X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Flmv%2Flmv_obd.c;h=ba1b6560399a06c4b99cbef4d38eba65bb28af7e;hp=19bbe64f569030c082b8af3f5b44f3c870915472;hb=002c2a80266b23c1df02d554fbdc7e5817c42d13;hpb=beadbad429aeffa66898bd62cc99e76290644ca8 diff --git a/lustre/lmv/lmv_obd.c b/lustre/lmv/lmv_obd.c index 19bbe64..ba1b656 100644 --- a/lustre/lmv/lmv_obd.c +++ b/lustre/lmv/lmv_obd.c @@ -23,21 +23,19 @@ * Copyright (c) 2004, 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/ - * Lustre is a trademark of Sun Microsystems, Inc. */ #define DEBUG_SUBSYSTEM S_LMV -#include + +#include #include #include #include -#ifdef HAVE_UIDGID_HEADER -# include -#endif +#include #include #include #include @@ -59,15 +57,15 @@ static int lmv_check_connect(struct obd_device *obd); -static void lmv_activate_target(struct lmv_obd *lmv, - struct lmv_tgt_desc *tgt, - int activate) +void lmv_activate_target(struct lmv_obd *lmv, struct lmv_tgt_desc *tgt, + int activate) { - if (tgt->ltd_active == activate) - return; + if (tgt->ltd_active == activate) + return; - tgt->ltd_active = activate; - lmv->desc.ld_active_tgt_count += (activate ? 1 : -1); + tgt->ltd_active = activate; + lmv->lmv_mdt_descs.ltd_lmv_desc.ld_active_tgt_count += + (activate ? 1 : -1); tgt->ltd_exp->exp_obd->obd_inactive = !activate; } @@ -83,62 +81,59 @@ static int lmv_set_mdc_active(struct lmv_obd *lmv, const struct obd_uuid *uuid, int activate) { - struct lmv_tgt_desc *tgt = NULL; - struct obd_device *obd; - __u32 i; - int rc = 0; + struct lu_tgt_desc *tgt = NULL; + struct obd_device *obd; + int rc = 0; + ENTRY; CDEBUG(D_INFO, "Searching in lmv %p for uuid %s (activate=%d)\n", lmv, uuid->uuid, activate); spin_lock(&lmv->lmv_lock); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - tgt = lmv->tgts[i]; - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; - - CDEBUG(D_INFO, "Target idx %d is %s conn %#llx\n", i, - tgt->ltd_uuid.uuid, tgt->ltd_exp->exp_handle.h_cookie); + lmv_foreach_connected_tgt(lmv, tgt) { + CDEBUG(D_INFO, "Target idx %d is %s conn %#llx\n", + tgt->ltd_index, tgt->ltd_uuid.uuid, + tgt->ltd_exp->exp_handle.h_cookie); if (obd_uuid_equals(uuid, &tgt->ltd_uuid)) break; } - if (i == lmv->desc.ld_tgt_count) - GOTO(out_lmv_lock, rc = -EINVAL); + if (!tgt) + GOTO(out_lmv_lock, rc = -EINVAL); - obd = class_exp2obd(tgt->ltd_exp); - if (obd == NULL) - GOTO(out_lmv_lock, rc = -ENOTCONN); + obd = class_exp2obd(tgt->ltd_exp); + if (obd == NULL) + GOTO(out_lmv_lock, rc = -ENOTCONN); - CDEBUG(D_INFO, "Found OBD %s=%s device %d (%p) type %s at LMV idx %d\n", - obd->obd_name, obd->obd_uuid.uuid, obd->obd_minor, obd, - obd->obd_type->typ_name, i); - LASSERT(strcmp(obd->obd_type->typ_name, LUSTRE_MDC_NAME) == 0); + CDEBUG(D_INFO, "Found OBD %s=%s device %d (%p) type %s at LMV idx %d\n", + obd->obd_name, obd->obd_uuid.uuid, obd->obd_minor, obd, + obd->obd_type->typ_name, tgt->ltd_index); + LASSERT(strcmp(obd->obd_type->typ_name, LUSTRE_MDC_NAME) == 0); - if (tgt->ltd_active == activate) { - CDEBUG(D_INFO, "OBD %p already %sactive!\n", obd, - activate ? "" : "in"); - GOTO(out_lmv_lock, rc); - } + if (tgt->ltd_active == activate) { + CDEBUG(D_INFO, "OBD %p already %sactive!\n", obd, + activate ? "" : "in"); + GOTO(out_lmv_lock, rc); + } - CDEBUG(D_INFO, "Marking OBD %p %sactive\n", obd, - activate ? "" : "in"); - lmv_activate_target(lmv, tgt, activate); - EXIT; + CDEBUG(D_INFO, "Marking OBD %p %sactive\n", obd, + activate ? "" : "in"); + lmv_activate_target(lmv, tgt, activate); + EXIT; out_lmv_lock: spin_unlock(&lmv->lmv_lock); return rc; } -struct obd_uuid *lmv_get_uuid(struct obd_export *exp) +static struct obd_uuid *lmv_get_uuid(struct obd_export *exp) { - struct lmv_obd *lmv = &exp->exp_obd->u.lmv; - struct lmv_tgt_desc *tgt = lmv->tgts[0]; + struct lmv_obd *lmv = &exp->exp_obd->u.lmv; + struct lmv_tgt_desc *tgt = lmv_tgt(lmv, 0); - return (tgt == NULL) ? NULL : obd_get_uuid(tgt->ltd_exp); + return tgt ? obd_get_uuid(tgt->ltd_exp) : NULL; } static int lmv_notify(struct obd_device *obd, struct obd_device *watched, @@ -210,33 +205,27 @@ static int lmv_connect(const struct lu_env *env, exp = class_conn2export(&conn); lmv->connected = 0; - lmv->cluuid = *cluuid; lmv->conn_data = *data; + lmv->lmv_cache = localdata; - if (lmv->targets_proc_entry == NULL) { - lmv->targets_proc_entry = lprocfs_register("target_obds", - obd->obd_proc_entry, - NULL, NULL); - if (IS_ERR(lmv->targets_proc_entry)) { - CERROR("%s: cannot register " - "/proc/fs/lustre/%s/%s/target_obds\n", - obd->obd_name, obd->obd_type->typ_name, - obd->obd_name); - lmv->targets_proc_entry = NULL; - } + lmv->lmv_tgts_kobj = kobject_create_and_add("target_obds", + &obd->obd_kset.kobj); + if (!lmv->lmv_tgts_kobj) { + CERROR("%s: cannot create /sys/fs/lustre/%s/%s/target_obds\n", + obd->obd_name, obd->obd_type->typ_name, obd->obd_name); } rc = lmv_check_connect(obd); if (rc != 0) - GOTO(out_proc, rc); + GOTO(out_sysfs, rc); *pexp = exp; RETURN(rc); -out_proc: - if (lmv->targets_proc_entry != NULL) - lprocfs_remove(&lmv->targets_proc_entry); +out_sysfs: + if (lmv->lmv_tgts_kobj) + kobject_put(lmv->lmv_tgts_kobj); class_disconnect(exp); @@ -246,21 +235,22 @@ out_proc: static int lmv_init_ea_size(struct obd_export *exp, __u32 easize, __u32 def_easize) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - __u32 i; - int rc = 0; - int change = 0; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int change = 0; + int rc = 0; + ENTRY; - if (lmv->max_easize < easize) { - lmv->max_easize = easize; - change = 1; - } - if (lmv->max_def_easize < def_easize) { - lmv->max_def_easize = def_easize; - change = 1; - } + if (lmv->max_easize < easize) { + lmv->max_easize = easize; + change = 1; + } + if (lmv->max_def_easize < def_easize) { + lmv->max_def_easize = def_easize; + change = 1; + } if (change == 0) RETURN(0); @@ -268,18 +258,14 @@ static int lmv_init_ea_size(struct obd_export *exp, __u32 easize, if (lmv->connected == 0) RETURN(0); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - struct lmv_tgt_desc *tgt = lmv->tgts[i]; - - if (tgt == NULL || tgt->ltd_exp == NULL || !tgt->ltd_active) { - CWARN("%s: NULL export for %d\n", obd->obd_name, i); + lmv_foreach_connected_tgt(lmv, tgt) { + if (!tgt->ltd_active) continue; - } rc = md_init_ea_size(tgt->ltd_exp, easize, def_easize); if (rc) { CERROR("%s: obd_init_ea_size() failed on MDT target %d:" - " rc = %d\n", obd->obd_name, i, rc); + " rc = %d\n", obd->obd_name, tgt->ltd_index, rc); break; } } @@ -288,40 +274,37 @@ static int lmv_init_ea_size(struct obd_export *exp, __u32 easize, #define MAX_STRING_SIZE 128 -int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) +static int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) { - struct lmv_obd *lmv = &obd->u.lmv; - struct obd_uuid *cluuid = &lmv->cluuid; - struct obd_uuid lmv_mdc_uuid = { "LMV_MDC_UUID" }; - struct obd_device *mdc_obd; - struct obd_export *mdc_exp; - struct lu_fld_target target; - int rc; - ENTRY; + struct lmv_obd *lmv = &obd->u.lmv; + struct obd_device *mdc_obd; + struct obd_export *mdc_exp; + struct lu_fld_target target; + int rc; + ENTRY; - mdc_obd = class_find_client_obd(&tgt->ltd_uuid, LUSTRE_MDC_NAME, - &obd->obd_uuid); - if (!mdc_obd) { - CERROR("target %s not attached\n", tgt->ltd_uuid.uuid); - RETURN(-EINVAL); - } + mdc_obd = class_find_client_obd(&tgt->ltd_uuid, LUSTRE_MDC_NAME, + &obd->obd_uuid); + if (!mdc_obd) { + CERROR("target %s not attached\n", tgt->ltd_uuid.uuid); + RETURN(-EINVAL); + } - CDEBUG(D_CONFIG, "connect to %s(%s) - %s, %s FOR %s\n", - mdc_obd->obd_name, mdc_obd->obd_uuid.uuid, - tgt->ltd_uuid.uuid, obd->obd_uuid.uuid, - cluuid->uuid); + CDEBUG(D_CONFIG, "connect to %s(%s) - %s, %s\n", + mdc_obd->obd_name, mdc_obd->obd_uuid.uuid, + tgt->ltd_uuid.uuid, obd->obd_uuid.uuid); - if (!mdc_obd->obd_set_up) { - CERROR("target %s is not set up\n", tgt->ltd_uuid.uuid); - RETURN(-EINVAL); - } + if (!mdc_obd->obd_set_up) { + CERROR("target %s is not set up\n", tgt->ltd_uuid.uuid); + RETURN(-EINVAL); + } - rc = obd_connect(NULL, &mdc_exp, mdc_obd, &lmv_mdc_uuid, - &lmv->conn_data, NULL); - if (rc) { - CERROR("target %s connect error %d\n", tgt->ltd_uuid.uuid, rc); - RETURN(rc); - } + rc = obd_connect(NULL, &mdc_exp, mdc_obd, &obd->obd_uuid, + &lmv->conn_data, lmv->lmv_cache); + if (rc) { + CERROR("target %s connect error %d\n", tgt->ltd_uuid.uuid, rc); + RETURN(rc); + } /* * Init fid sequence client for this mdc and add new fld target. @@ -330,19 +313,19 @@ int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) if (rc) RETURN(rc); - target.ft_srv = NULL; - target.ft_exp = mdc_exp; - target.ft_idx = tgt->ltd_idx; + target.ft_srv = NULL; + target.ft_exp = mdc_exp; + target.ft_idx = tgt->ltd_index; - fld_client_add_target(&lmv->lmv_fld, &target); + fld_client_add_target(&lmv->lmv_fld, &target); - rc = obd_register_observer(mdc_obd, obd); - if (rc) { - obd_disconnect(mdc_exp); - CERROR("target %s register_observer error %d\n", - tgt->ltd_uuid.uuid, rc); - RETURN(rc); - } + rc = obd_register_observer(mdc_obd, obd); + if (rc) { + obd_disconnect(mdc_exp); + CERROR("target %s register_observer error %d\n", + tgt->ltd_uuid.uuid, rc); + RETURN(rc); + } if (obd->obd_observer) { /* @@ -358,53 +341,47 @@ int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) tgt->ltd_active = 1; tgt->ltd_exp = mdc_exp; - lmv->desc.ld_active_tgt_count++; + lmv->lmv_mdt_descs.ltd_lmv_desc.ld_active_tgt_count++; md_init_ea_size(tgt->ltd_exp, lmv->max_easize, lmv->max_def_easize); - CDEBUG(D_CONFIG, "Connected to %s(%s) successfully (%d)\n", - mdc_obd->obd_name, mdc_obd->obd_uuid.uuid, - atomic_read(&obd->obd_refcount)); - - if (lmv->targets_proc_entry != NULL) { - struct proc_dir_entry *mdc_symlink; - - LASSERT(mdc_obd->obd_type != NULL); - LASSERT(mdc_obd->obd_type->typ_name != NULL); - mdc_symlink = lprocfs_add_symlink(mdc_obd->obd_name, - lmv->targets_proc_entry, - "../../../%s/%s", - mdc_obd->obd_type->typ_name, - mdc_obd->obd_name); - if (mdc_symlink == NULL) { - CERROR("cannot register LMV target " - "/proc/fs/lustre/%s/%s/target_obds/%s\n", - obd->obd_type->typ_name, obd->obd_name, - mdc_obd->obd_name); - } + rc = lu_qos_add_tgt(&lmv->lmv_qos, tgt); + if (rc) { + obd_disconnect(mdc_exp); + RETURN(rc); } + + CDEBUG(D_CONFIG, "Connected to %s(%s) successfully (%d)\n", + mdc_obd->obd_name, mdc_obd->obd_uuid.uuid, + atomic_read(&obd->obd_refcount)); + + lmv_statfs_check_update(obd, tgt); + + if (lmv->lmv_tgts_kobj) + /* Even if we failed to create the link, that's fine */ + rc = sysfs_create_link(lmv->lmv_tgts_kobj, + &mdc_obd->obd_kset.kobj, + mdc_obd->obd_name); RETURN(0); } -static void lmv_del_target(struct lmv_obd *lmv, int index) +static void lmv_del_target(struct lmv_obd *lmv, struct lu_tgt_desc *tgt) { - if (lmv->tgts[index] == NULL) - return; - - OBD_FREE_PTR(lmv->tgts[index]); - lmv->tgts[index] = NULL; - return; + LASSERT(tgt); + ltd_del_tgt(&lmv->lmv_mdt_descs, tgt); + OBD_FREE_PTR(tgt); } static int lmv_add_target(struct obd_device *obd, struct obd_uuid *uuidp, __u32 index, int gen) { struct obd_device *mdc_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int orig_tgt_count = 0; - int rc = 0; - ENTRY; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + struct lu_tgt_descs *ltd = &lmv->lmv_mdt_descs; + int rc = 0; + + ENTRY; CDEBUG(D_CONFIG, "Target uuid: %s. index %d\n", uuidp->uuid, index); mdc_obd = class_find_client_obd(uuidp, LUSTRE_MDC_NAME, @@ -415,237 +392,162 @@ static int lmv_add_target(struct obd_device *obd, struct obd_uuid *uuidp, RETURN(-EINVAL); } - mutex_lock(&lmv->lmv_init_mutex); - if ((index < lmv->tgts_size) && (lmv->tgts[index] != NULL)) { - tgt = lmv->tgts[index]; - CERROR("%s: UUID %s already assigned at LOV target index %d:" - " rc = %d\n", obd->obd_name, - obd_uuid2str(&tgt->ltd_uuid), index, -EEXIST); - mutex_unlock(&lmv->lmv_init_mutex); - RETURN(-EEXIST); - } - - if (index >= lmv->tgts_size) { - /* We need to reallocate the lmv target array. */ - struct lmv_tgt_desc **newtgts, **old = NULL; - __u32 newsize = 1; - __u32 oldsize = 0; - - while (newsize < index + 1) - newsize = newsize << 1; - OBD_ALLOC(newtgts, sizeof(*newtgts) * newsize); - if (newtgts == NULL) { - mutex_unlock(&lmv->lmv_init_mutex); - RETURN(-ENOMEM); - } - - if (lmv->tgts_size) { - memcpy(newtgts, lmv->tgts, - sizeof(*newtgts) * lmv->tgts_size); - old = lmv->tgts; - oldsize = lmv->tgts_size; - } - - lmv->tgts = newtgts; - lmv->tgts_size = newsize; - smp_rmb(); - if (old) - OBD_FREE(old, sizeof(*old) * oldsize); - - CDEBUG(D_CONFIG, "tgts: %p size: %d\n", lmv->tgts, - lmv->tgts_size); - } - OBD_ALLOC_PTR(tgt); - if (!tgt) { - mutex_unlock(&lmv->lmv_init_mutex); + if (!tgt) RETURN(-ENOMEM); - } mutex_init(&tgt->ltd_fid_mutex); - tgt->ltd_idx = index; + tgt->ltd_index = index; tgt->ltd_uuid = *uuidp; tgt->ltd_active = 0; - lmv->tgts[index] = tgt; - if (index >= lmv->desc.ld_tgt_count) { - orig_tgt_count = lmv->desc.ld_tgt_count; - lmv->desc.ld_tgt_count = index + 1; - } - if (lmv->connected == 0) { + mutex_lock(<d->ltd_mutex); + rc = ltd_add_tgt(ltd, tgt); + mutex_unlock(<d->ltd_mutex); + + if (rc) + GOTO(out_tgt, rc); + + if (!lmv->connected) /* lmv_check_connect() will connect this target. */ - mutex_unlock(&lmv->lmv_init_mutex); RETURN(0); - } - /* Otherwise let's connect it ourselves */ - mutex_unlock(&lmv->lmv_init_mutex); rc = lmv_connect_mdc(obd, tgt); - if (rc != 0) { - spin_lock(&lmv->lmv_lock); - if (lmv->desc.ld_tgt_count == index + 1) - lmv->desc.ld_tgt_count = orig_tgt_count; - memset(tgt, 0, sizeof(*tgt)); - spin_unlock(&lmv->lmv_lock); - } else { + if (!rc) { int easize = sizeof(struct lmv_stripe_md) + - lmv->desc.ld_tgt_count * sizeof(struct lu_fid); + lmv->lmv_mdt_count * sizeof(struct lu_fid); + lmv_init_ea_size(obd->obd_self_export, easize, 0); } RETURN(rc); + +out_tgt: + OBD_FREE_PTR(tgt); + return rc; } static int lmv_check_connect(struct obd_device *obd) { - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - __u32 i; - int rc; - int easize; - ENTRY; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int easize; + int rc; - if (lmv->connected) - RETURN(0); + ENTRY; - mutex_lock(&lmv->lmv_init_mutex); - if (lmv->connected) { - mutex_unlock(&lmv->lmv_init_mutex); - RETURN(0); - } + if (lmv->connected) + RETURN(0); - if (lmv->desc.ld_tgt_count == 0) { - mutex_unlock(&lmv->lmv_init_mutex); - CERROR("%s: no targets configured.\n", obd->obd_name); - RETURN(-EINVAL); - } + mutex_lock(&lmv->lmv_mdt_descs.ltd_mutex); + if (lmv->connected) + GOTO(unlock, rc = 0); - LASSERT(lmv->tgts != NULL); + if (!lmv->lmv_mdt_count) { + CERROR("%s: no targets configured: rc = -EINVAL\n", + obd->obd_name); + GOTO(unlock, rc = -EINVAL); + } - if (lmv->tgts[0] == NULL) { - mutex_unlock(&lmv->lmv_init_mutex); - CERROR("%s: no target configured for index 0.\n", + if (!lmv_mdt0_inited(lmv)) { + CERROR("%s: no target configured for index 0: rc = -EINVAL.\n", obd->obd_name); - RETURN(-EINVAL); + GOTO(unlock, rc = -EINVAL); } CDEBUG(D_CONFIG, "Time to connect %s to %s\n", - lmv->cluuid.uuid, obd->obd_name); + obd->obd_uuid.uuid, obd->obd_name); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - tgt = lmv->tgts[i]; - if (tgt == NULL) - continue; + lmv_foreach_tgt(lmv, tgt) { rc = lmv_connect_mdc(obd, tgt); if (rc) GOTO(out_disc, rc); } lmv->connected = 1; - easize = lmv_mds_md_size(lmv->desc.ld_tgt_count, LMV_MAGIC); + easize = lmv_mds_md_size(lmv->lmv_mdt_count, LMV_MAGIC); lmv_init_ea_size(obd->obd_self_export, easize, 0); - mutex_unlock(&lmv->lmv_init_mutex); - RETURN(0); + EXIT; +unlock: + mutex_unlock(&lmv->lmv_mdt_descs.ltd_mutex); + + return rc; - out_disc: - while (i-- > 0) { - int rc2; - tgt = lmv->tgts[i]; - if (tgt == NULL) +out_disc: + lmv_foreach_tgt(lmv, tgt) { + tgt->ltd_active = 0; + if (!tgt->ltd_exp) continue; - tgt->ltd_active = 0; - if (tgt->ltd_exp) { - --lmv->desc.ld_active_tgt_count; - rc2 = obd_disconnect(tgt->ltd_exp); - if (rc2) { - CERROR("LMV target %s disconnect on " - "MDC idx %d: error %d\n", - tgt->ltd_uuid.uuid, i, rc2); - } - } - } - mutex_unlock(&lmv->lmv_init_mutex); + --lmv->lmv_mdt_descs.ltd_lmv_desc.ld_active_tgt_count; + obd_disconnect(tgt->ltd_exp); + } - RETURN(rc); + goto unlock; } static int lmv_disconnect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) { - struct lmv_obd *lmv = &obd->u.lmv; - struct obd_device *mdc_obd; - int rc; - ENTRY; + struct lmv_obd *lmv = &obd->u.lmv; + struct obd_device *mdc_obd; + int rc; + ENTRY; - LASSERT(tgt != NULL); - LASSERT(obd != NULL); + LASSERT(tgt != NULL); + LASSERT(obd != NULL); - mdc_obd = class_exp2obd(tgt->ltd_exp); + mdc_obd = class_exp2obd(tgt->ltd_exp); - if (mdc_obd) { - mdc_obd->obd_force = obd->obd_force; - mdc_obd->obd_fail = obd->obd_fail; - mdc_obd->obd_no_recov = obd->obd_no_recov; + if (mdc_obd) { + mdc_obd->obd_force = obd->obd_force; + mdc_obd->obd_fail = obd->obd_fail; + mdc_obd->obd_no_recov = obd->obd_no_recov; - if (lmv->targets_proc_entry != NULL) - lprocfs_remove_proc_entry(mdc_obd->obd_name, - lmv->targets_proc_entry); + if (lmv->lmv_tgts_kobj) + sysfs_remove_link(lmv->lmv_tgts_kobj, + mdc_obd->obd_name); } rc = obd_fid_fini(tgt->ltd_exp->exp_obd); if (rc) - CERROR("Can't finanize fids factory\n"); - - CDEBUG(D_INFO, "Disconnected from %s(%s) successfully\n", - tgt->ltd_exp->exp_obd->obd_name, - tgt->ltd_exp->exp_obd->obd_uuid.uuid); - - obd_register_observer(tgt->ltd_exp->exp_obd, NULL); - rc = obd_disconnect(tgt->ltd_exp); - if (rc) { - if (tgt->ltd_active) { - CERROR("Target %s disconnect error %d\n", - tgt->ltd_uuid.uuid, rc); - } - } + CERROR("Can't finalize fids factory\n"); + + CDEBUG(D_INFO, "Disconnected from %s(%s) successfully\n", + tgt->ltd_exp->exp_obd->obd_name, + tgt->ltd_exp->exp_obd->obd_uuid.uuid); + + obd_register_observer(tgt->ltd_exp->exp_obd, NULL); + rc = obd_disconnect(tgt->ltd_exp); + if (rc) { + if (tgt->ltd_active) { + CERROR("Target %s disconnect error %d\n", + tgt->ltd_uuid.uuid, rc); + } + } - lmv_activate_target(lmv, tgt, 0); - tgt->ltd_exp = NULL; - RETURN(0); + lmv_activate_target(lmv, tgt, 0); + tgt->ltd_exp = NULL; + RETURN(0); } static int lmv_disconnect(struct obd_export *exp) { - struct obd_device *obd = class_exp2obd(exp); - struct lmv_obd *lmv = &obd->u.lmv; - int rc; - __u32 i; - ENTRY; + struct obd_device *obd = class_exp2obd(exp); + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; - if (!lmv->tgts) - goto out_local; + ENTRY; - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - if (lmv->tgts[i] == NULL || lmv->tgts[i]->ltd_exp == NULL) - continue; + lmv_foreach_connected_tgt(lmv, tgt) + lmv_disconnect_mdc(obd, tgt); - lmv_disconnect_mdc(obd, lmv->tgts[i]); - } + if (lmv->lmv_tgts_kobj) + kobject_put(lmv->lmv_tgts_kobj); - if (lmv->targets_proc_entry != NULL) - lprocfs_remove(&lmv->targets_proc_entry); - else - CERROR("/proc/fs/lustre/%s/%s/target_obds missing\n", - obd->obd_type->typ_name, obd->obd_name); - -out_local: - /* - * This is the case when no real connection is established by - * lmv_check_connect(). - */ - if (!lmv->connected) - class_export_put(exp); - rc = class_disconnect(exp); + if (!lmv->connected) + class_export_put(exp); + rc = class_disconnect(exp); lmv->connected = 0; RETURN(rc); @@ -654,17 +556,17 @@ out_local: static int lmv_fid2path(struct obd_export *exp, int len, void *karg, void __user *uarg) { - struct obd_device *obddev = class_exp2obd(exp); - struct lmv_obd *lmv = &obddev->u.lmv; + struct obd_device *obd = class_exp2obd(exp); + struct lmv_obd *lmv = &obd->u.lmv; struct getinfo_fid2path *gf; - struct lmv_tgt_desc *tgt; + struct lmv_tgt_desc *tgt; struct getinfo_fid2path *remote_gf = NULL; - struct lu_fid root_fid; - int remote_gf_size = 0; - int rc; + struct lu_fid root_fid; + int remote_gf_size = 0; + int rc; gf = karg; - tgt = lmv_find_target(lmv, &gf->gf_fid); + tgt = lmv_fid2tgt(lmv, &gf->gf_fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -681,6 +583,7 @@ repeat_fid2path: if (remote_gf != NULL) { struct getinfo_fid2path *ori_gf; char *ptr; + int len; ori_gf = (struct getinfo_fid2path *)karg; if (strlen(ori_gf->gf_u.gf_path) + 1 + @@ -689,13 +592,12 @@ repeat_fid2path: ptr = ori_gf->gf_u.gf_path; - memmove(ptr + strlen(gf->gf_u.gf_path) + 1, ptr, - strlen(ori_gf->gf_u.gf_path)); - - strncpy(ptr, gf->gf_u.gf_path, - strlen(gf->gf_u.gf_path)); - ptr += strlen(gf->gf_u.gf_path); - *ptr = '/'; + len = strlen(gf->gf_u.gf_path); + /* move the current path to the right to release space + * for closer-to-root part */ + memmove(ptr + len + 1, ptr, strlen(ori_gf->gf_u.gf_path)); + memcpy(ptr, gf->gf_u.gf_path, len); + ptr[len] = '/'; } CDEBUG(D_INFO, "%s: get path %s "DFID" rec: %llu ln: %u\n", @@ -722,7 +624,7 @@ repeat_fid2path: GOTO(out_fid2path, rc = -EINVAL); } - tgt = lmv_find_target(lmv, &gf->gf_fid); + tgt = lmv_fid2tgt(lmv, &gf->gf_fid); if (IS_ERR(tgt)) GOTO(out_fid2path, rc = -EINVAL); @@ -744,13 +646,13 @@ static int lmv_hsm_req_count(struct lmv_obd *lmv, const struct hsm_user_request *hur, const struct lmv_tgt_desc *tgt_mds) { - __u32 i; - int nr = 0; - struct lmv_tgt_desc *curr_tgt; + struct lmv_tgt_desc *curr_tgt; + __u32 i; + int nr = 0; /* count how many requests must be sent to the given target */ for (i = 0; i < hur->hur_request.hr_itemcount; i++) { - curr_tgt = lmv_find_target(lmv, &hur->hur_user_item[i].hui_fid); + curr_tgt = lmv_fid2tgt(lmv, &hur->hur_user_item[i].hui_fid); if (IS_ERR(curr_tgt)) RETURN(PTR_ERR(curr_tgt)); if (obd_uuid_equals(&curr_tgt->ltd_uuid, &tgt_mds->ltd_uuid)) @@ -764,15 +666,14 @@ static int lmv_hsm_req_build(struct lmv_obd *lmv, const struct lmv_tgt_desc *tgt_mds, struct hsm_user_request *hur_out) { - __u32 i, nr_out; - struct lmv_tgt_desc *curr_tgt; + __u32 i, nr_out; + struct lmv_tgt_desc *curr_tgt; /* build the hsm_user_request for the given target */ hur_out->hur_request = hur_in->hur_request; nr_out = 0; for (i = 0; i < hur_in->hur_request.hr_itemcount; i++) { - curr_tgt = lmv_find_target(lmv, - &hur_in->hur_user_item[i].hui_fid); + curr_tgt = lmv_fid2tgt(lmv, &hur_in->hur_user_item[i].hui_fid); if (IS_ERR(curr_tgt)) RETURN(PTR_ERR(curr_tgt)); if (obd_uuid_equals(&curr_tgt->ltd_uuid, &tgt_mds->ltd_uuid)) { @@ -788,52 +689,81 @@ static int lmv_hsm_req_build(struct lmv_obd *lmv, RETURN(0); } -static int lmv_hsm_ct_unregister(struct lmv_obd *lmv, unsigned int cmd, int len, - struct lustre_kernelcomm *lk, +static int lmv_hsm_ct_unregister(struct obd_device *obd, unsigned int cmd, + int len, struct lustre_kernelcomm *lk, void __user *uarg) { - __u32 i; - int rc; + struct lmv_obd *lmv = &obd->u.lmv; + struct lu_tgt_desc *tgt; + int rc; + ENTRY; /* unregister request (call from llapi_hsm_copytool_fini) */ - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - struct lmv_tgt_desc *tgt = lmv->tgts[i]; - - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; + lmv_foreach_connected_tgt(lmv, tgt) /* best effort: try to clean as much as possible * (continue on error) */ obd_iocontrol(cmd, tgt->ltd_exp, len, lk, uarg); - } /* Whatever the result, remove copytool from kuc groups. * Unreached coordinators will get EPIPE on next requests * and will unregister automatically. */ - rc = libcfs_kkuc_group_rem(lk->lk_uid, lk->lk_group); + rc = libcfs_kkuc_group_rem(&obd->obd_uuid, lk->lk_uid, lk->lk_group); RETURN(rc); } -static int lmv_hsm_ct_register(struct lmv_obd *lmv, unsigned int cmd, int len, - struct lustre_kernelcomm *lk, __user void *uarg) +static int lmv_hsm_ct_register(struct obd_device *obd, unsigned int cmd, + int len, struct lustre_kernelcomm *lk, + void __user *uarg) { - struct file *filp; - __u32 i, j; - int err, rc; - bool any_set = false; - struct kkuc_ct_data kcd = { 0 }; + struct lmv_obd *lmv = &obd->u.lmv; + struct file *filp; + bool any_set = false; + struct kkuc_ct_data *kcd; + size_t kcd_size; + struct lu_tgt_desc *tgt; + __u32 i; + int err; + int rc = 0; + ENTRY; + filp = fget(lk->lk_wfd); + if (!filp) + RETURN(-EBADF); + + if (lk->lk_flags & LK_FLG_DATANR) + kcd_size = offsetof(struct kkuc_ct_data, + kcd_archives[lk->lk_data_count]); + else + kcd_size = sizeof(*kcd); + + OBD_ALLOC(kcd, kcd_size); + if (kcd == NULL) + GOTO(err_fput, rc = -ENOMEM); + + kcd->kcd_nr_archives = lk->lk_data_count; + if (lk->lk_flags & LK_FLG_DATANR) { + kcd->kcd_magic = KKUC_CT_DATA_ARRAY_MAGIC; + if (lk->lk_data_count > 0) + memcpy(kcd->kcd_archives, lk->lk_data, + sizeof(*kcd->kcd_archives) * lk->lk_data_count); + } else { + kcd->kcd_magic = KKUC_CT_DATA_BITMAP_MAGIC; + } + + rc = libcfs_kkuc_group_add(filp, &obd->obd_uuid, lk->lk_uid, + lk->lk_group, kcd, kcd_size); + OBD_FREE(kcd, kcd_size); + if (rc) + GOTO(err_fput, rc); + /* All or nothing: try to register to all MDS. * In case of failure, unregister from previous MDS, * except if it because of inactive target. */ - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - struct lmv_tgt_desc *tgt = lmv->tgts[i]; - - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; + lmv_foreach_connected_tgt(lmv, tgt) { err = obd_iocontrol(cmd, tgt->ltd_exp, len, lk, uarg); if (err) { if (tgt->ltd_active) { @@ -841,18 +771,20 @@ static int lmv_hsm_ct_register(struct lmv_obd *lmv, unsigned int cmd, int len, CERROR("%s: iocontrol MDC %s on MDT" " idx %d cmd %x: err = %d\n", lmv2obd_dev(lmv)->obd_name, - tgt->ltd_uuid.uuid, i, cmd, err); + tgt->ltd_uuid.uuid, tgt->ltd_index, cmd, + err); rc = err; lk->lk_flags |= LK_FLG_STOP; + i = tgt->ltd_index; /* unregister from previous MDS */ - for (j = 0; j < i; j++) { - tgt = lmv->tgts[j]; - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; + lmv_foreach_connected_tgt(lmv, tgt) { + if (tgt->ltd_index >= i) + break; + obd_iocontrol(cmd, tgt->ltd_exp, len, lk, uarg); } - RETURN(rc); + GOTO(err_kkuc_rem, rc); } /* else: transient error. * kuc will register to the missing MDT @@ -864,56 +796,50 @@ static int lmv_hsm_ct_register(struct lmv_obd *lmv, unsigned int cmd, int len, if (!any_set) /* no registration done: return error */ - RETURN(-ENOTCONN); - - /* at least one registration done, with no failure */ - filp = fget(lk->lk_wfd); - if (filp == NULL) - RETURN(-EBADF); + GOTO(err_kkuc_rem, rc = -ENOTCONN); - kcd.kcd_magic = KKUC_CT_DATA_MAGIC; - kcd.kcd_uuid = lmv->cluuid; - kcd.kcd_archive = lk->lk_data; + RETURN(0); - rc = libcfs_kkuc_group_add(filp, lk->lk_uid, lk->lk_group, - &kcd, sizeof(kcd)); - if (rc != 0) - fput(filp); +err_kkuc_rem: + libcfs_kkuc_group_rem(&obd->obd_uuid, lk->lk_uid, lk->lk_group); - RETURN(rc); +err_fput: + fput(filp); + return rc; } - - - static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, int len, void *karg, void __user *uarg) { - struct obd_device *obddev = class_exp2obd(exp); - struct lmv_obd *lmv = &obddev->u.lmv; - struct lmv_tgt_desc *tgt = NULL; - __u32 i = 0; - int rc = 0; - int set = 0; - __u32 count = lmv->desc.ld_tgt_count; + struct obd_device *obd = class_exp2obd(exp); + struct lmv_obd *lmv = &obd->u.lmv; + struct lu_tgt_desc *tgt = NULL; + int set = 0; + __u32 count = lmv->lmv_mdt_count; + int rc = 0; + ENTRY; - if (count == 0) - RETURN(-ENOTTY); + if (count == 0) + RETURN(-ENOTTY); + + switch (cmd) { + case IOC_OBD_STATFS: { + struct obd_ioctl_data *data = karg; + struct obd_device *mdc_obd; + struct obd_statfs stat_buf = {0}; + __u32 index; - switch (cmd) { - case IOC_OBD_STATFS: { - struct obd_ioctl_data *data = karg; - struct obd_device *mdc_obd; - struct obd_statfs stat_buf = {0}; - __u32 index; + memcpy(&index, data->ioc_inlbuf2, sizeof(__u32)); - memcpy(&index, data->ioc_inlbuf2, sizeof(__u32)); - if ((index >= count)) - RETURN(-ENODEV); + if (index >= lmv->lmv_mdt_descs.ltd_tgts_size) + RETURN(-ENODEV); - tgt = lmv->tgts[index]; - if (tgt == NULL || !tgt->ltd_active) + tgt = lmv_tgt(lmv, index); + if (!tgt) + RETURN(-EAGAIN); + + if (!tgt->ltd_active) RETURN(-ENODATA); mdc_obd = class_exp2obd(tgt->ltd_exp); @@ -927,7 +853,7 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, RETURN(-EFAULT); rc = obd_statfs(NULL, tgt->ltd_exp, &stat_buf, - cfs_time_shift_64(-OBD_STATFS_CACHE_SECONDS), + ktime_get_seconds() - OBD_STATFS_CACHE_SECONDS, 0); if (rc) RETURN(rc); @@ -936,59 +862,50 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, (int) sizeof(stat_buf)))) RETURN(-EFAULT); break; - } - case OBD_IOC_QUOTACTL: { - struct if_quotactl *qctl = karg; - struct obd_quotactl *oqctl; + } + case OBD_IOC_QUOTACTL: { + struct if_quotactl *qctl = karg; + struct obd_quotactl *oqctl; if (qctl->qc_valid == QC_MDTIDX) { - if (count <= qctl->qc_idx) - RETURN(-EINVAL); - - tgt = lmv->tgts[qctl->qc_idx]; - if (tgt == NULL || tgt->ltd_exp == NULL) - RETURN(-EINVAL); + tgt = lmv_tgt(lmv, qctl->qc_idx); } else if (qctl->qc_valid == QC_UUID) { - for (i = 0; i < count; i++) { - tgt = lmv->tgts[i]; - if (tgt == NULL) - continue; + lmv_foreach_tgt(lmv, tgt) { if (!obd_uuid_equals(&tgt->ltd_uuid, &qctl->obd_uuid)) continue; - if (tgt->ltd_exp == NULL) - RETURN(-EINVAL); + if (!tgt->ltd_exp) + RETURN(-EINVAL); - break; - } - } else { - RETURN(-EINVAL); - } + break; + } + } else { + RETURN(-EINVAL); + } - if (i >= count) - RETURN(-EAGAIN); + if (!tgt || !tgt->ltd_exp) + RETURN(-EINVAL); - LASSERT(tgt != NULL && tgt->ltd_exp != NULL); - OBD_ALLOC_PTR(oqctl); - if (!oqctl) - RETURN(-ENOMEM); + OBD_ALLOC_PTR(oqctl); + if (!oqctl) + RETURN(-ENOMEM); - QCTL_COPY(oqctl, qctl); - rc = obd_quotactl(tgt->ltd_exp, oqctl); - if (rc == 0) { - QCTL_COPY(qctl, oqctl); - qctl->qc_valid = QC_MDTIDX; - qctl->obd_uuid = tgt->ltd_uuid; - } - OBD_FREE_PTR(oqctl); - break; - } + QCTL_COPY(oqctl, qctl); + rc = obd_quotactl(tgt->ltd_exp, oqctl); + if (rc == 0) { + QCTL_COPY(qctl, oqctl); + qctl->qc_valid = QC_MDTIDX; + qctl->obd_uuid = tgt->ltd_uuid; + } + OBD_FREE_PTR(oqctl); + break; + } case LL_IOC_GET_CONNECT_FLAGS: { - tgt = lmv->tgts[0]; - if (tgt == NULL || tgt->ltd_exp == NULL) - RETURN(-ENODATA); - rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg); + tgt = lmv_tgt(lmv, 0); + rc = -ENODATA; + if (tgt && tgt->ltd_exp) + rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg); break; } case LL_IOC_FID2MDTIDX: { @@ -1011,9 +928,9 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, case LL_IOC_HSM_STATE_GET: case LL_IOC_HSM_STATE_SET: case LL_IOC_HSM_ACTION: { - struct md_op_data *op_data = karg; + struct md_op_data *op_data = karg; - tgt = lmv_find_target(lmv, &op_data->op_fid1); + tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -1026,7 +943,7 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, case LL_IOC_HSM_PROGRESS: { const struct hsm_progress_kernel *hpk = karg; - tgt = lmv_find_target(lmv, &hpk->hpk_fid); + tgt = lmv_fid2tgt(lmv, &hpk->hpk_fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg); @@ -1043,22 +960,17 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, * or if there is a single MDS, no need to split * the request. */ if (reqcount == 1 || count == 1) { - tgt = lmv_find_target(lmv, - &hur->hur_user_item[0].hui_fid); + tgt = lmv_fid2tgt(lmv, &hur->hur_user_item[0].hui_fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); rc = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg); } else { /* split fid list to their respective MDS */ - for (i = 0; i < count; i++) { + lmv_foreach_connected_tgt(lmv, tgt) { int nr, rc1; size_t reqlen; struct hsm_user_request *req; - tgt = lmv->tgts[i]; - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; - nr = lmv_hsm_req_count(lmv, hur, tgt); if (nr < 0) RETURN(nr); @@ -1086,14 +998,14 @@ hsm_req_err: break; } case LL_IOC_LOV_SWAP_LAYOUTS: { - struct md_op_data *op_data = karg; - struct lmv_tgt_desc *tgt1, *tgt2; + struct md_op_data *op_data = karg; + struct lmv_tgt_desc *tgt1, *tgt2; - tgt1 = lmv_find_target(lmv, &op_data->op_fid1); + tgt1 = lmv_fid2tgt(lmv, &op_data->op_fid1); if (IS_ERR(tgt1)) RETURN(PTR_ERR(tgt1)); - tgt2 = lmv_find_target(lmv, &op_data->op_fid2); + tgt2 = lmv_fid2tgt(lmv, &op_data->op_fid2); if (IS_ERR(tgt2)) RETURN(PTR_ERR(tgt2)); @@ -1101,7 +1013,7 @@ hsm_req_err: RETURN(-EINVAL); /* only files on same MDT can have their layouts swapped */ - if (tgt1->ltd_idx != tgt2->ltd_idx) + if (tgt1->ltd_index != tgt2->ltd_index) RETURN(-EPERM); rc = obd_iocontrol(cmd, tgt1->ltd_exp, len, karg, uarg); @@ -1110,215 +1022,149 @@ hsm_req_err: case LL_IOC_HSM_CT_START: { struct lustre_kernelcomm *lk = karg; if (lk->lk_flags & LK_FLG_STOP) - rc = lmv_hsm_ct_unregister(lmv, cmd, len, lk, uarg); + rc = lmv_hsm_ct_unregister(obd, cmd, len, lk, uarg); else - rc = lmv_hsm_ct_register(lmv, cmd, len, lk, uarg); + rc = lmv_hsm_ct_register(obd, cmd, len, lk, uarg); break; } default: - for (i = 0; i < count; i++) { + lmv_foreach_connected_tgt(lmv, tgt) { struct obd_device *mdc_obd; int err; - tgt = lmv->tgts[i]; - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; /* ll_umount_begin() sets force flag but for lmv, not * mdc. Let's pass it through */ mdc_obd = class_exp2obd(tgt->ltd_exp); - mdc_obd->obd_force = obddev->obd_force; + mdc_obd->obd_force = obd->obd_force; err = obd_iocontrol(cmd, tgt->ltd_exp, len, karg, uarg); if (err) { if (tgt->ltd_active) { CERROR("error: iocontrol MDC %s on MDT" " idx %d cmd %x: err = %d\n", - tgt->ltd_uuid.uuid, i, cmd, err); + tgt->ltd_uuid.uuid, + tgt->ltd_index, cmd, err); if (!rc) rc = err; } } else set = 1; - } - if (!set && !rc) - rc = -EIO; - } - RETURN(rc); + } + if (!set && !rc) + rc = -EIO; + } + RETURN(rc); } -/** - * This is _inode_ placement policy function (not name). - */ -static int lmv_placement_policy(struct obd_device *obd, - struct md_op_data *op_data, u32 *mds) +int lmv_fid_alloc(const struct lu_env *env, struct obd_export *exp, + struct lu_fid *fid, struct md_op_data *op_data) { - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_user_md *lum; + struct obd_device *obd = class_exp2obd(exp); + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; ENTRY; - LASSERT(mds != NULL); - - if (lmv->desc.ld_tgt_count == 1) { - *mds = 0; - RETURN(0); - } - - lum = op_data->op_data; - /* Choose MDS by - * 1. See if the stripe offset is specified by lum. - * 2. Then check if there is default stripe offset. - * 3. Finally choose MDS by name hash if the parent - * is striped directory. (see lmv_locate_mds()). */ - if (op_data->op_cli_flags & CLI_SET_MEA && lum != NULL && - le32_to_cpu(lum->lum_stripe_offset) != (__u32)-1) { - *mds = le32_to_cpu(lum->lum_stripe_offset); - } else if (op_data->op_default_stripe_offset != (__u32)-1) { - *mds = op_data->op_default_stripe_offset; - op_data->op_mds = *mds; - /* Correct the stripe offset in lum */ - if (lum != NULL) - lum->lum_stripe_offset = cpu_to_le32(*mds); - } else { - *mds = op_data->op_mds; - } - - RETURN(0); -} + LASSERT(op_data); + LASSERT(fid); -int __lmv_fid_alloc(struct lmv_obd *lmv, struct lu_fid *fid, u32 mds) -{ - struct lmv_tgt_desc *tgt; - int rc; - ENTRY; + tgt = lmv_tgt(lmv, op_data->op_mds); + if (!tgt) + RETURN(-ENODEV); - tgt = lmv_get_target(lmv, mds, NULL); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + if (!tgt->ltd_active || !tgt->ltd_exp) + RETURN(-ENODEV); /* * New seq alloc and FLD setup should be atomic. Otherwise we may find * on server that seq in new allocated fid is not yet known. */ mutex_lock(&tgt->ltd_fid_mutex); - - if (tgt->ltd_active == 0 || tgt->ltd_exp == NULL) - GOTO(out, rc = -ENODEV); - - /* - * Asking underlying tgt layer to allocate new fid. - */ rc = obd_fid_alloc(NULL, tgt->ltd_exp, fid, NULL); + mutex_unlock(&tgt->ltd_fid_mutex); if (rc > 0) { LASSERT(fid_is_sane(fid)); rc = 0; } - EXIT; -out: - mutex_unlock(&tgt->ltd_fid_mutex); - return rc; -} - -int lmv_fid_alloc(const struct lu_env *env, struct obd_export *exp, - struct lu_fid *fid, struct md_op_data *op_data) -{ - struct obd_device *obd = class_exp2obd(exp); - struct lmv_obd *lmv = &obd->u.lmv; - u32 mds = 0; - int rc; - ENTRY; - - LASSERT(op_data != NULL); - LASSERT(fid != NULL); - - rc = lmv_placement_policy(obd, op_data, &mds); - if (rc) { - CERROR("Can't get target for allocating fid, " - "rc %d\n", rc); - RETURN(rc); - } - - rc = __lmv_fid_alloc(lmv, fid, mds); - if (rc) { - CERROR("Can't alloc new fid, rc %d\n", rc); - RETURN(rc); - } - - RETURN(rc); + RETURN(rc); } static int lmv_setup(struct obd_device *obd, struct lustre_cfg *lcfg) { - struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_obd *lmv = &obd->u.lmv; struct lmv_desc *desc; - int rc; - ENTRY; + struct lnet_process_id lnet_id; + int i = 0; + int rc; - if (LUSTRE_CFG_BUFLEN(lcfg, 1) < 1) { - CERROR("LMV setup requires a descriptor\n"); - RETURN(-EINVAL); - } + ENTRY; - desc = (struct lmv_desc *)lustre_cfg_buf(lcfg, 1); - if (sizeof(*desc) > LUSTRE_CFG_BUFLEN(lcfg, 1)) { - CERROR("Lmv descriptor size wrong: %d > %d\n", - (int)sizeof(*desc), LUSTRE_CFG_BUFLEN(lcfg, 1)); - RETURN(-EINVAL); - } + if (LUSTRE_CFG_BUFLEN(lcfg, 1) < 1) { + CERROR("LMV setup requires a descriptor\n"); + RETURN(-EINVAL); + } - lmv->tgts_size = 32U; - OBD_ALLOC(lmv->tgts, sizeof(*lmv->tgts) * lmv->tgts_size); - if (lmv->tgts == NULL) - RETURN(-ENOMEM); + desc = (struct lmv_desc *)lustre_cfg_buf(lcfg, 1); + if (sizeof(*desc) > LUSTRE_CFG_BUFLEN(lcfg, 1)) { + CERROR("Lmv descriptor size wrong: %d > %d\n", + (int)sizeof(*desc), LUSTRE_CFG_BUFLEN(lcfg, 1)); + RETURN(-EINVAL); + } - obd_str2uuid(&lmv->desc.ld_uuid, desc->ld_uuid.uuid); - lmv->desc.ld_tgt_count = 0; - lmv->desc.ld_active_tgt_count = 0; + obd_str2uuid(&lmv->lmv_mdt_descs.ltd_lmv_desc.ld_uuid, + desc->ld_uuid.uuid); + lmv->lmv_mdt_descs.ltd_lmv_desc.ld_tgt_count = 0; + lmv->lmv_mdt_descs.ltd_lmv_desc.ld_active_tgt_count = 0; + lmv->lmv_mdt_descs.ltd_lmv_desc.ld_qos_maxage = + LMV_DESC_QOS_MAXAGE_DEFAULT; lmv->max_def_easize = 0; lmv->max_easize = 0; spin_lock_init(&lmv->lmv_lock); - mutex_init(&lmv->lmv_init_mutex); - -#ifdef CONFIG_PROC_FS - obd->obd_vars = lprocfs_lmv_obd_vars; - lprocfs_obd_setup(obd, true); - lprocfs_alloc_md_stats(obd, 0); - rc = lprocfs_seq_create(obd->obd_proc_entry, "target_obd", - 0444, &lmv_proc_target_fops, obd); - if (rc) - CWARN("%s: error adding LMV target_obd file: rc = %d\n", - obd->obd_name, rc); -#endif - rc = fld_client_init(&lmv->lmv_fld, obd->obd_name, - LUSTRE_CLI_FLD_HASH_DHT); - if (rc) { - CERROR("Can't init FLD, err %d\n", rc); - GOTO(out, rc); + + /* + * initialize rr_index to lower 32bit of netid, so that client + * can distribute subdirs evenly from the beginning. + */ + while (LNetGetId(i++, &lnet_id) != -ENOENT) { + if (lnet_id.nid != LNET_NID_LO_0) { + lmv->lmv_qos_rr_index = (u32)lnet_id.nid; + break; + } } - RETURN(0); + rc = lmv_tunables_init(obd); + if (rc) + CWARN("%s: error adding LMV sysfs/debugfs files: rc = %d\n", + obd->obd_name, rc); + + rc = fld_client_init(&lmv->lmv_fld, obd->obd_name, + LUSTRE_CLI_FLD_HASH_DHT); + if (rc) + CERROR("Can't init FLD, err %d\n", rc); -out: - return rc; + rc = lu_tgt_descs_init(&lmv->lmv_mdt_descs, true); + if (rc) + CWARN("%s: error initialize target table: rc = %d\n", + obd->obd_name, rc); + + RETURN(rc); } static int lmv_cleanup(struct obd_device *obd) { - struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_obd *lmv = &obd->u.lmv; + struct lu_tgt_desc *tgt; + struct lu_tgt_desc *tmp; + ENTRY; fld_client_fini(&lmv->lmv_fld); - if (lmv->tgts != NULL) { - int i; - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - if (lmv->tgts[i] == NULL) - continue; - lmv_del_target(lmv, i); - } - OBD_FREE(lmv->tgts, sizeof(*lmv->tgts) * lmv->tgts_size); - lmv->tgts_size = 0; - } + lmv_foreach_tgt_safe(lmv, tgt, tmp) + lmv_del_target(lmv, tgt); + lu_tgt_descs_fini(&lmv->lmv_mdt_descs); + RETURN(0); } @@ -1354,278 +1200,621 @@ out: RETURN(rc); } +static int lmv_select_statfs_mdt(struct lmv_obd *lmv, __u32 flags) +{ + int i; + + if (flags & OBD_STATFS_FOR_MDT0) + return 0; + + if (lmv->lmv_statfs_start || lmv->lmv_mdt_count == 1) + return lmv->lmv_statfs_start; + + /* choose initial MDT for this client */ + for (i = 0;; i++) { + struct lnet_process_id lnet_id; + if (LNetGetId(i, &lnet_id) == -ENOENT) + break; + + if (lnet_id.nid != LNET_NID_LO_0) { + /* We dont need a full 64-bit modulus, just enough + * to distribute the requests across MDTs evenly. + */ + lmv->lmv_statfs_start = (u32)lnet_id.nid % + lmv->lmv_mdt_count; + break; + } + } + + return lmv->lmv_statfs_start; +} + static int lmv_statfs(const struct lu_env *env, struct obd_export *exp, - struct obd_statfs *osfs, __u64 max_age, __u32 flags) + struct obd_statfs *osfs, time64_t max_age, __u32 flags) { - struct obd_device *obd = class_exp2obd(exp); - struct lmv_obd *lmv = &obd->u.lmv; - struct obd_statfs *temp; - int rc = 0; - __u32 i; + struct obd_device *obd = class_exp2obd(exp); + struct lmv_obd *lmv = &obd->u.lmv; + struct obd_statfs *temp; + struct lu_tgt_desc *tgt; + __u32 i; + __u32 idx; + int rc = 0; + ENTRY; - OBD_ALLOC(temp, sizeof(*temp)); - if (temp == NULL) - RETURN(-ENOMEM); + OBD_ALLOC(temp, sizeof(*temp)); + if (temp == NULL) + RETURN(-ENOMEM); + + /* distribute statfs among MDTs */ + idx = lmv_select_statfs_mdt(lmv, flags); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - if (lmv->tgts[i] == NULL || lmv->tgts[i]->ltd_exp == NULL) + for (i = 0; i < lmv->lmv_mdt_descs.ltd_tgts_size; i++, idx++) { + idx = idx % lmv->lmv_mdt_descs.ltd_tgts_size; + tgt = lmv_tgt(lmv, idx); + if (!tgt || !tgt->ltd_exp) continue; - rc = obd_statfs(env, lmv->tgts[i]->ltd_exp, temp, - max_age, flags); + rc = obd_statfs(env, tgt->ltd_exp, temp, max_age, + flags | OBD_STATFS_NESTED); if (rc) { - CERROR("can't stat MDS #%d (%s), error %d\n", i, - lmv->tgts[i]->ltd_exp->exp_obd->obd_name, - rc); + CERROR("%s: can't stat MDS #%d: rc = %d\n", + tgt->ltd_exp->exp_obd->obd_name, i, rc); GOTO(out_free_temp, rc); } + if (temp->os_state & OS_STATFS_SUM || + flags == OBD_STATFS_FOR_MDT0) { + /* reset to the last aggregated values + * and don't sum with non-aggrated data */ + /* If the statfs is from mount, it needs to retrieve + * necessary information from MDT0. i.e. mount does + * not need the merged osfs from all of MDT. Also + * clients can be mounted as long as MDT0 is in + * service */ + *osfs = *temp; + break; + } + if (i == 0) { *osfs = *temp; - /* If the statfs is from mount, it will needs - * retrieve necessary information from MDT0. - * i.e. mount does not need the merged osfs - * from all of MDT. - * And also clients can be mounted as long as - * MDT0 is in service*/ - if (flags & OBD_STATFS_FOR_MDT0) - GOTO(out_free_temp, rc); - } else { - osfs->os_bavail += temp->os_bavail; - osfs->os_blocks += temp->os_blocks; - osfs->os_ffree += temp->os_ffree; - osfs->os_files += temp->os_files; - } - } + } else { + osfs->os_bavail += temp->os_bavail; + osfs->os_blocks += temp->os_blocks; + osfs->os_ffree += temp->os_ffree; + osfs->os_files += temp->os_files; + osfs->os_granted += temp->os_granted; + } + } - EXIT; + EXIT; out_free_temp: - OBD_FREE(temp, sizeof(*temp)); - return rc; + OBD_FREE(temp, sizeof(*temp)); + return rc; +} + +static int lmv_statfs_update(void *cookie, int rc) +{ + struct obd_info *oinfo = cookie; + struct obd_device *obd = oinfo->oi_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt = oinfo->oi_tgt; + struct obd_statfs *osfs = oinfo->oi_osfs; + + /* + * NB: don't deactivate TGT upon error, because we may not trigger async + * statfs any longer, then there is no chance to activate TGT. + */ + if (!rc) { + spin_lock(&lmv->lmv_lock); + tgt->ltd_statfs = *osfs; + tgt->ltd_statfs_age = ktime_get_seconds(); + spin_unlock(&lmv->lmv_lock); + set_bit(LQ_DIRTY, &lmv->lmv_qos.lq_flags); + } + + return rc; +} + +/* update tgt statfs async if it's ld_qos_maxage old */ +int lmv_statfs_check_update(struct obd_device *obd, struct lmv_tgt_desc *tgt) +{ + struct obd_info oinfo = { + .oi_obd = obd, + .oi_tgt = tgt, + .oi_cb_up = lmv_statfs_update, + }; + int rc; + + if (ktime_get_seconds() - tgt->ltd_statfs_age < + obd->u.lmv.lmv_mdt_descs.ltd_lmv_desc.ld_qos_maxage) + return 0; + + rc = obd_statfs_async(tgt->ltd_exp, &oinfo, 0, NULL); + + return rc; } static int lmv_get_root(struct obd_export *exp, const char *fileset, struct lu_fid *fid) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - int rc; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lu_tgt_desc *tgt = lmv_tgt(lmv, 0); + int rc; + + ENTRY; - rc = md_get_root(lmv->tgts[0]->ltd_exp, fileset, fid); + if (!tgt) + RETURN(-ENODEV); + + rc = md_get_root(tgt->ltd_exp, fileset, fid); RETURN(rc); } static int lmv_getxattr(struct obd_export *exp, const struct lu_fid *fid, - u64 valid, const char *name, - const char *input, int input_size, int output_size, - int flags, struct ptlrpc_request **request) -{ - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; - ENTRY; + u64 obd_md_valid, const char *name, size_t buf_size, + struct ptlrpc_request **req) +{ + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; - tgt = lmv_find_target(lmv, fid); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ENTRY; - rc = md_getxattr(tgt->ltd_exp, fid, valid, name, input, - input_size, output_size, flags, request); + tgt = lmv_fid2tgt(lmv, fid); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + rc = md_getxattr(tgt->ltd_exp, fid, obd_md_valid, name, buf_size, req); RETURN(rc); } static int lmv_setxattr(struct obd_export *exp, const struct lu_fid *fid, - u64 valid, const char *name, - const char *input, int input_size, int output_size, - int flags, __u32 suppgid, - struct ptlrpc_request **request) + u64 obd_md_valid, const char *name, + const void *value, size_t value_size, + unsigned int xattr_flags, u32 suppgid, + struct ptlrpc_request **req) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; - tgt = lmv_find_target(lmv, fid); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ENTRY; + + tgt = lmv_fid2tgt(lmv, fid); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); - rc = md_setxattr(tgt->ltd_exp, fid, valid, name, input, - input_size, output_size, flags, suppgid, - request); + rc = md_setxattr(tgt->ltd_exp, fid, obd_md_valid, name, + value, value_size, xattr_flags, suppgid, req); RETURN(rc); } static int lmv_getattr(struct obd_export *exp, struct md_op_data *op_data, - struct ptlrpc_request **request) + struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; - tgt = lmv_find_target(lmv, &op_data->op_fid1); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ENTRY; + + tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); if (op_data->op_flags & MF_GET_MDT_IDX) { - op_data->op_mds = tgt->ltd_idx; + op_data->op_mds = tgt->ltd_index; RETURN(0); } - rc = md_getattr(tgt->ltd_exp, op_data, request); + rc = md_getattr(tgt->ltd_exp, op_data, request); - RETURN(rc); + RETURN(rc); } static int lmv_null_inode(struct obd_export *exp, const struct lu_fid *fid) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - __u32 i; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lu_tgt_desc *tgt; + + ENTRY; - CDEBUG(D_INODE, "CBDATA for "DFID"\n", PFID(fid)); + CDEBUG(D_INODE, "CBDATA for "DFID"\n", PFID(fid)); /* * With DNE every object can have two locks in different namespaces: * lookup lock in space of MDT storing direntry and update/open lock in * space of MDT storing inode. */ - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - if (lmv->tgts[i] == NULL || lmv->tgts[i]->ltd_exp == NULL) - continue; - md_null_inode(lmv->tgts[i]->ltd_exp, fid); - } + lmv_foreach_connected_tgt(lmv, tgt) + md_null_inode(tgt->ltd_exp, fid); RETURN(0); } static int lmv_close(struct obd_export *exp, struct md_op_data *op_data, - struct md_open_data *mod, struct ptlrpc_request **request) + struct md_open_data *mod, struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; - tgt = lmv_find_target(lmv, &op_data->op_fid1); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ENTRY; - CDEBUG(D_INODE, "CLOSE "DFID"\n", PFID(&op_data->op_fid1)); - rc = md_close(tgt->ltd_exp, op_data, mod, request); - RETURN(rc); + tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + CDEBUG(D_INODE, "CLOSE "DFID"\n", PFID(&op_data->op_fid1)); + rc = md_close(tgt->ltd_exp, op_data, mod, request); + RETURN(rc); } -/** - * Choosing the MDT by name or FID in @op_data. - * For non-striped directory, it will locate MDT by fid. - * For striped-directory, it will locate MDT by name. And also - * it will reset op_fid1 with the FID of the choosen stripe. - **/ -struct lmv_tgt_desc * -lmv_locate_target_for_name(struct lmv_obd *lmv, struct lmv_stripe_md *lsm, - const char *name, int namelen, struct lu_fid *fid, - u32 *mds) +static struct lu_tgt_desc *lmv_locate_tgt_qos(struct lmv_obd *lmv, __u32 *mdt) +{ + struct lu_tgt_desc *tgt, *cur = NULL; + __u64 total_avail = 0; + __u64 total_weight = 0; + __u64 cur_weight = 0; + int total_usable = 0; + __u64 rand; + int rc; + + ENTRY; + + if (!ltd_qos_is_usable(&lmv->lmv_mdt_descs)) + RETURN(ERR_PTR(-EAGAIN)); + + down_write(&lmv->lmv_qos.lq_rw_sem); + + if (!ltd_qos_is_usable(&lmv->lmv_mdt_descs)) + GOTO(unlock, tgt = ERR_PTR(-EAGAIN)); + + rc = ltd_qos_penalties_calc(&lmv->lmv_mdt_descs); + if (rc) + GOTO(unlock, tgt = ERR_PTR(rc)); + + lmv_foreach_tgt(lmv, tgt) { + if (!tgt->ltd_exp || !tgt->ltd_active) { + tgt->ltd_qos.ltq_usable = 0; + continue; + } + + tgt->ltd_qos.ltq_usable = 1; + lu_tgt_qos_weight_calc(tgt); + if (tgt->ltd_index == *mdt) + cur = tgt; + total_avail += tgt->ltd_qos.ltq_avail; + total_weight += tgt->ltd_qos.ltq_weight; + total_usable++; + } + + /* if current MDT has above-average space, within range of the QOS + * threshold, stay on the same MDT to avoid creating needless remote + * MDT directories. + */ + rand = total_avail * (256 - lmv->lmv_qos.lq_threshold_rr) / + (total_usable * 256); + if (cur && cur->ltd_qos.ltq_avail >= rand) { + tgt = cur; + GOTO(unlock, rc = 0); + } + + rand = lu_prandom_u64_max(total_weight); + + lmv_foreach_connected_tgt(lmv, tgt) { + if (!tgt->ltd_qos.ltq_usable) + continue; + + cur_weight += tgt->ltd_qos.ltq_weight; + if (cur_weight < rand) + continue; + + *mdt = tgt->ltd_index; + ltd_qos_update(&lmv->lmv_mdt_descs, tgt, &total_weight); + GOTO(unlock, rc = 0); + } + + /* no proper target found */ + GOTO(unlock, tgt = ERR_PTR(-EAGAIN)); +unlock: + up_write(&lmv->lmv_qos.lq_rw_sem); + + return tgt; +} + +static struct lu_tgt_desc *lmv_locate_tgt_rr(struct lmv_obd *lmv, __u32 *mdt) { - struct lmv_tgt_desc *tgt; - const struct lmv_oinfo *oinfo; + struct lu_tgt_desc *tgt; + int i; + int index; + + ENTRY; + + spin_lock(&lmv->lmv_lock); + for (i = 0; i < lmv->lmv_mdt_descs.ltd_tgts_size; i++) { + index = (i + lmv->lmv_qos_rr_index) % + lmv->lmv_mdt_descs.ltd_tgts_size; + tgt = lmv_tgt(lmv, index); + if (!tgt || !tgt->ltd_exp || !tgt->ltd_active) + continue; + + *mdt = tgt->ltd_index; + lmv->lmv_qos_rr_index = (*mdt + 1) % + lmv->lmv_mdt_descs.ltd_tgts_size; + spin_unlock(&lmv->lmv_lock); + + RETURN(tgt); + } + spin_unlock(&lmv->lmv_lock); + + RETURN(ERR_PTR(-ENODEV)); +} + +/* locate MDT by file name, for striped directory, the file name hash decides + * which stripe its dirent is stored. + */ +static struct lmv_tgt_desc * +lmv_locate_tgt_by_name(struct lmv_obd *lmv, struct lmv_stripe_md *lsm, + const char *name, int namelen, struct lu_fid *fid, + __u32 *mds, bool new_layout) +{ + struct lmv_tgt_desc *tgt; + const struct lmv_oinfo *oinfo; + + if (!lmv_dir_striped(lsm) || !namelen) { + tgt = lmv_fid2tgt(lmv, fid); + if (IS_ERR(tgt)) + return tgt; + + *mds = tgt->ltd_index; + return tgt; + } if (OBD_FAIL_CHECK(OBD_FAIL_LFSCK_BAD_NAME_HASH)) { if (cfs_fail_val >= lsm->lsm_md_stripe_count) - RETURN(ERR_PTR(-EBADF)); + return ERR_PTR(-EBADF); oinfo = &lsm->lsm_md_oinfo[cfs_fail_val]; } else { - oinfo = lsm_name_to_stripe_info(lsm, name, namelen); + oinfo = lsm_name_to_stripe_info(lsm, name, namelen, new_layout); if (IS_ERR(oinfo)) - RETURN(ERR_CAST(oinfo)); + return ERR_CAST(oinfo); } - if (fid != NULL) - *fid = oinfo->lmo_fid; - if (mds != NULL) - *mds = oinfo->lmo_mds; + /* check stripe FID is sane */ + if (!fid_is_sane(&oinfo->lmo_fid)) + return ERR_PTR(-ENODEV); - tgt = lmv_get_target(lmv, oinfo->lmo_mds, NULL); + *fid = oinfo->lmo_fid; + *mds = oinfo->lmo_mds; + tgt = lmv_tgt(lmv, oinfo->lmo_mds); - CDEBUG(D_INFO, "locate on mds %u "DFID"\n", oinfo->lmo_mds, - PFID(&oinfo->lmo_fid)); - return tgt; + CDEBUG(D_INODE, "locate MDT %u parent "DFID"\n", *mds, PFID(fid)); + + return tgt ? tgt : ERR_PTR(-ENODEV); } /** - * Locate mds by fid or name + * Locate MDT of op_data->op_fid1 + * + * For striped directory, it will locate the stripe by name hash, if hash_type + * is unknown, it will return the stripe specified by 'op_data->op_stripe_index' + * which is set outside, and if dir is migrating, 'op_data->op_new_layout' + * indicates whether old or new layout is used to locate. * - * For striped directory (lsm != NULL), it will locate the stripe - * by name hash (see lsm_name_to_stripe_info()). Note: if the hash_type - * is unknown, it will return -EBADFD, and lmv_intent_lookup might need - * walk through all of stripes to locate the entry. + * For plain direcotry, it just locate the MDT of op_data->op_fid1. * - * For normal direcotry, it will locate MDS by FID directly. - * \param[in] lmv LMV device - * \param[in] op_data client MD stack parameters, name, namelen - * mds_num etc. - * \param[in] fid object FID used to locate MDS. + * \param[in] lmv LMV device + * \param[in/out] op_data client MD stack parameters, name, namelen etc, + * op_mds and op_fid1 will be updated if op_mea1 + * indicates fid1 represents a striped directory. * * retval pointer to the lmv_tgt_desc if succeed. * ERR_PTR(errno) if failed. */ -struct lmv_tgt_desc* -lmv_locate_mds(struct lmv_obd *lmv, struct md_op_data *op_data, - struct lu_fid *fid) +struct lmv_tgt_desc * +lmv_locate_tgt(struct lmv_obd *lmv, struct md_op_data *op_data) { - struct lmv_stripe_md *lsm = op_data->op_mea1; - struct lmv_tgt_desc *tgt; + struct lmv_stripe_md *lsm = op_data->op_mea1; + struct lmv_oinfo *oinfo; + struct lmv_tgt_desc *tgt; + + if (lmv_dir_foreign(lsm)) + return ERR_PTR(-ENODATA); /* During creating VOLATILE file, it should honor the mdt * index if the file under striped dir is being restored, see * ct_restore(). */ if (op_data->op_bias & MDS_CREATE_VOLATILE && - (int)op_data->op_mds != -1) { - int i; - tgt = lmv_get_target(lmv, op_data->op_mds, NULL); - if (IS_ERR(tgt)) - return tgt; + op_data->op_mds != LMV_OFFSET_DEFAULT) { + tgt = lmv_tgt(lmv, op_data->op_mds); + if (!tgt) + return ERR_PTR(-ENODEV); + + if (lmv_dir_striped(lsm)) { + int i; - if (lsm != NULL) { /* refill the right parent fid */ for (i = 0; i < lsm->lsm_md_stripe_count; i++) { - struct lmv_oinfo *oinfo; - oinfo = &lsm->lsm_md_oinfo[i]; if (oinfo->lmo_mds == op_data->op_mds) { - *fid = oinfo->lmo_fid; + op_data->op_fid1 = oinfo->lmo_fid; break; } } if (i == lsm->lsm_md_stripe_count) - *fid = lsm->lsm_md_oinfo[0].lmo_fid; + op_data->op_fid1 = lsm->lsm_md_oinfo[0].lmo_fid; } + } else if (lmv_dir_bad_hash(lsm)) { + LASSERT(op_data->op_stripe_index < lsm->lsm_md_stripe_count); + oinfo = &lsm->lsm_md_oinfo[op_data->op_stripe_index]; - return tgt; + op_data->op_fid1 = oinfo->lmo_fid; + op_data->op_mds = oinfo->lmo_mds; + tgt = lmv_tgt(lmv, oinfo->lmo_mds); + if (!tgt) + return ERR_PTR(-ENODEV); + } else { + tgt = lmv_locate_tgt_by_name(lmv, op_data->op_mea1, + op_data->op_name, op_data->op_namelen, + &op_data->op_fid1, &op_data->op_mds, + op_data->op_new_layout); } - if (lsm == NULL || op_data->op_namelen == 0) { - tgt = lmv_find_target(lmv, fid); + return tgt; +} + +/* Locate MDT of op_data->op_fid2 for link/rename */ +static struct lmv_tgt_desc * +lmv_locate_tgt2(struct lmv_obd *lmv, struct md_op_data *op_data) +{ + struct lmv_tgt_desc *tgt; + int rc; + + LASSERT(op_data->op_name); + if (lmv_dir_layout_changing(op_data->op_mea2)) { + struct lu_fid fid1 = op_data->op_fid1; + struct lmv_stripe_md *lsm1 = op_data->op_mea1; + struct ptlrpc_request *request = NULL; + + /* + * avoid creating new file under old layout of migrating + * directory, check it here. + */ + tgt = lmv_locate_tgt_by_name(lmv, op_data->op_mea2, + op_data->op_name, op_data->op_namelen, + &op_data->op_fid2, &op_data->op_mds, false); if (IS_ERR(tgt)) - return tgt; + RETURN(tgt); + + op_data->op_fid1 = op_data->op_fid2; + op_data->op_mea1 = op_data->op_mea2; + rc = md_getattr_name(tgt->ltd_exp, op_data, &request); + op_data->op_fid1 = fid1; + op_data->op_mea1 = lsm1; + if (!rc) { + ptlrpc_req_finished(request); + RETURN(ERR_PTR(-EEXIST)); + } - op_data->op_mds = tgt->ltd_idx; - return tgt; + if (rc != -ENOENT) + RETURN(ERR_PTR(rc)); + } + + return lmv_locate_tgt_by_name(lmv, op_data->op_mea2, op_data->op_name, + op_data->op_namelen, &op_data->op_fid2, + &op_data->op_mds, true); +} + +int lmv_old_layout_lookup(struct lmv_obd *lmv, struct md_op_data *op_data) +{ + struct lu_tgt_desc *tgt; + struct ptlrpc_request *request; + int rc; + + LASSERT(lmv_dir_layout_changing(op_data->op_mea1)); + LASSERT(!op_data->op_new_layout); + + tgt = lmv_locate_tgt(lmv, op_data); + if (IS_ERR(tgt)) + return PTR_ERR(tgt); + + rc = md_getattr_name(tgt->ltd_exp, op_data, &request); + if (!rc) { + ptlrpc_req_finished(request); + return -EEXIST; } - return lmv_locate_target_for_name(lmv, lsm, op_data->op_name, - op_data->op_namelen, fid, - &op_data->op_mds); + return rc; +} + +static inline bool lmv_op_user_qos_mkdir(const struct md_op_data *op_data) +{ + const struct lmv_user_md *lum = op_data->op_data; + + return (op_data->op_cli_flags & CLI_SET_MEA) && lum && + le32_to_cpu(lum->lum_magic) == LMV_USER_MAGIC && + le32_to_cpu(lum->lum_stripe_offset) == LMV_OFFSET_DEFAULT; +} + +static inline bool lmv_op_default_qos_mkdir(const struct md_op_data *op_data) +{ + const struct lmv_stripe_md *lsm = op_data->op_default_mea1; + + return lsm && lsm->lsm_md_master_mdt_index == LMV_OFFSET_DEFAULT; +} + +/* mkdir by QoS in two cases: + * 1. 'lfs mkdir -i -1' + * 2. parent default LMV master_mdt_index is -1 + * + * NB, mkdir by QoS only if parent is not striped, this is to avoid remote + * directories under striped directory. + */ +static inline bool lmv_op_qos_mkdir(const struct md_op_data *op_data) +{ + if (op_data->op_code != LUSTRE_OPC_MKDIR) + return false; + + if (lmv_dir_striped(op_data->op_mea1)) + return false; + + if (lmv_op_user_qos_mkdir(op_data)) + return true; + + if (lmv_op_default_qos_mkdir(op_data)) + return true; + + return false; +} + +/* if default LMV is set, and its index is LMV_OFFSET_DEFAULT, and + * 1. max_inherit_rr is set and is not LMV_INHERIT_RR_NONE + * 2. or parent is ROOT + * mkdir roundrobin. + * NB, this also needs to check server is balanced, which is checked by caller. + */ +static inline bool lmv_op_default_rr_mkdir(const struct md_op_data *op_data) +{ + const struct lmv_stripe_md *lsm = op_data->op_default_mea1; + + if (!lmv_op_default_qos_mkdir(op_data)) + return false; + + return lsm->lsm_md_max_inherit_rr != LMV_INHERIT_RR_NONE || + fid_is_root(&op_data->op_fid1); +} + +/* 'lfs mkdir -i ' */ +static inline bool lmv_op_user_specific_mkdir(const struct md_op_data *op_data) +{ + const struct lmv_user_md *lum = op_data->op_data; + + return op_data->op_code == LUSTRE_OPC_MKDIR && + op_data->op_cli_flags & CLI_SET_MEA && lum && + (le32_to_cpu(lum->lum_magic) == LMV_USER_MAGIC || + le32_to_cpu(lum->lum_magic) == LMV_USER_MAGIC_SPECIFIC) && + le32_to_cpu(lum->lum_stripe_offset) != LMV_OFFSET_DEFAULT; +} + +/* parent default LMV master_mdt_index is not -1. */ +static inline bool +lmv_op_default_specific_mkdir(const struct md_op_data *op_data) +{ + return op_data->op_code == LUSTRE_OPC_MKDIR && + op_data->op_default_mea1 && + op_data->op_default_mea1->lsm_md_master_mdt_index != + LMV_OFFSET_DEFAULT; } int lmv_create(struct obd_export *exp, struct md_op_data *op_data, @@ -1633,40 +1822,85 @@ int lmv_create(struct obd_export *exp, struct md_op_data *op_data, gid_t gid, cfs_cap_t cap_effective, __u64 rdev, struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + struct mdt_body *repbody; + int rc; + ENTRY; - if (!lmv->desc.ld_active_tgt_count) + if (!lmv->lmv_mdt_descs.ltd_lmv_desc.ld_active_tgt_count) RETURN(-EIO); - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); + if (lmv_dir_bad_hash(op_data->op_mea1)) + RETURN(-EBADF); + + if (lmv_dir_layout_changing(op_data->op_mea1)) { + /* + * if parent is migrating, create() needs to lookup existing + * name in both old and new layout, check old layout on client. + */ + rc = lmv_old_layout_lookup(lmv, op_data); + if (rc != -ENOENT) + RETURN(rc); + + op_data->op_new_layout = true; + } + + tgt = lmv_locate_tgt(lmv, op_data); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - CDEBUG(D_INODE, "CREATE name '%.*s' on "DFID" -> mds #%x\n", - (int)op_data->op_namelen, op_data->op_name, - PFID(&op_data->op_fid1), op_data->op_mds); - - rc = lmv_fid_alloc(NULL, exp, &op_data->op_fid2, op_data); - if (rc) - RETURN(rc); - if (exp_connect_flags(exp) & OBD_CONNECT_DIR_STRIPE) { - /* Send the create request to the MDT where the object - * will be located */ - tgt = lmv_find_target(lmv, &op_data->op_fid2); + if (lmv_op_user_specific_mkdir(op_data)) { + struct lmv_user_md *lum = op_data->op_data; + + op_data->op_mds = le32_to_cpu(lum->lum_stripe_offset); + tgt = lmv_tgt(lmv, op_data->op_mds); + if (!tgt) + RETURN(-ENODEV); + } else if (lmv_op_default_specific_mkdir(op_data)) { + op_data->op_mds = + op_data->op_default_mea1->lsm_md_master_mdt_index; + tgt = lmv_tgt(lmv, op_data->op_mds); + if (!tgt) + RETURN(-ENODEV); + } else if (lmv_op_qos_mkdir(op_data)) { + struct lmv_tgt_desc *tmp = tgt; + + tgt = lmv_locate_tgt_qos(lmv, &op_data->op_mds); + if (tgt == ERR_PTR(-EAGAIN)) { + if (ltd_qos_is_balanced(&lmv->lmv_mdt_descs) && + !lmv_op_default_rr_mkdir(op_data) && + !lmv_op_user_qos_mkdir(op_data)) + /* if it's not necessary, don't create remote + * directory. + */ + tgt = tmp; + else + tgt = lmv_locate_tgt_rr(lmv, &op_data->op_mds); + } if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - op_data->op_mds = tgt->ltd_idx; - } else { - CDEBUG(D_CONFIG, "Server doesn't support striped dirs\n"); + /* + * only update statfs after QoS mkdir, this means the cached + * statfs may be stale, and current mkdir may not follow QoS + * accurately, but it's not serious, and avoids periodic statfs + * when client doesn't mkdir by QoS. + */ + lmv_statfs_check_update(obd, tgt); } - CDEBUG(D_INODE, "CREATE obj "DFID" -> mds #%x\n", - PFID(&op_data->op_fid2), op_data->op_mds); +retry: + rc = lmv_fid_alloc(NULL, exp, &op_data->op_fid2, op_data); + if (rc) + RETURN(rc); + + CDEBUG(D_INODE, "CREATE name '%.*s' "DFID" on "DFID" -> mds #%x\n", + (int)op_data->op_namelen, op_data->op_name, + PFID(&op_data->op_fid2), PFID(&op_data->op_fid1), + op_data->op_mds); op_data->op_flags |= MF_MDC_CANCEL_FID1; rc = md_create(tgt->ltd_exp, op_data, data, datalen, mode, uid, gid, @@ -1676,7 +1910,30 @@ int lmv_create(struct obd_export *exp, struct md_op_data *op_data, RETURN(rc); CDEBUG(D_INODE, "Created - "DFID"\n", PFID(&op_data->op_fid2)); } - RETURN(rc); + + /* dir restripe needs to send to MDT where dir is located */ + if (rc != -EREMOTE || + !(exp_connect_flags2(exp) & OBD_CONNECT2_CRUSH)) + RETURN(rc); + + repbody = req_capsule_server_get(&(*request)->rq_pill, &RMF_MDT_BODY); + if (repbody == NULL) + RETURN(-EPROTO); + + /* Not cross-ref case, just get out of here. */ + if (likely(!(repbody->mbo_valid & OBD_MD_MDS))) + RETURN(rc); + + op_data->op_fid2 = repbody->mbo_fid1; + ptlrpc_req_finished(*request); + *request = NULL; + + tgt = lmv_fid2tgt(lmv, &op_data->op_fid2); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + op_data->op_mds = tgt->ltd_index; + goto retry; } static int @@ -1684,20 +1941,21 @@ lmv_enqueue(struct obd_export *exp, struct ldlm_enqueue_info *einfo, const union ldlm_policy_data *policy, struct md_op_data *op_data, struct lustre_handle *lockh, __u64 extra_lock_flags) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; + ENTRY; CDEBUG(D_INODE, "ENQUEUE on "DFID"\n", PFID(&op_data->op_fid1)); - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); + tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); CDEBUG(D_INODE, "ENQUEUE on "DFID" -> mds #%u\n", - PFID(&op_data->op_fid1), tgt->ltd_idx); + PFID(&op_data->op_fid1), tgt->ltd_index); rc = md_enqueue(tgt->ltd_exp, einfo, policy, op_data, lockh, extra_lock_flags); @@ -1705,52 +1963,50 @@ lmv_enqueue(struct obd_export *exp, struct ldlm_enqueue_info *einfo, RETURN(rc); } -static int +int lmv_getattr_name(struct obd_export *exp,struct md_op_data *op_data, struct ptlrpc_request **preq) { - struct ptlrpc_request *req = NULL; - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - struct mdt_body *body; - int rc; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + struct mdt_body *body; + int rc; + ENTRY; - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); +retry: + tgt = lmv_locate_tgt(lmv, op_data); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); CDEBUG(D_INODE, "GETATTR_NAME for %*s on "DFID" -> mds #%d\n", (int)op_data->op_namelen, op_data->op_name, - PFID(&op_data->op_fid1), tgt->ltd_idx); + PFID(&op_data->op_fid1), tgt->ltd_index); rc = md_getattr_name(tgt->ltd_exp, op_data, preq); - if (rc != 0) + if (rc == -ENOENT && lmv_dir_retry_check_update(op_data)) { + ptlrpc_req_finished(*preq); + *preq = NULL; + goto retry; + } + + if (rc) RETURN(rc); body = req_capsule_server_get(&(*preq)->rq_pill, &RMF_MDT_BODY); LASSERT(body != NULL); if (body->mbo_valid & OBD_MD_MDS) { - struct lu_fid rid = body->mbo_fid1; - CDEBUG(D_INODE, "Request attrs for "DFID"\n", - PFID(&rid)); - - tgt = lmv_find_target(lmv, &rid); - if (IS_ERR(tgt)) { - ptlrpc_req_finished(*preq); - preq = NULL; - RETURN(PTR_ERR(tgt)); - } - - op_data->op_fid1 = rid; + op_data->op_fid1 = body->mbo_fid1; op_data->op_valid |= OBD_MD_FLCROSSREF; op_data->op_namelen = 0; op_data->op_name = NULL; - rc = md_getattr_name(tgt->ltd_exp, op_data, &req); + ptlrpc_req_finished(*preq); - *preq = req; + *preq = NULL; + + goto retry; } RETURN(rc); @@ -1777,12 +2033,12 @@ static int lmv_early_cancel(struct obd_export *exp, struct lmv_tgt_desc *tgt, RETURN(0); if (tgt == NULL) { - tgt = lmv_find_target(lmv, fid); + tgt = lmv_fid2tgt(lmv, fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); } - if (tgt->ltd_idx != op_tgt) { + if (tgt->ltd_index != op_tgt) { CDEBUG(D_INODE, "EARLY_CANCEL on "DFID"\n", PFID(fid)); policy.l_inodebits.bits = bits; rc = md_cancel_unused(tgt->ltd_exp, fid, &policy, @@ -1820,19 +2076,8 @@ static int lmv_link(struct obd_export *exp, struct md_op_data *op_data, op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid()); op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid()); op_data->op_cap = cfs_curproc_cap_pack(); - if (op_data->op_mea2 != NULL) { - struct lmv_stripe_md *lsm = op_data->op_mea2; - const struct lmv_oinfo *oinfo; - - oinfo = lsm_name_to_stripe_info(lsm, op_data->op_name, - op_data->op_namelen); - if (IS_ERR(oinfo)) - RETURN(PTR_ERR(oinfo)); - - op_data->op_fid2 = oinfo->lmo_fid; - } - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid2); + tgt = lmv_locate_tgt2(lmv, op_data); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -1840,7 +2085,7 @@ static int lmv_link(struct obd_export *exp, struct md_op_data *op_data, * Cancel UPDATE lock on child (fid1). */ op_data->op_flags |= MF_MDC_CANCEL_FID2; - rc = lmv_early_cancel(exp, NULL, op_data, tgt->ltd_idx, LCK_EX, + rc = lmv_early_cancel(exp, NULL, op_data, tgt->ltd_index, LCK_EX, MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID1); if (rc != 0) RETURN(rc); @@ -1850,158 +2095,278 @@ static int lmv_link(struct obd_export *exp, struct md_op_data *op_data, RETURN(rc); } -static int lmv_rename(struct obd_export *exp, struct md_op_data *op_data, - const char *old, size_t oldlen, - const char *new, size_t newlen, - struct ptlrpc_request **request) +static int lmv_migrate(struct obd_export *exp, struct md_op_data *op_data, + const char *name, size_t namelen, + struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *src_tgt; - struct lmv_tgt_desc *tgt_tgt; - struct obd_export *target_exp; - struct mdt_body *body; - int rc; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_stripe_md *lsm = op_data->op_mea1; + struct lmv_tgt_desc *parent_tgt; + struct lmv_tgt_desc *sp_tgt; + struct lmv_tgt_desc *tp_tgt = NULL; + struct lmv_tgt_desc *child_tgt; + struct lmv_tgt_desc *tgt; + struct lu_fid target_fid = { 0 }; + int rc; + ENTRY; - LASSERT(oldlen != 0); + LASSERT(op_data->op_cli_flags & CLI_MIGRATE); - CDEBUG(D_INODE, "RENAME %.*s in "DFID":%d to %.*s in "DFID":%d\n", - (int)oldlen, old, PFID(&op_data->op_fid1), - op_data->op_mea1 ? op_data->op_mea1->lsm_md_stripe_count : 0, - (int)newlen, new, PFID(&op_data->op_fid2), - op_data->op_mea2 ? op_data->op_mea2->lsm_md_stripe_count : 0); + CDEBUG(D_INODE, "MIGRATE "DFID"/%.*s\n", + PFID(&op_data->op_fid1), (int)namelen, name); op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid()); op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid()); op_data->op_cap = cfs_curproc_cap_pack(); - if (op_data->op_cli_flags & CLI_MIGRATE) { - LASSERTF(fid_is_sane(&op_data->op_fid3), "invalid FID "DFID"\n", - PFID(&op_data->op_fid3)); - - if (op_data->op_mea1 != NULL) { - struct lmv_stripe_md *lsm = op_data->op_mea1; - struct lmv_tgt_desc *tmp; - - /* Fix the parent fid for striped dir */ - tmp = lmv_locate_target_for_name(lmv, lsm, old, - oldlen, - &op_data->op_fid1, - NULL); - if (IS_ERR(tmp)) - RETURN(PTR_ERR(tmp)); - } - - rc = lmv_fid_alloc(NULL, exp, &op_data->op_fid2, op_data); - if (rc != 0) - RETURN(rc); - src_tgt = lmv_find_target(lmv, &op_data->op_fid3); - if (IS_ERR(src_tgt)) - RETURN(PTR_ERR(src_tgt)); - - target_exp = src_tgt->ltd_exp; - } else { - if (op_data->op_mea1 != NULL) { - struct lmv_stripe_md *lsm = op_data->op_mea1; - - src_tgt = lmv_locate_target_for_name(lmv, lsm, old, - oldlen, - &op_data->op_fid1, - &op_data->op_mds); - } else { - src_tgt = lmv_find_target(lmv, &op_data->op_fid1); - } - if (IS_ERR(src_tgt)) - RETURN(PTR_ERR(src_tgt)); + parent_tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); + if (IS_ERR(parent_tgt)) + RETURN(PTR_ERR(parent_tgt)); + if (lmv_dir_striped(lsm)) { + const struct lmv_oinfo *oinfo; - if (op_data->op_mea2 != NULL) { - struct lmv_stripe_md *lsm = op_data->op_mea2; + oinfo = lsm_name_to_stripe_info(lsm, name, namelen, false); + if (IS_ERR(oinfo)) + RETURN(PTR_ERR(oinfo)); - tgt_tgt = lmv_locate_target_for_name(lmv, lsm, new, - newlen, - &op_data->op_fid2, - &op_data->op_mds); - } else { - tgt_tgt = lmv_find_target(lmv, &op_data->op_fid2); + /* save source stripe FID in fid4 temporarily for ELC */ + op_data->op_fid4 = oinfo->lmo_fid; + sp_tgt = lmv_tgt(lmv, oinfo->lmo_mds); + if (!sp_tgt) + RETURN(-ENODEV); + /* + * if parent is being migrated too, fill op_fid2 with target + * stripe fid, otherwise the target stripe is not created yet. + */ + if (lmv_dir_layout_changing(lsm)) { + oinfo = lsm_name_to_stripe_info(lsm, name, namelen, + true); + if (IS_ERR(oinfo)) + RETURN(PTR_ERR(oinfo)); + + op_data->op_fid2 = oinfo->lmo_fid; + tp_tgt = lmv_tgt(lmv, oinfo->lmo_mds); + if (!tp_tgt) + RETURN(-ENODEV); } - if (IS_ERR(tgt_tgt)) - RETURN(PTR_ERR(tgt_tgt)); - - target_exp = tgt_tgt->ltd_exp; + } else { + sp_tgt = parent_tgt; } - /* - * LOOKUP lock on src child (fid3) should also be cancelled for - * src_tgt in mdc_rename. - */ - op_data->op_flags |= MF_MDC_CANCEL_FID1 | MF_MDC_CANCEL_FID3; + child_tgt = lmv_fid2tgt(lmv, &op_data->op_fid3); + if (IS_ERR(child_tgt)) + RETURN(PTR_ERR(child_tgt)); - /* - * Cancel UPDATE locks on tgt parent (fid2), tgt_tgt is its - * own target. + /* for directory, migrate to MDT specified by lum_stripe_offset; + * otherwise migrate to the target stripe of parent, but parent + * directory may have finished migration (normally current file too), + * allocate FID on MDT lum_stripe_offset, and server will check + * whether file was migrated already. */ - rc = lmv_early_cancel(exp, NULL, op_data, src_tgt->ltd_idx, - LCK_EX, MDS_INODELOCK_UPDATE, - MF_MDC_CANCEL_FID2); + if (S_ISDIR(op_data->op_mode) || !tp_tgt) { + struct lmv_user_md *lum = op_data->op_data; - if (rc != 0) + op_data->op_mds = le32_to_cpu(lum->lum_stripe_offset); + } else { + op_data->op_mds = tp_tgt->ltd_index; + } + rc = lmv_fid_alloc(NULL, exp, &target_fid, op_data); + if (rc) RETURN(rc); + /* - * Cancel LOOKUP locks on source child (fid3) for parent tgt_tgt. + * for directory, send migrate request to the MDT where the object will + * be migrated to, because we can't create a striped directory remotely. + * + * otherwise, send to the MDT where source is located because regular + * file may open lease. + * + * NB. if MDT doesn't support DIR_MIGRATE, send to source MDT too for + * backward compatibility. */ - if (fid_is_sane(&op_data->op_fid3)) { - struct lmv_tgt_desc *tgt; - - tgt = lmv_find_target(lmv, &op_data->op_fid1); + if (S_ISDIR(op_data->op_mode) && + (exp_connect_flags2(exp) & OBD_CONNECT2_DIR_MIGRATE)) { + tgt = lmv_fid2tgt(lmv, &target_fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); + } else { + tgt = child_tgt; + } - /* Cancel LOOKUP lock on its parent */ - rc = lmv_early_cancel(exp, tgt, op_data, src_tgt->ltd_idx, - LCK_EX, MDS_INODELOCK_LOOKUP, - MF_MDC_CANCEL_FID3); - if (rc != 0) + /* cancel UPDATE lock of parent master object */ + rc = lmv_early_cancel(exp, parent_tgt, op_data, tgt->ltd_index, LCK_EX, + MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID1); + if (rc) + RETURN(rc); + + /* cancel UPDATE lock of source parent */ + if (sp_tgt != parent_tgt) { + /* + * migrate RPC packs master object FID, because we can only pack + * two FIDs in reint RPC, but MDS needs to know both source + * parent and target parent, and it will obtain them from master + * FID and LMV, the other FID in RPC is kept for target. + * + * since this FID is not passed to MDC, cancel it anyway. + */ + rc = lmv_early_cancel(exp, sp_tgt, op_data, -1, LCK_EX, + MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID4); + if (rc) RETURN(rc); - rc = lmv_early_cancel(exp, NULL, op_data, src_tgt->ltd_idx, - LCK_EX, MDS_INODELOCK_FULL, + op_data->op_flags &= ~MF_MDC_CANCEL_FID4; + } + op_data->op_fid4 = target_fid; + + /* cancel UPDATE locks of target parent */ + rc = lmv_early_cancel(exp, tp_tgt, op_data, tgt->ltd_index, LCK_EX, + MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID2); + if (rc) + RETURN(rc); + + /* cancel LOOKUP lock of source if source is remote object */ + if (child_tgt != sp_tgt) { + rc = lmv_early_cancel(exp, sp_tgt, op_data, tgt->ltd_index, + LCK_EX, MDS_INODELOCK_LOOKUP, MF_MDC_CANCEL_FID3); - if (rc != 0) + if (rc) RETURN(rc); } -retry_rename: - /* - * Cancel all the locks on tgt child (fid4). - */ + /* cancel ELC locks of source */ + rc = lmv_early_cancel(exp, child_tgt, op_data, tgt->ltd_index, LCK_EX, + MDS_INODELOCK_ELC, MF_MDC_CANCEL_FID3); + if (rc) + RETURN(rc); + + rc = md_rename(tgt->ltd_exp, op_data, name, namelen, NULL, 0, request); + + RETURN(rc); +} + +static int lmv_rename(struct obd_export *exp, struct md_op_data *op_data, + const char *old, size_t oldlen, + const char *new, size_t newlen, + struct ptlrpc_request **request) +{ + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *sp_tgt; + struct lmv_tgt_desc *tp_tgt = NULL; + struct lmv_tgt_desc *src_tgt = NULL; + struct lmv_tgt_desc *tgt; + struct mdt_body *body; + int rc; + + ENTRY; + + LASSERT(oldlen != 0); + + if (op_data->op_cli_flags & CLI_MIGRATE) { + rc = lmv_migrate(exp, op_data, old, oldlen, request); + RETURN(rc); + } + + op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid()); + op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid()); + op_data->op_cap = cfs_curproc_cap_pack(); + + op_data->op_name = new; + op_data->op_namelen = newlen; + + tp_tgt = lmv_locate_tgt2(lmv, op_data); + if (IS_ERR(tp_tgt)) + RETURN(PTR_ERR(tp_tgt)); + + /* Since the target child might be destroyed, and it might become + * orphan, and we can only check orphan on the local MDT right now, so + * we send rename request to the MDT where target child is located. If + * target child does not exist, then it will send the request to the + * target parent */ if (fid_is_sane(&op_data->op_fid4)) { - struct lmv_tgt_desc *tgt; + tgt = lmv_fid2tgt(lmv, &op_data->op_fid4); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + } else { + tgt = tp_tgt; + } + + op_data->op_flags |= MF_MDC_CANCEL_FID4; + + /* cancel UPDATE locks of target parent */ + rc = lmv_early_cancel(exp, tp_tgt, op_data, tgt->ltd_index, LCK_EX, + MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID2); + if (rc != 0) + RETURN(rc); + + if (fid_is_sane(&op_data->op_fid4)) { + /* cancel LOOKUP lock of target on target parent */ + if (tgt != tp_tgt) { + rc = lmv_early_cancel(exp, tp_tgt, op_data, + tgt->ltd_index, LCK_EX, + MDS_INODELOCK_LOOKUP, + MF_MDC_CANCEL_FID4); + if (rc != 0) + RETURN(rc); + } + } + + if (fid_is_sane(&op_data->op_fid3)) { + src_tgt = lmv_fid2tgt(lmv, &op_data->op_fid3); + if (IS_ERR(src_tgt)) + RETURN(PTR_ERR(src_tgt)); - rc = lmv_early_cancel(exp, NULL, op_data, src_tgt->ltd_idx, - LCK_EX, MDS_INODELOCK_FULL, - MF_MDC_CANCEL_FID4); + /* cancel ELC locks of source */ + rc = lmv_early_cancel(exp, src_tgt, op_data, tgt->ltd_index, + LCK_EX, MDS_INODELOCK_ELC, + MF_MDC_CANCEL_FID3); if (rc != 0) RETURN(rc); + } - tgt = lmv_find_target(lmv, &op_data->op_fid4); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + op_data->op_name = old; + op_data->op_namelen = oldlen; +retry: + sp_tgt = lmv_locate_tgt(lmv, op_data); + if (IS_ERR(sp_tgt)) + RETURN(PTR_ERR(sp_tgt)); + + /* cancel UPDATE locks of source parent */ + rc = lmv_early_cancel(exp, sp_tgt, op_data, tgt->ltd_index, LCK_EX, + MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID1); + if (rc != 0) + RETURN(rc); - /* Since the target child might be destroyed, and it might - * become orphan, and we can only check orphan on the local - * MDT right now, so we send rename request to the MDT where - * target child is located. If target child does not exist, - * then it will send the request to the target parent */ - target_exp = tgt->ltd_exp; + if (fid_is_sane(&op_data->op_fid3)) { + /* cancel LOOKUP lock of source on source parent */ + if (src_tgt != sp_tgt) { + rc = lmv_early_cancel(exp, sp_tgt, op_data, + tgt->ltd_index, LCK_EX, + MDS_INODELOCK_LOOKUP, + MF_MDC_CANCEL_FID3); + if (rc != 0) + RETURN(rc); + } } - rc = md_rename(target_exp, op_data, old, oldlen, new, newlen, - request); +rename: + CDEBUG(D_INODE, "RENAME "DFID"/%.*s to "DFID"/%.*s\n", + PFID(&op_data->op_fid1), (int)oldlen, old, + PFID(&op_data->op_fid2), (int)newlen, new); + + rc = md_rename(tgt->ltd_exp, op_data, old, oldlen, new, newlen, + request); + if (rc == -ENOENT && lmv_dir_retry_check_update(op_data)) { + ptlrpc_req_finished(*request); + *request = NULL; + goto retry; + } - if (rc != 0 && rc != -EXDEV) + if (rc && rc != -EXDEV) RETURN(rc); body = req_capsule_server_get(&(*request)->rq_pill, &RMF_MDT_BODY); @@ -2012,29 +2377,46 @@ retry_rename: if (likely(!(body->mbo_valid & OBD_MD_MDS))) RETURN(rc); - CDEBUG(D_INODE, "%s: try rename to another MDT for "DFID"\n", - exp->exp_obd->obd_name, PFID(&body->mbo_fid1)); - op_data->op_fid4 = body->mbo_fid1; + ptlrpc_req_finished(*request); *request = NULL; - goto retry_rename; + + tgt = lmv_fid2tgt(lmv, &op_data->op_fid4); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + if (fid_is_sane(&op_data->op_fid4)) { + /* cancel LOOKUP lock of target on target parent */ + if (tgt != tp_tgt) { + rc = lmv_early_cancel(exp, tp_tgt, op_data, + tgt->ltd_index, LCK_EX, + MDS_INODELOCK_LOOKUP, + MF_MDC_CANCEL_FID4); + if (rc != 0) + RETURN(rc); + } + } + + goto rename; } static int lmv_setattr(struct obd_export *exp, struct md_op_data *op_data, void *ea, size_t ealen, struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc = 0; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc = 0; + ENTRY; - CDEBUG(D_INODE, "SETATTR for "DFID", valid 0x%x\n", - PFID(&op_data->op_fid1), op_data->op_attr.ia_valid); + CDEBUG(D_INODE, "SETATTR for "DFID", valid 0x%x/0x%x\n", + PFID(&op_data->op_fid1), op_data->op_attr.ia_valid, + op_data->op_xvalid); op_data->op_flags |= MF_MDC_CANCEL_FID1; - tgt = lmv_find_target(lmv, &op_data->op_fid1); + tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -2046,13 +2428,14 @@ static int lmv_setattr(struct obd_export *exp, struct md_op_data *op_data, static int lmv_fsync(struct obd_export *exp, const struct lu_fid *fid, struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; + ENTRY; - tgt = lmv_find_target(lmv, fid); + tgt = lmv_fid2tgt(lmv, fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -2076,7 +2459,7 @@ struct lmv_dir_ctxt { struct stripe_dirent ldc_stripes[0]; }; -static inline void put_stripe_dirent(struct stripe_dirent *stripe) +static inline void stripe_dirent_unload(struct stripe_dirent *stripe) { if (stripe->sd_page) { kunmap(stripe->sd_page); @@ -2090,59 +2473,84 @@ static inline void put_lmv_dir_ctxt(struct lmv_dir_ctxt *ctxt) { int i; - for (i = 0; i < ctxt->ldc_count; i++) - put_stripe_dirent(&ctxt->ldc_stripes[i]); + for (i = 0; i < ctxt->ldc_count; i++) + stripe_dirent_unload(&ctxt->ldc_stripes[i]); +} + +/* if @ent is dummy, or . .., get next */ +static struct lu_dirent *stripe_dirent_get(struct lmv_dir_ctxt *ctxt, + struct lu_dirent *ent, + int stripe_index) +{ + for (; ent; ent = lu_dirent_next(ent)) { + /* Skip dummy entry */ + if (le16_to_cpu(ent->lde_namelen) == 0) + continue; + + /* skip . and .. for other stripes */ + if (stripe_index && + (strncmp(ent->lde_name, ".", + le16_to_cpu(ent->lde_namelen)) == 0 || + strncmp(ent->lde_name, "..", + le16_to_cpu(ent->lde_namelen)) == 0)) + continue; + + if (le64_to_cpu(ent->lde_hash) >= ctxt->ldc_hash) + break; + } + + return ent; } -static struct lu_dirent *stripe_dirent_next(struct lmv_dir_ctxt *ctxt, +static struct lu_dirent *stripe_dirent_load(struct lmv_dir_ctxt *ctxt, struct stripe_dirent *stripe, int stripe_index) { + struct md_op_data *op_data = ctxt->ldc_op_data; + struct lmv_oinfo *oinfo; + struct lu_fid fid = op_data->op_fid1; + struct inode *inode = op_data->op_data; + struct lmv_tgt_desc *tgt; struct lu_dirent *ent = stripe->sd_ent; __u64 hash = ctxt->ldc_hash; - __u64 end; int rc = 0; + ENTRY; LASSERT(stripe == &ctxt->ldc_stripes[stripe_index]); - - if (stripe->sd_eof) - RETURN(NULL); - - if (ent) { - ent = lu_dirent_next(ent); - if (!ent) { -check_eof: - end = le64_to_cpu(stripe->sd_dp->ldp_hash_end); - LASSERTF(hash <= end, "hash %llx end %llx\n", - hash, end); + LASSERT(!ent); + + do { + if (stripe->sd_page) { + __u64 end = le64_to_cpu(stripe->sd_dp->ldp_hash_end); + + /* @hash should be the last dirent hash */ + LASSERTF(hash <= end, + "ctxt@%p stripe@%p hash %llx end %llx\n", + ctxt, stripe, hash, end); + /* unload last page */ + stripe_dirent_unload(stripe); + /* eof */ if (end == MDS_DIR_END_OFF) { - stripe->sd_ent = NULL; stripe->sd_eof = true; - RETURN(NULL); + break; } - - put_stripe_dirent(stripe); hash = end; } - } - - if (!ent) { - struct md_op_data *op_data = ctxt->ldc_op_data; - struct lmv_oinfo *oinfo; - struct lu_fid fid = op_data->op_fid1; - struct inode *inode = op_data->op_data; - struct lmv_tgt_desc *tgt; - - LASSERT(!stripe->sd_page); oinfo = &op_data->op_mea1->lsm_md_oinfo[stripe_index]; - tgt = lmv_get_target(ctxt->ldc_lmv, oinfo->lmo_mds, NULL); - if (IS_ERR(tgt)) - GOTO(out, rc = PTR_ERR(tgt)); + if (!oinfo->lmo_root) { + rc = -ENOENT; + break; + } + + tgt = lmv_tgt(ctxt->ldc_lmv, oinfo->lmo_mds); + if (!tgt) { + rc = -ENODEV; + break; + } - /* op_data will be shared by each stripe, so we need - * reset these value for each stripe */ + /* op_data is shared by stripes, reset after use */ op_data->op_fid1 = oinfo->lmo_fid; op_data->op_fid2 = oinfo->lmo_fid; op_data->op_data = oinfo->lmo_root; @@ -2155,45 +2563,48 @@ check_eof: op_data->op_data = inode; if (rc) - GOTO(out, rc); - - stripe->sd_dp = page_address(stripe->sd_page); - ent = lu_dirent_start(stripe->sd_dp); - } - - for (; ent; ent = lu_dirent_next(ent)) { - /* Skip dummy entry */ - if (le16_to_cpu(ent->lde_namelen) == 0) - continue; - - /* skip . and .. for other stripes */ - if (stripe_index && - (strncmp(ent->lde_name, ".", - le16_to_cpu(ent->lde_namelen)) == 0 || - strncmp(ent->lde_name, "..", - le16_to_cpu(ent->lde_namelen)) == 0)) - continue; - - if (le64_to_cpu(ent->lde_hash) >= hash) break; - } - if (!ent) - goto check_eof; - EXIT; + stripe->sd_dp = page_address(stripe->sd_page); + ent = stripe_dirent_get(ctxt, lu_dirent_start(stripe->sd_dp), + stripe_index); + /* in case a page filled with ., .. and dummy, read next */ + } while (!ent); -out: stripe->sd_ent = ent; - /* treat error as eof, so dir can be partially accessed */ if (rc) { - put_stripe_dirent(stripe); + LASSERT(!ent); + /* treat error as eof, so dir can be partially accessed */ stripe->sd_eof = true; LCONSOLE_WARN("dir "DFID" stripe %d readdir failed: %d, " "directory is partially accessed!\n", PFID(&ctxt->ldc_op_data->op_fid1), stripe_index, rc); } - return ent; + + RETURN(ent); +} + +static int lmv_file_resync(struct obd_export *exp, struct md_op_data *data) +{ + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; + + ENTRY; + + rc = lmv_check_connect(obd); + if (rc != 0) + RETURN(rc); + + tgt = lmv_fid2tgt(lmv, &data->op_fid1); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + data->op_flags |= MF_MDC_CANCEL_FID1; + rc = md_file_resync(tgt->ltd_exp, data); + RETURN(rc); } /** @@ -2225,8 +2636,7 @@ static struct lu_dirent *lmv_dirent_next(struct lmv_dir_ctxt *ctxt) continue; if (!stripe->sd_ent) { - /* locate starting entry */ - stripe_dirent_next(ctxt, stripe, i); + stripe_dirent_load(ctxt, stripe, i); if (!stripe->sd_ent) { LASSERT(stripe->sd_eof); continue; @@ -2247,7 +2657,8 @@ static struct lu_dirent *lmv_dirent_next(struct lmv_dir_ctxt *ctxt) stripe = &ctxt->ldc_stripes[min]; ent = stripe->sd_ent; /* pop found dirent */ - stripe_dirent_next(ctxt, stripe, min); + stripe->sd_ent = stripe_dirent_get(ctxt, lu_dirent_next(ent), + min); } return ent; @@ -2389,23 +2800,26 @@ free_page: return rc; } -int lmv_read_page(struct obd_export *exp, struct md_op_data *op_data, - struct md_callback *cb_op, __u64 offset, - struct page **ppage) +static int lmv_read_page(struct obd_export *exp, struct md_op_data *op_data, + struct md_callback *cb_op, __u64 offset, + struct page **ppage) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_stripe_md *lsm = op_data->op_mea1; - struct lmv_tgt_desc *tgt; - int rc; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; + ENTRY; - if (unlikely(lsm != NULL)) { + if (unlikely(lmv_dir_foreign(op_data->op_mea1))) + RETURN(-ENODATA); + + if (unlikely(lmv_dir_striped(op_data->op_mea1))) { rc = lmv_striped_read_page(exp, op_data, cb_op, offset, ppage); RETURN(rc); } - tgt = lmv_find_target(lmv, &op_data->op_fid1); + tgt = lmv_fid2tgt(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -2439,68 +2853,34 @@ int lmv_read_page(struct obd_export *exp, struct md_op_data *op_data, * negative errno if failed. */ static int lmv_unlink(struct obd_export *exp, struct md_op_data *op_data, - struct ptlrpc_request **request) + struct ptlrpc_request **request) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt = NULL; - struct lmv_tgt_desc *parent_tgt = NULL; - struct mdt_body *body; - int rc; - int stripe_index = 0; - struct lmv_stripe_md *lsm = op_data->op_mea1; - ENTRY; - -retry_unlink: - /* For striped dir, we need to locate the parent as well */ - if (lsm != NULL) { - struct lmv_tgt_desc *tmp; - - LASSERT(op_data->op_name != NULL && - op_data->op_namelen != 0); - - tmp = lmv_locate_target_for_name(lmv, lsm, - op_data->op_name, - op_data->op_namelen, - &op_data->op_fid1, - &op_data->op_mds); - - /* return -EBADFD means unknown hash type, might - * need try all sub-stripe here */ - if (IS_ERR(tmp) && PTR_ERR(tmp) != -EBADFD) - RETURN(PTR_ERR(tmp)); - - /* Note: both migrating dir and unknown hash dir need to - * try all of sub-stripes, so we need start search the - * name from stripe 0, but migrating dir is already handled - * inside lmv_locate_target_for_name(), so we only check - * unknown hash type directory here */ - if (!lmv_is_known_hash_type(lsm->lsm_md_hash_type)) { - struct lmv_oinfo *oinfo; - - oinfo = &lsm->lsm_md_oinfo[stripe_index]; - - op_data->op_fid1 = oinfo->lmo_fid; - op_data->op_mds = oinfo->lmo_mds; - } - } - -try_next_stripe: - /* Send unlink requests to the MDT where the child is located */ - if (likely(!fid_is_zero(&op_data->op_fid2))) - tgt = lmv_find_target(lmv, &op_data->op_fid2); - else if (lsm != NULL) - tgt = lmv_get_target(lmv, op_data->op_mds, NULL); - else - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + struct lmv_tgt_desc *parent_tgt; + struct mdt_body *body; + int rc; - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ENTRY; op_data->op_fsuid = from_kuid(&init_user_ns, current_fsuid()); op_data->op_fsgid = from_kgid(&init_user_ns, current_fsgid()); op_data->op_cap = cfs_curproc_cap_pack(); +retry: + parent_tgt = lmv_locate_tgt(lmv, op_data); + if (IS_ERR(parent_tgt)) + RETURN(PTR_ERR(parent_tgt)); + + if (likely(!fid_is_zero(&op_data->op_fid2))) { + tgt = lmv_fid2tgt(lmv, &op_data->op_fid2); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + } else { + tgt = parent_tgt; + } + /* * If child's fid is given, cancel unused locks for it if it is from * another export than parent. @@ -2510,50 +2890,30 @@ try_next_stripe: */ op_data->op_flags |= MF_MDC_CANCEL_FID1 | MF_MDC_CANCEL_FID3; - /* - * Cancel FULL locks on child (fid3). - */ - parent_tgt = lmv_find_target(lmv, &op_data->op_fid1); - if (IS_ERR(parent_tgt)) - RETURN(PTR_ERR(parent_tgt)); - - if (parent_tgt != tgt) { - rc = lmv_early_cancel(exp, parent_tgt, op_data, tgt->ltd_idx, + if (parent_tgt != tgt) + rc = lmv_early_cancel(exp, parent_tgt, op_data, tgt->ltd_index, LCK_EX, MDS_INODELOCK_LOOKUP, MF_MDC_CANCEL_FID3); - } - rc = lmv_early_cancel(exp, NULL, op_data, tgt->ltd_idx, LCK_EX, - MDS_INODELOCK_FULL, MF_MDC_CANCEL_FID3); - if (rc != 0) + rc = lmv_early_cancel(exp, NULL, op_data, tgt->ltd_index, LCK_EX, + MDS_INODELOCK_ELC, MF_MDC_CANCEL_FID3); + if (rc) RETURN(rc); CDEBUG(D_INODE, "unlink with fid="DFID"/"DFID" -> mds #%u\n", - PFID(&op_data->op_fid1), PFID(&op_data->op_fid2), tgt->ltd_idx); + PFID(&op_data->op_fid1), PFID(&op_data->op_fid2), + tgt->ltd_index); rc = md_unlink(tgt->ltd_exp, op_data, request); - if (rc != 0 && rc != -EREMOTE && rc != -ENOENT) - RETURN(rc); - - /* Try next stripe if it is needed. */ - if (rc == -ENOENT && lsm != NULL && lmv_need_try_all_stripes(lsm)) { - struct lmv_oinfo *oinfo; - - stripe_index++; - if (stripe_index >= lsm->lsm_md_stripe_count) - RETURN(rc); - - oinfo = &lsm->lsm_md_oinfo[stripe_index]; - - op_data->op_fid1 = oinfo->lmo_fid; - op_data->op_mds = oinfo->lmo_mds; - + if (rc == -ENOENT && lmv_dir_retry_check_update(op_data)) { ptlrpc_req_finished(*request); *request = NULL; - - goto try_next_stripe; + goto retry; } + if (rc != -EREMOTE) + RETURN(rc); + body = req_capsule_server_get(&(*request)->rq_pill, &RMF_MDT_BODY); if (body == NULL) RETURN(-EPROTO); @@ -2562,39 +2922,23 @@ try_next_stripe: if (likely(!(body->mbo_valid & OBD_MD_MDS))) RETURN(rc); - CDEBUG(D_INODE, "%s: try unlink to another MDT for "DFID"\n", - exp->exp_obd->obd_name, PFID(&body->mbo_fid1)); - - /* This is a remote object, try remote MDT, Note: it may - * try more than 1 time here, Considering following case - * /mnt/lustre is root on MDT0, remote1 is on MDT1 - * 1. Initially A does not know where remote1 is, it send - * unlink RPC to MDT0, MDT0 return -EREMOTE, it will - * resend unlink RPC to MDT1 (retry 1st time). - * - * 2. During the unlink RPC in flight, - * client B mv /mnt/lustre/remote1 /mnt/lustre/remote2 - * and create new remote1, but on MDT0 - * - * 3. MDT1 get unlink RPC(from A), then do remote lock on - * /mnt/lustre, then lookup get fid of remote1, and find - * it is remote dir again, and replay -EREMOTE again. - * - * 4. Then A will resend unlink RPC to MDT0. (retry 2nd times). - * - * In theory, it might try unlimited time here, but it should - * be very rare case. */ + /* This is a remote object, try remote MDT. */ op_data->op_fid2 = body->mbo_fid1; ptlrpc_req_finished(*request); *request = NULL; - goto retry_unlink; + tgt = lmv_fid2tgt(lmv, &op_data->op_fid2); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + goto retry; } static int lmv_precleanup(struct obd_device *obd) { ENTRY; - fld_client_proc_fini(&obd->u.lmv.lmv_fld); + libcfs_kkuc_group_rem(&obd->obd_uuid, 0, KUC_GRP_HSM); + fld_client_debugfs_fini(&obd->u.lmv.lmv_fld); lprocfs_obd_cleanup(obd); lprocfs_free_md_stats(obd); RETURN(0); @@ -2619,31 +2963,24 @@ static int lmv_precleanup(struct obd_device *obd) static int lmv_get_info(const struct lu_env *env, struct obd_export *exp, __u32 keylen, void *key, __u32 *vallen, void *val) { - struct obd_device *obd; - struct lmv_obd *lmv; - int rc = 0; - ENTRY; - - obd = class_exp2obd(exp); - if (obd == NULL) { - CDEBUG(D_IOCTL, "Invalid client cookie %#llx\n", - exp->exp_handle.h_cookie); - RETURN(-EINVAL); - } + struct obd_device *obd; + struct lmv_obd *lmv; + struct lu_tgt_desc *tgt; + int rc = 0; - lmv = &obd->u.lmv; - if (keylen >= strlen("remote_flag") && !strcmp(key, "remote_flag")) { - int i; + ENTRY; - LASSERT(*vallen == sizeof(__u32)); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - struct lmv_tgt_desc *tgt = lmv->tgts[i]; - /* - * All tgts should be connected when this gets called. - */ - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; + obd = class_exp2obd(exp); + if (obd == NULL) { + CDEBUG(D_IOCTL, "Invalid client cookie %#llx\n", + exp->exp_handle.h_cookie); + RETURN(-EINVAL); + } + lmv = &obd->u.lmv; + if (keylen >= strlen("remote_flag") && !strcmp(key, "remote_flag")) { + LASSERT(*vallen == sizeof(__u32)); + lmv_foreach_connected_tgt(lmv, tgt) { if (!obd_get_info(env, tgt->ltd_exp, keylen, key, vallen, val)) RETURN(0); @@ -2656,18 +2993,112 @@ static int lmv_get_info(const struct lu_env *env, struct obd_export *exp, * Forwarding this request to first MDS, it should know LOV * desc. */ - rc = obd_get_info(env, lmv->tgts[0]->ltd_exp, keylen, key, - vallen, val); + tgt = lmv_tgt(lmv, 0); + if (!tgt) + RETURN(-ENODEV); + + rc = obd_get_info(env, tgt->ltd_exp, keylen, key, vallen, val); if (!rc && KEY_IS(KEY_CONN_DATA)) exp->exp_connect_data = *(struct obd_connect_data *)val; - RETURN(rc); - } else if (KEY_IS(KEY_TGT_COUNT)) { - *((int *)val) = lmv->desc.ld_tgt_count; - RETURN(0); - } + RETURN(rc); + } else if (KEY_IS(KEY_TGT_COUNT)) { + *((int *)val) = lmv->lmv_mdt_descs.ltd_lmv_desc.ld_tgt_count; + RETURN(0); + } + + CDEBUG(D_IOCTL, "Invalid key\n"); + RETURN(-EINVAL); +} + +static int lmv_rmfid(struct obd_export *exp, struct fid_array *fa, + int *__rcs, struct ptlrpc_request_set *_set) +{ + struct obd_device *obd = class_exp2obd(exp); + struct ptlrpc_request_set *set = _set; + struct lmv_obd *lmv = &obd->u.lmv; + int tgt_count = lmv->lmv_mdt_count; + struct lu_tgt_desc *tgt; + struct fid_array *fat, **fas = NULL; + int i, rc, **rcs = NULL; + + if (!set) { + set = ptlrpc_prep_set(); + if (!set) + RETURN(-ENOMEM); + } + + /* split FIDs by targets */ + OBD_ALLOC_PTR_ARRAY(fas, tgt_count); + if (fas == NULL) + GOTO(out, rc = -ENOMEM); + OBD_ALLOC_PTR_ARRAY(rcs, tgt_count); + if (rcs == NULL) + GOTO(out_fas, rc = -ENOMEM); + + for (i = 0; i < fa->fa_nr; i++) { + unsigned int idx; + + rc = lmv_fld_lookup(lmv, &fa->fa_fids[i], &idx); + if (rc) { + CDEBUG(D_OTHER, "can't lookup "DFID": rc = %d\n", + PFID(&fa->fa_fids[i]), rc); + continue; + } + LASSERT(idx < tgt_count); + if (!fas[idx]) + OBD_ALLOC(fas[idx], offsetof(struct fid_array, + fa_fids[fa->fa_nr])); + if (!fas[idx]) + GOTO(out, rc = -ENOMEM); + if (!rcs[idx]) + OBD_ALLOC_PTR_ARRAY(rcs[idx], fa->fa_nr); + if (!rcs[idx]) + GOTO(out, rc = -ENOMEM); + + fat = fas[idx]; + fat->fa_fids[fat->fa_nr++] = fa->fa_fids[i]; + } + + lmv_foreach_connected_tgt(lmv, tgt) { + fat = fas[tgt->ltd_index]; + if (!fat || fat->fa_nr == 0) + continue; + rc = md_rmfid(tgt->ltd_exp, fat, rcs[tgt->ltd_index], set); + } + + rc = ptlrpc_set_wait(NULL, set); + if (rc == 0) { + int j = 0; + for (i = 0; i < tgt_count; i++) { + fat = fas[i]; + if (!fat || fat->fa_nr == 0) + continue; + /* copy FIDs back */ + memcpy(fa->fa_fids + j, fat->fa_fids, + fat->fa_nr * sizeof(struct lu_fid)); + /* copy rcs back */ + memcpy(__rcs + j, rcs[i], fat->fa_nr * sizeof(**rcs)); + j += fat->fa_nr; + } + } + if (set != _set) + ptlrpc_set_destroy(set); + +out: + for (i = 0; i < tgt_count; i++) { + if (fas && fas[i]) + OBD_FREE(fas[i], offsetof(struct fid_array, + fa_fids[fa->fa_nr])); + if (rcs && rcs[i]) + OBD_FREE_PTR_ARRAY(rcs[i], fa->fa_nr); + } + if (rcs) + OBD_FREE_PTR_ARRAY(rcs, tgt_count); +out_fas: + if (fas) + OBD_FREE_PTR_ARRAY(fas, tgt_count); - CDEBUG(D_IOCTL, "Invalid key\n"); - RETURN(-EINVAL); + RETURN(rc); } /** @@ -2686,13 +3117,13 @@ static int lmv_get_info(const struct lu_env *env, struct obd_export *exp, * \retval 0 on success * \retval negative negated errno on failure */ -int lmv_set_info_async(const struct lu_env *env, struct obd_export *exp, - __u32 keylen, void *key, __u32 vallen, void *val, - struct ptlrpc_request_set *set) +static int lmv_set_info_async(const struct lu_env *env, struct obd_export *exp, + __u32 keylen, void *key, __u32 vallen, void *val, + struct ptlrpc_request_set *set) { - struct lmv_tgt_desc *tgt = NULL; - struct obd_device *obd; - struct lmv_obd *lmv; + struct lmv_tgt_desc *tgt; + struct obd_device *obd; + struct lmv_obd *lmv; int rc = 0; ENTRY; @@ -2706,14 +3137,9 @@ int lmv_set_info_async(const struct lu_env *env, struct obd_export *exp, if (KEY_IS(KEY_READ_ONLY) || KEY_IS(KEY_FLUSH_CTX) || KEY_IS(KEY_DEFAULT_EASIZE)) { - int i, err = 0; - - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - tgt = lmv->tgts[i]; - - if (tgt == NULL || tgt->ltd_exp == NULL) - continue; + int err = 0; + lmv_foreach_connected_tgt(lmv, tgt) { err = obd_set_info_async(env, tgt->ltd_exp, keylen, key, vallen, val, set); if (err && rc == 0) @@ -2744,25 +3170,40 @@ static int lmv_unpack_md_v1(struct obd_export *exp, struct lmv_stripe_md *lsm, else lsm->lsm_md_hash_type = le32_to_cpu(lmm1->lmv_hash_type); lsm->lsm_md_layout_version = le32_to_cpu(lmm1->lmv_layout_version); + lsm->lsm_md_migrate_offset = le32_to_cpu(lmm1->lmv_migrate_offset); + lsm->lsm_md_migrate_hash = le32_to_cpu(lmm1->lmv_migrate_hash); cplen = strlcpy(lsm->lsm_md_pool_name, lmm1->lmv_pool_name, sizeof(lsm->lsm_md_pool_name)); if (cplen >= sizeof(lsm->lsm_md_pool_name)) RETURN(-E2BIG); - CDEBUG(D_INFO, "unpack lsm count %d, master %d hash_type %d" + CDEBUG(D_INFO, "unpack lsm count %d/%d, master %d hash_type %#x/%#x " "layout_version %d\n", lsm->lsm_md_stripe_count, - lsm->lsm_md_master_mdt_index, lsm->lsm_md_hash_type, + lsm->lsm_md_migrate_offset, lsm->lsm_md_master_mdt_index, + lsm->lsm_md_hash_type, lsm->lsm_md_migrate_hash, lsm->lsm_md_layout_version); stripe_count = le32_to_cpu(lmm1->lmv_stripe_count); for (i = 0; i < stripe_count; i++) { fid_le_to_cpu(&lsm->lsm_md_oinfo[i].lmo_fid, &lmm1->lmv_stripe_fids[i]); + /* + * set default value -1, so lmv_locate_tgt() knows this stripe + * target is not initialized. + */ + lsm->lsm_md_oinfo[i].lmo_mds = LMV_OFFSET_DEFAULT; + if (!fid_is_sane(&lsm->lsm_md_oinfo[i].lmo_fid)) + continue; + rc = lmv_fld_lookup(lmv, &lsm->lsm_md_oinfo[i].lmo_fid, &lsm->lsm_md_oinfo[i].lmo_mds); - if (rc != 0) + if (rc == -ENOENT) + continue; + + if (rc) RETURN(rc); + CDEBUG(D_INFO, "unpack fid #%d "DFID"\n", i, PFID(&lsm->lsm_md_oinfo[i].lmo_fid)); } @@ -2770,6 +3211,21 @@ static int lmv_unpack_md_v1(struct obd_export *exp, struct lmv_stripe_md *lsm, RETURN(rc); } +static inline int lmv_unpack_user_md(struct obd_export *exp, + struct lmv_stripe_md *lsm, + const struct lmv_user_md *lmu) +{ + lsm->lsm_md_magic = le32_to_cpu(lmu->lum_magic); + lsm->lsm_md_stripe_count = le32_to_cpu(lmu->lum_stripe_count); + lsm->lsm_md_master_mdt_index = le32_to_cpu(lmu->lum_stripe_offset); + lsm->lsm_md_hash_type = le32_to_cpu(lmu->lum_hash_type); + lsm->lsm_md_max_inherit = lmu->lum_max_inherit; + lsm->lsm_md_max_inherit_rr = lmu->lum_max_inherit_rr; + lsm->lsm_md_pool_name[LOV_MAXPOOLNAME] = 0; + + return 0; +} + static int lmv_unpackmd(struct obd_export *exp, struct lmv_stripe_md **lsmp, const union lmv_mds_md *lmm, size_t lmm_size) { @@ -2785,20 +3241,48 @@ static int lmv_unpackmd(struct obd_export *exp, struct lmv_stripe_md **lsmp, /* Free memmd */ if (lsm != NULL && lmm == NULL) { int i; - for (i = 0; i < lsm->lsm_md_stripe_count; i++) { - /* For migrating inode, the master stripe and master - * object will be the same, so do not need iput, see - * ll_update_lsm_md */ - if (!(lsm->lsm_md_hash_type & LMV_HASH_FLAG_MIGRATION && - i == 0) && lsm->lsm_md_oinfo[i].lmo_root != NULL) + struct lmv_foreign_md *lfm = (struct lmv_foreign_md *)lsm; + + if (lfm->lfm_magic == LMV_MAGIC_FOREIGN) { + size_t lfm_size; + + lfm_size = lfm->lfm_length + offsetof(typeof(*lfm), + lfm_value[0]); + OBD_FREE_LARGE(lfm, lfm_size); + RETURN(0); + } + + if (lmv_dir_striped(lsm)) { + for (i = 0; i < lsm->lsm_md_stripe_count; i++) iput(lsm->lsm_md_oinfo[i].lmo_root); + lsm_size = lmv_stripe_md_size(lsm->lsm_md_stripe_count); + } else { + lsm_size = lmv_stripe_md_size(0); } - lsm_size = lmv_stripe_md_size(lsm->lsm_md_stripe_count); OBD_FREE(lsm, lsm_size); *lsmp = NULL; RETURN(0); } + /* foreign lmv case */ + if (le32_to_cpu(lmm->lmv_magic) == LMV_MAGIC_FOREIGN) { + struct lmv_foreign_md *lfm = (struct lmv_foreign_md *)lsm; + + if (lfm == NULL) { + OBD_ALLOC_LARGE(lfm, lmm_size); + if (lfm == NULL) + RETURN(-ENOMEM); + *lsmp = (struct lmv_stripe_md *)lfm; + } + lfm->lfm_magic = le32_to_cpu(lmm->lmv_foreign_md.lfm_magic); + lfm->lfm_length = le32_to_cpu(lmm->lmv_foreign_md.lfm_length); + lfm->lfm_type = le32_to_cpu(lmm->lmv_foreign_md.lfm_type); + lfm->lfm_flags = le32_to_cpu(lmm->lmv_foreign_md.lfm_flags); + memcpy(&lfm->lfm_value, &lmm->lmv_foreign_md.lfm_value, + lfm->lfm_length); + RETURN(lmm_size); + } + if (le32_to_cpu(lmm->lmv_magic) == LMV_MAGIC_STRIPE) RETURN(-EPERM); @@ -2820,7 +3304,6 @@ static int lmv_unpackmd(struct obd_export *exp, struct lmv_stripe_md **lsmp, */ lsm_size = lmv_stripe_md_size(0); - lsm_size = lmv_stripe_md_size(lmv_mds_md_stripe_count_get(lmm)); if (lsm == NULL) { OBD_ALLOC(lsm, lsm_size); if (lsm == NULL) @@ -2833,6 +3316,9 @@ static int lmv_unpackmd(struct obd_export *exp, struct lmv_stripe_md **lsmp, case LMV_MAGIC_V1: rc = lmv_unpack_md_v1(exp, lsm, &lmm->lmv_md_v1); break; + case LMV_USER_MAGIC: + rc = lmv_unpack_user_md(exp, lsm, &lmm->lmv_user_md); + break; default: CERROR("%s: unrecognized magic %x\n", exp->exp_obd->obd_name, le32_to_cpu(lmm->lmv_magic)); @@ -2860,17 +3346,16 @@ static int lmv_cancel_unused(struct obd_export *exp, const struct lu_fid *fid, void *opaque) { struct lmv_obd *lmv = &exp->exp_obd->u.lmv; + struct lu_tgt_desc *tgt; + int err; int rc = 0; - __u32 i; + ENTRY; LASSERT(fid != NULL); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - struct lmv_tgt_desc *tgt = lmv->tgts[i]; - int err; - - if (tgt == NULL || tgt->ltd_exp == NULL || !tgt->ltd_active) + lmv_foreach_connected_tgt(lmv, tgt) { + if (!tgt->ltd_active) continue; err = md_cancel_unused(tgt->ltd_exp, fid, policy, mode, flags, @@ -2885,9 +3370,10 @@ static int lmv_set_lock_data(struct obd_export *exp, const struct lustre_handle *lockh, void *data, __u64 *bits) { - struct lmv_obd *lmv = &exp->exp_obd->u.lmv; - struct lmv_tgt_desc *tgt = lmv->tgts[0]; - int rc; + struct lmv_obd *lmv = &exp->exp_obd->u.lmv; + struct lmv_tgt_desc *tgt = lmv_tgt(lmv, 0); + int rc; + ENTRY; if (tgt == NULL || tgt->ltd_exp == NULL) @@ -2896,42 +3382,44 @@ static int lmv_set_lock_data(struct obd_export *exp, RETURN(rc); } -enum ldlm_mode lmv_lock_match(struct obd_export *exp, __u64 flags, - const struct lu_fid *fid, enum ldlm_type type, - union ldlm_policy_data *policy, - enum ldlm_mode mode, struct lustre_handle *lockh) +static enum ldlm_mode +lmv_lock_match(struct obd_export *exp, __u64 flags, + const struct lu_fid *fid, enum ldlm_type type, + union ldlm_policy_data *policy, + enum ldlm_mode mode, struct lustre_handle *lockh) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - enum ldlm_mode rc; - int tgt; - int i; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + enum ldlm_mode rc; + struct lu_tgt_desc *tgt; + int i; + int index; + ENTRY; CDEBUG(D_INODE, "Lock match for "DFID"\n", PFID(fid)); - /* + /* * With DNE every object can have two locks in different namespaces: * lookup lock in space of MDT storing direntry and update/open lock in * space of MDT storing inode. Try the MDT that the FID maps to first, * since this can be easily found, and only try others if that fails. */ - for (i = 0, tgt = lmv_find_target_index(lmv, fid); - i < lmv->desc.ld_tgt_count; - i++, tgt = (tgt + 1) % lmv->desc.ld_tgt_count) { - if (tgt < 0) { + for (i = 0, index = lmv_fid2tgt_index(lmv, fid); + i < lmv->lmv_mdt_descs.ltd_tgts_size; + i++, index = (index + 1) % lmv->lmv_mdt_descs.ltd_tgts_size) { + if (index < 0) { CDEBUG(D_HA, "%s: "DFID" is inaccessible: rc = %d\n", - obd->obd_name, PFID(fid), tgt); - tgt = 0; + obd->obd_name, PFID(fid), index); + index = 0; } - if (lmv->tgts[tgt] == NULL || - lmv->tgts[tgt]->ltd_exp == NULL || - lmv->tgts[tgt]->ltd_active == 0) + tgt = lmv_tgt(lmv, index); + if (!tgt || !tgt->ltd_exp || !tgt->ltd_active) continue; - rc = md_lock_match(lmv->tgts[tgt]->ltd_exp, flags, fid, - type, policy, mode, lockh); + rc = md_lock_match(tgt->ltd_exp, flags, fid, type, policy, mode, + lockh); if (rc) RETURN(rc); } @@ -2939,121 +3427,137 @@ enum ldlm_mode lmv_lock_match(struct obd_export *exp, __u64 flags, RETURN(0); } -int lmv_get_lustre_md(struct obd_export *exp, struct ptlrpc_request *req, - struct obd_export *dt_exp, struct obd_export *md_exp, - struct lustre_md *md) +static int +lmv_get_lustre_md(struct obd_export *exp, struct req_capsule *pill, + struct obd_export *dt_exp, struct obd_export *md_exp, + struct lustre_md *md) { - struct lmv_obd *lmv = &exp->exp_obd->u.lmv; - struct lmv_tgt_desc *tgt = lmv->tgts[0]; + struct lmv_obd *lmv = &exp->exp_obd->u.lmv; + struct lmv_tgt_desc *tgt = lmv_tgt(lmv, 0); - if (tgt == NULL || tgt->ltd_exp == NULL) - RETURN(-EINVAL); + if (!tgt || !tgt->ltd_exp) + return -EINVAL; - return md_get_lustre_md(lmv->tgts[0]->ltd_exp, req, dt_exp, md_exp, md); + return md_get_lustre_md(tgt->ltd_exp, pill, dt_exp, md_exp, md); } -int lmv_free_lustre_md(struct obd_export *exp, struct lustre_md *md) +static int lmv_free_lustre_md(struct obd_export *exp, struct lustre_md *md) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt = lmv->tgts[0]; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt = lmv_tgt(lmv, 0); + ENTRY; + if (md->default_lmv) { + lmv_free_memmd(md->default_lmv); + md->default_lmv = NULL; + } if (md->lmv != NULL) { lmv_free_memmd(md->lmv); md->lmv = NULL; } - if (tgt == NULL || tgt->ltd_exp == NULL) + if (!tgt || !tgt->ltd_exp) RETURN(-EINVAL); - RETURN(md_free_lustre_md(lmv->tgts[0]->ltd_exp, md)); + RETURN(md_free_lustre_md(tgt->ltd_exp, md)); } -int lmv_set_open_replay_data(struct obd_export *exp, - struct obd_client_handle *och, - struct lookup_intent *it) +static int lmv_set_open_replay_data(struct obd_export *exp, + struct obd_client_handle *och, + struct lookup_intent *it) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + ENTRY; - tgt = lmv_find_target(lmv, &och->och_fid); + tgt = lmv_fid2tgt(lmv, &och->och_fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); RETURN(md_set_open_replay_data(tgt->ltd_exp, och, it)); } -int lmv_clear_open_replay_data(struct obd_export *exp, - struct obd_client_handle *och) +static int lmv_clear_open_replay_data(struct obd_export *exp, + struct obd_client_handle *och) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + + ENTRY; - tgt = lmv_find_target(lmv, &och->och_fid); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + tgt = lmv_fid2tgt(lmv, &och->och_fid); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); - RETURN(md_clear_open_replay_data(tgt->ltd_exp, och)); + RETURN(md_clear_open_replay_data(tgt->ltd_exp, och)); } -int lmv_intent_getattr_async(struct obd_export *exp, - struct md_enqueue_info *minfo) +static int lmv_intent_getattr_async(struct obd_export *exp, + struct md_op_item *item) { - struct md_op_data *op_data = &minfo->mi_data; + struct md_op_data *op_data = &item->mop_data; struct obd_device *obd = exp->exp_obd; struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt = NULL; + struct lmv_tgt_desc *ptgt; + struct lmv_tgt_desc *ctgt; int rc; + ENTRY; if (!fid_is_sane(&op_data->op_fid2)) RETURN(-EINVAL); - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ptgt = lmv_locate_tgt(lmv, op_data); + if (IS_ERR(ptgt)) + RETURN(PTR_ERR(ptgt)); + + ctgt = lmv_fid2tgt(lmv, &op_data->op_fid2); + if (IS_ERR(ctgt)) + RETURN(PTR_ERR(ctgt)); /* - * no special handle for remote dir, which needs to fetch both LOOKUP - * lock on parent, and then UPDATE lock on child MDT, which makes all - * complicated because this is done async. So only LOOKUP lock is - * fetched for remote dir, but considering remote dir is rare case, - * and not supporting it in statahead won't cause any issue, just leave - * it as is. + * remote object needs two RPCs to lookup and getattr, considering the + * complexity don't support statahead for now. */ + if (ctgt != ptgt) + RETURN(-EREMOTE); + + rc = md_intent_getattr_async(ptgt->ltd_exp, item); - rc = md_intent_getattr_async(tgt->ltd_exp, minfo); RETURN(rc); } -int lmv_revalidate_lock(struct obd_export *exp, struct lookup_intent *it, - struct lu_fid *fid, __u64 *bits) +static int lmv_revalidate_lock(struct obd_export *exp, struct lookup_intent *it, + struct lu_fid *fid, __u64 *bits) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt; - int rc; - ENTRY; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt; + int rc; - tgt = lmv_find_target(lmv, fid); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); + ENTRY; + + tgt = lmv_fid2tgt(lmv, fid); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); - rc = md_revalidate_lock(tgt->ltd_exp, it, fid, bits); - RETURN(rc); + rc = md_revalidate_lock(tgt->ltd_exp, it, fid, bits); + RETURN(rc); } -int lmv_get_fid_from_lsm(struct obd_export *exp, - const struct lmv_stripe_md *lsm, - const char *name, int namelen, struct lu_fid *fid) +static int lmv_get_fid_from_lsm(struct obd_export *exp, + const struct lmv_stripe_md *lsm, + const char *name, int namelen, + struct lu_fid *fid) { const struct lmv_oinfo *oinfo; - LASSERT(lsm != NULL); - oinfo = lsm_name_to_stripe_info(lsm, name, namelen); + LASSERT(lmv_dir_striped(lsm)); + + oinfo = lsm_name_to_stripe_info(lsm, name, namelen, false); if (IS_ERR(oinfo)) return PTR_ERR(oinfo); @@ -3067,52 +3571,49 @@ int lmv_get_fid_from_lsm(struct obd_export *exp, * process with other slave MDTs. The only exception is Q_GETOQUOTA for which * we directly fetch data from the slave MDTs. */ -int lmv_quotactl(struct obd_device *unused, struct obd_export *exp, - struct obd_quotactl *oqctl) -{ - struct obd_device *obd = class_exp2obd(exp); - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt = lmv->tgts[0]; - int rc = 0; - __u32 i; - __u64 curspace, curinodes; +static int lmv_quotactl(struct obd_device *unused, struct obd_export *exp, + struct obd_quotactl *oqctl) +{ + struct obd_device *obd = class_exp2obd(exp); + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt = lmv_tgt(lmv, 0); + __u64 curspace, curinodes; + int rc = 0; + ENTRY; - if (tgt == NULL || - tgt->ltd_exp == NULL || - !tgt->ltd_active || - lmv->desc.ld_tgt_count == 0) { + if (!tgt || !tgt->ltd_exp || !tgt->ltd_active) { CERROR("master lmv inactive\n"); RETURN(-EIO); } - if (oqctl->qc_cmd != Q_GETOQUOTA) { - rc = obd_quotactl(tgt->ltd_exp, oqctl); - RETURN(rc); - } + if (oqctl->qc_cmd != Q_GETOQUOTA) { + rc = obd_quotactl(tgt->ltd_exp, oqctl); + RETURN(rc); + } - curspace = curinodes = 0; - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { + curspace = curinodes = 0; + lmv_foreach_connected_tgt(lmv, tgt) { int err; - tgt = lmv->tgts[i]; - if (tgt == NULL || tgt->ltd_exp == NULL || !tgt->ltd_active) + if (!tgt->ltd_active) continue; - err = obd_quotactl(tgt->ltd_exp, oqctl); - if (err) { - CERROR("getquota on mdt %d failed. %d\n", i, err); - if (!rc) - rc = err; - } else { - curspace += oqctl->qc_dqblk.dqb_curspace; - curinodes += oqctl->qc_dqblk.dqb_curinodes; - } - } - oqctl->qc_dqblk.dqb_curspace = curspace; - oqctl->qc_dqblk.dqb_curinodes = curinodes; + err = obd_quotactl(tgt->ltd_exp, oqctl); + if (err) { + CERROR("getquota on mdt %d failed. %d\n", + tgt->ltd_index, err); + if (!rc) + rc = err; + } else { + curspace += oqctl->qc_dqblk.dqb_curspace; + curinodes += oqctl->qc_dqblk.dqb_curinodes; + } + } + oqctl->qc_dqblk.dqb_curspace = curspace; + oqctl->qc_dqblk.dqb_curinodes = curinodes; - RETURN(rc); + RETURN(rc); } static int lmv_merge_attr(struct obd_export *exp, @@ -3123,6 +3624,9 @@ static int lmv_merge_attr(struct obd_export *exp, int rc; int i; + if (!lmv_dir_striped(lsm)) + return 0; + rc = lmv_revalidate_slaves(exp, lsm, cb_blocking, 0); if (rc < 0) return rc; @@ -3130,12 +3634,15 @@ static int lmv_merge_attr(struct obd_export *exp, for (i = 0; i < lsm->lsm_md_stripe_count; i++) { struct inode *inode = lsm->lsm_md_oinfo[i].lmo_root; - CDEBUG(D_INFO, ""DFID" size %llu, blocks %llu nlink %u," - " atime %lu ctime %lu, mtime %lu.\n", + if (!inode) + continue; + + CDEBUG(D_INFO, + "" DFID " size %llu, blocks %llu nlink %u, atime %lld ctime %lld, mtime %lld.\n", PFID(&lsm->lsm_md_oinfo[i].lmo_fid), i_size_read(inode), (unsigned long long)inode->i_blocks, - inode->i_nlink, LTIME_S(inode->i_atime), - LTIME_S(inode->i_ctime), LTIME_S(inode->i_mtime)); + inode->i_nlink, (s64)inode->i_atime.tv_sec, + (s64)inode->i_ctime.tv_sec, (s64)inode->i_mtime.tv_sec); /* for slave stripe, it needs to subtract nlink for . and .. */ if (i != 0) @@ -3146,19 +3653,19 @@ static int lmv_merge_attr(struct obd_export *exp, attr->cat_size += i_size_read(inode); attr->cat_blocks += inode->i_blocks; - if (attr->cat_atime < LTIME_S(inode->i_atime)) - attr->cat_atime = LTIME_S(inode->i_atime); + if (attr->cat_atime < inode->i_atime.tv_sec) + attr->cat_atime = inode->i_atime.tv_sec; - if (attr->cat_ctime < LTIME_S(inode->i_ctime)) - attr->cat_ctime = LTIME_S(inode->i_ctime); + if (attr->cat_ctime < inode->i_ctime.tv_sec) + attr->cat_ctime = inode->i_ctime.tv_sec; - if (attr->cat_mtime < LTIME_S(inode->i_mtime)) - attr->cat_mtime = LTIME_S(inode->i_mtime); + if (attr->cat_mtime < inode->i_mtime.tv_sec) + attr->cat_mtime = inode->i_mtime.tv_sec; } return 0; } -struct obd_ops lmv_obd_ops = { +static const struct obd_ops lmv_obd_ops = { .o_owner = THIS_MODULE, .o_setup = lmv_setup, .o_cleanup = lmv_cleanup, @@ -3171,11 +3678,12 @@ struct obd_ops lmv_obd_ops = { .o_set_info_async = lmv_set_info_async, .o_notify = lmv_notify, .o_get_uuid = lmv_get_uuid, + .o_fid_alloc = lmv_fid_alloc, .o_iocontrol = lmv_iocontrol, .o_quotactl = lmv_quotactl }; -struct md_ops lmv_md_ops = { +static const struct md_ops lmv_md_ops = { .m_get_root = lmv_get_root, .m_null_inode = lmv_null_inode, .m_close = lmv_close, @@ -3190,6 +3698,7 @@ struct md_ops lmv_md_ops = { .m_setattr = lmv_setattr, .m_setxattr = lmv_setxattr, .m_fsync = lmv_fsync, + .m_file_resync = lmv_file_resync, .m_read_page = lmv_read_page, .m_unlink = lmv_unlink, .m_init_ea_size = lmv_init_ea_size, @@ -3205,11 +3714,12 @@ struct md_ops lmv_md_ops = { .m_revalidate_lock = lmv_revalidate_lock, .m_get_fid_from_lsm = lmv_get_fid_from_lsm, .m_unpackmd = lmv_unpackmd, + .m_rmfid = lmv_rmfid, }; static int __init lmv_init(void) { - return class_register_type(&lmv_obd_ops, &lmv_md_ops, true, NULL, + return class_register_type(&lmv_obd_ops, &lmv_md_ops, true, LUSTRE_LMV_NAME, NULL); }