X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Fllite%2Fnamei.c;h=edcaade2c1d0873daacbc797549fd06c6b7d49b9;hp=f53818a9acb09e69d2b03b72b3ff496c82bf9f20;hb=e07d4a58206756252c28b51048057c87a7cdb3c6;hpb=c8d56a664306c3ceb4e598999c41eb72ea46a68f diff --git a/lustre/llite/namei.c b/lustre/llite/namei.c index f53818a..edcaade 100644 --- a/lustre/llite/namei.c +++ b/lustre/llite/namei.c @@ -23,7 +23,7 @@ * Copyright (c) 2002, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. * - * Copyright (c) 2011, 2016, Intel Corporation. + * Copyright (c) 2011, 2017, Intel Corporation. */ /* * This file is part of Lustre, http://www.lustre.org/ @@ -38,16 +38,13 @@ #include #include #include -#ifdef HAVE_UIDGID_HEADER -# include -#endif +#include #define DEBUG_SUBSYSTEM S_LLITE #include #include #include -#include #include "llite_internal.h" static int ll_create_it(struct inode *dir, struct dentry *dentry, @@ -139,6 +136,9 @@ struct inode *ll_iget(struct super_block *sb, ino_t hash, inode_has_no_xattr(inode); unlock_new_inode(inode); } + } else if (is_bad_inode(inode)) { + iput(inode); + inode = ERR_PTR(-ESTALE); } else if (!(inode->i_state & (I_FREEING | I_CLEAR))) { rc = ll_update_inode(inode, md); CDEBUG(D_VFSTRACE, "got inode: "DFID"(%p): rc = %d\n", @@ -159,7 +159,7 @@ static void ll_invalidate_negative_children(struct inode *dir) struct dentry *dentry, *tmp_subdir; DECLARE_LL_D_HLIST_NODE_PTR(p); - ll_lock_dcache(dir); + spin_lock(&dir->i_lock); ll_d_hlist_for_each_entry(dentry, p, &dir->i_dentry) { spin_lock(&dentry->d_lock); if (!list_empty(&dentry->d_subdirs)) { @@ -174,7 +174,7 @@ static void ll_invalidate_negative_children(struct inode *dir) } spin_unlock(&dentry->d_lock); } - ll_unlock_dcache(dir); + spin_unlock(&dir->i_lock); } int ll_test_inode_by_fid(struct inode *inode, void *opaque) @@ -182,168 +182,317 @@ int ll_test_inode_by_fid(struct inode *inode, void *opaque) return lu_fid_eq(&ll_i2info(inode)->lli_fid, opaque); } -int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc, - void *data, int flag) +static int ll_dom_lock_cancel(struct inode *inode, struct ldlm_lock *lock) { - struct lustre_handle lockh; + struct lu_env *env; + struct ll_inode_info *lli = ll_i2info(inode); + __u16 refcheck; int rc; ENTRY; - switch (flag) { - case LDLM_CB_BLOCKING: - ldlm_lock2handle(lock, &lockh); - rc = ldlm_cli_cancel(&lockh, LCF_ASYNC); - if (rc < 0) { - CDEBUG(D_INODE, "ldlm_cli_cancel: rc = %d\n", rc); - RETURN(rc); - } - break; - case LDLM_CB_CANCELING: { - struct inode *inode = ll_inode_from_resource_lock(lock); - __u64 bits = lock->l_policy_data.l_inodebits.bits; + if (!lli->lli_clob) { + /* due to DoM read on open, there may exist pages for Lustre + * regular file even though cl_object is not set up yet. */ + truncate_inode_pages(inode->i_mapping, 0); + RETURN(0); + } - /* Inode is set to lock->l_resource->lr_lvb_inode - * for mdc - bug 24555 */ - LASSERT(lock->l_ast_data == NULL); + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + RETURN(PTR_ERR(env)); + + /* reach MDC layer to flush data under the DoM ldlm lock */ + rc = cl_object_flush(env, lli->lli_clob, lock); + if (rc == -ENODATA) { + CDEBUG(D_INODE, "inode "DFID" layout has no DoM stripe\n", + PFID(ll_inode2fid(inode))); + /* most likely result of layout change, do nothing */ + rc = 0; + } - if (inode == NULL) - break; + cl_env_put(env, &refcheck); + RETURN(rc); +} + +static void ll_lock_cancel_bits(struct ldlm_lock *lock, __u64 to_cancel) +{ + struct inode *inode = ll_inode_from_resource_lock(lock); + struct ll_inode_info *lli; + __u64 bits = to_cancel; + int rc; + + ENTRY; + + if (!inode) { + /* That means the inode is evicted most likely and may cause + * the skipping of lock cleanups below, so print the message + * about that in log. + */ + if (lock->l_resource->lr_lvb_inode) + LDLM_DEBUG(lock, + "can't take inode for the lock (%sevicted)\n", + lock->l_resource->lr_lvb_inode->i_state & + I_FREEING ? "" : "not "); + RETURN_EXIT; + } + + if (!fid_res_name_eq(ll_inode2fid(inode), + &lock->l_resource->lr_name)) { + LDLM_ERROR(lock, "data mismatch with object "DFID"(%p)", + PFID(ll_inode2fid(inode)), inode); + LBUG(); + } + + if (bits & MDS_INODELOCK_XATTR) { + ll_xattr_cache_destroy(inode); + bits &= ~MDS_INODELOCK_XATTR; + } - /* Invalidate all dentries associated with this inode */ - LASSERT(ldlm_is_canceling(lock)); + /* For OPEN locks we differentiate between lock modes + * LCK_CR, LCK_CW, LCK_PR - bug 22891 */ + if (bits & MDS_INODELOCK_OPEN) + ll_have_md_lock(inode, &bits, lock->l_req_mode); - if (!fid_res_name_eq(ll_inode2fid(inode), - &lock->l_resource->lr_name)) { - LDLM_ERROR(lock, "data mismatch with object "DFID"(%p)", - PFID(ll_inode2fid(inode)), inode); + if (bits & MDS_INODELOCK_OPEN) { + fmode_t fmode; + + switch (lock->l_req_mode) { + case LCK_CW: + fmode = FMODE_WRITE; + break; + case LCK_PR: + fmode = FMODE_EXEC; + break; + case LCK_CR: + fmode = FMODE_READ; + break; + default: + LDLM_ERROR(lock, "bad lock mode for OPEN lock"); LBUG(); } - if (bits & MDS_INODELOCK_XATTR) { - if (S_ISDIR(inode->i_mode)) - ll_i2info(inode)->lli_def_stripe_offset = -1; - ll_xattr_cache_destroy(inode); - bits &= ~MDS_INODELOCK_XATTR; - } + ll_md_real_close(inode, fmode); - /* For OPEN locks we differentiate between lock modes - * LCK_CR, LCK_CW, LCK_PR - bug 22891 */ - if (bits & MDS_INODELOCK_OPEN) - ll_have_md_lock(inode, &bits, lock->l_req_mode); - - if (bits & MDS_INODELOCK_OPEN) { - fmode_t fmode; - - switch (lock->l_req_mode) { - case LCK_CW: - fmode = FMODE_WRITE; - break; - case LCK_PR: - fmode = FMODE_EXEC; - break; - case LCK_CR: - fmode = FMODE_READ; - break; - default: - LDLM_ERROR(lock, "bad lock mode for OPEN lock"); - LBUG(); - } + bits &= ~MDS_INODELOCK_OPEN; + } - ll_md_real_close(inode, fmode); + if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE | + MDS_INODELOCK_LAYOUT | MDS_INODELOCK_PERM | + MDS_INODELOCK_DOM)) + ll_have_md_lock(inode, &bits, LCK_MINMODE); - bits &= ~MDS_INODELOCK_OPEN; - } + if (bits & MDS_INODELOCK_DOM) { + rc = ll_dom_lock_cancel(inode, lock); + if (rc < 0) + CDEBUG(D_INODE, "cannot flush DoM data " + DFID": rc = %d\n", + PFID(ll_inode2fid(inode)), rc); + } - if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_UPDATE | - MDS_INODELOCK_LAYOUT | MDS_INODELOCK_PERM)) - ll_have_md_lock(inode, &bits, LCK_MINMODE); - - if (bits & MDS_INODELOCK_LAYOUT) { - struct cl_object_conf conf = { - .coc_opc = OBJECT_CONF_INVALIDATE, - .coc_inode = inode, - }; - - rc = ll_layout_conf(inode, &conf); - if (rc < 0) - CDEBUG(D_INODE, "cannot invalidate layout of " - DFID": rc = %d\n", - PFID(ll_inode2fid(inode)), rc); - } + if (bits & MDS_INODELOCK_LAYOUT) { + struct cl_object_conf conf = { + .coc_opc = OBJECT_CONF_INVALIDATE, + .coc_inode = inode, + }; - if (bits & MDS_INODELOCK_UPDATE) { - struct ll_inode_info *lli = ll_i2info(inode); - lli->lli_update_atime = 1; - } + rc = ll_layout_conf(inode, &conf); + if (rc < 0) + CDEBUG(D_INODE, "cannot invalidate layout of " + DFID": rc = %d\n", + PFID(ll_inode2fid(inode)), rc); + } + + lli = ll_i2info(inode); - if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) { - struct ll_inode_info *lli = ll_i2info(inode); + if (bits & MDS_INODELOCK_UPDATE) + ll_file_set_flag(lli, LLIF_UPDATE_ATIME); - CDEBUG(D_INODE, "invalidating inode "DFID" lli = %p, " - "pfid = "DFID"\n", PFID(ll_inode2fid(inode)), - lli, PFID(&lli->lli_pfid)); - truncate_inode_pages(inode->i_mapping, 0); + if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) { + CDEBUG(D_INODE, "invalidating inode "DFID" lli = %p, " + "pfid = "DFID"\n", PFID(ll_inode2fid(inode)), + lli, PFID(&lli->lli_pfid)); + truncate_inode_pages(inode->i_mapping, 0); - if (unlikely(!fid_is_zero(&lli->lli_pfid))) { - struct inode *master_inode = NULL; - unsigned long hash; + if (unlikely(!fid_is_zero(&lli->lli_pfid))) { + struct inode *master_inode = NULL; + unsigned long hash; - /* This is slave inode, since all of the child - * dentry is connected on the master inode, so - * we have to invalidate the negative children - * on master inode */ - CDEBUG(D_INODE, "Invalidate s"DFID" m"DFID"\n", - PFID(ll_inode2fid(inode)), - PFID(&lli->lli_pfid)); + /* This is slave inode, since all of the child dentry + * is connected on the master inode, so we have to + * invalidate the negative children on master inode */ + CDEBUG(D_INODE, "Invalidate s"DFID" m"DFID"\n", + PFID(ll_inode2fid(inode)), PFID(&lli->lli_pfid)); - hash = cl_fid_build_ino(&lli->lli_pfid, + hash = cl_fid_build_ino(&lli->lli_pfid, ll_need_32bit_api(ll_i2sbi(inode))); - /* Do not lookup the inode with ilookup5, - * otherwise it will cause dead lock, - * - * 1. Client1 send chmod req to the MDT0, then - * on MDT0, it enqueues master and all of its - * slaves lock, (mdt_attr_set() -> - * mdt_lock_slaves()), after gets master and - * stripe0 lock, it will send the enqueue req - * (for stripe1) to MDT1, then MDT1 finds the - * lock has been granted to client2. Then MDT1 - * sends blocking ast to client2. - * - * 2. At the same time, client2 tries to unlink - * the striped dir (rm -rf striped_dir), and - * during lookup, it will hold the master inode - * of the striped directory, whose inode state - * is NEW, then tries to revalidate all of its - * slaves, (ll_prep_inode()->ll_iget()-> - * ll_read_inode2()-> ll_update_inode().). And - * it will be blocked on the server side because - * of 1. - * - * 3. Then the client get the blocking_ast req, - * cancel the lock, but being blocked if using - * ->ilookup5()), because master inode state is - * NEW. */ - master_inode = ilookup5_nowait(inode->i_sb, - hash, ll_test_inode_by_fid, + /* Do not lookup the inode with ilookup5, otherwise + * it will cause dead lock, + * 1. Client1 send chmod req to the MDT0, then on MDT0, + * it enqueues master and all of its slaves lock, + * (mdt_attr_set() -> mdt_lock_slaves()), after gets + * master and stripe0 lock, it will send the enqueue + * req (for stripe1) to MDT1, then MDT1 finds the lock + * has been granted to client2. Then MDT1 sends blocking + * ast to client2. + * 2. At the same time, client2 tries to unlink + * the striped dir (rm -rf striped_dir), and during + * lookup, it will hold the master inode of the striped + * directory, whose inode state is NEW, then tries to + * revalidate all of its slaves, (ll_prep_inode()-> + * ll_iget()->ll_read_inode2()-> ll_update_inode().). + * And it will be blocked on the server side because + * of 1. + * 3. Then the client get the blocking_ast req, cancel + * the lock, but being blocked if using ->ilookup5()), + * because master inode state is NEW. */ + master_inode = ilookup5_nowait(inode->i_sb, hash, + ll_test_inode_by_fid, (void *)&lli->lli_pfid); - if (master_inode) { - ll_invalidate_negative_children( - master_inode); - iput(master_inode); - } - } else { - ll_invalidate_negative_children(inode); + if (master_inode) { + ll_invalidate_negative_children(master_inode); + iput(master_inode); } + } else { + ll_invalidate_negative_children(inode); } + } - if ((bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM)) && - inode->i_sb->s_root != NULL && - inode != inode->i_sb->s_root->d_inode) - ll_invalidate_aliases(inode); + if ((bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM)) && + inode->i_sb->s_root != NULL && + inode != inode->i_sb->s_root->d_inode) + ll_invalidate_aliases(inode); - iput(inode); + if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM)) + forget_all_cached_acls(inode); + + iput(inode); + RETURN_EXIT; +} + +/* Check if the given lock may be downgraded instead of canceling and + * that convert is really needed. */ +int ll_md_need_convert(struct ldlm_lock *lock) +{ + struct ldlm_namespace *ns = ldlm_lock_to_ns(lock); + struct inode *inode; + __u64 wanted = lock->l_policy_data.l_inodebits.cancel_bits; + __u64 bits = lock->l_policy_data.l_inodebits.bits & ~wanted; + enum ldlm_mode mode = LCK_MINMODE; + + if (!lock->l_conn_export || + !exp_connect_lock_convert(lock->l_conn_export)) + return 0; + + if (!wanted || !bits || ldlm_is_cancel(lock)) + return 0; + + /* do not convert locks other than DOM for now */ + if (!((bits | wanted) & MDS_INODELOCK_DOM)) + return 0; + + /* We may have already remaining bits in some other lock so + * lock convert will leave us just extra lock for the same bit. + * Check if client has other lock with the same bits and the same + * or lower mode and don't convert if any. + */ + switch (lock->l_req_mode) { + case LCK_PR: + mode = LCK_PR; + /* fallthrough */ + case LCK_PW: + mode |= LCK_CR; + break; + case LCK_CW: + mode = LCK_CW; + /* fallthrough */ + case LCK_CR: + mode |= LCK_CR; + break; + default: + /* do not convert other modes */ + return 0; + } + + /* is lock is too old to be converted? */ + lock_res_and_lock(lock); + if (ktime_after(ktime_get(), + ktime_add(lock->l_last_used, + ktime_set(ns->ns_dirty_age_limit, 0)))) { + unlock_res_and_lock(lock); + return 0; + } + unlock_res_and_lock(lock); + + inode = ll_inode_from_resource_lock(lock); + ll_have_md_lock(inode, &bits, mode); + iput(inode); + return !!(bits); +} + +int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *ld, + void *data, int flag) +{ + struct lustre_handle lockh; + int rc; + + ENTRY; + + switch (flag) { + case LDLM_CB_BLOCKING: + { + __u64 cancel_flags = LCF_ASYNC; + + /* if lock convert is not needed then still have to + * pass lock via ldlm_cli_convert() to keep all states + * correct, set cancel_bits to full lock bits to cause + * full cancel to happen. + */ + if (!ll_md_need_convert(lock)) { + lock_res_and_lock(lock); + lock->l_policy_data.l_inodebits.cancel_bits = + lock->l_policy_data.l_inodebits.bits; + unlock_res_and_lock(lock); + } + rc = ldlm_cli_convert(lock, cancel_flags); + if (!rc) + RETURN(0); + /* continue with cancel otherwise */ + ldlm_lock2handle(lock, &lockh); + rc = ldlm_cli_cancel(&lockh, cancel_flags); + if (rc < 0) { + CDEBUG(D_INODE, "ldlm_cli_cancel: rc = %d\n", rc); + RETURN(rc); + } + break; + } + case LDLM_CB_CANCELING: + { + __u64 to_cancel = lock->l_policy_data.l_inodebits.bits; + + /* Nothing to do for non-granted locks */ + if (!ldlm_is_granted(lock)) + break; + + /* If 'ld' is supplied then bits to be cancelled are passed + * implicitly by lock converting and cancel_bits from 'ld' + * should be used. Otherwise full cancel is being performed + * and lock inodebits are used. + * + * Note: we cannot rely on cancel_bits in lock itself at this + * moment because they can be changed by concurrent thread, + * so ldlm_cli_inodebits_convert() pass cancel bits implicitly + * in 'ld' parameter. + */ + if (ld) { + /* partial bits cancel allowed only during convert */ + LASSERT(ldlm_is_converting(lock)); + /* mask cancel bits by lock bits so only no any unused + * bits are passed to ll_lock_cancel_bits() + */ + to_cancel &= ld->l_policy_data.l_inodebits.cancel_bits; + } + ll_lock_cancel_bits(lock, to_cancel); break; } default: @@ -396,7 +545,7 @@ static struct dentry *ll_find_alias(struct inode *inode, struct dentry *dentry) discon_alias = invalid_alias = NULL; - ll_lock_dcache(inode); + spin_lock(&inode->i_lock); ll_d_hlist_for_each_entry(alias, p, &inode->i_dentry) { LASSERT(alias != dentry); @@ -422,7 +571,7 @@ static struct dentry *ll_find_alias(struct inode *inode, struct dentry *dentry) dget_dlock(alias); spin_unlock(&alias->d_lock); } - ll_unlock_dcache(inode); + spin_unlock(&inode->i_lock); return alias; } @@ -463,7 +612,9 @@ struct dentry *ll_splice_alias(struct inode *inode, struct dentry *de) static int ll_lookup_it_finish(struct ptlrpc_request *request, struct lookup_intent *it, - struct inode *parent, struct dentry **de) + struct inode *parent, struct dentry **de, + void *secctx, __u32 secctxlen, + ktime_t kstart) { struct inode *inode = NULL; __u64 bits = 0; @@ -476,20 +627,55 @@ static int ll_lookup_it_finish(struct ptlrpc_request *request, CDEBUG(D_DENTRY, "it %p it_disposition %x\n", it, it->it_disposition); if (!it_disposition(it, DISP_LOOKUP_NEG)) { - rc = ll_prep_inode(&inode, request, (*de)->d_sb, it); - if (rc) - RETURN(rc); - - ll_set_lock_data(ll_i2sbi(parent)->ll_md_exp, inode, it, &bits); - - /* We used to query real size from OSTs here, but actually - this is not needed. For stat() calls size would be updated - from subsequent do_revalidate()->ll_inode_revalidate_it() in - 2.4 and - vfs_getattr_it->ll_getattr()->ll_inode_revalidate_it() in 2.6 - Everybody else who needs correct file size would call - ll_glimpse_size or some equivalent themselves anyway. - Also see bug 7198. */ + struct req_capsule *pill = &request->rq_pill; + struct mdt_body *body = req_capsule_server_get(pill, + &RMF_MDT_BODY); + + rc = ll_prep_inode(&inode, request, (*de)->d_sb, it); + if (rc) + RETURN(rc); + + if (it->it_op & IT_OPEN) + ll_dom_finish_open(inode, request, it); + + ll_set_lock_data(ll_i2sbi(parent)->ll_md_exp, inode, it, &bits); + + /* We used to query real size from OSTs here, but actually + * this is not needed. For stat() calls size would be updated + * from subsequent do_revalidate()->ll_inode_revalidate_it() in + * 2.4 and + * vfs_getattr_it->ll_getattr()->ll_inode_revalidate_it() in 2.6 + * Everybody else who needs correct file size would call + * ll_glimpse_size or some equivalent themselves anyway. + * Also see bug 7198. + */ + + /* If security context was returned by MDT, put it in + * inode now to save an extra getxattr from security hooks, + * and avoid deadlock. + */ + if (body->mbo_valid & OBD_MD_SECCTX) { + secctx = req_capsule_server_get(pill, &RMF_FILE_SECCTX); + secctxlen = req_capsule_get_size(pill, + &RMF_FILE_SECCTX, + RCL_SERVER); + + if (secctxlen) + CDEBUG(D_SEC, "server returned security context" + " for "DFID"\n", + PFID(ll_inode2fid(inode))); + } + + if (secctx != NULL && secctxlen != 0) { + inode_lock(inode); + rc = security_inode_notifysecctx(inode, secctx, + secctxlen); + inode_unlock(inode); + if (rc) + CWARN("cannot set security context for " + DFID": rc = %d\n", + PFID(ll_inode2fid(inode)), rc); + } } /* Only hash *de if it is unhashed (new dentry). @@ -506,9 +692,9 @@ static int ll_lookup_it_finish(struct ptlrpc_request *request, if (bits & MDS_INODELOCK_LOOKUP) d_lustre_revalidate(*de); } else if (!it_disposition(it, DISP_OPEN_CREATE)) { - /* If file created on server, don't depend on parent UPDATE - * lock to unhide it. It is left hidden and next lookup can - * find it in ll_splice_alias. + /* + * If file was created on the server, the dentry is revalidated + * in ll_create_it if the lock allows for it. */ /* Check that parent has UPDATE lock. */ struct lookup_intent parent_it = { @@ -517,7 +703,7 @@ static int ll_lookup_it_finish(struct ptlrpc_request *request, struct lu_fid fid = ll_i2info(parent)->lli_fid; /* If it is striped directory, get the real stripe parent */ - if (unlikely(ll_i2info(parent)->lli_lsm_md != NULL)) { + if (unlikely(ll_dir_striped(parent))) { rc = md_get_fid_from_lsm(ll_i2mdexp(parent), ll_i2info(parent)->lli_lsm_md, (*de)->d_name.name, @@ -533,26 +719,38 @@ static int ll_lookup_it_finish(struct ptlrpc_request *request, } } + if (it_disposition(it, DISP_OPEN_CREATE)) { + ll_stats_ops_tally(ll_i2sbi(parent), LPROC_LL_MKNOD, + ktime_us_delta(ktime_get(), kstart)); + } + GOTO(out, rc = 0); out: - if (rc != 0 && it->it_op & IT_OPEN) + if (rc != 0 && it->it_op & IT_OPEN) { + ll_intent_drop_lock(it); ll_open_cleanup((*de)->d_sb, request); + } return rc; } static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry, struct lookup_intent *it, - void **secctx, __u32 *secctxlen) + void **secctx, __u32 *secctxlen, + struct pcc_create_attach *pca) { + ktime_t kstart = ktime_get(); struct lookup_intent lookup_it = { .it_op = IT_LOOKUP }; struct dentry *save = dentry, *retval; struct ptlrpc_request *req = NULL; struct md_op_data *op_data = NULL; - __u32 opc; - int rc; - ENTRY; + struct lov_user_md *lum = NULL; + __u32 opc; + int rc; + char secctx_name[XATTR_NAME_MAX + 1]; + + ENTRY; if (dentry->d_name.len > ll_i2sbi(parent)->ll_namelen) RETURN(ERR_PTR(-ENAMETOOLONG)); @@ -574,7 +772,7 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry, } if (it->it_op & IT_OPEN && it->it_flags & FMODE_WRITE && - dentry->d_sb->s_flags & MS_RDONLY) + dentry->d_sb->s_flags & SB_RDONLY) RETURN(ERR_PTR(-EROFS)); if (it->it_op & IT_CREAT) @@ -604,6 +802,54 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry, *secctx = op_data->op_file_secctx; if (secctxlen != NULL) *secctxlen = op_data->op_file_secctx_size; + } else { + if (secctx != NULL) + *secctx = NULL; + if (secctxlen != NULL) + *secctxlen = 0; + } + + /* ask for security context upon intent */ + if (it->it_op & (IT_LOOKUP | IT_GETATTR | IT_OPEN)) { + /* get name of security xattr to request to server */ + rc = ll_listsecurity(parent, secctx_name, + sizeof(secctx_name)); + if (rc < 0) { + CDEBUG(D_SEC, "cannot get security xattr name for " + DFID": rc = %d\n", + PFID(ll_inode2fid(parent)), rc); + } else if (rc > 0) { + op_data->op_file_secctx_name = secctx_name; + op_data->op_file_secctx_name_size = rc; + CDEBUG(D_SEC, "'%.*s' is security xattr for "DFID"\n", + rc, secctx_name, PFID(ll_inode2fid(parent))); + } + } + + if (pca && pca->pca_dataset) { + struct pcc_dataset *dataset = pca->pca_dataset; + + OBD_ALLOC_PTR(lum); + if (lum == NULL) + GOTO(out, retval = ERR_PTR(-ENOMEM)); + + lum->lmm_magic = LOV_USER_MAGIC_V1; + lum->lmm_pattern = LOV_PATTERN_F_RELEASED | LOV_PATTERN_RAID0; + op_data->op_data = lum; + op_data->op_data_size = sizeof(*lum); + op_data->op_archive_id = dataset->pccd_rwid; + + rc = obd_fid_alloc(NULL, ll_i2mdexp(parent), &op_data->op_fid2, + op_data); + if (rc) + GOTO(out, retval = ERR_PTR(rc)); + + rc = pcc_inode_create(parent->i_sb, dataset, &op_data->op_fid2, + &pca->pca_dentry); + if (rc) + GOTO(out, retval = ERR_PTR(rc)); + + it->it_flags |= MDS_OPEN_PCC; } rc = md_intent_lock(ll_i2mdexp(parent), op_data, it, &req, @@ -637,7 +883,12 @@ static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry, if (rc < 0) GOTO(out, retval = ERR_PTR(rc)); - rc = ll_lookup_it_finish(req, it, parent, &dentry); + /* dir layout may change */ + ll_unlock_md_op_lsm(op_data); + rc = ll_lookup_it_finish(req, it, parent, &dentry, + secctx != NULL ? *secctx : NULL, + secctxlen != NULL ? *secctxlen : 0, + kstart); if (rc != 0) { ll_intent_release(it); GOTO(out, retval = ERR_PTR(rc)); @@ -663,6 +914,9 @@ out: ll_finish_md_op_data(op_data); } + if (lum != NULL) + OBD_FREE_PTR(lum); + ptlrpc_req_finished(req); return retval; } @@ -691,7 +945,7 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry, itp = NULL; else itp = ⁢ - de = ll_lookup_it(parent, dentry, itp, NULL, NULL); + de = ll_lookup_it(parent, dentry, itp, NULL, NULL, NULL); if (itp != NULL) ll_intent_release(itp); @@ -699,26 +953,49 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry, return de; } +#ifdef FMODE_CREATED /* added in Linux v4.18-rc1-20-g73a09dd */ +# define ll_is_opened(o, f) ((f)->f_mode & FMODE_OPENED) +# define ll_finish_open(f, d, o) finish_open((f), (d), NULL) +# define ll_last_arg +# define ll_set_created(o, f) \ +do { \ + (f)->f_mode |= FMODE_CREATED; \ +} while (0) + +#else +# define ll_is_opened(o, f) (*(o)) +# define ll_finish_open(f, d, o) finish_open((f), (d), NULL, (o)) +# define ll_last_arg , int *opened +# define ll_set_created(o, f) \ +do { \ + *(o) |= FILE_CREATED; \ +} while (0) + +#endif + /* * For cached negative dentry and new dentry, handle lookup/create/open * together. */ static int ll_atomic_open(struct inode *dir, struct dentry *dentry, struct file *file, unsigned open_flags, - umode_t mode, int *opened) + umode_t mode ll_last_arg) { struct lookup_intent *it; struct dentry *de; long long lookup_flags = LOOKUP_OPEN; void *secctx = NULL; __u32 secctxlen = 0; + struct ll_sb_info *sbi; + struct pcc_create_attach pca = { NULL, NULL }; int rc = 0; ENTRY; CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p), file %p," "open_flags %x, mode %x opened %d\n", dentry->d_name.len, dentry->d_name.name, - PFID(ll_inode2fid(dir)), dir, file, open_flags, mode, *opened); + PFID(ll_inode2fid(dir)), dir, file, open_flags, mode, + ll_is_opened(opened, file)); /* Only negative dentries enter here */ LASSERT(dentry->d_inode == NULL); @@ -746,13 +1023,28 @@ static int ll_atomic_open(struct inode *dir, struct dentry *dentry, if (open_flags & O_CREAT) { it->it_op |= IT_CREAT; lookup_flags |= LOOKUP_CREATE; + sbi = ll_i2sbi(dir); + /* Volatile file is used for HSM restore, so do not use PCC */ + if (!filename_is_volatile(dentry->d_name.name, + dentry->d_name.len, NULL)) { + struct pcc_matcher item; + struct pcc_dataset *dataset; + + item.pm_uid = from_kuid(&init_user_ns, current_uid()); + item.pm_gid = from_kgid(&init_user_ns, current_gid()); + item.pm_projid = ll_i2info(dir)->lli_projid; + item.pm_name = &dentry->d_name; + dataset = pcc_dataset_match_get(&sbi->ll_pcc_super, + &item); + pca.pca_dataset = dataset; + } } it->it_create_mode = (mode & S_IALLUGO) | S_IFREG; it->it_flags = (open_flags & ~O_ACCMODE) | OPEN_FMODE(open_flags); it->it_flags &= ~MDS_OPEN_FL_INTERNAL; /* Dentry added to dcache tree in ll_lookup_it */ - de = ll_lookup_it(dir, dentry, it, &secctx, &secctxlen); + de = ll_lookup_it(dir, dentry, it, &secctx, &secctxlen, &pca); if (IS_ERR(de)) rc = PTR_ERR(de); else if (de != NULL) @@ -772,8 +1064,31 @@ static int ll_atomic_open(struct inode *dir, struct dentry *dentry, goto out_release; } - *opened |= FILE_CREATED; + rc = pcc_inode_create_fini(dentry->d_inode, &pca); + if (rc) { + if (de != NULL) + dput(de); + GOTO(out_release, rc); + } + + ll_set_created(opened, file); + } else { + /* Open the file with O_CREAT, but the file already + * existed on MDT. This may happend in the case that + * the LOOKUP ibits lock is revoked and the + * corresponding dentry cache is deleted. + * i.e. In the current Lustre, the truncate operation + * will revoke the LOOKUP ibits lock, and the file + * dentry cache will be invalidated. The following open + * with O_CREAT flag will call into ->atomic_open, the + * file was wrongly though as newly created file and + * try to auto cache the file. So after client knows it + * is not a DISP_OPEN_CREATE, it should cleanup the + * already created PCC copy. + */ + pcc_create_attach_cleanup(dir->i_sb, &pca); } + if (dentry->d_inode && it_disposition(it, DISP_OPEN_OPEN)) { /* Open dentry. */ if (S_ISFIFO(dentry->d_inode->i_mode)) { @@ -783,7 +1098,7 @@ static int ll_atomic_open(struct inode *dir, struct dentry *dentry, rc = finish_no_open(file, de); } else { file->private_data = it; - rc = finish_open(file, dentry, NULL, opened); + rc = ll_finish_open(file, dentry, opened); /* We dget in ll_splice_alias. finish_open takes * care of dget for fd open. */ @@ -793,6 +1108,8 @@ static int ll_atomic_open(struct inode *dir, struct dentry *dentry, } else { rc = finish_no_open(file, de); } + } else { + pcc_create_attach_cleanup(dir->i_sb, &pca); } out_release: @@ -855,12 +1172,12 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry, it = ll_convert_intent(&nd->intent.open, nd->flags, (nd->path.mnt->mnt_flags & MNT_READONLY) || - (nd->path.mnt->mnt_sb->s_flags & MS_RDONLY)); + (nd->path.mnt->mnt_sb->s_flags & SB_RDONLY)); if (IS_ERR(it)) RETURN((struct dentry *)it); } - de = ll_lookup_it(parent, dentry, it, NULL, NULL); + de = ll_lookup_it(parent, dentry, it, NULL, NULL, NULL); if (de) dentry = de; if ((nd->flags & LOOKUP_OPEN) && !IS_ERR(dentry)) { /* Open */ @@ -900,7 +1217,7 @@ static struct dentry *ll_lookup_nd(struct inode *parent, struct dentry *dentry, OBD_FREE(it, sizeof(*it)); } } else { - de = ll_lookup_it(parent, dentry, NULL, NULL, NULL); + de = ll_lookup_it(parent, dentry, NULL, NULL, NULL, NULL); } RETURN(de); @@ -959,6 +1276,7 @@ static int ll_create_it(struct inode *dir, struct dentry *dentry, void *secctx, __u32 secctxlen) { struct inode *inode; + __u64 bits = 0; int rc = 0; ENTRY; @@ -994,6 +1312,10 @@ static int ll_create_it(struct inode *dir, struct dentry *dentry, RETURN(rc); } + ll_set_lock_data(ll_i2sbi(dir)->ll_md_exp, inode, it, &bits); + if (bits & MDS_INODELOCK_LOOKUP) + d_lustre_revalidate(dentry); + RETURN(0); } @@ -1004,16 +1326,17 @@ void ll_update_times(struct ptlrpc_request *request, struct inode *inode) LASSERT(body); if (body->mbo_valid & OBD_MD_FLMTIME && - body->mbo_mtime > LTIME_S(inode->i_mtime)) { - CDEBUG(D_INODE, "setting fid "DFID" mtime from %lu to %llu" - "\n", PFID(ll_inode2fid(inode)), - LTIME_S(inode->i_mtime), body->mbo_mtime); - LTIME_S(inode->i_mtime) = body->mbo_mtime; + body->mbo_mtime > inode->i_mtime.tv_sec) { + CDEBUG(D_INODE, + "setting fid " DFID " mtime from %lld to %llu\n", + PFID(ll_inode2fid(inode)), + (s64)inode->i_mtime.tv_sec, body->mbo_mtime); + inode->i_mtime.tv_sec = body->mbo_mtime; } if (body->mbo_valid & OBD_MD_FLCTIME && - body->mbo_ctime > LTIME_S(inode->i_ctime)) - LTIME_S(inode->i_ctime) = body->mbo_ctime; + body->mbo_ctime > inode->i_ctime.tv_sec) + inode->i_ctime.tv_sec = body->mbo_ctime; } static int ll_new_node(struct inode *dir, struct dentry *dchild, @@ -1050,13 +1373,11 @@ again: from_kuid(&init_user_ns, current_fsuid()), from_kgid(&init_user_ns, current_fsgid()), cfs_curproc_cap_pack(), rdev, &request); - if (err < 0 && err != -EREMOTE) - GOTO(err_exit, err); - - /* If the client doesn't know where to create a subdirectory (or - * in case of a race that sends the RPC to the wrong MDS), the - * MDS will return -EREMOTE and the client will fetch the layout - * of the directory, then create the directory on the right MDT. */ +#if LUSTRE_VERSION_CODE < OBD_OCD_VERSION(2, 14, 58, 0) + /* + * server < 2.12.58 doesn't pack default LMV in intent_getattr reply, + * fetch default LMV here. + */ if (unlikely(err == -EREMOTE)) { struct ll_inode_info *lli = ll_i2info(dir); struct lmv_user_md *lum; @@ -1068,24 +1389,54 @@ again: err2 = ll_dir_getstripe(dir, (void **)&lum, &lumsize, &request, OBD_MD_DEFAULT_MEA); + ll_finish_md_op_data(op_data); + op_data = NULL; if (err2 == 0) { - /* Update stripe_offset and retry */ - lli->lli_def_stripe_offset = lum->lum_stripe_offset; - } else if (err2 == -ENODATA && - lli->lli_def_stripe_offset != -1) { - /* If there are no default stripe EA on the MDT, but the + struct lustre_md md = { NULL }; + + md.body = req_capsule_server_get(&request->rq_pill, + &RMF_MDT_BODY); + if (!md.body) + GOTO(err_exit, err = -EPROTO); + + OBD_ALLOC_PTR(md.default_lmv); + if (!md.default_lmv) + GOTO(err_exit, err = -ENOMEM); + + md.default_lmv->lsm_md_magic = lum->lum_magic; + md.default_lmv->lsm_md_stripe_count = + lum->lum_stripe_count; + md.default_lmv->lsm_md_master_mdt_index = + lum->lum_stripe_offset; + md.default_lmv->lsm_md_hash_type = lum->lum_hash_type; + + err = ll_update_inode(dir, &md); + md_free_lustre_md(sbi->ll_md_exp, &md); + if (err) + GOTO(err_exit, err); + } else if (err2 == -ENODATA && lli->lli_default_lsm_md) { + /* + * If there are no default stripe EA on the MDT, but the * client has default stripe, then it probably means - * default stripe EA has just been deleted. */ - lli->lli_def_stripe_offset = -1; + * default stripe EA has just been deleted. + */ + down_write(&lli->lli_lsm_sem); + if (lli->lli_default_lsm_md) + OBD_FREE_PTR(lli->lli_default_lsm_md); + lli->lli_default_lsm_md = NULL; + up_write(&lli->lli_lsm_sem); } else { GOTO(err_exit, err); } ptlrpc_req_finished(request); request = NULL; - ll_finish_md_op_data(op_data); goto again; } +#endif + + if (err < 0) + GOTO(err_exit, err); ll_update_times(request, dir); @@ -1131,38 +1482,41 @@ static int ll_mknod(struct inode *dir, struct dentry *dchild, ll_umode_t mode, dev_t rdev) { struct qstr *name = &dchild->d_name; + ktime_t kstart = ktime_get(); int err; - ENTRY; + ENTRY; CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p) mode %o dev %x\n", name->len, name->name, PFID(ll_inode2fid(dir)), dir, - mode, rdev); + mode, rdev); if (!IS_POSIXACL(dir) || !exp_connect_umask(ll_i2mdexp(dir))) mode &= ~current_umask(); - switch (mode & S_IFMT) { - case 0: - mode |= S_IFREG; /* for mode = 0 case, fallthrough */ - case S_IFREG: - case S_IFCHR: - case S_IFBLK: - case S_IFIFO: - case S_IFSOCK: + switch (mode & S_IFMT) { + case 0: + mode |= S_IFREG; + /* fallthrough */ + case S_IFREG: + case S_IFCHR: + case S_IFBLK: + case S_IFIFO: + case S_IFSOCK: err = ll_new_node(dir, dchild, NULL, mode, old_encode_dev(rdev), LUSTRE_OPC_MKNOD); - break; - case S_IFDIR: - err = -EPERM; - break; - default: - err = -EINVAL; - } + break; + case S_IFDIR: + err = -EPERM; + break; + default: + err = -EINVAL; + } - if (!err) - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKNOD, 1); + if (!err) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKNOD, + ktime_us_delta(ktime_get(), kstart)); - RETURN(err); + RETURN(err); } #ifdef HAVE_IOP_ATOMIC_OPEN @@ -1172,6 +1526,7 @@ static int ll_mknod(struct inode *dir, struct dentry *dchild, ll_umode_t mode, static int ll_create_nd(struct inode *dir, struct dentry *dentry, umode_t mode, bool want_excl) { + ktime_t kstart = ktime_get(); int rc; CFS_FAIL_TIMEOUT(OBD_FAIL_LLITE_CREATE_FILE_PAUSE, cfs_fail_val); @@ -1185,11 +1540,13 @@ static int ll_create_nd(struct inode *dir, struct dentry *dentry, * volatile file name, so we use ll_mknod() here. */ rc = ll_mknod(dir, dentry, mode, 0); - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE, 1); - CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, unhashed %d\n", dentry->d_name.len, dentry->d_name.name, d_unhashed(dentry)); + if (!rc) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE, + ktime_us_delta(ktime_get(), kstart)); + return rc; } #else /* !HAVE_IOP_ATOMIC_OPEN */ @@ -1198,6 +1555,7 @@ static int ll_create_nd(struct inode *dir, struct dentry *dentry, { struct ll_dentry_data *lld = ll_d2d(dentry); struct lookup_intent *it = NULL; + ktime_t kstart = ktime_get(); int rc; CFS_FAIL_TIMEOUT(OBD_FAIL_LLITE_CREATE_FILE_PAUSE, cfs_fail_val); @@ -1231,16 +1589,17 @@ static int ll_create_nd(struct inode *dir, struct dentry *dentry, filp = lookup_instantiate_filp(nd, dentry, NULL); if (IS_ERR(filp)) rc = PTR_ERR(filp); - } + } out: - ll_intent_release(it); - OBD_FREE(it, sizeof(*it)); + ll_intent_release(it); + OBD_FREE(it, sizeof(*it)); - if (!rc) - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE, 1); + if (!rc) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE, + ktime_us_delta(ktime_get(), kstart)); - return rc; + return rc; } #endif /* HAVE_IOP_ATOMIC_OPEN */ @@ -1248,6 +1607,7 @@ static int ll_symlink(struct inode *dir, struct dentry *dchild, const char *oldpath) { struct qstr *name = &dchild->d_name; + ktime_t kstart = ktime_get(); int err; ENTRY; @@ -1258,10 +1618,11 @@ static int ll_symlink(struct inode *dir, struct dentry *dchild, err = ll_new_node(dir, dchild, oldpath, S_IFLNK | S_IRWXUGO, 0, LUSTRE_OPC_SYMLINK); - if (!err) - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_SYMLINK, 1); + if (!err) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_SYMLINK, + ktime_us_delta(ktime_get(), kstart)); - RETURN(err); + RETURN(err); } static int ll_link(struct dentry *old_dentry, struct inode *dir, @@ -1272,6 +1633,7 @@ static int ll_link(struct dentry *old_dentry, struct inode *dir, struct ll_sb_info *sbi = ll_i2sbi(dir); struct ptlrpc_request *request = NULL; struct md_op_data *op_data; + ktime_t kstart = ktime_get(); int err; ENTRY; @@ -1279,29 +1641,31 @@ static int ll_link(struct dentry *old_dentry, struct inode *dir, "target=%.*s\n", PFID(ll_inode2fid(src)), src, PFID(ll_inode2fid(dir)), dir, name->len, name->name); - op_data = ll_prep_md_op_data(NULL, src, dir, name->name, name->len, - 0, LUSTRE_OPC_ANY, NULL); - if (IS_ERR(op_data)) - RETURN(PTR_ERR(op_data)); + op_data = ll_prep_md_op_data(NULL, src, dir, name->name, name->len, + 0, LUSTRE_OPC_ANY, NULL); + if (IS_ERR(op_data)) + RETURN(PTR_ERR(op_data)); - err = md_link(sbi->ll_md_exp, op_data, &request); - ll_finish_md_op_data(op_data); - if (err) - GOTO(out, err); + err = md_link(sbi->ll_md_exp, op_data, &request); + ll_finish_md_op_data(op_data); + if (err) + GOTO(out, err); - ll_update_times(request, dir); - ll_stats_ops_tally(sbi, LPROC_LL_LINK, 1); - EXIT; + ll_update_times(request, dir); + ll_stats_ops_tally(sbi, LPROC_LL_LINK, + ktime_us_delta(ktime_get(), kstart)); + EXIT; out: - ptlrpc_req_finished(request); - RETURN(err); + ptlrpc_req_finished(request); + RETURN(err); } static int ll_mkdir(struct inode *dir, struct dentry *dchild, ll_umode_t mode) { struct qstr *name = &dchild->d_name; - int err; - ENTRY; + ktime_t kstart = ktime_get(); + int err; + ENTRY; CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n", name->len, name->name, PFID(ll_inode2fid(dir)), dir); @@ -1313,7 +1677,8 @@ static int ll_mkdir(struct inode *dir, struct dentry *dchild, ll_umode_t mode) err = ll_new_node(dir, dchild, NULL, mode, 0, LUSTRE_OPC_MKDIR); if (err == 0) - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKDIR, 1); + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKDIR, + ktime_us_delta(ktime_get(), kstart)); RETURN(err); } @@ -1321,10 +1686,12 @@ static int ll_mkdir(struct inode *dir, struct dentry *dchild, ll_umode_t mode) static int ll_rmdir(struct inode *dir, struct dentry *dchild) { struct qstr *name = &dchild->d_name; - struct ptlrpc_request *request = NULL; - struct md_op_data *op_data; - int rc; - ENTRY; + struct ptlrpc_request *request = NULL; + struct md_op_data *op_data; + ktime_t kstart = ktime_get(); + int rc; + + ENTRY; CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n", name->len, name->name, PFID(ll_inode2fid(dir)), dir); @@ -1341,15 +1708,16 @@ static int ll_rmdir(struct inode *dir, struct dentry *dchild) op_data->op_fid3 = *ll_inode2fid(dchild->d_inode); op_data->op_fid2 = op_data->op_fid3; - rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request); - ll_finish_md_op_data(op_data); - if (rc == 0) { - ll_update_times(request, dir); - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, 1); - } + rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request); + ll_finish_md_op_data(op_data); + if (!rc) + ll_update_times(request, dir); - ptlrpc_req_finished(request); - RETURN(rc); + ptlrpc_req_finished(request); + if (!rc) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, + ktime_us_delta(ktime_get(), kstart)); + RETURN(rc); } /** @@ -1359,6 +1727,7 @@ int ll_rmdir_entry(struct inode *dir, char *name, int namelen) { struct ptlrpc_request *request = NULL; struct md_op_data *op_data; + ktime_t kstart = ktime_get(); int rc; ENTRY; @@ -1372,29 +1741,34 @@ int ll_rmdir_entry(struct inode *dir, char *name, int namelen) op_data->op_cli_flags |= CLI_RM_ENTRY; rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request); ll_finish_md_op_data(op_data); - if (rc == 0) { + if (!rc) ll_update_times(request, dir); - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, 1); - } ptlrpc_req_finished(request); + if (!rc) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, + ktime_us_delta(ktime_get(), kstart)); RETURN(rc); } static int ll_unlink(struct inode *dir, struct dentry *dchild) { struct qstr *name = &dchild->d_name; - struct ptlrpc_request *request = NULL; - struct md_op_data *op_data; - int rc; - ENTRY; + struct ptlrpc_request *request = NULL; + struct md_op_data *op_data; + struct mdt_body *body; + ktime_t kstart = ktime_get(); + int rc; + + ENTRY; + CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n", name->len, name->name, PFID(ll_inode2fid(dir)), dir); - /* - * XXX: unlink bind mountpoint maybe call to here, - * just check it as vfs_unlink does. - */ + /* + * XXX: unlink bind mountpoint maybe call to here, + * just check it as vfs_unlink does. + */ if (unlikely(d_mountpoint(dchild))) RETURN(-EBUSY); @@ -1403,21 +1777,34 @@ static int ll_unlink(struct inode *dir, struct dentry *dchild) if (IS_ERR(op_data)) RETURN(PTR_ERR(op_data)); - if (dchild->d_inode != NULL) - op_data->op_fid3 = *ll_inode2fid(dchild->d_inode); - + op_data->op_fid3 = *ll_inode2fid(dchild->d_inode); + /* notify lower layer if inode has dirty pages */ + if (S_ISREG(dchild->d_inode->i_mode) && + ll_i2info(dchild->d_inode)->lli_clob && + dirty_cnt(dchild->d_inode)) + op_data->op_cli_flags |= CLI_DIRTY_DATA; op_data->op_fid2 = op_data->op_fid3; rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request); ll_finish_md_op_data(op_data); if (rc) GOTO(out, rc); - ll_update_times(request, dir); - ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_UNLINK, 1); + /* + * The server puts attributes in on the last unlink, use them to update + * the link count so the inode can be freed immediately. + */ + body = req_capsule_server_get(&request->rq_pill, &RMF_MDT_BODY); + if (body->mbo_valid & OBD_MD_FLNLINK) + set_nlink(dchild->d_inode, body->mbo_nlink); + + ll_update_times(request, dir); - out: - ptlrpc_req_finished(request); - RETURN(rc); +out: + ptlrpc_req_finished(request); + if (!rc) + ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_UNLINK, + ktime_us_delta(ktime_get(), kstart)); + RETURN(rc); } static int ll_rename(struct inode *src, struct dentry *src_dchild, @@ -1432,6 +1819,7 @@ static int ll_rename(struct inode *src, struct dentry *src_dchild, struct ptlrpc_request *request = NULL; struct ll_sb_info *sbi = ll_i2sbi(src); struct md_op_data *op_data; + ktime_t kstart = ktime_get(); int err; ENTRY; @@ -1460,20 +1848,22 @@ static int ll_rename(struct inode *src, struct dentry *src_dchild, if (tgt_dchild->d_inode != NULL) op_data->op_fid4 = *ll_inode2fid(tgt_dchild->d_inode); - err = md_rename(sbi->ll_md_exp, op_data, - src_name->name, src_name->len, - tgt_name->name, tgt_name->len, &request); - ll_finish_md_op_data(op_data); - if (!err) { - ll_update_times(request, src); - ll_update_times(request, tgt); - ll_stats_ops_tally(sbi, LPROC_LL_RENAME, 1); - } + err = md_rename(sbi->ll_md_exp, op_data, + src_name->name, src_name->len, + tgt_name->name, tgt_name->len, &request); + ll_finish_md_op_data(op_data); + if (!err) { + ll_update_times(request, src); + ll_update_times(request, tgt); + } - ptlrpc_req_finished(request); + ptlrpc_req_finished(request); - if (err == 0) + if (!err) { d_move(src_dchild, tgt_dchild); + ll_stats_ops_tally(sbi, LPROC_LL_RENAME, + ktime_us_delta(ktime_get(), kstart)); + } RETURN(err); }