diff options
Diffstat (limited to 'fs')
-rw-r--r-- | fs/nfsd/nfs4proc.c | 24 | ||||
-rw-r--r-- | fs/nfsd/nfs4state.c | 30 |
2 files changed, 30 insertions, 24 deletions
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index e08edc17c6a..361b4007d4a 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -162,7 +162,7 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ static inline int -nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) +nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, struct nfs4_stateowner **replay_owner) { int status; dprintk("NFSD: nfsd4_open filename %.*s op_stateowner %p\n", @@ -238,8 +238,10 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open */ status = nfsd4_process_open2(rqstp, current_fh, open); out: - if (open->op_stateowner) + if (open->op_stateowner) { nfs4_get_stateowner(open->op_stateowner); + *replay_owner = open->op_stateowner; + } nfs4_unlock_state(); return status; } @@ -809,8 +811,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, op->status = nfsd4_access(rqstp, current_fh, &op->u.access); break; case OP_CLOSE: - op->status = nfsd4_close(rqstp, current_fh, &op->u.close); - replay_owner = op->u.close.cl_stateowner; + op->status = nfsd4_close(rqstp, current_fh, &op->u.close, &replay_owner); break; case OP_COMMIT: op->status = nfsd4_commit(rqstp, current_fh, &op->u.commit); @@ -831,15 +832,13 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, op->status = nfsd4_link(rqstp, current_fh, save_fh, &op->u.link); break; case OP_LOCK: - op->status = nfsd4_lock(rqstp, current_fh, &op->u.lock); - replay_owner = op->u.lock.lk_stateowner; + op->status = nfsd4_lock(rqstp, current_fh, &op->u.lock, &replay_owner); break; case OP_LOCKT: op->status = nfsd4_lockt(rqstp, current_fh, &op->u.lockt); break; case OP_LOCKU: - op->status = nfsd4_locku(rqstp, current_fh, &op->u.locku); - replay_owner = op->u.locku.lu_stateowner; + op->status = nfsd4_locku(rqstp, current_fh, &op->u.locku, &replay_owner); break; case OP_LOOKUP: op->status = nfsd4_lookup(rqstp, current_fh, &op->u.lookup); @@ -853,16 +852,13 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, op->status = nfs_ok; break; case OP_OPEN: - op->status = nfsd4_open(rqstp, current_fh, &op->u.open); - replay_owner = op->u.open.op_stateowner; + op->status = nfsd4_open(rqstp, current_fh, &op->u.open, &replay_owner); break; case OP_OPEN_CONFIRM: - op->status = nfsd4_open_confirm(rqstp, current_fh, &op->u.open_confirm); - replay_owner = op->u.open_confirm.oc_stateowner; + op->status = nfsd4_open_confirm(rqstp, current_fh, &op->u.open_confirm, &replay_owner); break; case OP_OPEN_DOWNGRADE: - op->status = nfsd4_open_downgrade(rqstp, current_fh, &op->u.open_downgrade); - replay_owner = op->u.open_downgrade.od_stateowner; + op->status = nfsd4_open_downgrade(rqstp, current_fh, &op->u.open_downgrade, &replay_owner); break; case OP_PUTFH: op->status = nfsd4_putfh(rqstp, current_fh, &op->u.putfh); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 109fd8c3d4f..11405e530e7 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2286,7 +2286,7 @@ check_replay: } int -nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc) +nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc, struct nfs4_stateowner **replay_owner) { int status; struct nfs4_stateowner *sop; @@ -2320,8 +2320,10 @@ nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfs nfsd4_create_clid_dir(sop->so_client); out: - if (oc->oc_stateowner) + if (oc->oc_stateowner) { nfs4_get_stateowner(oc->oc_stateowner); + *replay_owner = oc->oc_stateowner; + } nfs4_unlock_state(); return status; } @@ -2352,7 +2354,7 @@ reset_union_bmap_deny(unsigned long deny, unsigned long *bmap) } int -nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od) +nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od, struct nfs4_stateowner **replay_owner) { int status; struct nfs4_stateid *stp; @@ -2394,8 +2396,10 @@ nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct n memcpy(&od->od_stateid, &stp->st_stateid, sizeof(stateid_t)); status = nfs_ok; out: - if (od->od_stateowner) + if (od->od_stateowner) { nfs4_get_stateowner(od->od_stateowner); + *replay_owner = od->od_stateowner; + } nfs4_unlock_state(); return status; } @@ -2404,7 +2408,7 @@ out: * nfs4_unlock_state() called after encode */ int -nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close) +nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close, struct nfs4_stateowner **replay_owner) { int status; struct nfs4_stateid *stp; @@ -2430,8 +2434,10 @@ nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_clos /* release_state_owner() calls nfsd_close() if needed */ release_state_owner(stp, OPEN_STATE); out: - if (close->cl_stateowner) + if (close->cl_stateowner) { nfs4_get_stateowner(close->cl_stateowner); + *replay_owner = close->cl_stateowner; + } nfs4_unlock_state(); return status; } @@ -2675,7 +2681,7 @@ check_lock_length(u64 offset, u64 length) * LOCK operation */ int -nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock) +nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock, struct nfs4_stateowner **replay_owner) { struct nfs4_stateowner *open_sop = NULL; struct nfs4_stateid *lock_stp; @@ -2835,8 +2841,10 @@ out_destroy_new_stateid: release_state_owner(lock_stp, LOCK_STATE); } out: - if (lock->lk_stateowner) + if (lock->lk_stateowner) { nfs4_get_stateowner(lock->lk_stateowner); + *replay_owner = lock->lk_stateowner; + } nfs4_unlock_state(); return status; } @@ -2925,7 +2933,7 @@ out: } int -nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku) +nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku, struct nfs4_stateowner **replay_owner) { struct nfs4_stateid *stp; struct file *filp = NULL; @@ -2981,8 +2989,10 @@ nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock memcpy(&locku->lu_stateid, &stp->st_stateid, sizeof(stateid_t)); out: - if (locku->lu_stateowner) + if (locku->lu_stateowner) { nfs4_get_stateowner(locku->lu_stateowner); + *replay_owner = locku->lu_stateowner; + } nfs4_unlock_state(); return status; |