lock_for_replay(struct ldlm_lock *lock, void *closure) > * bug 17614: locks being actively cancelled. Get a reference > * on a lock so that it does not disappear under us (e.g. due to cancel) > */ > - if (!(lock->l_flags & (LDLM_FL_FAILED | LDLM_FL_CANCELING))) { > + if (!(lock->l_flags & (LDLM_FL_FAILED | LDLM_FL_BL_DONE))) { > list_add(&lock->l_pending_chain, list); > LDLM_LOCK_GET(lock); > } > @@ -1915,7 +1915,7 @@ static int replay_one_lock(struct obd_import *imp, struct ldlm_lock *lock) > int flags; > > /* Bug 11974: Do not replay a lock which is actively being canceled */ > - if (ldlm_is_canceling(lock)) { > + if (ldlm_is_bl_done(lock)) { > LDLM_DEBUG(lock, "Not replaying canceled lock:"); > return 0; > } > diff --git a/drivers/staging/lustre/lustre/osc/osc_request.c b/drivers/staging/lustre/lustre/osc/osc_request.c > index 091558e..8023561 100644 > --- a/drivers/staging/lustre/lustre/osc/osc_request.c > +++ b/drivers/staging/lustre/lustre/osc/osc_request.c > @@ -1823,6 +1823,7 @@ int osc_build_rpc(const struct lu_env *env, struct client_obd *cli, > DEBUG_REQ(D_INODE, req, "%d pages, aa %p. now %ur/%dw in flight", > page_count, aa, cli->cl_r_in_flight, > cli->cl_w_in_flight); > + OBD_FAIL_TIMEOUT(OBD_FAIL_OSC_DELAY_IO, cfs_fail_val); > > ptlrpcd_add_req(req); > rc = 0; > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 04/14] staging/lustre/llite: clear inode timestamps after losing UPDATE lock 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (2 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 03/14] staging/lustre: conflicting PW & PR extent locks on a client Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:38 ` James Simmons 2016-11-03 1:24 ` [PATCH 05/14] staging/lustre: Get rid of cl_env hash table Oleg Drokin ` (9 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Niu Yawei, Jinshan Xiong, Oleg Drokin From: Niu Yawei <yawei.niu@intel.com> Otherwise, those leftovers would interfere with new timestamps especially when the timestamps are set back in time on the other clients. Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com> Signed-off-by: Niu Yawei <yawei.niu@intel.com> Reviewed-on: http://review.whamcloud.com/22623 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8446 Reviewed-by: Bobi Jam <bobijam@hotmail.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/llite/namei.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/staging/lustre/lustre/llite/namei.c b/drivers/staging/lustre/lustre/llite/namei.c index 74d9b73..c268f32 100644 --- a/drivers/staging/lustre/lustre/llite/namei.c +++ b/drivers/staging/lustre/lustre/llite/namei.c @@ -251,6 +251,16 @@ int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc, PFID(ll_inode2fid(inode)), rc); } + if (bits & MDS_INODELOCK_UPDATE) { + struct ll_inode_info *lli = ll_i2info(inode); + + spin_lock(&lli->lli_lock); + LTIME_S(inode->i_mtime) = 0; + LTIME_S(inode->i_atime) = 0; + LTIME_S(inode->i_ctime) = 0; + spin_unlock(&lli->lli_lock); + } + if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) { struct ll_inode_info *lli = ll_i2info(inode); -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 04/14] staging/lustre/llite: clear inode timestamps after losing UPDATE lock 2016-11-03 1:24 ` [PATCH 04/14] staging/lustre/llite: clear inode timestamps after losing UPDATE lock Oleg Drokin @ 2016-11-07 1:38 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:38 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Niu Yawei, Jinshan Xiong > From: Niu Yawei <yawei.niu@intel.com> > > Otherwise, those leftovers would interfere with new timestamps > especially when the timestamps are set back in time on the other > clients. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com> > Signed-off-by: Niu Yawei <yawei.niu@intel.com> > Reviewed-on: http://review.whamcloud.com/22623 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8446 > Reviewed-by: Bobi Jam <bobijam@hotmail.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/llite/namei.c | 10 ++++++++++ > 1 file changed, 10 insertions(+) > > diff --git a/drivers/staging/lustre/lustre/llite/namei.c b/drivers/staging/lustre/lustre/llite/namei.c > index 74d9b73..c268f32 100644 > --- a/drivers/staging/lustre/lustre/llite/namei.c > +++ b/drivers/staging/lustre/lustre/llite/namei.c > @@ -251,6 +251,16 @@ int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc, > PFID(ll_inode2fid(inode)), rc); > } > > + if (bits & MDS_INODELOCK_UPDATE) { > + struct ll_inode_info *lli = ll_i2info(inode); > + > + spin_lock(&lli->lli_lock); > + LTIME_S(inode->i_mtime) = 0; > + LTIME_S(inode->i_atime) = 0; > + LTIME_S(inode->i_ctime) = 0; > + spin_unlock(&lli->lli_lock); > + } > + > if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) { > struct ll_inode_info *lli = ll_i2info(inode); > > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 05/14] staging/lustre: Get rid of cl_env hash table 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (3 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 04/14] staging/lustre/llite: clear inode timestamps after losing UPDATE lock Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:53 ` James Simmons 2016-11-03 1:24 ` [PATCH 06/14] staging/lustre/llite: drop_caches hangs in cl_inode_fini() Oleg Drokin ` (8 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Jinshan Xiong, Oleg Drokin From: Jinshan Xiong <jinshan.xiong@intel.com> cl_env hash table is under heavy contention when there are lots of processes doing IO at the same time; reduce lock contention by replacing cl_env cache with percpu array; remove cl_env_nested_get() and cl_env_nested_put(); remove cl_env_reenter() and cl_env_reexit(); Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-on: http://review.whamcloud.com/20254 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4257 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Bobi Jam <bobijam@hotmail.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/include/cl_object.h | 22 -- drivers/staging/lustre/lustre/ldlm/ldlm_pool.c | 9 - drivers/staging/lustre/lustre/llite/file.c | 18 +- drivers/staging/lustre/lustre/llite/lcommon_cl.c | 11 +- drivers/staging/lustre/lustre/llite/lcommon_misc.c | 8 +- drivers/staging/lustre/lustre/llite/llite_mmap.c | 59 ++-- drivers/staging/lustre/lustre/llite/rw.c | 6 +- drivers/staging/lustre/lustre/llite/rw26.c | 16 +- drivers/staging/lustre/lustre/lov/lov_io.c | 5 - drivers/staging/lustre/lustre/lov/lov_object.c | 7 +- .../staging/lustre/lustre/obdclass/cl_internal.h | 23 -- drivers/staging/lustre/lustre/obdclass/cl_io.c | 1 - drivers/staging/lustre/lustre/obdclass/cl_object.c | 389 ++++----------------- drivers/staging/lustre/lustre/obdclass/lu_object.c | 4 - .../staging/lustre/lustre/obdecho/echo_client.c | 14 +- drivers/staging/lustre/lustre/osc/osc_cache.c | 6 +- drivers/staging/lustre/lustre/osc/osc_lock.c | 116 +++--- drivers/staging/lustre/lustre/osc/osc_page.c | 6 +- 18 files changed, 183 insertions(+), 537 deletions(-) diff --git a/drivers/staging/lustre/lustre/include/cl_object.h b/drivers/staging/lustre/lustre/include/cl_object.h index 514d650..3fe26e7 100644 --- a/drivers/staging/lustre/lustre/include/cl_object.h +++ b/drivers/staging/lustre/lustre/include/cl_object.h @@ -2640,35 +2640,13 @@ void cl_sync_io_end(const struct lu_env *env, struct cl_sync_io *anchor); * - allocation and destruction of environment is amortized by caching no * longer used environments instead of destroying them; * - * - there is a notion of "current" environment, attached to the kernel - * data structure representing current thread Top-level lustre code - * allocates an environment and makes it current, then calls into - * non-lustre code, that in turn calls lustre back. Low-level lustre - * code thus called can fetch environment created by the top-level code - * and reuse it, avoiding additional environment allocation. - * Right now, three interfaces can attach the cl_env to running thread: - * - cl_env_get - * - cl_env_implant - * - cl_env_reexit(cl_env_reenter had to be called priorly) - * * \see lu_env, lu_context, lu_context_key * @{ */ -struct cl_env_nest { - int cen_refcheck; - void *cen_cookie; -}; - struct lu_env *cl_env_get(int *refcheck); struct lu_env *cl_env_alloc(int *refcheck, __u32 tags); -struct lu_env *cl_env_nested_get(struct cl_env_nest *nest); void cl_env_put(struct lu_env *env, int *refcheck); -void cl_env_nested_put(struct cl_env_nest *nest, struct lu_env *env); -void *cl_env_reenter(void); -void cl_env_reexit(void *cookie); -void cl_env_implant(struct lu_env *env, int *refcheck); -void cl_env_unplant(struct lu_env *env, int *refcheck); unsigned int cl_env_cache_purge(unsigned int nr); struct lu_env *cl_env_percpu_get(void); void cl_env_percpu_put(struct lu_env *env); diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c b/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c index b29c9561..19831c5 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c @@ -794,7 +794,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) int nr_ns; struct ldlm_namespace *ns; struct ldlm_namespace *ns_old = NULL; /* loop detection */ - void *cookie; if (client == LDLM_NAMESPACE_CLIENT && !(gfp_mask & __GFP_FS)) return 0; @@ -802,8 +801,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) CDEBUG(D_DLMTRACE, "Request to count %s locks from all pools\n", client == LDLM_NAMESPACE_CLIENT ? "client" : "server"); - cookie = cl_env_reenter(); - /* * Find out how many resources we may release. */ @@ -812,7 +809,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) mutex_lock(ldlm_namespace_lock(client)); if (list_empty(ldlm_namespace_list(client))) { mutex_unlock(ldlm_namespace_lock(client)); - cl_env_reexit(cookie); return 0; } ns = ldlm_namespace_first_locked(client); @@ -838,7 +834,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) ldlm_namespace_put(ns); } - cl_env_reexit(cookie); return total; } @@ -847,13 +842,10 @@ static unsigned long ldlm_pools_scan(ldlm_side_t client, int nr, gfp_t gfp_mask) unsigned long freed = 0; int tmp, nr_ns; struct ldlm_namespace *ns; - void *cookie; if (client == LDLM_NAMESPACE_CLIENT && !(gfp_mask & __GFP_FS)) return -1; - cookie = cl_env_reenter(); - /* * Shrink at least ldlm_namespace_nr_read(client) namespaces. */ @@ -883,7 +875,6 @@ static unsigned long ldlm_pools_scan(ldlm_side_t client, int nr, gfp_t gfp_mask) freed += ldlm_pool_shrink(&ns->ns_pool, cancel, gfp_mask); ldlm_namespace_put(ns); } - cl_env_reexit(cookie); /* * we only decrease the SLV in server pools shrinker, return * SHRINK_STOP to kernel to avoid needless loop. LU-1128 diff --git a/drivers/staging/lustre/lustre/llite/file.c b/drivers/staging/lustre/lustre/llite/file.c index 0ae78c7..c1c7551 100644 --- a/drivers/staging/lustre/lustre/llite/file.c +++ b/drivers/staging/lustre/lustre/llite/file.c @@ -1584,11 +1584,11 @@ int ll_data_version(struct inode *inode, __u64 *data_version, int flags) */ int ll_hsm_release(struct inode *inode) { - struct cl_env_nest nest; struct lu_env *env; struct obd_client_handle *och = NULL; __u64 data_version = 0; int rc; + int refcheck; CDEBUG(D_INODE, "%s: Releasing file "DFID".\n", ll_get_fsname(inode->i_sb, NULL, 0), @@ -1605,14 +1605,14 @@ int ll_hsm_release(struct inode *inode) if (rc != 0) goto out; - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) { rc = PTR_ERR(env); goto out; } ll_merge_attr(env, inode); - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); /* Release the file. * NB: lease lock handle is released in mdc_hsm_release_pack() because @@ -2268,17 +2268,17 @@ static int ll_flush(struct file *file, fl_owner_t id) int cl_sync_file_range(struct inode *inode, loff_t start, loff_t end, enum cl_fsync_mode mode, int ignore_layout) { - struct cl_env_nest nest; struct lu_env *env; struct cl_io *io; struct cl_fsync_io *fio; int result; + int refcheck; if (mode != CL_FSYNC_NONE && mode != CL_FSYNC_LOCAL && mode != CL_FSYNC_DISCARD && mode != CL_FSYNC_ALL) return -EINVAL; - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) return PTR_ERR(env); @@ -2301,7 +2301,7 @@ int cl_sync_file_range(struct inode *inode, loff_t start, loff_t end, if (result == 0) result = fio->fi_nr_written; cl_io_fini(env, io); - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return result; } @@ -3149,14 +3149,14 @@ int ll_layout_conf(struct inode *inode, const struct cl_object_conf *conf) { struct ll_inode_info *lli = ll_i2info(inode); struct cl_object *obj = lli->lli_clob; - struct cl_env_nest nest; struct lu_env *env; int rc; + int refcheck; if (!obj) return 0; - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) return PTR_ERR(env); @@ -3190,7 +3190,7 @@ int ll_layout_conf(struct inode *inode, const struct cl_object_conf *conf) ll_layout_version_set(lli, cl.cl_layout_gen); } out: - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return rc; } diff --git a/drivers/staging/lustre/lustre/llite/lcommon_cl.c b/drivers/staging/lustre/lustre/llite/lcommon_cl.c index 56e414b..dd1cfd8 100644 --- a/drivers/staging/lustre/lustre/llite/lcommon_cl.c +++ b/drivers/staging/lustre/lustre/llite/lcommon_cl.c @@ -246,15 +246,11 @@ void cl_inode_fini(struct inode *inode) int emergency; if (clob) { - void *cookie; - - cookie = cl_env_reenter(); env = cl_env_get(&refcheck); emergency = IS_ERR(env); if (emergency) { mutex_lock(&cl_inode_fini_guard); LASSERT(cl_inode_fini_env); - cl_env_implant(cl_inode_fini_env, &refcheck); env = cl_inode_fini_env; } /* @@ -266,13 +262,10 @@ void cl_inode_fini(struct inode *inode) lu_object_ref_del(&clob->co_lu, "inode", inode); cl_object_put_last(env, clob); lli->lli_clob = NULL; - if (emergency) { - cl_env_unplant(cl_inode_fini_env, &refcheck); + if (emergency) mutex_unlock(&cl_inode_fini_guard); - } else { + else cl_env_put(env, &refcheck); - } - cl_env_reexit(cookie); } } diff --git a/drivers/staging/lustre/lustre/llite/lcommon_misc.c b/drivers/staging/lustre/lustre/llite/lcommon_misc.c index 1558b55..f48660e 100644 --- a/drivers/staging/lustre/lustre/llite/lcommon_misc.c +++ b/drivers/staging/lustre/lustre/llite/lcommon_misc.c @@ -162,13 +162,11 @@ int cl_get_grouplock(struct cl_object *obj, unsigned long gid, int nonblock, return rc; } - cg->lg_env = cl_env_get(&refcheck); + cg->lg_env = env; cg->lg_io = io; cg->lg_lock = lock; cg->lg_gid = gid; - LASSERT(cg->lg_env == env); - cl_env_unplant(env, &refcheck); return 0; } @@ -177,14 +175,10 @@ void cl_put_grouplock(struct ll_grouplock *cg) struct lu_env *env = cg->lg_env; struct cl_io *io = cg->lg_io; struct cl_lock *lock = cg->lg_lock; - int refcheck; LASSERT(cg->lg_env); LASSERT(cg->lg_gid); - cl_env_implant(env, &refcheck); - cl_env_put(env, &refcheck); - cl_lock_release(env, lock); cl_io_fini(env, io); cl_env_put(env, NULL); diff --git a/drivers/staging/lustre/lustre/llite/llite_mmap.c b/drivers/staging/lustre/lustre/llite/llite_mmap.c index 4366918..ad2a699 100644 --- a/drivers/staging/lustre/lustre/llite/llite_mmap.c +++ b/drivers/staging/lustre/lustre/llite/llite_mmap.c @@ -80,43 +80,24 @@ struct vm_area_struct *our_vma(struct mm_struct *mm, unsigned long addr, * API independent part for page fault initialization. * \param vma - virtual memory area addressed to page fault * \param env - corespondent lu_env to processing - * \param nest - nested level * \param index - page index corespondent to fault. * \parm ra_flags - vma readahead flags. * - * \return allocated and initialized env for fault operation. - * \retval EINVAL if env can't allocated - * \return other error codes from cl_io_init. + * \return error codes from cl_io_init. */ static struct cl_io * -ll_fault_io_init(struct vm_area_struct *vma, struct lu_env **env_ret, - struct cl_env_nest *nest, pgoff_t index, - unsigned long *ra_flags) +ll_fault_io_init(struct lu_env *env, struct vm_area_struct *vma, + pgoff_t index, unsigned long *ra_flags) { struct file *file = vma->vm_file; struct inode *inode = file_inode(file); struct cl_io *io; struct cl_fault_io *fio; - struct lu_env *env; int rc; - *env_ret = NULL; if (ll_file_nolock(file)) return ERR_PTR(-EOPNOTSUPP); - /* - * page fault can be called when lustre IO is - * already active for the current thread, e.g., when doing read/write - * against user level buffer mapped from Lustre buffer. To avoid - * stomping on existing context, optionally force an allocation of a new - * one. - */ - env = cl_env_nested_get(nest); - if (IS_ERR(env)) - return ERR_PTR(-EINVAL); - - *env_ret = env; - restart: io = vvp_env_thread_io(env); io->ci_obj = ll_i2info(inode)->lli_clob; @@ -155,7 +136,6 @@ ll_fault_io_init(struct vm_area_struct *vma, struct lu_env **env_ret, if (io->ci_need_restart) goto restart; - cl_env_nested_put(nest, env); io = ERR_PTR(rc); } @@ -169,13 +149,17 @@ static int ll_page_mkwrite0(struct vm_area_struct *vma, struct page *vmpage, struct lu_env *env; struct cl_io *io; struct vvp_io *vio; - struct cl_env_nest nest; int result; + int refcheck; sigset_t set; struct inode *inode; struct ll_inode_info *lli; - io = ll_fault_io_init(vma, &env, &nest, vmpage->index, NULL); + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + return PTR_ERR(env); + + io = ll_fault_io_init(env, vma, vmpage->index, NULL); if (IS_ERR(io)) { result = PTR_ERR(io); goto out; @@ -240,8 +224,8 @@ static int ll_page_mkwrite0(struct vm_area_struct *vma, struct page *vmpage, out_io: cl_io_fini(env, io); - cl_env_nested_put(&nest, env); out: + cl_env_put(env, &refcheck); CDEBUG(D_MMAP, "%s mkwrite with %d\n", current->comm, result); LASSERT(ergo(result == 0, PageLocked(vmpage))); @@ -285,13 +269,19 @@ static int ll_fault0(struct vm_area_struct *vma, struct vm_fault *vmf) struct vvp_io *vio = NULL; struct page *vmpage; unsigned long ra_flags; - struct cl_env_nest nest; - int result; + int result = 0; int fault_ret = 0; + int refcheck; - io = ll_fault_io_init(vma, &env, &nest, vmf->pgoff, &ra_flags); - if (IS_ERR(io)) - return to_fault_error(PTR_ERR(io)); + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + return PTR_ERR(env); + + io = ll_fault_io_init(env, vma, vmf->pgoff, &ra_flags); + if (IS_ERR(io)) { + result = to_fault_error(PTR_ERR(io)); + goto out; + } result = io->ci_result; if (result == 0) { @@ -322,14 +312,15 @@ static int ll_fault0(struct vm_area_struct *vma, struct vm_fault *vmf) } } cl_io_fini(env, io); - cl_env_nested_put(&nest, env); vma->vm_flags |= ra_flags; + +out: + cl_env_put(env, &refcheck); if (result != 0 && !(fault_ret & VM_FAULT_RETRY)) fault_ret |= to_fault_error(result); - CDEBUG(D_MMAP, "%s fault %d/%d\n", - current->comm, fault_ret, result); + CDEBUG(D_MMAP, "%s fault %d/%d\n", current->comm, fault_ret, result); return fault_ret; } diff --git a/drivers/staging/lustre/lustre/llite/rw.c b/drivers/staging/lustre/lustre/llite/rw.c index 80cb8e0..d2515a8 100644 --- a/drivers/staging/lustre/lustre/llite/rw.c +++ b/drivers/staging/lustre/lustre/llite/rw.c @@ -896,17 +896,17 @@ int ll_writepage(struct page *vmpage, struct writeback_control *wbc) struct cl_io *io; struct cl_page *page; struct cl_object *clob; - struct cl_env_nest nest; bool redirtied = false; bool unlocked = false; int result; + int refcheck; LASSERT(PageLocked(vmpage)); LASSERT(!PageWriteback(vmpage)); LASSERT(ll_i2dtexp(inode)); - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) { result = PTR_ERR(env); goto out; @@ -971,7 +971,7 @@ int ll_writepage(struct page *vmpage, struct writeback_control *wbc) } } - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); goto out; out: diff --git a/drivers/staging/lustre/lustre/llite/rw26.c b/drivers/staging/lustre/lustre/llite/rw26.c index 67010be..1a08a9d 100644 --- a/drivers/staging/lustre/lustre/llite/rw26.c +++ b/drivers/staging/lustre/lustre/llite/rw26.c @@ -103,7 +103,6 @@ static void ll_invalidatepage(struct page *vmpage, unsigned int offset, static int ll_releasepage(struct page *vmpage, gfp_t gfp_mask) { struct lu_env *env; - void *cookie; struct cl_object *obj; struct cl_page *page; struct address_space *mapping; @@ -129,7 +128,6 @@ static int ll_releasepage(struct page *vmpage, gfp_t gfp_mask) if (!page) return 1; - cookie = cl_env_reenter(); env = cl_env_percpu_get(); LASSERT(!IS_ERR(env)); @@ -155,7 +153,6 @@ static int ll_releasepage(struct page *vmpage, gfp_t gfp_mask) cl_page_put(env, page); cl_env_percpu_put(env); - cl_env_reexit(cookie); return result; } @@ -340,7 +337,8 @@ static ssize_t ll_direct_IO_26_seg(const struct lu_env *env, struct cl_io *io, PAGE_SIZE) & ~(DT_MAX_BRW_SIZE - 1)) static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) { - struct lu_env *env; + struct ll_cl_context *lcc; + const struct lu_env *env; struct cl_io *io; struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -348,7 +346,6 @@ static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) ssize_t count = iov_iter_count(iter); ssize_t tot_bytes = 0, result = 0; long size = MAX_DIO_SIZE; - int refcheck; /* FIXME: io smaller than PAGE_SIZE is broken on ia64 ??? */ if ((file_offset & ~PAGE_MASK) || (count & ~PAGE_MASK)) @@ -363,9 +360,13 @@ static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) if (iov_iter_alignment(iter) & ~PAGE_MASK) return -EINVAL; - env = cl_env_get(&refcheck); + lcc = ll_cl_find(file); + if (!lcc) + return -EIO; + + env = lcc->lcc_env; LASSERT(!IS_ERR(env)); - io = vvp_env_io(env)->vui_cl.cis_io; + io = lcc->lcc_io; LASSERT(io); while (iov_iter_count(iter)) { @@ -422,7 +423,6 @@ static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) vio->u.write.vui_written += tot_bytes; } - cl_env_put(env, &refcheck); return tot_bytes ? tot_bytes : result; } diff --git a/drivers/staging/lustre/lustre/lov/lov_io.c b/drivers/staging/lustre/lustre/lov/lov_io.c index 011ab0f..002326c 100644 --- a/drivers/staging/lustre/lustre/lov/lov_io.c +++ b/drivers/staging/lustre/lustre/lov/lov_io.c @@ -167,12 +167,7 @@ static int lov_io_sub_init(const struct lu_env *env, struct lov_io *lio, sub->sub_env = ld->ld_emrg[stripe]->emrg_env; sub->sub_borrowed = 1; } else { - void *cookie; - - /* obtain new environment */ - cookie = cl_env_reenter(); sub->sub_env = cl_env_get(&sub->sub_refcheck); - cl_env_reexit(cookie); if (IS_ERR(sub->sub_env)) result = PTR_ERR(sub->sub_env); diff --git a/drivers/staging/lustre/lustre/lov/lov_object.c b/drivers/staging/lustre/lustre/lov/lov_object.c index 711a3b9..317311c 100644 --- a/drivers/staging/lustre/lustre/lov/lov_object.c +++ b/drivers/staging/lustre/lustre/lov/lov_object.c @@ -729,19 +729,15 @@ static int lov_layout_change(const struct lu_env *unused, union lov_layout_state *state = &lov->u; const struct lov_layout_operations *old_ops; const struct lov_layout_operations *new_ops; - void *cookie; struct lu_env *env; int refcheck; int rc; LASSERT(0 <= lov->lo_type && lov->lo_type < ARRAY_SIZE(lov_dispatch)); - cookie = cl_env_reenter(); env = cl_env_get(&refcheck); - if (IS_ERR(env)) { - cl_env_reexit(cookie); + if (IS_ERR(env)) return PTR_ERR(env); - } LASSERT(0 <= llt && llt < ARRAY_SIZE(lov_dispatch)); @@ -782,7 +778,6 @@ static int lov_layout_change(const struct lu_env *unused, lov->lo_type = llt; out: cl_env_put(env, &refcheck); - cl_env_reexit(cookie); return rc; } diff --git a/drivers/staging/lustre/lustre/obdclass/cl_internal.h b/drivers/staging/lustre/lustre/obdclass/cl_internal.h index e866754..7b403fb 100644 --- a/drivers/staging/lustre/lustre/obdclass/cl_internal.h +++ b/drivers/staging/lustre/lustre/obdclass/cl_internal.h @@ -50,25 +50,6 @@ enum clt_nesting_level { }; /** - * Counters used to check correctness of cl_lock interface usage. - */ -struct cl_thread_counters { - /** - * Number of outstanding calls to cl_lock_mutex_get() made by the - * current thread. For debugging. - */ - int ctc_nr_locks_locked; - /** List of locked locks. */ - struct lu_ref ctc_locks_locked; - /** Number of outstanding holds on locks. */ - int ctc_nr_held; - /** Number of outstanding uses on locks. */ - int ctc_nr_used; - /** Number of held extent locks. */ - int ctc_nr_locks_acquired; -}; - -/** * Thread local state internal for generic cl-code. */ struct cl_thread_info { @@ -83,10 +64,6 @@ struct cl_thread_info { */ struct cl_lock_descr clt_descr; struct cl_page_list clt_list; - /** - * Counters for every level of lock nesting. - */ - struct cl_thread_counters clt_counters[CNL_NR]; /** @} debugging */ /* diff --git a/drivers/staging/lustre/lustre/obdclass/cl_io.c b/drivers/staging/lustre/lustre/obdclass/cl_io.c index c5621ad..af8e4d6 100644 --- a/drivers/staging/lustre/lustre/obdclass/cl_io.c +++ b/drivers/staging/lustre/lustre/obdclass/cl_io.c @@ -412,7 +412,6 @@ void cl_io_unlock(const struct lu_env *env, struct cl_io *io) scan->cis_iop->op[io->ci_type].cio_unlock(env, scan); } io->ci_state = CIS_UNLOCKED; - LASSERT(!cl_env_info(env)->clt_counters[CNL_TOP].ctc_nr_locks_acquired); } EXPORT_SYMBOL(cl_io_unlock); diff --git a/drivers/staging/lustre/lustre/obdclass/cl_object.c b/drivers/staging/lustre/lustre/obdclass/cl_object.c index b4600c4..f5d4e23 100644 --- a/drivers/staging/lustre/lustre/obdclass/cl_object.c +++ b/drivers/staging/lustre/lustre/obdclass/cl_object.c @@ -544,36 +544,20 @@ EXPORT_SYMBOL(cl_site_stats_print); * bz20044, bz22683. */ -static LIST_HEAD(cl_envs); -static unsigned int cl_envs_cached_nr; -static unsigned int cl_envs_cached_max = 128; /* XXX: prototype: arbitrary limit - * for now. - */ -static DEFINE_SPINLOCK(cl_envs_guard); +static unsigned int cl_envs_cached_max = 32; /* XXX: prototype: arbitrary limit + * for now. + */ +static struct cl_env_cache { + rwlock_t cec_guard; + unsigned int cec_count; + struct list_head cec_envs; +} *cl_envs = NULL; struct cl_env { void *ce_magic; struct lu_env ce_lu; struct lu_context ce_ses; - /** - * This allows cl_env to be entered into cl_env_hash which implements - * the current thread -> client environment lookup. - */ - struct hlist_node ce_node; - /** - * Owner for the current cl_env. - * - * If LL_TASK_CL_ENV is defined, this point to the owning current, - * only for debugging purpose ; - * Otherwise hash is used, and this is the key for cfs_hash. - * Now current thread pid is stored. Note using thread pointer would - * lead to unbalanced hash because of its specific allocation locality - * and could be varied for different platforms and OSes, even different - * OS versions. - */ - void *ce_owner; - /* * Linkage into global list of all client environments. Used for * garbage collection. @@ -597,122 +581,13 @@ static void cl_env_init0(struct cl_env *cle, void *debug) { LASSERT(cle->ce_ref == 0); LASSERT(cle->ce_magic == &cl_env_init0); - LASSERT(!cle->ce_debug && !cle->ce_owner); + LASSERT(!cle->ce_debug); cle->ce_ref = 1; cle->ce_debug = debug; CL_ENV_INC(busy); } -/* - * The implementation of using hash table to connect cl_env and thread - */ - -static struct cfs_hash *cl_env_hash; - -static unsigned cl_env_hops_hash(struct cfs_hash *lh, - const void *key, unsigned mask) -{ -#if BITS_PER_LONG == 64 - return cfs_hash_u64_hash((__u64)key, mask); -#else - return cfs_hash_u32_hash((__u32)key, mask); -#endif -} - -static void *cl_env_hops_obj(struct hlist_node *hn) -{ - struct cl_env *cle = hlist_entry(hn, struct cl_env, ce_node); - - LASSERT(cle->ce_magic == &cl_env_init0); - return (void *)cle; -} - -static int cl_env_hops_keycmp(const void *key, struct hlist_node *hn) -{ - struct cl_env *cle = cl_env_hops_obj(hn); - - LASSERT(cle->ce_owner); - return (key == cle->ce_owner); -} - -static void cl_env_hops_noop(struct cfs_hash *hs, struct hlist_node *hn) -{ - struct cl_env *cle = hlist_entry(hn, struct cl_env, ce_node); - - LASSERT(cle->ce_magic == &cl_env_init0); -} - -static struct cfs_hash_ops cl_env_hops = { - .hs_hash = cl_env_hops_hash, - .hs_key = cl_env_hops_obj, - .hs_keycmp = cl_env_hops_keycmp, - .hs_object = cl_env_hops_obj, - .hs_get = cl_env_hops_noop, - .hs_put_locked = cl_env_hops_noop, -}; - -static inline struct cl_env *cl_env_fetch(void) -{ - struct cl_env *cle; - - cle = cfs_hash_lookup(cl_env_hash, (void *)(long)current->pid); - LASSERT(ergo(cle, cle->ce_magic == &cl_env_init0)); - return cle; -} - -static inline void cl_env_attach(struct cl_env *cle) -{ - if (cle) { - int rc; - - LASSERT(!cle->ce_owner); - cle->ce_owner = (void *)(long)current->pid; - rc = cfs_hash_add_unique(cl_env_hash, cle->ce_owner, - &cle->ce_node); - LASSERT(rc == 0); - } -} - -static inline void cl_env_do_detach(struct cl_env *cle) -{ - void *cookie; - - LASSERT(cle->ce_owner == (void *)(long)current->pid); - cookie = cfs_hash_del(cl_env_hash, cle->ce_owner, - &cle->ce_node); - LASSERT(cookie == cle); - cle->ce_owner = NULL; -} - -static int cl_env_store_init(void) -{ - cl_env_hash = cfs_hash_create("cl_env", - HASH_CL_ENV_BITS, HASH_CL_ENV_BITS, - HASH_CL_ENV_BKT_BITS, 0, - CFS_HASH_MIN_THETA, - CFS_HASH_MAX_THETA, - &cl_env_hops, - CFS_HASH_RW_BKTLOCK); - return cl_env_hash ? 0 : -ENOMEM; -} - -static void cl_env_store_fini(void) -{ - cfs_hash_putref(cl_env_hash); -} - -static inline struct cl_env *cl_env_detach(struct cl_env *cle) -{ - if (!cle) - cle = cl_env_fetch(); - - if (cle && cle->ce_owner) - cl_env_do_detach(cle); - - return cle; -} - static struct lu_env *cl_env_new(__u32 ctx_tags, __u32 ses_tags, void *debug) { struct lu_env *env; @@ -762,16 +637,20 @@ static struct lu_env *cl_env_obtain(void *debug) { struct cl_env *cle; struct lu_env *env; + int cpu = get_cpu(); - spin_lock(&cl_envs_guard); - LASSERT(equi(cl_envs_cached_nr == 0, list_empty(&cl_envs))); - if (cl_envs_cached_nr > 0) { + read_lock(&cl_envs[cpu].cec_guard); + LASSERT(equi(cl_envs[cpu].cec_count == 0, + list_empty(&cl_envs[cpu].cec_envs))); + if (cl_envs[cpu].cec_count > 0) { int rc; - cle = container_of(cl_envs.next, struct cl_env, ce_linkage); + cle = container_of(cl_envs[cpu].cec_envs.next, struct cl_env, + ce_linkage); list_del_init(&cle->ce_linkage); - cl_envs_cached_nr--; - spin_unlock(&cl_envs_guard); + cl_envs[cpu].cec_count--; + read_unlock(&cl_envs[cpu].cec_guard); + put_cpu(); env = &cle->ce_lu; rc = lu_env_refill(env); @@ -784,7 +663,8 @@ static struct lu_env *cl_env_obtain(void *debug) env = ERR_PTR(rc); } } else { - spin_unlock(&cl_envs_guard); + read_unlock(&cl_envs[cpu].cec_guard); + put_cpu(); env = cl_env_new(lu_context_tags_default, lu_session_tags_default, debug); } @@ -796,27 +676,6 @@ static inline struct cl_env *cl_env_container(struct lu_env *env) return container_of(env, struct cl_env, ce_lu); } -static struct lu_env *cl_env_peek(int *refcheck) -{ - struct lu_env *env; - struct cl_env *cle; - - CL_ENV_INC(lookup); - - /* check that we don't go far from untrusted pointer */ - CLASSERT(offsetof(struct cl_env, ce_magic) == 0); - - env = NULL; - cle = cl_env_fetch(); - if (cle) { - CL_ENV_INC(hit); - env = &cle->ce_lu; - *refcheck = ++cle->ce_ref; - } - CDEBUG(D_OTHER, "%d@%p\n", cle ? cle->ce_ref : 0, cle); - return env; -} - /** * Returns lu_env: if there already is an environment associated with the * current thread, it is returned, otherwise, new environment is allocated. @@ -834,17 +693,13 @@ struct lu_env *cl_env_get(int *refcheck) { struct lu_env *env; - env = cl_env_peek(refcheck); - if (!env) { - env = cl_env_obtain(__builtin_return_address(0)); - if (!IS_ERR(env)) { - struct cl_env *cle; + env = cl_env_obtain(__builtin_return_address(0)); + if (!IS_ERR(env)) { + struct cl_env *cle; - cle = cl_env_container(env); - cl_env_attach(cle); - *refcheck = cle->ce_ref; - CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); - } + cle = cl_env_container(env); + *refcheck = cle->ce_ref; + CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); } return env; } @@ -859,7 +714,6 @@ struct lu_env *cl_env_alloc(int *refcheck, __u32 tags) { struct lu_env *env; - LASSERT(!cl_env_peek(refcheck)); env = cl_env_new(tags, tags, __builtin_return_address(0)); if (!IS_ERR(env)) { struct cl_env *cle; @@ -874,7 +728,6 @@ EXPORT_SYMBOL(cl_env_alloc); static void cl_env_exit(struct cl_env *cle) { - LASSERT(!cle->ce_owner); lu_context_exit(&cle->ce_lu.le_ctx); lu_context_exit(&cle->ce_ses); } @@ -887,20 +740,25 @@ static void cl_env_exit(struct cl_env *cle) unsigned int cl_env_cache_purge(unsigned int nr) { struct cl_env *cle; + unsigned int i; - spin_lock(&cl_envs_guard); - for (; !list_empty(&cl_envs) && nr > 0; --nr) { - cle = container_of(cl_envs.next, struct cl_env, ce_linkage); - list_del_init(&cle->ce_linkage); - LASSERT(cl_envs_cached_nr > 0); - cl_envs_cached_nr--; - spin_unlock(&cl_envs_guard); + for_each_possible_cpu(i) { + write_lock(&cl_envs[i].cec_guard); + for (; !list_empty(&cl_envs[i].cec_envs) && nr > 0; --nr) { + cle = container_of(cl_envs[i].cec_envs.next, + struct cl_env, ce_linkage); + list_del_init(&cle->ce_linkage); + LASSERT(cl_envs[i].cec_count > 0); + cl_envs[i].cec_count--; + write_unlock(&cl_envs[i].cec_guard); - cl_env_fini(cle); - spin_lock(&cl_envs_guard); + cl_env_fini(cle); + write_lock(&cl_envs[i].cec_guard); + } + LASSERT(equi(cl_envs[i].cec_count == 0, + list_empty(&cl_envs[i].cec_envs))); + write_unlock(&cl_envs[i].cec_guard); } - LASSERT(equi(cl_envs_cached_nr == 0, list_empty(&cl_envs))); - spin_unlock(&cl_envs_guard); return nr; } EXPORT_SYMBOL(cl_env_cache_purge); @@ -923,8 +781,9 @@ void cl_env_put(struct lu_env *env, int *refcheck) CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); if (--cle->ce_ref == 0) { + int cpu = get_cpu(); + CL_ENV_DEC(busy); - cl_env_detach(cle); cle->ce_debug = NULL; cl_env_exit(cle); /* @@ -933,107 +792,22 @@ void cl_env_put(struct lu_env *env, int *refcheck) * Return environment to the cache only when it was allocated * with the standard tags. */ - if (cl_envs_cached_nr < cl_envs_cached_max && + if (cl_envs[cpu].cec_count < cl_envs_cached_max && (env->le_ctx.lc_tags & ~LCT_HAS_EXIT) == LCT_CL_THREAD && (env->le_ses->lc_tags & ~LCT_HAS_EXIT) == LCT_SESSION) { - spin_lock(&cl_envs_guard); - list_add(&cle->ce_linkage, &cl_envs); - cl_envs_cached_nr++; - spin_unlock(&cl_envs_guard); + read_lock(&cl_envs[cpu].cec_guard); + list_add(&cle->ce_linkage, &cl_envs[cpu].cec_envs); + cl_envs[cpu].cec_count++; + read_unlock(&cl_envs[cpu].cec_guard); } else { cl_env_fini(cle); } + put_cpu(); } } EXPORT_SYMBOL(cl_env_put); /** - * Declares a point of re-entrancy. - * - * \see cl_env_reexit() - */ -void *cl_env_reenter(void) -{ - return cl_env_detach(NULL); -} -EXPORT_SYMBOL(cl_env_reenter); - -/** - * Exits re-entrancy. - */ -void cl_env_reexit(void *cookie) -{ - cl_env_detach(NULL); - cl_env_attach(cookie); -} -EXPORT_SYMBOL(cl_env_reexit); - -/** - * Setup user-supplied \a env as a current environment. This is to be used to - * guaranteed that environment exists even when cl_env_get() fails. It is up - * to user to ensure proper concurrency control. - * - * \see cl_env_unplant() - */ -void cl_env_implant(struct lu_env *env, int *refcheck) -{ - struct cl_env *cle = cl_env_container(env); - - LASSERT(cle->ce_ref > 0); - - cl_env_attach(cle); - cl_env_get(refcheck); - CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); -} -EXPORT_SYMBOL(cl_env_implant); - -/** - * Detach environment installed earlier by cl_env_implant(). - */ -void cl_env_unplant(struct lu_env *env, int *refcheck) -{ - struct cl_env *cle = cl_env_container(env); - - LASSERT(cle->ce_ref > 1); - - CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); - - cl_env_detach(cle); - cl_env_put(env, refcheck); -} -EXPORT_SYMBOL(cl_env_unplant); - -struct lu_env *cl_env_nested_get(struct cl_env_nest *nest) -{ - struct lu_env *env; - - nest->cen_cookie = NULL; - env = cl_env_peek(&nest->cen_refcheck); - if (env) { - if (!cl_io_is_going(env)) - return env; - cl_env_put(env, &nest->cen_refcheck); - nest->cen_cookie = cl_env_reenter(); - } - env = cl_env_get(&nest->cen_refcheck); - if (IS_ERR(env)) { - cl_env_reexit(nest->cen_cookie); - return env; - } - - LASSERT(!cl_io_is_going(env)); - return env; -} -EXPORT_SYMBOL(cl_env_nested_get); - -void cl_env_nested_put(struct cl_env_nest *nest, struct lu_env *env) -{ - cl_env_put(env, &nest->cen_refcheck); - cl_env_reexit(nest->cen_cookie); -} -EXPORT_SYMBOL(cl_env_nested_put); - -/** * Converts struct ost_lvb to struct cl_attr. * * \see cl_attr2lvb @@ -1060,6 +834,10 @@ static int cl_env_percpu_init(void) for_each_possible_cpu(i) { struct lu_env *env; + rwlock_init(&cl_envs[i].cec_guard); + INIT_LIST_HEAD(&cl_envs[i].cec_envs); + cl_envs[i].cec_count = 0; + cle = &cl_env_percpu[i]; env = &cle->ce_lu; @@ -1127,7 +905,6 @@ void cl_env_percpu_put(struct lu_env *env) LASSERT(cle->ce_ref == 0); CL_ENV_DEC(busy); - cl_env_detach(cle); cle->ce_debug = NULL; put_cpu(); @@ -1141,7 +918,6 @@ struct lu_env *cl_env_percpu_get(void) cle = &cl_env_percpu[get_cpu()]; cl_env_init0(cle, __builtin_return_address(0)); - cl_env_attach(cle); return &cle->ce_lu; } EXPORT_SYMBOL(cl_env_percpu_get); @@ -1205,51 +981,19 @@ LU_KEY_INIT_FINI(cl0, struct cl_thread_info); static void *cl_key_init(const struct lu_context *ctx, struct lu_context_key *key) { - struct cl_thread_info *info; - - info = cl0_key_init(ctx, key); - if (!IS_ERR(info)) { - size_t i; - - for (i = 0; i < ARRAY_SIZE(info->clt_counters); ++i) - lu_ref_init(&info->clt_counters[i].ctc_locks_locked); - } - return info; + return cl0_key_init(ctx, key); } static void cl_key_fini(const struct lu_context *ctx, struct lu_context_key *key, void *data) { - struct cl_thread_info *info; - size_t i; - - info = data; - for (i = 0; i < ARRAY_SIZE(info->clt_counters); ++i) - lu_ref_fini(&info->clt_counters[i].ctc_locks_locked); cl0_key_fini(ctx, key, data); } -static void cl_key_exit(const struct lu_context *ctx, - struct lu_context_key *key, void *data) -{ - struct cl_thread_info *info = data; - size_t i; - - for (i = 0; i < ARRAY_SIZE(info->clt_counters); ++i) { - LASSERT(info->clt_counters[i].ctc_nr_held == 0); - LASSERT(info->clt_counters[i].ctc_nr_used == 0); - LASSERT(info->clt_counters[i].ctc_nr_locks_acquired == 0); - LASSERT(info->clt_counters[i].ctc_nr_locks_locked == 0); - lu_ref_fini(&info->clt_counters[i].ctc_locks_locked); - lu_ref_init(&info->clt_counters[i].ctc_locks_locked); - } -} - static struct lu_context_key cl_key = { .lct_tags = LCT_CL_THREAD, .lct_init = cl_key_init, .lct_fini = cl_key_fini, - .lct_exit = cl_key_exit }; static struct lu_kmem_descr cl_object_caches[] = { @@ -1273,13 +1017,15 @@ int cl_global_init(void) { int result; - result = cl_env_store_init(); - if (result) - return result; + cl_envs = kzalloc(sizeof(*cl_envs) * num_possible_cpus(), GFP_KERNEL); + if (!cl_envs) { + result = -ENOMEM; + goto out; + } result = lu_kmem_init(cl_object_caches); if (result) - goto out_store; + goto out_envs; LU_CONTEXT_KEY_INIT(&cl_key); result = lu_context_key_register(&cl_key); @@ -1289,16 +1035,17 @@ int cl_global_init(void) result = cl_env_percpu_init(); if (result) /* no cl_env_percpu_fini on error */ - goto out_context; + goto out_keys; return 0; -out_context: +out_keys: lu_context_key_degister(&cl_key); out_kmem: lu_kmem_fini(cl_object_caches); -out_store: - cl_env_store_fini(); +out_envs: + kfree(cl_envs); +out: return result; } @@ -1310,5 +1057,5 @@ void cl_global_fini(void) cl_env_percpu_fini(); lu_context_key_degister(&cl_key); lu_kmem_fini(cl_object_caches); - cl_env_store_fini(); + kfree(cl_envs); } diff --git a/drivers/staging/lustre/lustre/obdclass/lu_object.c b/drivers/staging/lustre/lustre/obdclass/lu_object.c index e031fd2..14ca82d 100644 --- a/drivers/staging/lustre/lustre/obdclass/lu_object.c +++ b/drivers/staging/lustre/lustre/obdclass/lu_object.c @@ -1521,10 +1521,6 @@ void lu_context_key_quiesce(struct lu_context_key *key) if (!(key->lct_tags & LCT_QUIESCENT)) { /* - * XXX layering violation. - */ - cl_env_cache_purge(~0); - /* * XXX memory barrier has to go here. */ spin_lock(&lu_keys_guard); diff --git a/drivers/staging/lustre/lustre/obdecho/echo_client.c b/drivers/staging/lustre/lustre/obdecho/echo_client.c index 30caf5f..25ecfb3 100644 --- a/drivers/staging/lustre/lustre/obdecho/echo_client.c +++ b/drivers/staging/lustre/lustre/obdecho/echo_client.c @@ -563,16 +563,10 @@ static void echo_thread_key_fini(const struct lu_context *ctx, kmem_cache_free(echo_thread_kmem, info); } -static void echo_thread_key_exit(const struct lu_context *ctx, - struct lu_context_key *key, void *data) -{ -} - static struct lu_context_key echo_thread_key = { .lct_tags = LCT_CL_THREAD, .lct_init = echo_thread_key_init, .lct_fini = echo_thread_key_fini, - .lct_exit = echo_thread_key_exit }; static void *echo_session_key_init(const struct lu_context *ctx, @@ -594,16 +588,10 @@ static void echo_session_key_fini(const struct lu_context *ctx, kmem_cache_free(echo_session_kmem, session); } -static void echo_session_key_exit(const struct lu_context *ctx, - struct lu_context_key *key, void *data) -{ -} - static struct lu_context_key echo_session_key = { .lct_tags = LCT_SESSION, .lct_init = echo_session_key_init, .lct_fini = echo_session_key_fini, - .lct_exit = echo_session_key_exit }; LU_TYPE_INIT_FINI(echo, &echo_thread_key, &echo_session_key); @@ -787,6 +775,8 @@ static struct lu_device *echo_device_free(const struct lu_env *env, cl_device_fini(&ed->ed_cl); kfree(ed); + cl_env_cache_purge(~0); + return NULL; } diff --git a/drivers/staging/lustre/lustre/osc/osc_cache.c b/drivers/staging/lustre/lustre/osc/osc_cache.c index b645957..cbbe85e 100644 --- a/drivers/staging/lustre/lustre/osc/osc_cache.c +++ b/drivers/staging/lustre/lustre/osc/osc_cache.c @@ -977,7 +977,6 @@ static int osc_extent_wait(const struct lu_env *env, struct osc_extent *ext, static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, bool partial) { - struct cl_env_nest nest; struct lu_env *env; struct cl_io *io; struct osc_object *obj = ext->oe_obj; @@ -990,6 +989,7 @@ static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, int grants = 0; int nr_pages = 0; int rc = 0; + int refcheck; LASSERT(sanity_check(ext) == 0); EASSERT(ext->oe_state == OES_TRUNC, ext); @@ -999,7 +999,7 @@ static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, * We can't use that env from osc_cache_truncate_start() because * it's from lov_io_sub and not fully initialized. */ - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); io = &osc_env_info(env)->oti_io; io->ci_obj = cl_object_top(osc2cl(obj)); rc = cl_io_init(env, io, CIT_MISC, io->ci_obj); @@ -1085,7 +1085,7 @@ static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, out: cl_io_fini(env, io); - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return rc; } diff --git a/drivers/staging/lustre/lustre/osc/osc_lock.c b/drivers/staging/lustre/lustre/osc/osc_lock.c index 45d5e6d..e337e87 100644 --- a/drivers/staging/lustre/lustre/osc/osc_lock.c +++ b/drivers/staging/lustre/lustre/osc/osc_lock.c @@ -294,10 +294,10 @@ static int osc_lock_upcall(void *cookie, struct lustre_handle *lockh, struct osc_lock *oscl = cookie; struct cl_lock_slice *slice = &oscl->ols_cl; struct lu_env *env; - struct cl_env_nest nest; int rc; + int refcheck; - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); /* should never happen, similar to osc_ldlm_blocking_ast(). */ LASSERT(!IS_ERR(env)); @@ -336,7 +336,7 @@ static int osc_lock_upcall(void *cookie, struct lustre_handle *lockh, if (oscl->ols_owner) cl_sync_io_note(env, oscl->ols_owner, rc); - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return rc; } @@ -347,9 +347,9 @@ static int osc_lock_upcall_agl(void *cookie, struct lustre_handle *lockh, struct osc_object *osc = cookie; struct ldlm_lock *dlmlock; struct lu_env *env; - struct cl_env_nest nest; + int refcheck; - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); LASSERT(!IS_ERR(env)); if (errcode == ELDLM_LOCK_MATCHED) { @@ -374,7 +374,7 @@ static int osc_lock_upcall_agl(void *cookie, struct lustre_handle *lockh, out: cl_object_put(env, osc2cl(osc)); - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return ldlm_error2errno(errcode); } @@ -382,11 +382,11 @@ static int osc_lock_flush(struct osc_object *obj, pgoff_t start, pgoff_t end, enum cl_lock_mode mode, int discard) { struct lu_env *env; - struct cl_env_nest nest; + int refcheck; int rc = 0; int rc2 = 0; - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) return PTR_ERR(env); @@ -404,7 +404,7 @@ static int osc_lock_flush(struct osc_object *obj, pgoff_t start, pgoff_t end, if (rc == 0 && rc2 < 0) rc = rc2; - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return rc; } @@ -536,7 +536,7 @@ static int osc_ldlm_blocking_ast(struct ldlm_lock *dlmlock, } case LDLM_CB_CANCELING: { struct lu_env *env; - struct cl_env_nest nest; + int refcheck; /* * This can be called in the context of outer IO, e.g., @@ -549,14 +549,14 @@ static int osc_ldlm_blocking_ast(struct ldlm_lock *dlmlock, * new environment has to be created to not corrupt outer * context. */ - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) { result = PTR_ERR(env); break; } result = osc_dlm_blocking_ast0(env, dlmlock, data, flag); - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); break; } default: @@ -568,61 +568,61 @@ static int osc_ldlm_blocking_ast(struct ldlm_lock *dlmlock, static int osc_ldlm_glimpse_ast(struct ldlm_lock *dlmlock, void *data) { struct ptlrpc_request *req = data; - struct cl_env_nest nest; struct lu_env *env; struct ost_lvb *lvb; struct req_capsule *cap; + struct cl_object *obj = NULL; int result; + int refcheck; LASSERT(lustre_msg_get_opc(req->rq_reqmsg) == LDLM_GL_CALLBACK); - env = cl_env_nested_get(&nest); - if (!IS_ERR(env)) { - struct cl_object *obj = NULL; + env = cl_env_get(&refcheck); + if (IS_ERR(env)) { + result = PTR_ERR(env); + goto out; + } - lock_res_and_lock(dlmlock); - if (dlmlock->l_ast_data) { - obj = osc2cl(dlmlock->l_ast_data); - cl_object_get(obj); - } - unlock_res_and_lock(dlmlock); + lock_res_and_lock(dlmlock); + if (dlmlock->l_ast_data) { + obj = osc2cl(dlmlock->l_ast_data); + cl_object_get(obj); + } + unlock_res_and_lock(dlmlock); - if (obj) { - /* Do not grab the mutex of cl_lock for glimpse. - * See LU-1274 for details. - * BTW, it's okay for cl_lock to be cancelled during - * this period because server can handle this race. - * See ldlm_server_glimpse_ast() for details. - * cl_lock_mutex_get(env, lock); - */ - cap = &req->rq_pill; - req_capsule_extend(cap, &RQF_LDLM_GL_CALLBACK); - req_capsule_set_size(cap, &RMF_DLM_LVB, RCL_SERVER, - sizeof(*lvb)); - result = req_capsule_server_pack(cap); - if (result == 0) { - lvb = req_capsule_server_get(cap, &RMF_DLM_LVB); - result = cl_object_glimpse(env, obj, lvb); - } - if (!exp_connect_lvb_type(req->rq_export)) - req_capsule_shrink(&req->rq_pill, - &RMF_DLM_LVB, - sizeof(struct ost_lvb_v1), - RCL_SERVER); - cl_object_put(env, obj); - } else { - /* - * These errors are normal races, so we don't want to - * fill the console with messages by calling - * ptlrpc_error() - */ - lustre_pack_reply(req, 1, NULL, NULL); - result = -ELDLM_NO_LOCK_DATA; + if (obj) { + /* Do not grab the mutex of cl_lock for glimpse. + * See LU-1274 for details. + * BTW, it's okay for cl_lock to be cancelled during + * this period because server can handle this race. + * See ldlm_server_glimpse_ast() for details. + * cl_lock_mutex_get(env, lock); + */ + cap = &req->rq_pill; + req_capsule_extend(cap, &RQF_LDLM_GL_CALLBACK); + req_capsule_set_size(cap, &RMF_DLM_LVB, RCL_SERVER, + sizeof(*lvb)); + result = req_capsule_server_pack(cap); + if (result == 0) { + lvb = req_capsule_server_get(cap, &RMF_DLM_LVB); + result = cl_object_glimpse(env, obj, lvb); } - cl_env_nested_put(&nest, env); + if (!exp_connect_lvb_type(req->rq_export)) + req_capsule_shrink(&req->rq_pill, &RMF_DLM_LVB, + sizeof(struct ost_lvb_v1), RCL_SERVER); + cl_object_put(env, obj); } else { - result = PTR_ERR(env); + /* + * These errors are normal races, so we don't want to + * fill the console with messages by calling + * ptlrpc_error() + */ + lustre_pack_reply(req, 1, NULL, NULL); + result = -ELDLM_NO_LOCK_DATA; } + cl_env_put(env, &refcheck); + +out: req->rq_status = result; return result; } @@ -677,12 +677,12 @@ static unsigned long osc_lock_weight(const struct lu_env *env, */ unsigned long osc_ldlm_weigh_ast(struct ldlm_lock *dlmlock) { - struct cl_env_nest nest; struct lu_env *env; struct osc_object *obj; struct osc_lock *oscl; unsigned long weight; bool found = false; + int refcheck; might_sleep(); /* @@ -692,7 +692,7 @@ unsigned long osc_ldlm_weigh_ast(struct ldlm_lock *dlmlock) * the upper context because cl_lock_put don't modify environment * variables. But just in case .. */ - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) /* Mostly because lack of memory, do not eliminate this lock */ return 1; @@ -722,7 +722,7 @@ unsigned long osc_ldlm_weigh_ast(struct ldlm_lock *dlmlock) weight = osc_lock_weight(env, obj, &dlmlock->l_policy_data.l_extent); out: - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); return weight; } diff --git a/drivers/staging/lustre/lustre/osc/osc_page.c b/drivers/staging/lustre/lustre/osc/osc_page.c index 2fb0e53..93248d1 100644 --- a/drivers/staging/lustre/lustre/osc/osc_page.c +++ b/drivers/staging/lustre/lustre/osc/osc_page.c @@ -644,15 +644,15 @@ long osc_lru_shrink(const struct lu_env *env, struct client_obd *cli, long osc_lru_reclaim(struct client_obd *cli) { - struct cl_env_nest nest; struct lu_env *env; struct cl_client_cache *cache = cli->cl_cache; int max_scans; + int refcheck; long rc = 0; LASSERT(cache); - env = cl_env_nested_get(&nest); + env = cl_env_get(&refcheck); if (IS_ERR(env)) return 0; @@ -704,7 +704,7 @@ long osc_lru_reclaim(struct client_obd *cli) spin_unlock(&cache->ccc_lru_lock); out: - cl_env_nested_put(&nest, env); + cl_env_put(env, &refcheck); CDEBUG(D_CACHE, "%s: cli %p freed %ld pages.\n", cli->cl_import->imp_obd->obd_name, cli, rc); return rc; -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 05/14] staging/lustre: Get rid of cl_env hash table 2016-11-03 1:24 ` [PATCH 05/14] staging/lustre: Get rid of cl_env hash table Oleg Drokin @ 2016-11-07 1:53 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:53 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Jinshan Xiong > From: Jinshan Xiong <jinshan.xiong@intel.com> > > cl_env hash table is under heavy contention when there are lots of > processes doing IO at the same time; > reduce lock contention by replacing cl_env cache with percpu array; > remove cl_env_nested_get() and cl_env_nested_put(); > remove cl_env_reenter() and cl_env_reexit(); Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com> > Reviewed-on: http://review.whamcloud.com/20254 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-4257 > Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> > Reviewed-by: Bobi Jam <bobijam@hotmail.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/include/cl_object.h | 22 -- > drivers/staging/lustre/lustre/ldlm/ldlm_pool.c | 9 - > drivers/staging/lustre/lustre/llite/file.c | 18 +- > drivers/staging/lustre/lustre/llite/lcommon_cl.c | 11 +- > drivers/staging/lustre/lustre/llite/lcommon_misc.c | 8 +- > drivers/staging/lustre/lustre/llite/llite_mmap.c | 59 ++-- > drivers/staging/lustre/lustre/llite/rw.c | 6 +- > drivers/staging/lustre/lustre/llite/rw26.c | 16 +- > drivers/staging/lustre/lustre/lov/lov_io.c | 5 - > drivers/staging/lustre/lustre/lov/lov_object.c | 7 +- > .../staging/lustre/lustre/obdclass/cl_internal.h | 23 -- > drivers/staging/lustre/lustre/obdclass/cl_io.c | 1 - > drivers/staging/lustre/lustre/obdclass/cl_object.c | 389 ++++----------------- > drivers/staging/lustre/lustre/obdclass/lu_object.c | 4 - > .../staging/lustre/lustre/obdecho/echo_client.c | 14 +- > drivers/staging/lustre/lustre/osc/osc_cache.c | 6 +- > drivers/staging/lustre/lustre/osc/osc_lock.c | 116 +++--- > drivers/staging/lustre/lustre/osc/osc_page.c | 6 +- > 18 files changed, 183 insertions(+), 537 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/include/cl_object.h b/drivers/staging/lustre/lustre/include/cl_object.h > index 514d650..3fe26e7 100644 > --- a/drivers/staging/lustre/lustre/include/cl_object.h > +++ b/drivers/staging/lustre/lustre/include/cl_object.h > @@ -2640,35 +2640,13 @@ void cl_sync_io_end(const struct lu_env *env, struct cl_sync_io *anchor); > * - allocation and destruction of environment is amortized by caching no > * longer used environments instead of destroying them; > * > - * - there is a notion of "current" environment, attached to the kernel > - * data structure representing current thread Top-level lustre code > - * allocates an environment and makes it current, then calls into > - * non-lustre code, that in turn calls lustre back. Low-level lustre > - * code thus called can fetch environment created by the top-level code > - * and reuse it, avoiding additional environment allocation. > - * Right now, three interfaces can attach the cl_env to running thread: > - * - cl_env_get > - * - cl_env_implant > - * - cl_env_reexit(cl_env_reenter had to be called priorly) > - * > * \see lu_env, lu_context, lu_context_key > * @{ > */ > > -struct cl_env_nest { > - int cen_refcheck; > - void *cen_cookie; > -}; > - > struct lu_env *cl_env_get(int *refcheck); > struct lu_env *cl_env_alloc(int *refcheck, __u32 tags); > -struct lu_env *cl_env_nested_get(struct cl_env_nest *nest); > void cl_env_put(struct lu_env *env, int *refcheck); > -void cl_env_nested_put(struct cl_env_nest *nest, struct lu_env *env); > -void *cl_env_reenter(void); > -void cl_env_reexit(void *cookie); > -void cl_env_implant(struct lu_env *env, int *refcheck); > -void cl_env_unplant(struct lu_env *env, int *refcheck); > unsigned int cl_env_cache_purge(unsigned int nr); > struct lu_env *cl_env_percpu_get(void); > void cl_env_percpu_put(struct lu_env *env); > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c b/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c > index b29c9561..19831c5 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_pool.c > @@ -794,7 +794,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) > int nr_ns; > struct ldlm_namespace *ns; > struct ldlm_namespace *ns_old = NULL; /* loop detection */ > - void *cookie; > > if (client == LDLM_NAMESPACE_CLIENT && !(gfp_mask & __GFP_FS)) > return 0; > @@ -802,8 +801,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) > CDEBUG(D_DLMTRACE, "Request to count %s locks from all pools\n", > client == LDLM_NAMESPACE_CLIENT ? "client" : "server"); > > - cookie = cl_env_reenter(); > - > /* > * Find out how many resources we may release. > */ > @@ -812,7 +809,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) > mutex_lock(ldlm_namespace_lock(client)); > if (list_empty(ldlm_namespace_list(client))) { > mutex_unlock(ldlm_namespace_lock(client)); > - cl_env_reexit(cookie); > return 0; > } > ns = ldlm_namespace_first_locked(client); > @@ -838,7 +834,6 @@ static unsigned long ldlm_pools_count(ldlm_side_t client, gfp_t gfp_mask) > ldlm_namespace_put(ns); > } > > - cl_env_reexit(cookie); > return total; > } > > @@ -847,13 +842,10 @@ static unsigned long ldlm_pools_scan(ldlm_side_t client, int nr, gfp_t gfp_mask) > unsigned long freed = 0; > int tmp, nr_ns; > struct ldlm_namespace *ns; > - void *cookie; > > if (client == LDLM_NAMESPACE_CLIENT && !(gfp_mask & __GFP_FS)) > return -1; > > - cookie = cl_env_reenter(); > - > /* > * Shrink at least ldlm_namespace_nr_read(client) namespaces. > */ > @@ -883,7 +875,6 @@ static unsigned long ldlm_pools_scan(ldlm_side_t client, int nr, gfp_t gfp_mask) > freed += ldlm_pool_shrink(&ns->ns_pool, cancel, gfp_mask); > ldlm_namespace_put(ns); > } > - cl_env_reexit(cookie); > /* > * we only decrease the SLV in server pools shrinker, return > * SHRINK_STOP to kernel to avoid needless loop. LU-1128 > diff --git a/drivers/staging/lustre/lustre/llite/file.c b/drivers/staging/lustre/lustre/llite/file.c > index 0ae78c7..c1c7551 100644 > --- a/drivers/staging/lustre/lustre/llite/file.c > +++ b/drivers/staging/lustre/lustre/llite/file.c > @@ -1584,11 +1584,11 @@ int ll_data_version(struct inode *inode, __u64 *data_version, int flags) > */ > int ll_hsm_release(struct inode *inode) > { > - struct cl_env_nest nest; > struct lu_env *env; > struct obd_client_handle *och = NULL; > __u64 data_version = 0; > int rc; > + int refcheck; > > CDEBUG(D_INODE, "%s: Releasing file "DFID".\n", > ll_get_fsname(inode->i_sb, NULL, 0), > @@ -1605,14 +1605,14 @@ int ll_hsm_release(struct inode *inode) > if (rc != 0) > goto out; > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) { > rc = PTR_ERR(env); > goto out; > } > > ll_merge_attr(env, inode); > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > > /* Release the file. > * NB: lease lock handle is released in mdc_hsm_release_pack() because > @@ -2268,17 +2268,17 @@ static int ll_flush(struct file *file, fl_owner_t id) > int cl_sync_file_range(struct inode *inode, loff_t start, loff_t end, > enum cl_fsync_mode mode, int ignore_layout) > { > - struct cl_env_nest nest; > struct lu_env *env; > struct cl_io *io; > struct cl_fsync_io *fio; > int result; > + int refcheck; > > if (mode != CL_FSYNC_NONE && mode != CL_FSYNC_LOCAL && > mode != CL_FSYNC_DISCARD && mode != CL_FSYNC_ALL) > return -EINVAL; > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) > return PTR_ERR(env); > > @@ -2301,7 +2301,7 @@ int cl_sync_file_range(struct inode *inode, loff_t start, loff_t end, > if (result == 0) > result = fio->fi_nr_written; > cl_io_fini(env, io); > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > > return result; > } > @@ -3149,14 +3149,14 @@ int ll_layout_conf(struct inode *inode, const struct cl_object_conf *conf) > { > struct ll_inode_info *lli = ll_i2info(inode); > struct cl_object *obj = lli->lli_clob; > - struct cl_env_nest nest; > struct lu_env *env; > int rc; > + int refcheck; > > if (!obj) > return 0; > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) > return PTR_ERR(env); > > @@ -3190,7 +3190,7 @@ int ll_layout_conf(struct inode *inode, const struct cl_object_conf *conf) > ll_layout_version_set(lli, cl.cl_layout_gen); > } > out: > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > return rc; > } > > diff --git a/drivers/staging/lustre/lustre/llite/lcommon_cl.c b/drivers/staging/lustre/lustre/llite/lcommon_cl.c > index 56e414b..dd1cfd8 100644 > --- a/drivers/staging/lustre/lustre/llite/lcommon_cl.c > +++ b/drivers/staging/lustre/lustre/llite/lcommon_cl.c > @@ -246,15 +246,11 @@ void cl_inode_fini(struct inode *inode) > int emergency; > > if (clob) { > - void *cookie; > - > - cookie = cl_env_reenter(); > env = cl_env_get(&refcheck); > emergency = IS_ERR(env); > if (emergency) { > mutex_lock(&cl_inode_fini_guard); > LASSERT(cl_inode_fini_env); > - cl_env_implant(cl_inode_fini_env, &refcheck); > env = cl_inode_fini_env; > } > /* > @@ -266,13 +262,10 @@ void cl_inode_fini(struct inode *inode) > lu_object_ref_del(&clob->co_lu, "inode", inode); > cl_object_put_last(env, clob); > lli->lli_clob = NULL; > - if (emergency) { > - cl_env_unplant(cl_inode_fini_env, &refcheck); > + if (emergency) > mutex_unlock(&cl_inode_fini_guard); > - } else { > + else > cl_env_put(env, &refcheck); > - } > - cl_env_reexit(cookie); > } > } > > diff --git a/drivers/staging/lustre/lustre/llite/lcommon_misc.c b/drivers/staging/lustre/lustre/llite/lcommon_misc.c > index 1558b55..f48660e 100644 > --- a/drivers/staging/lustre/lustre/llite/lcommon_misc.c > +++ b/drivers/staging/lustre/lustre/llite/lcommon_misc.c > @@ -162,13 +162,11 @@ int cl_get_grouplock(struct cl_object *obj, unsigned long gid, int nonblock, > return rc; > } > > - cg->lg_env = cl_env_get(&refcheck); > + cg->lg_env = env; > cg->lg_io = io; > cg->lg_lock = lock; > cg->lg_gid = gid; > - LASSERT(cg->lg_env == env); > > - cl_env_unplant(env, &refcheck); > return 0; > } > > @@ -177,14 +175,10 @@ void cl_put_grouplock(struct ll_grouplock *cg) > struct lu_env *env = cg->lg_env; > struct cl_io *io = cg->lg_io; > struct cl_lock *lock = cg->lg_lock; > - int refcheck; > > LASSERT(cg->lg_env); > LASSERT(cg->lg_gid); > > - cl_env_implant(env, &refcheck); > - cl_env_put(env, &refcheck); > - > cl_lock_release(env, lock); > cl_io_fini(env, io); > cl_env_put(env, NULL); > diff --git a/drivers/staging/lustre/lustre/llite/llite_mmap.c b/drivers/staging/lustre/lustre/llite/llite_mmap.c > index 4366918..ad2a699 100644 > --- a/drivers/staging/lustre/lustre/llite/llite_mmap.c > +++ b/drivers/staging/lustre/lustre/llite/llite_mmap.c > @@ -80,43 +80,24 @@ struct vm_area_struct *our_vma(struct mm_struct *mm, unsigned long addr, > * API independent part for page fault initialization. > * \param vma - virtual memory area addressed to page fault > * \param env - corespondent lu_env to processing > - * \param nest - nested level > * \param index - page index corespondent to fault. > * \parm ra_flags - vma readahead flags. > * > - * \return allocated and initialized env for fault operation. > - * \retval EINVAL if env can't allocated > - * \return other error codes from cl_io_init. > + * \return error codes from cl_io_init. > */ > static struct cl_io * > -ll_fault_io_init(struct vm_area_struct *vma, struct lu_env **env_ret, > - struct cl_env_nest *nest, pgoff_t index, > - unsigned long *ra_flags) > +ll_fault_io_init(struct lu_env *env, struct vm_area_struct *vma, > + pgoff_t index, unsigned long *ra_flags) > { > struct file *file = vma->vm_file; > struct inode *inode = file_inode(file); > struct cl_io *io; > struct cl_fault_io *fio; > - struct lu_env *env; > int rc; > > - *env_ret = NULL; > if (ll_file_nolock(file)) > return ERR_PTR(-EOPNOTSUPP); > > - /* > - * page fault can be called when lustre IO is > - * already active for the current thread, e.g., when doing read/write > - * against user level buffer mapped from Lustre buffer. To avoid > - * stomping on existing context, optionally force an allocation of a new > - * one. > - */ > - env = cl_env_nested_get(nest); > - if (IS_ERR(env)) > - return ERR_PTR(-EINVAL); > - > - *env_ret = env; > - > restart: > io = vvp_env_thread_io(env); > io->ci_obj = ll_i2info(inode)->lli_clob; > @@ -155,7 +136,6 @@ ll_fault_io_init(struct vm_area_struct *vma, struct lu_env **env_ret, > if (io->ci_need_restart) > goto restart; > > - cl_env_nested_put(nest, env); > io = ERR_PTR(rc); > } > > @@ -169,13 +149,17 @@ static int ll_page_mkwrite0(struct vm_area_struct *vma, struct page *vmpage, > struct lu_env *env; > struct cl_io *io; > struct vvp_io *vio; > - struct cl_env_nest nest; > int result; > + int refcheck; > sigset_t set; > struct inode *inode; > struct ll_inode_info *lli; > > - io = ll_fault_io_init(vma, &env, &nest, vmpage->index, NULL); > + env = cl_env_get(&refcheck); > + if (IS_ERR(env)) > + return PTR_ERR(env); > + > + io = ll_fault_io_init(env, vma, vmpage->index, NULL); > if (IS_ERR(io)) { > result = PTR_ERR(io); > goto out; > @@ -240,8 +224,8 @@ static int ll_page_mkwrite0(struct vm_area_struct *vma, struct page *vmpage, > > out_io: > cl_io_fini(env, io); > - cl_env_nested_put(&nest, env); > out: > + cl_env_put(env, &refcheck); > CDEBUG(D_MMAP, "%s mkwrite with %d\n", current->comm, result); > LASSERT(ergo(result == 0, PageLocked(vmpage))); > > @@ -285,13 +269,19 @@ static int ll_fault0(struct vm_area_struct *vma, struct vm_fault *vmf) > struct vvp_io *vio = NULL; > struct page *vmpage; > unsigned long ra_flags; > - struct cl_env_nest nest; > - int result; > + int result = 0; > int fault_ret = 0; > + int refcheck; > > - io = ll_fault_io_init(vma, &env, &nest, vmf->pgoff, &ra_flags); > - if (IS_ERR(io)) > - return to_fault_error(PTR_ERR(io)); > + env = cl_env_get(&refcheck); > + if (IS_ERR(env)) > + return PTR_ERR(env); > + > + io = ll_fault_io_init(env, vma, vmf->pgoff, &ra_flags); > + if (IS_ERR(io)) { > + result = to_fault_error(PTR_ERR(io)); > + goto out; > + } > > result = io->ci_result; > if (result == 0) { > @@ -322,14 +312,15 @@ static int ll_fault0(struct vm_area_struct *vma, struct vm_fault *vmf) > } > } > cl_io_fini(env, io); > - cl_env_nested_put(&nest, env); > > vma->vm_flags |= ra_flags; > + > +out: > + cl_env_put(env, &refcheck); > if (result != 0 && !(fault_ret & VM_FAULT_RETRY)) > fault_ret |= to_fault_error(result); > > - CDEBUG(D_MMAP, "%s fault %d/%d\n", > - current->comm, fault_ret, result); > + CDEBUG(D_MMAP, "%s fault %d/%d\n", current->comm, fault_ret, result); > return fault_ret; > } > > diff --git a/drivers/staging/lustre/lustre/llite/rw.c b/drivers/staging/lustre/lustre/llite/rw.c > index 80cb8e0..d2515a8 100644 > --- a/drivers/staging/lustre/lustre/llite/rw.c > +++ b/drivers/staging/lustre/lustre/llite/rw.c > @@ -896,17 +896,17 @@ int ll_writepage(struct page *vmpage, struct writeback_control *wbc) > struct cl_io *io; > struct cl_page *page; > struct cl_object *clob; > - struct cl_env_nest nest; > bool redirtied = false; > bool unlocked = false; > int result; > + int refcheck; > > LASSERT(PageLocked(vmpage)); > LASSERT(!PageWriteback(vmpage)); > > LASSERT(ll_i2dtexp(inode)); > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) { > result = PTR_ERR(env); > goto out; > @@ -971,7 +971,7 @@ int ll_writepage(struct page *vmpage, struct writeback_control *wbc) > } > } > > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > goto out; > > out: > diff --git a/drivers/staging/lustre/lustre/llite/rw26.c b/drivers/staging/lustre/lustre/llite/rw26.c > index 67010be..1a08a9d 100644 > --- a/drivers/staging/lustre/lustre/llite/rw26.c > +++ b/drivers/staging/lustre/lustre/llite/rw26.c > @@ -103,7 +103,6 @@ static void ll_invalidatepage(struct page *vmpage, unsigned int offset, > static int ll_releasepage(struct page *vmpage, gfp_t gfp_mask) > { > struct lu_env *env; > - void *cookie; > struct cl_object *obj; > struct cl_page *page; > struct address_space *mapping; > @@ -129,7 +128,6 @@ static int ll_releasepage(struct page *vmpage, gfp_t gfp_mask) > if (!page) > return 1; > > - cookie = cl_env_reenter(); > env = cl_env_percpu_get(); > LASSERT(!IS_ERR(env)); > > @@ -155,7 +153,6 @@ static int ll_releasepage(struct page *vmpage, gfp_t gfp_mask) > cl_page_put(env, page); > > cl_env_percpu_put(env); > - cl_env_reexit(cookie); > return result; > } > > @@ -340,7 +337,8 @@ static ssize_t ll_direct_IO_26_seg(const struct lu_env *env, struct cl_io *io, > PAGE_SIZE) & ~(DT_MAX_BRW_SIZE - 1)) > static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) > { > - struct lu_env *env; > + struct ll_cl_context *lcc; > + const struct lu_env *env; > struct cl_io *io; > struct file *file = iocb->ki_filp; > struct inode *inode = file->f_mapping->host; > @@ -348,7 +346,6 @@ static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) > ssize_t count = iov_iter_count(iter); > ssize_t tot_bytes = 0, result = 0; > long size = MAX_DIO_SIZE; > - int refcheck; > > /* FIXME: io smaller than PAGE_SIZE is broken on ia64 ??? */ > if ((file_offset & ~PAGE_MASK) || (count & ~PAGE_MASK)) > @@ -363,9 +360,13 @@ static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) > if (iov_iter_alignment(iter) & ~PAGE_MASK) > return -EINVAL; > > - env = cl_env_get(&refcheck); > + lcc = ll_cl_find(file); > + if (!lcc) > + return -EIO; > + > + env = lcc->lcc_env; > LASSERT(!IS_ERR(env)); > - io = vvp_env_io(env)->vui_cl.cis_io; > + io = lcc->lcc_io; > LASSERT(io); > > while (iov_iter_count(iter)) { > @@ -422,7 +423,6 @@ static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter) > vio->u.write.vui_written += tot_bytes; > } > > - cl_env_put(env, &refcheck); > return tot_bytes ? tot_bytes : result; > } > > diff --git a/drivers/staging/lustre/lustre/lov/lov_io.c b/drivers/staging/lustre/lustre/lov/lov_io.c > index 011ab0f..002326c 100644 > --- a/drivers/staging/lustre/lustre/lov/lov_io.c > +++ b/drivers/staging/lustre/lustre/lov/lov_io.c > @@ -167,12 +167,7 @@ static int lov_io_sub_init(const struct lu_env *env, struct lov_io *lio, > sub->sub_env = ld->ld_emrg[stripe]->emrg_env; > sub->sub_borrowed = 1; > } else { > - void *cookie; > - > - /* obtain new environment */ > - cookie = cl_env_reenter(); > sub->sub_env = cl_env_get(&sub->sub_refcheck); > - cl_env_reexit(cookie); > if (IS_ERR(sub->sub_env)) > result = PTR_ERR(sub->sub_env); > > diff --git a/drivers/staging/lustre/lustre/lov/lov_object.c b/drivers/staging/lustre/lustre/lov/lov_object.c > index 711a3b9..317311c 100644 > --- a/drivers/staging/lustre/lustre/lov/lov_object.c > +++ b/drivers/staging/lustre/lustre/lov/lov_object.c > @@ -729,19 +729,15 @@ static int lov_layout_change(const struct lu_env *unused, > union lov_layout_state *state = &lov->u; > const struct lov_layout_operations *old_ops; > const struct lov_layout_operations *new_ops; > - void *cookie; > struct lu_env *env; > int refcheck; > int rc; > > LASSERT(0 <= lov->lo_type && lov->lo_type < ARRAY_SIZE(lov_dispatch)); > > - cookie = cl_env_reenter(); > env = cl_env_get(&refcheck); > - if (IS_ERR(env)) { > - cl_env_reexit(cookie); > + if (IS_ERR(env)) > return PTR_ERR(env); > - } > > LASSERT(0 <= llt && llt < ARRAY_SIZE(lov_dispatch)); > > @@ -782,7 +778,6 @@ static int lov_layout_change(const struct lu_env *unused, > lov->lo_type = llt; > out: > cl_env_put(env, &refcheck); > - cl_env_reexit(cookie); > return rc; > } > > diff --git a/drivers/staging/lustre/lustre/obdclass/cl_internal.h b/drivers/staging/lustre/lustre/obdclass/cl_internal.h > index e866754..7b403fb 100644 > --- a/drivers/staging/lustre/lustre/obdclass/cl_internal.h > +++ b/drivers/staging/lustre/lustre/obdclass/cl_internal.h > @@ -50,25 +50,6 @@ enum clt_nesting_level { > }; > > /** > - * Counters used to check correctness of cl_lock interface usage. > - */ > -struct cl_thread_counters { > - /** > - * Number of outstanding calls to cl_lock_mutex_get() made by the > - * current thread. For debugging. > - */ > - int ctc_nr_locks_locked; > - /** List of locked locks. */ > - struct lu_ref ctc_locks_locked; > - /** Number of outstanding holds on locks. */ > - int ctc_nr_held; > - /** Number of outstanding uses on locks. */ > - int ctc_nr_used; > - /** Number of held extent locks. */ > - int ctc_nr_locks_acquired; > -}; > - > -/** > * Thread local state internal for generic cl-code. > */ > struct cl_thread_info { > @@ -83,10 +64,6 @@ struct cl_thread_info { > */ > struct cl_lock_descr clt_descr; > struct cl_page_list clt_list; > - /** > - * Counters for every level of lock nesting. > - */ > - struct cl_thread_counters clt_counters[CNL_NR]; > /** @} debugging */ > > /* > diff --git a/drivers/staging/lustre/lustre/obdclass/cl_io.c b/drivers/staging/lustre/lustre/obdclass/cl_io.c > index c5621ad..af8e4d6 100644 > --- a/drivers/staging/lustre/lustre/obdclass/cl_io.c > +++ b/drivers/staging/lustre/lustre/obdclass/cl_io.c > @@ -412,7 +412,6 @@ void cl_io_unlock(const struct lu_env *env, struct cl_io *io) > scan->cis_iop->op[io->ci_type].cio_unlock(env, scan); > } > io->ci_state = CIS_UNLOCKED; > - LASSERT(!cl_env_info(env)->clt_counters[CNL_TOP].ctc_nr_locks_acquired); > } > EXPORT_SYMBOL(cl_io_unlock); > > diff --git a/drivers/staging/lustre/lustre/obdclass/cl_object.c b/drivers/staging/lustre/lustre/obdclass/cl_object.c > index b4600c4..f5d4e23 100644 > --- a/drivers/staging/lustre/lustre/obdclass/cl_object.c > +++ b/drivers/staging/lustre/lustre/obdclass/cl_object.c > @@ -544,36 +544,20 @@ EXPORT_SYMBOL(cl_site_stats_print); > * bz20044, bz22683. > */ > > -static LIST_HEAD(cl_envs); > -static unsigned int cl_envs_cached_nr; > -static unsigned int cl_envs_cached_max = 128; /* XXX: prototype: arbitrary limit > - * for now. > - */ > -static DEFINE_SPINLOCK(cl_envs_guard); > +static unsigned int cl_envs_cached_max = 32; /* XXX: prototype: arbitrary limit > + * for now. > + */ > +static struct cl_env_cache { > + rwlock_t cec_guard; > + unsigned int cec_count; > + struct list_head cec_envs; > +} *cl_envs = NULL; > > struct cl_env { > void *ce_magic; > struct lu_env ce_lu; > struct lu_context ce_ses; > > - /** > - * This allows cl_env to be entered into cl_env_hash which implements > - * the current thread -> client environment lookup. > - */ > - struct hlist_node ce_node; > - /** > - * Owner for the current cl_env. > - * > - * If LL_TASK_CL_ENV is defined, this point to the owning current, > - * only for debugging purpose ; > - * Otherwise hash is used, and this is the key for cfs_hash. > - * Now current thread pid is stored. Note using thread pointer would > - * lead to unbalanced hash because of its specific allocation locality > - * and could be varied for different platforms and OSes, even different > - * OS versions. > - */ > - void *ce_owner; > - > /* > * Linkage into global list of all client environments. Used for > * garbage collection. > @@ -597,122 +581,13 @@ static void cl_env_init0(struct cl_env *cle, void *debug) > { > LASSERT(cle->ce_ref == 0); > LASSERT(cle->ce_magic == &cl_env_init0); > - LASSERT(!cle->ce_debug && !cle->ce_owner); > + LASSERT(!cle->ce_debug); > > cle->ce_ref = 1; > cle->ce_debug = debug; > CL_ENV_INC(busy); > } > > -/* > - * The implementation of using hash table to connect cl_env and thread > - */ > - > -static struct cfs_hash *cl_env_hash; > - > -static unsigned cl_env_hops_hash(struct cfs_hash *lh, > - const void *key, unsigned mask) > -{ > -#if BITS_PER_LONG == 64 > - return cfs_hash_u64_hash((__u64)key, mask); > -#else > - return cfs_hash_u32_hash((__u32)key, mask); > -#endif > -} > - > -static void *cl_env_hops_obj(struct hlist_node *hn) > -{ > - struct cl_env *cle = hlist_entry(hn, struct cl_env, ce_node); > - > - LASSERT(cle->ce_magic == &cl_env_init0); > - return (void *)cle; > -} > - > -static int cl_env_hops_keycmp(const void *key, struct hlist_node *hn) > -{ > - struct cl_env *cle = cl_env_hops_obj(hn); > - > - LASSERT(cle->ce_owner); > - return (key == cle->ce_owner); > -} > - > -static void cl_env_hops_noop(struct cfs_hash *hs, struct hlist_node *hn) > -{ > - struct cl_env *cle = hlist_entry(hn, struct cl_env, ce_node); > - > - LASSERT(cle->ce_magic == &cl_env_init0); > -} > - > -static struct cfs_hash_ops cl_env_hops = { > - .hs_hash = cl_env_hops_hash, > - .hs_key = cl_env_hops_obj, > - .hs_keycmp = cl_env_hops_keycmp, > - .hs_object = cl_env_hops_obj, > - .hs_get = cl_env_hops_noop, > - .hs_put_locked = cl_env_hops_noop, > -}; > - > -static inline struct cl_env *cl_env_fetch(void) > -{ > - struct cl_env *cle; > - > - cle = cfs_hash_lookup(cl_env_hash, (void *)(long)current->pid); > - LASSERT(ergo(cle, cle->ce_magic == &cl_env_init0)); > - return cle; > -} > - > -static inline void cl_env_attach(struct cl_env *cle) > -{ > - if (cle) { > - int rc; > - > - LASSERT(!cle->ce_owner); > - cle->ce_owner = (void *)(long)current->pid; > - rc = cfs_hash_add_unique(cl_env_hash, cle->ce_owner, > - &cle->ce_node); > - LASSERT(rc == 0); > - } > -} > - > -static inline void cl_env_do_detach(struct cl_env *cle) > -{ > - void *cookie; > - > - LASSERT(cle->ce_owner == (void *)(long)current->pid); > - cookie = cfs_hash_del(cl_env_hash, cle->ce_owner, > - &cle->ce_node); > - LASSERT(cookie == cle); > - cle->ce_owner = NULL; > -} > - > -static int cl_env_store_init(void) > -{ > - cl_env_hash = cfs_hash_create("cl_env", > - HASH_CL_ENV_BITS, HASH_CL_ENV_BITS, > - HASH_CL_ENV_BKT_BITS, 0, > - CFS_HASH_MIN_THETA, > - CFS_HASH_MAX_THETA, > - &cl_env_hops, > - CFS_HASH_RW_BKTLOCK); > - return cl_env_hash ? 0 : -ENOMEM; > -} > - > -static void cl_env_store_fini(void) > -{ > - cfs_hash_putref(cl_env_hash); > -} > - > -static inline struct cl_env *cl_env_detach(struct cl_env *cle) > -{ > - if (!cle) > - cle = cl_env_fetch(); > - > - if (cle && cle->ce_owner) > - cl_env_do_detach(cle); > - > - return cle; > -} > - > static struct lu_env *cl_env_new(__u32 ctx_tags, __u32 ses_tags, void *debug) > { > struct lu_env *env; > @@ -762,16 +637,20 @@ static struct lu_env *cl_env_obtain(void *debug) > { > struct cl_env *cle; > struct lu_env *env; > + int cpu = get_cpu(); > > - spin_lock(&cl_envs_guard); > - LASSERT(equi(cl_envs_cached_nr == 0, list_empty(&cl_envs))); > - if (cl_envs_cached_nr > 0) { > + read_lock(&cl_envs[cpu].cec_guard); > + LASSERT(equi(cl_envs[cpu].cec_count == 0, > + list_empty(&cl_envs[cpu].cec_envs))); > + if (cl_envs[cpu].cec_count > 0) { > int rc; > > - cle = container_of(cl_envs.next, struct cl_env, ce_linkage); > + cle = container_of(cl_envs[cpu].cec_envs.next, struct cl_env, > + ce_linkage); > list_del_init(&cle->ce_linkage); > - cl_envs_cached_nr--; > - spin_unlock(&cl_envs_guard); > + cl_envs[cpu].cec_count--; > + read_unlock(&cl_envs[cpu].cec_guard); > + put_cpu(); > > env = &cle->ce_lu; > rc = lu_env_refill(env); > @@ -784,7 +663,8 @@ static struct lu_env *cl_env_obtain(void *debug) > env = ERR_PTR(rc); > } > } else { > - spin_unlock(&cl_envs_guard); > + read_unlock(&cl_envs[cpu].cec_guard); > + put_cpu(); > env = cl_env_new(lu_context_tags_default, > lu_session_tags_default, debug); > } > @@ -796,27 +676,6 @@ static inline struct cl_env *cl_env_container(struct lu_env *env) > return container_of(env, struct cl_env, ce_lu); > } > > -static struct lu_env *cl_env_peek(int *refcheck) > -{ > - struct lu_env *env; > - struct cl_env *cle; > - > - CL_ENV_INC(lookup); > - > - /* check that we don't go far from untrusted pointer */ > - CLASSERT(offsetof(struct cl_env, ce_magic) == 0); > - > - env = NULL; > - cle = cl_env_fetch(); > - if (cle) { > - CL_ENV_INC(hit); > - env = &cle->ce_lu; > - *refcheck = ++cle->ce_ref; > - } > - CDEBUG(D_OTHER, "%d@%p\n", cle ? cle->ce_ref : 0, cle); > - return env; > -} > - > /** > * Returns lu_env: if there already is an environment associated with the > * current thread, it is returned, otherwise, new environment is allocated. > @@ -834,17 +693,13 @@ struct lu_env *cl_env_get(int *refcheck) > { > struct lu_env *env; > > - env = cl_env_peek(refcheck); > - if (!env) { > - env = cl_env_obtain(__builtin_return_address(0)); > - if (!IS_ERR(env)) { > - struct cl_env *cle; > + env = cl_env_obtain(__builtin_return_address(0)); > + if (!IS_ERR(env)) { > + struct cl_env *cle; > > - cle = cl_env_container(env); > - cl_env_attach(cle); > - *refcheck = cle->ce_ref; > - CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); > - } > + cle = cl_env_container(env); > + *refcheck = cle->ce_ref; > + CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); > } > return env; > } > @@ -859,7 +714,6 @@ struct lu_env *cl_env_alloc(int *refcheck, __u32 tags) > { > struct lu_env *env; > > - LASSERT(!cl_env_peek(refcheck)); > env = cl_env_new(tags, tags, __builtin_return_address(0)); > if (!IS_ERR(env)) { > struct cl_env *cle; > @@ -874,7 +728,6 @@ EXPORT_SYMBOL(cl_env_alloc); > > static void cl_env_exit(struct cl_env *cle) > { > - LASSERT(!cle->ce_owner); > lu_context_exit(&cle->ce_lu.le_ctx); > lu_context_exit(&cle->ce_ses); > } > @@ -887,20 +740,25 @@ static void cl_env_exit(struct cl_env *cle) > unsigned int cl_env_cache_purge(unsigned int nr) > { > struct cl_env *cle; > + unsigned int i; > > - spin_lock(&cl_envs_guard); > - for (; !list_empty(&cl_envs) && nr > 0; --nr) { > - cle = container_of(cl_envs.next, struct cl_env, ce_linkage); > - list_del_init(&cle->ce_linkage); > - LASSERT(cl_envs_cached_nr > 0); > - cl_envs_cached_nr--; > - spin_unlock(&cl_envs_guard); > + for_each_possible_cpu(i) { > + write_lock(&cl_envs[i].cec_guard); > + for (; !list_empty(&cl_envs[i].cec_envs) && nr > 0; --nr) { > + cle = container_of(cl_envs[i].cec_envs.next, > + struct cl_env, ce_linkage); > + list_del_init(&cle->ce_linkage); > + LASSERT(cl_envs[i].cec_count > 0); > + cl_envs[i].cec_count--; > + write_unlock(&cl_envs[i].cec_guard); > > - cl_env_fini(cle); > - spin_lock(&cl_envs_guard); > + cl_env_fini(cle); > + write_lock(&cl_envs[i].cec_guard); > + } > + LASSERT(equi(cl_envs[i].cec_count == 0, > + list_empty(&cl_envs[i].cec_envs))); > + write_unlock(&cl_envs[i].cec_guard); > } > - LASSERT(equi(cl_envs_cached_nr == 0, list_empty(&cl_envs))); > - spin_unlock(&cl_envs_guard); > return nr; > } > EXPORT_SYMBOL(cl_env_cache_purge); > @@ -923,8 +781,9 @@ void cl_env_put(struct lu_env *env, int *refcheck) > > CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); > if (--cle->ce_ref == 0) { > + int cpu = get_cpu(); > + > CL_ENV_DEC(busy); > - cl_env_detach(cle); > cle->ce_debug = NULL; > cl_env_exit(cle); > /* > @@ -933,107 +792,22 @@ void cl_env_put(struct lu_env *env, int *refcheck) > * Return environment to the cache only when it was allocated > * with the standard tags. > */ > - if (cl_envs_cached_nr < cl_envs_cached_max && > + if (cl_envs[cpu].cec_count < cl_envs_cached_max && > (env->le_ctx.lc_tags & ~LCT_HAS_EXIT) == LCT_CL_THREAD && > (env->le_ses->lc_tags & ~LCT_HAS_EXIT) == LCT_SESSION) { > - spin_lock(&cl_envs_guard); > - list_add(&cle->ce_linkage, &cl_envs); > - cl_envs_cached_nr++; > - spin_unlock(&cl_envs_guard); > + read_lock(&cl_envs[cpu].cec_guard); > + list_add(&cle->ce_linkage, &cl_envs[cpu].cec_envs); > + cl_envs[cpu].cec_count++; > + read_unlock(&cl_envs[cpu].cec_guard); > } else { > cl_env_fini(cle); > } > + put_cpu(); > } > } > EXPORT_SYMBOL(cl_env_put); > > /** > - * Declares a point of re-entrancy. > - * > - * \see cl_env_reexit() > - */ > -void *cl_env_reenter(void) > -{ > - return cl_env_detach(NULL); > -} > -EXPORT_SYMBOL(cl_env_reenter); > - > -/** > - * Exits re-entrancy. > - */ > -void cl_env_reexit(void *cookie) > -{ > - cl_env_detach(NULL); > - cl_env_attach(cookie); > -} > -EXPORT_SYMBOL(cl_env_reexit); > - > -/** > - * Setup user-supplied \a env as a current environment. This is to be used to > - * guaranteed that environment exists even when cl_env_get() fails. It is up > - * to user to ensure proper concurrency control. > - * > - * \see cl_env_unplant() > - */ > -void cl_env_implant(struct lu_env *env, int *refcheck) > -{ > - struct cl_env *cle = cl_env_container(env); > - > - LASSERT(cle->ce_ref > 0); > - > - cl_env_attach(cle); > - cl_env_get(refcheck); > - CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); > -} > -EXPORT_SYMBOL(cl_env_implant); > - > -/** > - * Detach environment installed earlier by cl_env_implant(). > - */ > -void cl_env_unplant(struct lu_env *env, int *refcheck) > -{ > - struct cl_env *cle = cl_env_container(env); > - > - LASSERT(cle->ce_ref > 1); > - > - CDEBUG(D_OTHER, "%d@%p\n", cle->ce_ref, cle); > - > - cl_env_detach(cle); > - cl_env_put(env, refcheck); > -} > -EXPORT_SYMBOL(cl_env_unplant); > - > -struct lu_env *cl_env_nested_get(struct cl_env_nest *nest) > -{ > - struct lu_env *env; > - > - nest->cen_cookie = NULL; > - env = cl_env_peek(&nest->cen_refcheck); > - if (env) { > - if (!cl_io_is_going(env)) > - return env; > - cl_env_put(env, &nest->cen_refcheck); > - nest->cen_cookie = cl_env_reenter(); > - } > - env = cl_env_get(&nest->cen_refcheck); > - if (IS_ERR(env)) { > - cl_env_reexit(nest->cen_cookie); > - return env; > - } > - > - LASSERT(!cl_io_is_going(env)); > - return env; > -} > -EXPORT_SYMBOL(cl_env_nested_get); > - > -void cl_env_nested_put(struct cl_env_nest *nest, struct lu_env *env) > -{ > - cl_env_put(env, &nest->cen_refcheck); > - cl_env_reexit(nest->cen_cookie); > -} > -EXPORT_SYMBOL(cl_env_nested_put); > - > -/** > * Converts struct ost_lvb to struct cl_attr. > * > * \see cl_attr2lvb > @@ -1060,6 +834,10 @@ static int cl_env_percpu_init(void) > for_each_possible_cpu(i) { > struct lu_env *env; > > + rwlock_init(&cl_envs[i].cec_guard); > + INIT_LIST_HEAD(&cl_envs[i].cec_envs); > + cl_envs[i].cec_count = 0; > + > cle = &cl_env_percpu[i]; > env = &cle->ce_lu; > > @@ -1127,7 +905,6 @@ void cl_env_percpu_put(struct lu_env *env) > LASSERT(cle->ce_ref == 0); > > CL_ENV_DEC(busy); > - cl_env_detach(cle); > cle->ce_debug = NULL; > > put_cpu(); > @@ -1141,7 +918,6 @@ struct lu_env *cl_env_percpu_get(void) > cle = &cl_env_percpu[get_cpu()]; > cl_env_init0(cle, __builtin_return_address(0)); > > - cl_env_attach(cle); > return &cle->ce_lu; > } > EXPORT_SYMBOL(cl_env_percpu_get); > @@ -1205,51 +981,19 @@ LU_KEY_INIT_FINI(cl0, struct cl_thread_info); > static void *cl_key_init(const struct lu_context *ctx, > struct lu_context_key *key) > { > - struct cl_thread_info *info; > - > - info = cl0_key_init(ctx, key); > - if (!IS_ERR(info)) { > - size_t i; > - > - for (i = 0; i < ARRAY_SIZE(info->clt_counters); ++i) > - lu_ref_init(&info->clt_counters[i].ctc_locks_locked); > - } > - return info; > + return cl0_key_init(ctx, key); > } > > static void cl_key_fini(const struct lu_context *ctx, > struct lu_context_key *key, void *data) > { > - struct cl_thread_info *info; > - size_t i; > - > - info = data; > - for (i = 0; i < ARRAY_SIZE(info->clt_counters); ++i) > - lu_ref_fini(&info->clt_counters[i].ctc_locks_locked); > cl0_key_fini(ctx, key, data); > } > > -static void cl_key_exit(const struct lu_context *ctx, > - struct lu_context_key *key, void *data) > -{ > - struct cl_thread_info *info = data; > - size_t i; > - > - for (i = 0; i < ARRAY_SIZE(info->clt_counters); ++i) { > - LASSERT(info->clt_counters[i].ctc_nr_held == 0); > - LASSERT(info->clt_counters[i].ctc_nr_used == 0); > - LASSERT(info->clt_counters[i].ctc_nr_locks_acquired == 0); > - LASSERT(info->clt_counters[i].ctc_nr_locks_locked == 0); > - lu_ref_fini(&info->clt_counters[i].ctc_locks_locked); > - lu_ref_init(&info->clt_counters[i].ctc_locks_locked); > - } > -} > - > static struct lu_context_key cl_key = { > .lct_tags = LCT_CL_THREAD, > .lct_init = cl_key_init, > .lct_fini = cl_key_fini, > - .lct_exit = cl_key_exit > }; > > static struct lu_kmem_descr cl_object_caches[] = { > @@ -1273,13 +1017,15 @@ int cl_global_init(void) > { > int result; > > - result = cl_env_store_init(); > - if (result) > - return result; > + cl_envs = kzalloc(sizeof(*cl_envs) * num_possible_cpus(), GFP_KERNEL); > + if (!cl_envs) { > + result = -ENOMEM; > + goto out; > + } > > result = lu_kmem_init(cl_object_caches); > if (result) > - goto out_store; > + goto out_envs; > > LU_CONTEXT_KEY_INIT(&cl_key); > result = lu_context_key_register(&cl_key); > @@ -1289,16 +1035,17 @@ int cl_global_init(void) > result = cl_env_percpu_init(); > if (result) > /* no cl_env_percpu_fini on error */ > - goto out_context; > + goto out_keys; > > return 0; > > -out_context: > +out_keys: > lu_context_key_degister(&cl_key); > out_kmem: > lu_kmem_fini(cl_object_caches); > -out_store: > - cl_env_store_fini(); > +out_envs: > + kfree(cl_envs); > +out: > return result; > } > > @@ -1310,5 +1057,5 @@ void cl_global_fini(void) > cl_env_percpu_fini(); > lu_context_key_degister(&cl_key); > lu_kmem_fini(cl_object_caches); > - cl_env_store_fini(); > + kfree(cl_envs); > } > diff --git a/drivers/staging/lustre/lustre/obdclass/lu_object.c b/drivers/staging/lustre/lustre/obdclass/lu_object.c > index e031fd2..14ca82d 100644 > --- a/drivers/staging/lustre/lustre/obdclass/lu_object.c > +++ b/drivers/staging/lustre/lustre/obdclass/lu_object.c > @@ -1521,10 +1521,6 @@ void lu_context_key_quiesce(struct lu_context_key *key) > > if (!(key->lct_tags & LCT_QUIESCENT)) { > /* > - * XXX layering violation. > - */ > - cl_env_cache_purge(~0); > - /* > * XXX memory barrier has to go here. > */ > spin_lock(&lu_keys_guard); > diff --git a/drivers/staging/lustre/lustre/obdecho/echo_client.c b/drivers/staging/lustre/lustre/obdecho/echo_client.c > index 30caf5f..25ecfb3 100644 > --- a/drivers/staging/lustre/lustre/obdecho/echo_client.c > +++ b/drivers/staging/lustre/lustre/obdecho/echo_client.c > @@ -563,16 +563,10 @@ static void echo_thread_key_fini(const struct lu_context *ctx, > kmem_cache_free(echo_thread_kmem, info); > } > > -static void echo_thread_key_exit(const struct lu_context *ctx, > - struct lu_context_key *key, void *data) > -{ > -} > - > static struct lu_context_key echo_thread_key = { > .lct_tags = LCT_CL_THREAD, > .lct_init = echo_thread_key_init, > .lct_fini = echo_thread_key_fini, > - .lct_exit = echo_thread_key_exit > }; > > static void *echo_session_key_init(const struct lu_context *ctx, > @@ -594,16 +588,10 @@ static void echo_session_key_fini(const struct lu_context *ctx, > kmem_cache_free(echo_session_kmem, session); > } > > -static void echo_session_key_exit(const struct lu_context *ctx, > - struct lu_context_key *key, void *data) > -{ > -} > - > static struct lu_context_key echo_session_key = { > .lct_tags = LCT_SESSION, > .lct_init = echo_session_key_init, > .lct_fini = echo_session_key_fini, > - .lct_exit = echo_session_key_exit > }; > > LU_TYPE_INIT_FINI(echo, &echo_thread_key, &echo_session_key); > @@ -787,6 +775,8 @@ static struct lu_device *echo_device_free(const struct lu_env *env, > cl_device_fini(&ed->ed_cl); > kfree(ed); > > + cl_env_cache_purge(~0); > + > return NULL; > } > > diff --git a/drivers/staging/lustre/lustre/osc/osc_cache.c b/drivers/staging/lustre/lustre/osc/osc_cache.c > index b645957..cbbe85e 100644 > --- a/drivers/staging/lustre/lustre/osc/osc_cache.c > +++ b/drivers/staging/lustre/lustre/osc/osc_cache.c > @@ -977,7 +977,6 @@ static int osc_extent_wait(const struct lu_env *env, struct osc_extent *ext, > static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, > bool partial) > { > - struct cl_env_nest nest; > struct lu_env *env; > struct cl_io *io; > struct osc_object *obj = ext->oe_obj; > @@ -990,6 +989,7 @@ static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, > int grants = 0; > int nr_pages = 0; > int rc = 0; > + int refcheck; > > LASSERT(sanity_check(ext) == 0); > EASSERT(ext->oe_state == OES_TRUNC, ext); > @@ -999,7 +999,7 @@ static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, > * We can't use that env from osc_cache_truncate_start() because > * it's from lov_io_sub and not fully initialized. > */ > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > io = &osc_env_info(env)->oti_io; > io->ci_obj = cl_object_top(osc2cl(obj)); > rc = cl_io_init(env, io, CIT_MISC, io->ci_obj); > @@ -1085,7 +1085,7 @@ static int osc_extent_truncate(struct osc_extent *ext, pgoff_t trunc_index, > > out: > cl_io_fini(env, io); > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > return rc; > } > > diff --git a/drivers/staging/lustre/lustre/osc/osc_lock.c b/drivers/staging/lustre/lustre/osc/osc_lock.c > index 45d5e6d..e337e87 100644 > --- a/drivers/staging/lustre/lustre/osc/osc_lock.c > +++ b/drivers/staging/lustre/lustre/osc/osc_lock.c > @@ -294,10 +294,10 @@ static int osc_lock_upcall(void *cookie, struct lustre_handle *lockh, > struct osc_lock *oscl = cookie; > struct cl_lock_slice *slice = &oscl->ols_cl; > struct lu_env *env; > - struct cl_env_nest nest; > int rc; > + int refcheck; > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > /* should never happen, similar to osc_ldlm_blocking_ast(). */ > LASSERT(!IS_ERR(env)); > > @@ -336,7 +336,7 @@ static int osc_lock_upcall(void *cookie, struct lustre_handle *lockh, > > if (oscl->ols_owner) > cl_sync_io_note(env, oscl->ols_owner, rc); > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > > return rc; > } > @@ -347,9 +347,9 @@ static int osc_lock_upcall_agl(void *cookie, struct lustre_handle *lockh, > struct osc_object *osc = cookie; > struct ldlm_lock *dlmlock; > struct lu_env *env; > - struct cl_env_nest nest; > + int refcheck; > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > LASSERT(!IS_ERR(env)); > > if (errcode == ELDLM_LOCK_MATCHED) { > @@ -374,7 +374,7 @@ static int osc_lock_upcall_agl(void *cookie, struct lustre_handle *lockh, > > out: > cl_object_put(env, osc2cl(osc)); > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > return ldlm_error2errno(errcode); > } > > @@ -382,11 +382,11 @@ static int osc_lock_flush(struct osc_object *obj, pgoff_t start, pgoff_t end, > enum cl_lock_mode mode, int discard) > { > struct lu_env *env; > - struct cl_env_nest nest; > + int refcheck; > int rc = 0; > int rc2 = 0; > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) > return PTR_ERR(env); > > @@ -404,7 +404,7 @@ static int osc_lock_flush(struct osc_object *obj, pgoff_t start, pgoff_t end, > if (rc == 0 && rc2 < 0) > rc = rc2; > > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > return rc; > } > > @@ -536,7 +536,7 @@ static int osc_ldlm_blocking_ast(struct ldlm_lock *dlmlock, > } > case LDLM_CB_CANCELING: { > struct lu_env *env; > - struct cl_env_nest nest; > + int refcheck; > > /* > * This can be called in the context of outer IO, e.g., > @@ -549,14 +549,14 @@ static int osc_ldlm_blocking_ast(struct ldlm_lock *dlmlock, > * new environment has to be created to not corrupt outer > * context. > */ > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) { > result = PTR_ERR(env); > break; > } > > result = osc_dlm_blocking_ast0(env, dlmlock, data, flag); > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > break; > } > default: > @@ -568,61 +568,61 @@ static int osc_ldlm_blocking_ast(struct ldlm_lock *dlmlock, > static int osc_ldlm_glimpse_ast(struct ldlm_lock *dlmlock, void *data) > { > struct ptlrpc_request *req = data; > - struct cl_env_nest nest; > struct lu_env *env; > struct ost_lvb *lvb; > struct req_capsule *cap; > + struct cl_object *obj = NULL; > int result; > + int refcheck; > > LASSERT(lustre_msg_get_opc(req->rq_reqmsg) == LDLM_GL_CALLBACK); > > - env = cl_env_nested_get(&nest); > - if (!IS_ERR(env)) { > - struct cl_object *obj = NULL; > + env = cl_env_get(&refcheck); > + if (IS_ERR(env)) { > + result = PTR_ERR(env); > + goto out; > + } > > - lock_res_and_lock(dlmlock); > - if (dlmlock->l_ast_data) { > - obj = osc2cl(dlmlock->l_ast_data); > - cl_object_get(obj); > - } > - unlock_res_and_lock(dlmlock); > + lock_res_and_lock(dlmlock); > + if (dlmlock->l_ast_data) { > + obj = osc2cl(dlmlock->l_ast_data); > + cl_object_get(obj); > + } > + unlock_res_and_lock(dlmlock); > > - if (obj) { > - /* Do not grab the mutex of cl_lock for glimpse. > - * See LU-1274 for details. > - * BTW, it's okay for cl_lock to be cancelled during > - * this period because server can handle this race. > - * See ldlm_server_glimpse_ast() for details. > - * cl_lock_mutex_get(env, lock); > - */ > - cap = &req->rq_pill; > - req_capsule_extend(cap, &RQF_LDLM_GL_CALLBACK); > - req_capsule_set_size(cap, &RMF_DLM_LVB, RCL_SERVER, > - sizeof(*lvb)); > - result = req_capsule_server_pack(cap); > - if (result == 0) { > - lvb = req_capsule_server_get(cap, &RMF_DLM_LVB); > - result = cl_object_glimpse(env, obj, lvb); > - } > - if (!exp_connect_lvb_type(req->rq_export)) > - req_capsule_shrink(&req->rq_pill, > - &RMF_DLM_LVB, > - sizeof(struct ost_lvb_v1), > - RCL_SERVER); > - cl_object_put(env, obj); > - } else { > - /* > - * These errors are normal races, so we don't want to > - * fill the console with messages by calling > - * ptlrpc_error() > - */ > - lustre_pack_reply(req, 1, NULL, NULL); > - result = -ELDLM_NO_LOCK_DATA; > + if (obj) { > + /* Do not grab the mutex of cl_lock for glimpse. > + * See LU-1274 for details. > + * BTW, it's okay for cl_lock to be cancelled during > + * this period because server can handle this race. > + * See ldlm_server_glimpse_ast() for details. > + * cl_lock_mutex_get(env, lock); > + */ > + cap = &req->rq_pill; > + req_capsule_extend(cap, &RQF_LDLM_GL_CALLBACK); > + req_capsule_set_size(cap, &RMF_DLM_LVB, RCL_SERVER, > + sizeof(*lvb)); > + result = req_capsule_server_pack(cap); > + if (result == 0) { > + lvb = req_capsule_server_get(cap, &RMF_DLM_LVB); > + result = cl_object_glimpse(env, obj, lvb); > } > - cl_env_nested_put(&nest, env); > + if (!exp_connect_lvb_type(req->rq_export)) > + req_capsule_shrink(&req->rq_pill, &RMF_DLM_LVB, > + sizeof(struct ost_lvb_v1), RCL_SERVER); > + cl_object_put(env, obj); > } else { > - result = PTR_ERR(env); > + /* > + * These errors are normal races, so we don't want to > + * fill the console with messages by calling > + * ptlrpc_error() > + */ > + lustre_pack_reply(req, 1, NULL, NULL); > + result = -ELDLM_NO_LOCK_DATA; > } > + cl_env_put(env, &refcheck); > + > +out: > req->rq_status = result; > return result; > } > @@ -677,12 +677,12 @@ static unsigned long osc_lock_weight(const struct lu_env *env, > */ > unsigned long osc_ldlm_weigh_ast(struct ldlm_lock *dlmlock) > { > - struct cl_env_nest nest; > struct lu_env *env; > struct osc_object *obj; > struct osc_lock *oscl; > unsigned long weight; > bool found = false; > + int refcheck; > > might_sleep(); > /* > @@ -692,7 +692,7 @@ unsigned long osc_ldlm_weigh_ast(struct ldlm_lock *dlmlock) > * the upper context because cl_lock_put don't modify environment > * variables. But just in case .. > */ > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) > /* Mostly because lack of memory, do not eliminate this lock */ > return 1; > @@ -722,7 +722,7 @@ unsigned long osc_ldlm_weigh_ast(struct ldlm_lock *dlmlock) > weight = osc_lock_weight(env, obj, &dlmlock->l_policy_data.l_extent); > > out: > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > return weight; > } > > diff --git a/drivers/staging/lustre/lustre/osc/osc_page.c b/drivers/staging/lustre/lustre/osc/osc_page.c > index 2fb0e53..93248d1 100644 > --- a/drivers/staging/lustre/lustre/osc/osc_page.c > +++ b/drivers/staging/lustre/lustre/osc/osc_page.c > @@ -644,15 +644,15 @@ long osc_lru_shrink(const struct lu_env *env, struct client_obd *cli, > > long osc_lru_reclaim(struct client_obd *cli) > { > - struct cl_env_nest nest; > struct lu_env *env; > struct cl_client_cache *cache = cli->cl_cache; > int max_scans; > + int refcheck; > long rc = 0; > > LASSERT(cache); > > - env = cl_env_nested_get(&nest); > + env = cl_env_get(&refcheck); > if (IS_ERR(env)) > return 0; > > @@ -704,7 +704,7 @@ long osc_lru_reclaim(struct client_obd *cli) > spin_unlock(&cache->ccc_lru_lock); > > out: > - cl_env_nested_put(&nest, env); > + cl_env_put(env, &refcheck); > CDEBUG(D_CACHE, "%s: cli %p freed %ld pages.\n", > cli->cl_import->imp_obd->obd_name, cli, rc); > return rc; > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 06/14] staging/lustre/llite: drop_caches hangs in cl_inode_fini() 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (4 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 05/14] staging/lustre: Get rid of cl_env hash table Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:54 ` James Simmons 2016-11-03 1:24 ` [PATCH 07/14] staging/lustre/ldlm: Reinstate ldlm_enqueue_pack() Oleg Drokin ` (7 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Andrew Perepechko, Oleg Drokin From: Andrew Perepechko <andrew.perepechko@seagate.com> This patch releases cl_pages on error in ll_write_begin() to avoid memory and object reference leaks. Also, it reuses per-cpu lu_env in ll_invalidatepage() in the same way as done in ll_releasepage(). Signed-off-by: Andrew Perepechko <andrew.perepechko@seagate.com> Seagate-bug-id: MRP-3504 Reviewed-on: http://review.whamcloud.com/22745 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8509 Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by: Bobi Jam <bobijam@hotmail.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/llite/rw26.c | 36 ++++++++++++++++-------------- 1 file changed, 19 insertions(+), 17 deletions(-) diff --git a/drivers/staging/lustre/lustre/llite/rw26.c b/drivers/staging/lustre/lustre/llite/rw26.c index 1a08a9d..ca45b44 100644 --- a/drivers/staging/lustre/lustre/llite/rw26.c +++ b/drivers/staging/lustre/lustre/llite/rw26.c @@ -71,8 +71,6 @@ static void ll_invalidatepage(struct page *vmpage, unsigned int offset, struct cl_page *page; struct cl_object *obj; - int refcheck; - LASSERT(PageLocked(vmpage)); LASSERT(!PageWriteback(vmpage)); @@ -82,21 +80,21 @@ static void ll_invalidatepage(struct page *vmpage, unsigned int offset, * happening with locked page too */ if (offset == 0 && length == PAGE_SIZE) { - env = cl_env_get(&refcheck); - if (!IS_ERR(env)) { - inode = vmpage->mapping->host; - obj = ll_i2info(inode)->lli_clob; - if (obj) { - page = cl_vmpage_page(vmpage, obj); - if (page) { - cl_page_delete(env, page); - cl_page_put(env, page); - } - } else { - LASSERT(vmpage->private == 0); + /* See the comment in ll_releasepage() */ + env = cl_env_percpu_get(); + LASSERT(!IS_ERR(env)); + inode = vmpage->mapping->host; + obj = ll_i2info(inode)->lli_clob; + if (obj) { + page = cl_vmpage_page(vmpage, obj); + if (page) { + cl_page_delete(env, page); + cl_page_put(env, page); } - cl_env_put(env, &refcheck); + } else { + LASSERT(vmpage->private == 0); } + cl_env_percpu_put(env); } } @@ -466,9 +464,9 @@ static int ll_write_begin(struct file *file, struct address_space *mapping, struct page **pagep, void **fsdata) { struct ll_cl_context *lcc; - const struct lu_env *env; + const struct lu_env *env = NULL; struct cl_io *io; - struct cl_page *page; + struct cl_page *page = NULL; struct cl_object *clob = ll_i2info(mapping->host)->lli_clob; pgoff_t index = pos >> PAGE_SHIFT; struct page *vmpage = NULL; @@ -556,6 +554,10 @@ static int ll_write_begin(struct file *file, struct address_space *mapping, unlock_page(vmpage); put_page(vmpage); } + if (!IS_ERR_OR_NULL(page)) { + lu_ref_del(&page->cp_reference, "cl_io", io); + cl_page_put(env, page); + } } else { *pagep = vmpage; *fsdata = lcc; -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 06/14] staging/lustre/llite: drop_caches hangs in cl_inode_fini() 2016-11-03 1:24 ` [PATCH 06/14] staging/lustre/llite: drop_caches hangs in cl_inode_fini() Oleg Drokin @ 2016-11-07 1:54 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:54 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Andrew Perepechko > From: Andrew Perepechko <andrew.perepechko@seagate.com> > > This patch releases cl_pages on error in ll_write_begin() > to avoid memory and object reference leaks. Also, it > reuses per-cpu lu_env in ll_invalidatepage() in the same > way as done in ll_releasepage(). Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Andrew Perepechko <andrew.perepechko@seagate.com> > Seagate-bug-id: MRP-3504 > Reviewed-on: http://review.whamcloud.com/22745 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8509 > Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> > Reviewed-by: Bobi Jam <bobijam@hotmail.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/llite/rw26.c | 36 ++++++++++++++++-------------- > 1 file changed, 19 insertions(+), 17 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/llite/rw26.c b/drivers/staging/lustre/lustre/llite/rw26.c > index 1a08a9d..ca45b44 100644 > --- a/drivers/staging/lustre/lustre/llite/rw26.c > +++ b/drivers/staging/lustre/lustre/llite/rw26.c > @@ -71,8 +71,6 @@ static void ll_invalidatepage(struct page *vmpage, unsigned int offset, > struct cl_page *page; > struct cl_object *obj; > > - int refcheck; > - > LASSERT(PageLocked(vmpage)); > LASSERT(!PageWriteback(vmpage)); > > @@ -82,21 +80,21 @@ static void ll_invalidatepage(struct page *vmpage, unsigned int offset, > * happening with locked page too > */ > if (offset == 0 && length == PAGE_SIZE) { > - env = cl_env_get(&refcheck); > - if (!IS_ERR(env)) { > - inode = vmpage->mapping->host; > - obj = ll_i2info(inode)->lli_clob; > - if (obj) { > - page = cl_vmpage_page(vmpage, obj); > - if (page) { > - cl_page_delete(env, page); > - cl_page_put(env, page); > - } > - } else { > - LASSERT(vmpage->private == 0); > + /* See the comment in ll_releasepage() */ > + env = cl_env_percpu_get(); > + LASSERT(!IS_ERR(env)); > + inode = vmpage->mapping->host; > + obj = ll_i2info(inode)->lli_clob; > + if (obj) { > + page = cl_vmpage_page(vmpage, obj); > + if (page) { > + cl_page_delete(env, page); > + cl_page_put(env, page); > } > - cl_env_put(env, &refcheck); > + } else { > + LASSERT(vmpage->private == 0); > } > + cl_env_percpu_put(env); > } > } > > @@ -466,9 +464,9 @@ static int ll_write_begin(struct file *file, struct address_space *mapping, > struct page **pagep, void **fsdata) > { > struct ll_cl_context *lcc; > - const struct lu_env *env; > + const struct lu_env *env = NULL; > struct cl_io *io; > - struct cl_page *page; > + struct cl_page *page = NULL; > struct cl_object *clob = ll_i2info(mapping->host)->lli_clob; > pgoff_t index = pos >> PAGE_SHIFT; > struct page *vmpage = NULL; > @@ -556,6 +554,10 @@ static int ll_write_begin(struct file *file, struct address_space *mapping, > unlock_page(vmpage); > put_page(vmpage); > } > + if (!IS_ERR_OR_NULL(page)) { > + lu_ref_del(&page->cp_reference, "cl_io", io); > + cl_page_put(env, page); > + } > } else { > *pagep = vmpage; > *fsdata = lcc; > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 07/14] staging/lustre/ldlm: Reinstate ldlm_enqueue_pack() 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (5 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 06/14] staging/lustre/llite: drop_caches hangs in cl_inode_fini() Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:53 ` James Simmons 2016-11-03 1:24 ` [PATCH 08/14] staging/lustre/ldlm: engage ELC for all ldlm enqueue req Oleg Drokin ` (6 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Oleg Drokin The function becomes used again with the next patch, so bring it back from dead, only this time make it static. Reverts: bf2a033360f7 ("staging/lustre/ldlm: Remove unused ldlm_enqueue_pack()") Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c index 6e704c7..1b9ae77 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c @@ -657,6 +657,27 @@ int ldlm_prep_enqueue_req(struct obd_export *exp, struct ptlrpc_request *req, } EXPORT_SYMBOL(ldlm_prep_enqueue_req); +static struct ptlrpc_request *ldlm_enqueue_pack(struct obd_export *exp, + int lvb_len) +{ + struct ptlrpc_request *req; + int rc; + + req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_LDLM_ENQUEUE); + if (!req) + return ERR_PTR(-ENOMEM); + + rc = ldlm_prep_enqueue_req(exp, req, NULL, 0); + if (rc) { + ptlrpc_request_free(req); + return ERR_PTR(rc); + } + + req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER, lvb_len); + ptlrpc_request_set_replen(req); + return req; +} + /** * Client-side lock enqueue. * -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 07/14] staging/lustre/ldlm: Reinstate ldlm_enqueue_pack() 2016-11-03 1:24 ` [PATCH 07/14] staging/lustre/ldlm: Reinstate ldlm_enqueue_pack() Oleg Drokin @ 2016-11-07 1:53 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:53 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List > The function becomes used again with the next patch, so bring it back > from dead, only this time make it static. Reviewed-by: James Simmons <jsimmons@infradead.org> > Reverts: bf2a033360f7 ("staging/lustre/ldlm: Remove unused ldlm_enqueue_pack()") > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 21 +++++++++++++++++++++ > 1 file changed, 21 insertions(+) > > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > index 6e704c7..1b9ae77 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > @@ -657,6 +657,27 @@ int ldlm_prep_enqueue_req(struct obd_export *exp, struct ptlrpc_request *req, > } > EXPORT_SYMBOL(ldlm_prep_enqueue_req); > > +static struct ptlrpc_request *ldlm_enqueue_pack(struct obd_export *exp, > + int lvb_len) > +{ > + struct ptlrpc_request *req; > + int rc; > + > + req = ptlrpc_request_alloc(class_exp2cliimp(exp), &RQF_LDLM_ENQUEUE); > + if (!req) > + return ERR_PTR(-ENOMEM); > + > + rc = ldlm_prep_enqueue_req(exp, req, NULL, 0); > + if (rc) { > + ptlrpc_request_free(req); > + return ERR_PTR(rc); > + } > + > + req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER, lvb_len); > + ptlrpc_request_set_replen(req); > + return req; > +} > + > /** > * Client-side lock enqueue. > * > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 08/14] staging/lustre/ldlm: engage ELC for all ldlm enqueue req 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (6 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 07/14] staging/lustre/ldlm: Reinstate ldlm_enqueue_pack() Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:55 ` James Simmons 2016-11-03 1:24 ` [PATCH 09/14] staging/lustre/ptlrpc: Suppress error for flock requests Oleg Drokin ` (5 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Hongchao Zhang, Oleg Drokin From: Hongchao Zhang <hongchao.zhang@intel.com> If there is no request passed into ldlm_cli_enqueue, the enqueue request will not engage ELC to drop unneeded locks. currently, this kind of request is mainly related to EXTENT locks enqueue requests (except for glimpse EXTENT lock for it has an intent). Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com> Reviewed-on: http://review.whamcloud.com/21739 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8209 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Vitaly Fertman <vitaly.fertman@seagate.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 21 ++++----------------- 1 file changed, 4 insertions(+), 17 deletions(-) diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c index 1b9ae77..c5d00d1 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c @@ -748,17 +748,14 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp, lock->l_last_activity = ktime_get_real_seconds(); /* lock not sent to server yet */ - if (!reqp || !*reqp) { - req = ptlrpc_request_alloc_pack(class_exp2cliimp(exp), - &RQF_LDLM_ENQUEUE, - LUSTRE_DLM_VERSION, - LDLM_ENQUEUE); - if (!req) { + req = ldlm_enqueue_pack(exp, lvb_len); + if (IS_ERR(req)) { failed_lock_cleanup(ns, lock, einfo->ei_mode); LDLM_LOCK_RELEASE(lock); - return -ENOMEM; + return PTR_ERR(req); } + req_passed_in = 0; if (reqp) *reqp = req; @@ -778,16 +775,6 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp, body->lock_flags = ldlm_flags_to_wire(*flags); body->lock_handle[0] = *lockh; - /* Continue as normal. */ - if (!req_passed_in) { - if (lvb_len > 0) - req_capsule_extend(&req->rq_pill, - &RQF_LDLM_ENQUEUE_LVB); - req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER, - lvb_len); - ptlrpc_request_set_replen(req); - } - /* * Liblustre client doesn't get extent locks, except for O_APPEND case * where [0, OBD_OBJECT_EOF] lock is taken, or truncate, where -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 08/14] staging/lustre/ldlm: engage ELC for all ldlm enqueue req 2016-11-03 1:24 ` [PATCH 08/14] staging/lustre/ldlm: engage ELC for all ldlm enqueue req Oleg Drokin @ 2016-11-07 1:55 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:55 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Hongchao Zhang > From: Hongchao Zhang <hongchao.zhang@intel.com> > > If there is no request passed into ldlm_cli_enqueue, the enqueue > request will not engage ELC to drop unneeded locks. currently, > this kind of request is mainly related to EXTENT locks enqueue > requests (except for glimpse EXTENT lock for it has an intent). Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Hongchao Zhang <hongchao.zhang@intel.com> > Reviewed-on: http://review.whamcloud.com/21739 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8209 > Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> > Reviewed-by: Vitaly Fertman <vitaly.fertman@seagate.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 21 ++++----------------- > 1 file changed, 4 insertions(+), 17 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > index 1b9ae77..c5d00d1 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > @@ -748,17 +748,14 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp, > lock->l_last_activity = ktime_get_real_seconds(); > > /* lock not sent to server yet */ > - > if (!reqp || !*reqp) { > - req = ptlrpc_request_alloc_pack(class_exp2cliimp(exp), > - &RQF_LDLM_ENQUEUE, > - LUSTRE_DLM_VERSION, > - LDLM_ENQUEUE); > - if (!req) { > + req = ldlm_enqueue_pack(exp, lvb_len); > + if (IS_ERR(req)) { > failed_lock_cleanup(ns, lock, einfo->ei_mode); > LDLM_LOCK_RELEASE(lock); > - return -ENOMEM; > + return PTR_ERR(req); > } > + > req_passed_in = 0; > if (reqp) > *reqp = req; > @@ -778,16 +775,6 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp, > body->lock_flags = ldlm_flags_to_wire(*flags); > body->lock_handle[0] = *lockh; > > - /* Continue as normal. */ > - if (!req_passed_in) { > - if (lvb_len > 0) > - req_capsule_extend(&req->rq_pill, > - &RQF_LDLM_ENQUEUE_LVB); > - req_capsule_set_size(&req->rq_pill, &RMF_DLM_LVB, RCL_SERVER, > - lvb_len); > - ptlrpc_request_set_replen(req); > - } > - > /* > * Liblustre client doesn't get extent locks, except for O_APPEND case > * where [0, OBD_OBJECT_EOF] lock is taken, or truncate, where > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 09/14] staging/lustre/ptlrpc: Suppress error for flock requests 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (7 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 08/14] staging/lustre/ldlm: engage ELC for all ldlm enqueue req Oleg Drokin @ 2016-11-03 1:24 ` Oleg Drokin 2016-11-07 1:57 ` James Simmons 2016-11-03 1:25 ` [PATCH 10/14] staging/lustre/llite: protect from accessing NULL lli_clob Oleg Drokin ` (4 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:24 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Patrick Farrell, Oleg Drokin From: Patrick Farrell <paf@cray.com> -EAGAIN is a normal return when requesting POSIX flocks. We can't recognize exactly that case here, but it's the only case that should result in -EAGAIN on LDLM_ENQUEUE, so don't print to console in that case. Signed-off-by: Patrick Farrell <paf@cray.com> Reviewed-on: http://review.whamcloud.com/22856 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8658 Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> Reviewed-by: Bob Glossman <bob.glossman@intel.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/ptlrpc/client.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/staging/lustre/lustre/ptlrpc/client.c b/drivers/staging/lustre/lustre/ptlrpc/client.c index 7cbfb4c..bb7ae4e 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/client.c +++ b/drivers/staging/lustre/lustre/ptlrpc/client.c @@ -1191,7 +1191,9 @@ static int ptlrpc_check_status(struct ptlrpc_request *req) lnet_nid_t nid = imp->imp_connection->c_peer.nid; __u32 opc = lustre_msg_get_opc(req->rq_reqmsg); - if (ptlrpc_console_allow(req)) + /* -EAGAIN is normal when using POSIX flocks */ + if (ptlrpc_console_allow(req) && + !(opc == LDLM_ENQUEUE && err == -EAGAIN)) LCONSOLE_ERROR_MSG(0x011, "%s: operation %s to node %s failed: rc = %d\n", imp->imp_obd->obd_name, ll_opcode2str(opc), -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 09/14] staging/lustre/ptlrpc: Suppress error for flock requests 2016-11-03 1:24 ` [PATCH 09/14] staging/lustre/ptlrpc: Suppress error for flock requests Oleg Drokin @ 2016-11-07 1:57 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:57 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Patrick Farrell > From: Patrick Farrell <paf@cray.com> > > -EAGAIN is a normal return when requesting POSIX flocks. > We can't recognize exactly that case here, but it's the > only case that should result in -EAGAIN on LDLM_ENQUEUE, so > don't print to console in that case. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Patrick Farrell <paf@cray.com> > Reviewed-on: http://review.whamcloud.com/22856 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8658 > Reviewed-by: Andreas Dilger <andreas.dilger@intel.com> > Reviewed-by: Bob Glossman <bob.glossman@intel.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/ptlrpc/client.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/drivers/staging/lustre/lustre/ptlrpc/client.c b/drivers/staging/lustre/lustre/ptlrpc/client.c > index 7cbfb4c..bb7ae4e 100644 > --- a/drivers/staging/lustre/lustre/ptlrpc/client.c > +++ b/drivers/staging/lustre/lustre/ptlrpc/client.c > @@ -1191,7 +1191,9 @@ static int ptlrpc_check_status(struct ptlrpc_request *req) > lnet_nid_t nid = imp->imp_connection->c_peer.nid; > __u32 opc = lustre_msg_get_opc(req->rq_reqmsg); > > - if (ptlrpc_console_allow(req)) > + /* -EAGAIN is normal when using POSIX flocks */ > + if (ptlrpc_console_allow(req) && > + !(opc == LDLM_ENQUEUE && err == -EAGAIN)) > LCONSOLE_ERROR_MSG(0x011, "%s: operation %s to node %s failed: rc = %d\n", > imp->imp_obd->obd_name, > ll_opcode2str(opc), > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 10/14] staging/lustre/llite: protect from accessing NULL lli_clob 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (8 preceding siblings ...) 2016-11-03 1:24 ` [PATCH 09/14] staging/lustre/ptlrpc: Suppress error for flock requests Oleg Drokin @ 2016-11-03 1:25 ` Oleg Drokin 2016-11-07 1:55 ` James Simmons 2016-11-03 1:25 ` [PATCH 11/14] staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs Oleg Drokin ` (3 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:25 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Bobi Jam, Oleg Drokin From: Bobi Jam <bobijam.xu@intel.com> Need to check file's lli_clob object before calling lov_read_and_clear_async_rc(). Signed-off-by: Bobi Jam <bobijam.xu@intel.com> Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> Reviewed-on: http://review.whamcloud.com/23031 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8682 Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/llite/file.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/staging/lustre/lustre/llite/file.c b/drivers/staging/lustre/lustre/llite/file.c index c1c7551..7886840 100644 --- a/drivers/staging/lustre/lustre/llite/file.c +++ b/drivers/staging/lustre/lustre/llite/file.c @@ -2328,9 +2328,11 @@ int ll_fsync(struct file *file, loff_t start, loff_t end, int datasync) lli->lli_async_rc = 0; if (rc == 0) rc = err; - err = lov_read_and_clear_async_rc(lli->lli_clob); - if (rc == 0) - rc = err; + if (lli->lli_clob) { + err = lov_read_and_clear_async_rc(lli->lli_clob); + if (rc == 0) + rc = err; + } } err = md_sync(ll_i2sbi(inode)->ll_md_exp, ll_inode2fid(inode), &req); -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 10/14] staging/lustre/llite: protect from accessing NULL lli_clob 2016-11-03 1:25 ` [PATCH 10/14] staging/lustre/llite: protect from accessing NULL lli_clob Oleg Drokin @ 2016-11-07 1:55 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:55 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Bobi Jam > From: Bobi Jam <bobijam.xu@intel.com> > > Need to check file's lli_clob object before calling > lov_read_and_clear_async_rc(). Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Bobi Jam <bobijam.xu@intel.com> > Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> > Reviewed-by: Oleg Drokin <oleg.drokin@intel.com> > Reviewed-on: http://review.whamcloud.com/23031 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8682 > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/llite/file.c | 8 +++++--- > 1 file changed, 5 insertions(+), 3 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/llite/file.c b/drivers/staging/lustre/lustre/llite/file.c > index c1c7551..7886840 100644 > --- a/drivers/staging/lustre/lustre/llite/file.c > +++ b/drivers/staging/lustre/lustre/llite/file.c > @@ -2328,9 +2328,11 @@ int ll_fsync(struct file *file, loff_t start, loff_t end, int datasync) > lli->lli_async_rc = 0; > if (rc == 0) > rc = err; > - err = lov_read_and_clear_async_rc(lli->lli_clob); > - if (rc == 0) > - rc = err; > + if (lli->lli_clob) { > + err = lov_read_and_clear_async_rc(lli->lli_clob); > + if (rc == 0) > + rc = err; > + } > } > > err = md_sync(ll_i2sbi(inode)->ll_md_exp, ll_inode2fid(inode), &req); > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 11/14] staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (9 preceding siblings ...) 2016-11-03 1:25 ` [PATCH 10/14] staging/lustre/llite: protect from accessing NULL lli_clob Oleg Drokin @ 2016-11-03 1:25 ` Oleg Drokin 2016-11-07 1:58 ` James Simmons 2016-11-03 1:25 ` [PATCH 12/14] staging/lustre/llite: update ras window correctly Oleg Drokin ` (2 subsequent siblings) 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:25 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Amir Shehata, Oleg Drokin From: Amir Shehata <amir.shehata@intel.com> cpu_pattern can specify exactly 1 cpu in a partition: "0[0]". That means CPT0 will have CPU 0. CPU 0 can have hyperthreading enabled. This combination would result in weight = cfs_cpu_ht_nsiblings(0); hrp->hrp_nthrs = cfs_cpt_weight(ptlrpc_hr.hr_cpt_table, i); hrp->hrp_nthrs /= weight; evaluating to 0. Where cfs_cpt_weight(ptlrpc_hr.hr_cpt_table, i) == 1 weight == 2 Therefore, if hrp_nthrs becomes zero, just set it to 1. Signed-off-by: Amir Shehata <amir.shehata@intel.com> Reviewed-on: http://review.whamcloud.com/19106 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8492 Reviewed-by: Liang Zhen <liang.zhen@intel.com> Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com> Reviewed-by: James Simmons <uja.ornl@yahoo.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/ptlrpc/service.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/staging/lustre/lustre/ptlrpc/service.c b/drivers/staging/lustre/lustre/ptlrpc/service.c index 72f3930..fc754e7 100644 --- a/drivers/staging/lustre/lustre/ptlrpc/service.c +++ b/drivers/staging/lustre/lustre/ptlrpc/service.c @@ -2541,8 +2541,9 @@ int ptlrpc_hr_init(void) hrp->hrp_nthrs = cfs_cpt_weight(ptlrpc_hr.hr_cpt_table, i); hrp->hrp_nthrs /= weight; + if (hrp->hrp_nthrs == 0) + hrp->hrp_nthrs = 1; - LASSERT(hrp->hrp_nthrs > 0); hrp->hrp_thrs = kzalloc_node(hrp->hrp_nthrs * sizeof(*hrt), GFP_NOFS, cfs_cpt_spread_node(ptlrpc_hr.hr_cpt_table, -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 11/14] staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs 2016-11-03 1:25 ` [PATCH 11/14] staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs Oleg Drokin @ 2016-11-07 1:58 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:58 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Amir Shehata On Wed, 2 Nov 2016, Oleg Drokin wrote: > From: Amir Shehata <amir.shehata@intel.com> > > cpu_pattern can specify exactly 1 cpu in a partition: > "0[0]". That means CPT0 will have CPU 0. CPU 0 can have > hyperthreading enabled. This combination would result in > > weight = cfs_cpu_ht_nsiblings(0); > hrp->hrp_nthrs = cfs_cpt_weight(ptlrpc_hr.hr_cpt_table, i); > hrp->hrp_nthrs /= weight; > > evaluating to 0. Where > cfs_cpt_weight(ptlrpc_hr.hr_cpt_table, i) == 1 > weight == 2 > > Therefore, if hrp_nthrs becomes zero, just set it to 1. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Amir Shehata <amir.shehata@intel.com> > Reviewed-on: http://review.whamcloud.com/19106 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8492 > Reviewed-by: Liang Zhen <liang.zhen@intel.com> > Reviewed-by: Doug Oucharek <doug.s.oucharek@intel.com> > Reviewed-by: James Simmons <uja.ornl@yahoo.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/ptlrpc/service.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/drivers/staging/lustre/lustre/ptlrpc/service.c b/drivers/staging/lustre/lustre/ptlrpc/service.c > index 72f3930..fc754e7 100644 > --- a/drivers/staging/lustre/lustre/ptlrpc/service.c > +++ b/drivers/staging/lustre/lustre/ptlrpc/service.c > @@ -2541,8 +2541,9 @@ int ptlrpc_hr_init(void) > > hrp->hrp_nthrs = cfs_cpt_weight(ptlrpc_hr.hr_cpt_table, i); > hrp->hrp_nthrs /= weight; > + if (hrp->hrp_nthrs == 0) > + hrp->hrp_nthrs = 1; > > - LASSERT(hrp->hrp_nthrs > 0); > hrp->hrp_thrs = > kzalloc_node(hrp->hrp_nthrs * sizeof(*hrt), GFP_NOFS, > cfs_cpt_spread_node(ptlrpc_hr.hr_cpt_table, > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 12/14] staging/lustre/llite: update ras window correctly 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (10 preceding siblings ...) 2016-11-03 1:25 ` [PATCH 11/14] staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs Oleg Drokin @ 2016-11-03 1:25 ` Oleg Drokin 2016-11-07 1:57 ` James Simmons 2016-11-03 1:25 ` [PATCH 13/14] staging/lustre/llite: do not clear uptodate bit in page delete Oleg Drokin 2016-11-03 1:25 ` [PATCH 14/14] staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users Oleg Drokin 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:25 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Bobi Jam, Oleg Drokin From: Bobi Jam <bobijam.xu@intel.com> When stride-RA hit case miss, we only reset normal sequential read-ahead window, but not reset the stride IO to avoid the overhead of re-detecting stride IO. While when the normal RA window is set to not insect with the stride-RA window, when we try to increase the stride-RA window length later, the presumption does not hold. This patch resets the stride IO as well in this case. Signed-off-by: Bobi Jam <bobijam.xu@intel.com> Reviewed-on: http://review.whamcloud.com/23032 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8683 Reviewed-by: wangdi <di.wang@intel.com> Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/llite/rw.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/staging/lustre/lustre/llite/rw.c b/drivers/staging/lustre/lustre/llite/rw.c index d2515a8..e34017d 100644 --- a/drivers/staging/lustre/lustre/llite/rw.c +++ b/drivers/staging/lustre/lustre/llite/rw.c @@ -809,13 +809,20 @@ static void ras_update(struct ll_sb_info *sbi, struct inode *inode, if (ra_miss) { if (index_in_stride_window(ras, index) && stride_io_mode(ras)) { - /*If stride-RA hit cache miss, the stride dector - *will not be reset to avoid the overhead of - *redetecting read-ahead mode - */ if (index != ras->ras_last_readpage + 1) ras->ras_consecutive_pages = 0; ras_reset(inode, ras, index); + + /* If stride-RA hit cache miss, the stride + * detector will not be reset to avoid the + * overhead of redetecting read-ahead mode, + * but on the condition that the stride window + * is still intersect with normal sequential + * read-ahead window. + */ + if (ras->ras_window_start < + ras->ras_stride_offset) + ras_stride_reset(ras); RAS_CDEBUG(ras); } else { /* Reset both stride window and normal RA -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 12/14] staging/lustre/llite: update ras window correctly 2016-11-03 1:25 ` [PATCH 12/14] staging/lustre/llite: update ras window correctly Oleg Drokin @ 2016-11-07 1:57 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:57 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Bobi Jam > From: Bobi Jam <bobijam.xu@intel.com> > > When stride-RA hit case miss, we only reset normal sequential > read-ahead window, but not reset the stride IO to avoid the overhead > of re-detecting stride IO. While when the normal RA window is set > to not insect with the stride-RA window, when we try to increase > the stride-RA window length later, the presumption does not hold. > > This patch resets the stride IO as well in this case. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Bobi Jam <bobijam.xu@intel.com> > Reviewed-on: http://review.whamcloud.com/23032 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8683 > Reviewed-by: wangdi <di.wang@intel.com> > Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/llite/rw.c | 15 +++++++++++---- > 1 file changed, 11 insertions(+), 4 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/llite/rw.c b/drivers/staging/lustre/lustre/llite/rw.c > index d2515a8..e34017d 100644 > --- a/drivers/staging/lustre/lustre/llite/rw.c > +++ b/drivers/staging/lustre/lustre/llite/rw.c > @@ -809,13 +809,20 @@ static void ras_update(struct ll_sb_info *sbi, struct inode *inode, > if (ra_miss) { > if (index_in_stride_window(ras, index) && > stride_io_mode(ras)) { > - /*If stride-RA hit cache miss, the stride dector > - *will not be reset to avoid the overhead of > - *redetecting read-ahead mode > - */ > if (index != ras->ras_last_readpage + 1) > ras->ras_consecutive_pages = 0; > ras_reset(inode, ras, index); > + > + /* If stride-RA hit cache miss, the stride > + * detector will not be reset to avoid the > + * overhead of redetecting read-ahead mode, > + * but on the condition that the stride window > + * is still intersect with normal sequential > + * read-ahead window. > + */ > + if (ras->ras_window_start < > + ras->ras_stride_offset) > + ras_stride_reset(ras); > RAS_CDEBUG(ras); > } else { > /* Reset both stride window and normal RA > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 13/14] staging/lustre/llite: do not clear uptodate bit in page delete 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (11 preceding siblings ...) 2016-11-03 1:25 ` [PATCH 12/14] staging/lustre/llite: update ras window correctly Oleg Drokin @ 2016-11-03 1:25 ` Oleg Drokin 2016-11-07 1:56 ` James Simmons 2016-11-03 1:25 ` [PATCH 14/14] staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users Oleg Drokin 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:25 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Jinshan Xiong, Oleg Drokin From: Jinshan Xiong <jinshan.xiong@intel.com> Otherwise, if the race between page fault and truncate occurs, it will cause the page fault routine to return an EIO error. In filemap_fault() { page_not_uptodate: ... ClearPageError(page); error = mapping->a_ops->readpage(file, page); if (!error) { wait_on_page_locked(page); if (!PageUptodate(page)) error = -EIO; } ... } However, I tend to think this is a defect in kernel implementation, because it assumes PageUptodate shouldn't be cleared but file read routine doesn't make the same assumption. Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com> Reviewed-on: http://review.whamcloud.com/22827 Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8633 Reviewed-by: Li Dongyang <dongyang.li@anu.edu.au> Reviewed-by: Bobi Jam <bobijam@hotmail.com> Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/llite/vvp_page.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/staging/lustre/lustre/llite/vvp_page.c b/drivers/staging/lustre/lustre/llite/vvp_page.c index 25490a5..23d6630 100644 --- a/drivers/staging/lustre/lustre/llite/vvp_page.c +++ b/drivers/staging/lustre/lustre/llite/vvp_page.c @@ -166,7 +166,6 @@ static void vvp_page_delete(const struct lu_env *env, refc = atomic_dec_return(&page->cp_ref); LASSERTF(refc >= 1, "page = %p, refc = %d\n", page, refc); - ClearPageUptodate(vmpage); ClearPagePrivate(vmpage); vmpage->private = 0; /* -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 13/14] staging/lustre/llite: do not clear uptodate bit in page delete 2016-11-03 1:25 ` [PATCH 13/14] staging/lustre/llite: do not clear uptodate bit in page delete Oleg Drokin @ 2016-11-07 1:56 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:56 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List, Jinshan Xiong > From: Jinshan Xiong <jinshan.xiong@intel.com> > > Otherwise, if the race between page fault and truncate occurs, it > will cause the page fault routine to return an EIO error. > > In filemap_fault() { > page_not_uptodate: > ... > ClearPageError(page); > error = mapping->a_ops->readpage(file, page); > if (!error) { > wait_on_page_locked(page); > if (!PageUptodate(page)) > error = -EIO; > } > ... > } > > However, I tend to think this is a defect in kernel implementation, > because it assumes PageUptodate shouldn't be cleared but file read > routine doesn't make the same assumption. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Jinshan Xiong <jinshan.xiong@intel.com> > Reviewed-on: http://review.whamcloud.com/22827 > Intel-bug-id: https://jira.hpdd.intel.com/browse/LU-8633 > Reviewed-by: Li Dongyang <dongyang.li@anu.edu.au> > Reviewed-by: Bobi Jam <bobijam@hotmail.com> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/llite/vvp_page.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/drivers/staging/lustre/lustre/llite/vvp_page.c b/drivers/staging/lustre/lustre/llite/vvp_page.c > index 25490a5..23d6630 100644 > --- a/drivers/staging/lustre/lustre/llite/vvp_page.c > +++ b/drivers/staging/lustre/lustre/llite/vvp_page.c > @@ -166,7 +166,6 @@ static void vvp_page_delete(const struct lu_env *env, > refc = atomic_dec_return(&page->cp_ref); > LASSERTF(refc >= 1, "page = %p, refc = %d\n", page, refc); > > - ClearPageUptodate(vmpage); > ClearPagePrivate(vmpage); > vmpage->private = 0; > /* > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH 14/14] staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin ` (12 preceding siblings ...) 2016-11-03 1:25 ` [PATCH 13/14] staging/lustre/llite: do not clear uptodate bit in page delete Oleg Drokin @ 2016-11-03 1:25 ` Oleg Drokin 2016-11-07 1:56 ` James Simmons 13 siblings, 1 reply; 29+ messages in thread From: Oleg Drokin @ 2016-11-03 1:25 UTC (permalink / raw) To: Greg Kroah-Hartman, devel, Andreas Dilger, James Simmons Cc: Linux Kernel Mailing List, Lustre Development List, Oleg Drokin This define only made sense in a userspace library client, not in the kernel. Signed-off-by: Oleg Drokin <green@linuxhacker.ru> --- drivers/staging/lustre/lustre/include/lustre_lib.h | 2 -- drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 15 +-------------- 2 files changed, 1 insertion(+), 16 deletions(-) diff --git a/drivers/staging/lustre/lustre/include/lustre_lib.h b/drivers/staging/lustre/lustre/include/lustre_lib.h index 6b23191..27f3148 100644 --- a/drivers/staging/lustre/lustre/include/lustre_lib.h +++ b/drivers/staging/lustre/lustre/include/lustre_lib.h @@ -350,8 +350,6 @@ do { \ l_wait_event_exclusive_head(wq, condition, &lwi); \ }) -#define LIBLUSTRE_CLIENT (0) - /** @} lib */ #endif /* _LUSTRE_LIB_H */ diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c index c5d00d1..6a96f2c 100644 --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c @@ -475,12 +475,7 @@ int ldlm_cli_enqueue_fini(struct obd_export *exp, struct ptlrpc_request *req, "client-side enqueue, new policy data"); } - if ((*flags) & LDLM_FL_AST_SENT || - /* Cancel extent locks as soon as possible on a liblustre client, - * because it cannot handle asynchronous ASTs robustly (see - * bug 7311). - */ - (LIBLUSTRE_CLIENT && type == LDLM_EXTENT)) { + if ((*flags) & LDLM_FL_AST_SENT) { lock_res_and_lock(lock); lock->l_flags |= LDLM_FL_CBPENDING | LDLM_FL_BL_AST; unlock_res_and_lock(lock); @@ -775,14 +770,6 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp, body->lock_flags = ldlm_flags_to_wire(*flags); body->lock_handle[0] = *lockh; - /* - * Liblustre client doesn't get extent locks, except for O_APPEND case - * where [0, OBD_OBJECT_EOF] lock is taken, or truncate, where - * [i_size, OBD_OBJECT_EOF] lock is taken. - */ - LASSERT(ergo(LIBLUSTRE_CLIENT, einfo->ei_type != LDLM_EXTENT || - policy->l_extent.end == OBD_OBJECT_EOF)); - if (async) { LASSERT(reqp); return 0; -- 2.7.4 ^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH 14/14] staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users 2016-11-03 1:25 ` [PATCH 14/14] staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users Oleg Drokin @ 2016-11-07 1:56 ` James Simmons 0 siblings, 0 replies; 29+ messages in thread From: James Simmons @ 2016-11-07 1:56 UTC (permalink / raw) To: Oleg Drokin Cc: Greg Kroah-Hartman, devel, Andreas Dilger, Linux Kernel Mailing List, Lustre Development List > This define only made sense in a userspace library client, not in the kernel. Reviewed-by: James Simmons <jsimmons@infradead.org> > Signed-off-by: Oleg Drokin <green@linuxhacker.ru> > --- > drivers/staging/lustre/lustre/include/lustre_lib.h | 2 -- > drivers/staging/lustre/lustre/ldlm/ldlm_request.c | 15 +-------------- > 2 files changed, 1 insertion(+), 16 deletions(-) > > diff --git a/drivers/staging/lustre/lustre/include/lustre_lib.h b/drivers/staging/lustre/lustre/include/lustre_lib.h > index 6b23191..27f3148 100644 > --- a/drivers/staging/lustre/lustre/include/lustre_lib.h > +++ b/drivers/staging/lustre/lustre/include/lustre_lib.h > @@ -350,8 +350,6 @@ do { \ > l_wait_event_exclusive_head(wq, condition, &lwi); \ > }) > > -#define LIBLUSTRE_CLIENT (0) > - > /** @} lib */ > > #endif /* _LUSTRE_LIB_H */ > diff --git a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > index c5d00d1..6a96f2c 100644 > --- a/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > +++ b/drivers/staging/lustre/lustre/ldlm/ldlm_request.c > @@ -475,12 +475,7 @@ int ldlm_cli_enqueue_fini(struct obd_export *exp, struct ptlrpc_request *req, > "client-side enqueue, new policy data"); > } > > - if ((*flags) & LDLM_FL_AST_SENT || > - /* Cancel extent locks as soon as possible on a liblustre client, > - * because it cannot handle asynchronous ASTs robustly (see > - * bug 7311). > - */ > - (LIBLUSTRE_CLIENT && type == LDLM_EXTENT)) { > + if ((*flags) & LDLM_FL_AST_SENT) { > lock_res_and_lock(lock); > lock->l_flags |= LDLM_FL_CBPENDING | LDLM_FL_BL_AST; > unlock_res_and_lock(lock); > @@ -775,14 +770,6 @@ int ldlm_cli_enqueue(struct obd_export *exp, struct ptlrpc_request **reqp, > body->lock_flags = ldlm_flags_to_wire(*flags); > body->lock_handle[0] = *lockh; > > - /* > - * Liblustre client doesn't get extent locks, except for O_APPEND case > - * where [0, OBD_OBJECT_EOF] lock is taken, or truncate, where > - * [i_size, OBD_OBJECT_EOF] lock is taken. > - */ > - LASSERT(ergo(LIBLUSTRE_CLIENT, einfo->ei_type != LDLM_EXTENT || > - policy->l_extent.end == OBD_OBJECT_EOF)); > - > if (async) { > LASSERT(reqp); > return 0; > -- > 2.7.4 > > ^ permalink raw reply [flat|nested] 29+ messages in thread
end of thread, other threads:[~2016-11-07 1:58 UTC | newest] Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2016-11-03 1:24 [PATCH 00/14] Lustre fixes Oleg Drokin 2016-11-03 1:24 ` [PATCH 01/14] staging/lustre/ldlm: Drop unused blocking_refs flock field Oleg Drokin 2016-11-07 1:38 ` James Simmons 2016-11-03 1:24 ` [PATCH 02/14] staging/lustre/ldlm: fix export reference problem Oleg Drokin 2016-11-07 1:37 ` James Simmons 2016-11-03 1:24 ` [PATCH 03/14] staging/lustre: conflicting PW & PR extent locks on a client Oleg Drokin 2016-11-07 1:52 ` James Simmons 2016-11-03 1:24 ` [PATCH 04/14] staging/lustre/llite: clear inode timestamps after losing UPDATE lock Oleg Drokin 2016-11-07 1:38 ` James Simmons 2016-11-03 1:24 ` [PATCH 05/14] staging/lustre: Get rid of cl_env hash table Oleg Drokin 2016-11-07 1:53 ` James Simmons 2016-11-03 1:24 ` [PATCH 06/14] staging/lustre/llite: drop_caches hangs in cl_inode_fini() Oleg Drokin 2016-11-07 1:54 ` James Simmons 2016-11-03 1:24 ` [PATCH 07/14] staging/lustre/ldlm: Reinstate ldlm_enqueue_pack() Oleg Drokin 2016-11-07 1:53 ` James Simmons 2016-11-03 1:24 ` [PATCH 08/14] staging/lustre/ldlm: engage ELC for all ldlm enqueue req Oleg Drokin 2016-11-07 1:55 ` James Simmons 2016-11-03 1:24 ` [PATCH 09/14] staging/lustre/ptlrpc: Suppress error for flock requests Oleg Drokin 2016-11-07 1:57 ` James Simmons 2016-11-03 1:25 ` [PATCH 10/14] staging/lustre/llite: protect from accessing NULL lli_clob Oleg Drokin 2016-11-07 1:55 ` James Simmons 2016-11-03 1:25 ` [PATCH 11/14] staging/lustre/ptlrpc: Correctly calculate hrp->hrp_nthrs Oleg Drokin 2016-11-07 1:58 ` James Simmons 2016-11-03 1:25 ` [PATCH 12/14] staging/lustre/llite: update ras window correctly Oleg Drokin 2016-11-07 1:57 ` James Simmons 2016-11-03 1:25 ` [PATCH 13/14] staging/lustre/llite: do not clear uptodate bit in page delete Oleg Drokin 2016-11-07 1:56 ` James Simmons 2016-11-03 1:25 ` [PATCH 14/14] staging/lustre: Get rid of LIBLUSTRE_CLIENT and its users Oleg Drokin 2016-11-07 1:56 ` James Simmons
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox
Powered by JetHome