X-Git-Url: https://git.whamcloud.com/?p=fs%2Flustre-release.git;a=blobdiff_plain;f=lustre%2Flmv%2Flmv_obd.c;h=329965e67479dc3f83a677b1f3c5922370b0dad9;hp=300710d970c1c4f99ab84ca442fe472c709b9246;hb=a857446dc6480841ab1e832970d3958f3962a885;hpb=d136d6bda8bb59d5055d2f64bef2abe6fbbfceda diff --git a/lustre/lmv/lmv_obd.c b/lustre/lmv/lmv_obd.c index 300710d..329965e 100644 --- a/lustre/lmv/lmv_obd.c +++ b/lustre/lmv/lmv_obd.c @@ -15,11 +15,7 @@ * * You should have received a copy of the GNU General Public License * version 2 along with this program; If not, see - * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf - * - * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, - * CA 95054 USA or visit www.sun.com if you need additional information or - * have any questions. + * http://www.gnu.org/licenses/gpl-2.0.html * * GPL HEADER END */ @@ -27,7 +23,7 @@ * Copyright (c) 2004, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. * - * Copyright (c) 2011, 2014, Intel Corporation. + * Copyright (c) 2011, 2017, Intel Corporation. */ /* * This file is part of Lustre, http://www.lustre.org/ @@ -35,7 +31,8 @@ */ #define DEBUG_SUBSYSTEM S_LMV -#include + +#include #include #include #include @@ -49,7 +46,6 @@ #include #include -#include #include #include #include @@ -58,10 +54,12 @@ #include #include #include -#include +#include #include #include "lmv_internal.h" +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) @@ -71,6 +69,8 @@ static void lmv_activate_target(struct lmv_obd *lmv, tgt->ltd_active = activate; lmv->desc.ld_active_tgt_count += (activate ? 1 : -1); + + tgt->ltd_exp->exp_obd->obd_inactive = !activate; } /** @@ -99,7 +99,7 @@ static int lmv_set_mdc_active(struct lmv_obd *lmv, if (tgt == NULL || tgt->ltd_exp == NULL) continue; - CDEBUG(D_INFO, "Target idx %d is %s conn "LPX64"\n", i, + CDEBUG(D_INFO, "Target idx %d is %s conn %#llx\n", i, tgt->ltd_uuid.uuid, tgt->ltd_exp->exp_handle.h_cookie); if (obd_uuid_equals(uuid, &tgt->ltd_uuid)) @@ -143,7 +143,7 @@ struct obd_uuid *lmv_get_uuid(struct obd_export *exp) } static int lmv_notify(struct obd_device *obd, struct obd_device *watched, - enum obd_notify_event ev, void *data) + enum obd_notify_event ev) { struct obd_connect_data *conn_data; struct lmv_obd *lmv = &obd->u.lmv; @@ -181,88 +181,60 @@ static int lmv_notify(struct obd_device *obd, struct obd_device *watched, */ obd->obd_self_export->exp_connect_data = *conn_data; } -#if 0 - else if (ev == OBD_NOTIFY_DISCON) { - /* - * For disconnect event, flush fld cache for failout MDS case. - */ - fld_client_flush(&lmv->lmv_fld); - } -#endif - /* - * Pass the notification up the chain. - */ - if (obd->obd_observer) - rc = obd_notify(obd->obd_observer, watched, ev, data); - RETURN(rc); + /* + * Pass the notification up the chain. + */ + if (obd->obd_observer) + rc = obd_notify(obd->obd_observer, watched, ev); + + RETURN(rc); } -/** - * This is fake connect function. Its purpose is to initialize lmv and say - * caller that everything is okay. Real connection will be performed later. - */ static int lmv_connect(const struct lu_env *env, - struct obd_export **exp, struct obd_device *obd, - struct obd_uuid *cluuid, struct obd_connect_data *data, - void *localdata) + struct obd_export **pexp, struct obd_device *obd, + struct obd_uuid *cluuid, struct obd_connect_data *data, + void *localdata) { - struct lmv_obd *lmv = &obd->u.lmv; - struct lustre_handle conn = { 0 }; - int rc = 0; - ENTRY; + struct lmv_obd *lmv = &obd->u.lmv; + struct lustre_handle conn = { 0 }; + struct obd_export *exp; + int rc; + ENTRY; - /* - * We don't want to actually do the underlying connections more than - * once, so keep track. - */ - lmv->refcount++; - if (lmv->refcount > 1) { - *exp = NULL; - RETURN(0); - } + rc = class_connect(&conn, obd, cluuid); + if (rc) { + CERROR("class_connection() returned %d\n", rc); + RETURN(rc); + } - rc = class_connect(&conn, obd, cluuid); - if (rc) { - CERROR("class_connection() returned %d\n", rc); - RETURN(rc); - } + exp = class_conn2export(&conn); - *exp = class_conn2export(&conn); - class_export_get(*exp); - - lmv->exp = *exp; - lmv->connected = 0; - lmv->cluuid = *cluuid; - - if (data) - lmv->conn_data = *data; - - 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->connected = 0; + lmv->conn_data = *data; + + 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); } - /* - * All real clients should perform actual connection right away, because - * it is possible, that LMV will not have opportunity to connect targets - * and MDC stuff will be called directly, for instance while reading - * ../mdc/../kbytesfree procfs file, etc. - */ - if (data != NULL && (data->ocd_connect_flags & OBD_CONNECT_REAL)) - rc = lmv_check_connect(obd); + rc = lmv_check_connect(obd); + if (rc != 0) + GOTO(out_sysfs, rc); + + *pexp = exp; - if (rc && lmv->targets_proc_entry != NULL) - lprocfs_remove(&lmv->targets_proc_entry); RETURN(rc); + +out_sysfs: + if (lmv->lmv_tgts_kobj) + kobject_put(lmv->lmv_tgts_kobj); + + class_disconnect(exp); + + return rc; } static int lmv_init_ea_size(struct obd_export *exp, __u32 easize, @@ -293,10 +265,12 @@ static int lmv_init_ea_size(struct obd_export *exp, __u32 easize, 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) { + if (tgt == NULL || tgt->ltd_exp == NULL) { CWARN("%s: NULL export for %d\n", obd->obd_name, i); continue; } + if (!tgt->ltd_active) + continue; rc = md_init_ea_size(tgt->ltd_exp, easize, def_easize); if (rc) { @@ -313,8 +287,6 @@ static int lmv_init_ea_size(struct obd_export *exp, __u32 easize, 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; @@ -328,18 +300,17 @@ int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) 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); } - rc = obd_connect(NULL, &mdc_exp, mdc_obd, &lmv_mdc_uuid, - &lmv->conn_data, NULL); + rc = obd_connect(NULL, &mdc_exp, mdc_obd, &obd->obd_uuid, + &lmv->conn_data, NULL); if (rc) { CERROR("target %s connect error %d\n", tgt->ltd_uuid.uuid, rc); RETURN(rc); @@ -366,18 +337,17 @@ int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) RETURN(rc); } - if (obd->obd_observer) { - /* - * Tell the observer about the new target. - */ + if (obd->obd_observer) { + /* + * Tell the observer about the new target. + */ rc = obd_notify(obd->obd_observer, mdc_exp->exp_obd, - OBD_NOTIFY_ACTIVE, - (void *)(tgt - lmv->tgts[0])); + OBD_NOTIFY_ACTIVE); if (rc) { obd_disconnect(mdc_exp); RETURN(rc); } - } + } tgt->ltd_active = 1; tgt->ltd_exp = mdc_exp; @@ -389,23 +359,11 @@ int lmv_connect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) 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); - } - } + 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); } @@ -422,6 +380,7 @@ static void lmv_del_target(struct lmv_obd *lmv, int index) 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; @@ -429,25 +388,18 @@ static int lmv_add_target(struct obd_device *obd, struct obd_uuid *uuidp, ENTRY; CDEBUG(D_CONFIG, "Target uuid: %s. index %d\n", uuidp->uuid, index); - - mutex_lock(&lmv->lmv_init_mutex); - - if (lmv->desc.ld_tgt_count == 0) { - struct obd_device *mdc_obd; - - mdc_obd = class_find_client_obd(uuidp, LUSTRE_MDC_NAME, - &obd->obd_uuid); - if (!mdc_obd) { - mutex_unlock(&lmv->lmv_init_mutex); - CERROR("%s: Target %s not attached: rc = %d\n", - obd->obd_name, uuidp->uuid, -EINVAL); - RETURN(-EINVAL); - } + mdc_obd = class_find_client_obd(uuidp, LUSTRE_MDC_NAME, + &obd->obd_uuid); + if (!mdc_obd) { + CERROR("%s: Target %s not attached: rc = %d\n", + obd->obd_name, uuidp->uuid, -EINVAL); + 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:" + CERROR("%s: UUID %s already assigned at LMV target index %d:" " rc = %d\n", obd->obd_name, obd_uuid2str(&tgt->ltd_uuid), index, -EEXIST); mutex_unlock(&lmv->lmv_init_mutex); @@ -501,26 +453,31 @@ static int lmv_add_target(struct obd_device *obd, struct obd_uuid *uuidp, lmv->desc.ld_tgt_count = index + 1; } - if (lmv->connected) { - 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 { - int easize = sizeof(struct lmv_stripe_md) + - lmv->desc.ld_tgt_count * sizeof(struct lu_fid); - lmv_init_ea_size(obd->obd_self_export, easize, 0); - } + if (lmv->connected == 0) { + /* 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 { + int easize = sizeof(struct lmv_stripe_md) + + lmv->desc.ld_tgt_count * sizeof(struct lu_fid); + lmv_init_ea_size(obd->obd_self_export, easize, 0); + } + RETURN(rc); } -int lmv_check_connect(struct obd_device *obd) +static int lmv_check_connect(struct obd_device *obd) { struct lmv_obd *lmv = &obd->u.lmv; struct lmv_tgt_desc *tgt; @@ -554,7 +511,7 @@ int lmv_check_connect(struct obd_device *obd) } 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]; @@ -565,7 +522,6 @@ int lmv_check_connect(struct obd_device *obd) GOTO(out_disc, rc); } - class_export_put(lmv->exp); lmv->connected = 1; easize = lmv_mds_md_size(lmv->desc.ld_tgt_count, LMV_MAGIC); lmv_init_ea_size(obd->obd_self_export, easize, 0); @@ -589,9 +545,10 @@ int lmv_check_connect(struct obd_device *obd) } } } - class_disconnect(lmv->exp); + mutex_unlock(&lmv->lmv_init_mutex); - RETURN(rc); + + RETURN(rc); } static int lmv_disconnect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) @@ -611,9 +568,9 @@ static int lmv_disconnect_mdc(struct obd_device *obd, struct lmv_tgt_desc *tgt) 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); @@ -649,13 +606,6 @@ static int lmv_disconnect(struct obd_export *exp) if (!lmv->tgts) goto out_local; - /* - * Only disconnect the underlying layers on the final disconnect. - */ - lmv->refcount--; - if (lmv->refcount != 0) - goto out_local; - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { if (lmv->tgts[i] == NULL || lmv->tgts[i]->ltd_exp == NULL) continue; @@ -663,11 +613,8 @@ static int lmv_disconnect(struct obd_export *exp) lmv_disconnect_mdc(obd, lmv->tgts[i]); } - 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); + if (lmv->lmv_tgts_kobj) + kobject_put(lmv->lmv_tgts_kobj); out_local: /* @@ -677,26 +624,31 @@ out_local: if (!lmv->connected) class_export_put(exp); rc = class_disconnect(exp); - if (lmv->refcount == 0) - lmv->connected = 0; - RETURN(rc); + lmv->connected = 0; + + RETURN(rc); } -static int lmv_fid2path(struct obd_export *exp, int len, void *karg, void *uarg) +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 getinfo_fid2path *gf; struct lmv_tgt_desc *tgt; struct getinfo_fid2path *remote_gf = NULL; + struct lu_fid root_fid; int remote_gf_size = 0; int rc; - gf = (struct getinfo_fid2path *)karg; + gf = karg; tgt = lmv_find_target(lmv, &gf->gf_fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); + root_fid = *gf->gf_u.gf_root_fid; + LASSERT(fid_is_sane(&root_fid)); + repeat_fid2path: rc = obd_iocontrol(OBD_IOC_FID2PATH, tgt->ltd_exp, len, gf, uarg); if (rc != 0 && rc != -EREMOTE) @@ -709,23 +661,24 @@ repeat_fid2path: char *ptr; ori_gf = (struct getinfo_fid2path *)karg; - if (strlen(ori_gf->gf_path) + - strlen(gf->gf_path) > ori_gf->gf_pathlen) + if (strlen(ori_gf->gf_u.gf_path) + 1 + + strlen(gf->gf_u.gf_path) + 1 > ori_gf->gf_pathlen) GOTO(out_fid2path, rc = -EOVERFLOW); - ptr = ori_gf->gf_path; + ptr = ori_gf->gf_u.gf_path; - memmove(ptr + strlen(gf->gf_path) + 1, ptr, - strlen(ori_gf->gf_path)); + memmove(ptr + strlen(gf->gf_u.gf_path) + 1, ptr, + strlen(ori_gf->gf_u.gf_path)); - strncpy(ptr, gf->gf_path, strlen(gf->gf_path)); - ptr += strlen(gf->gf_path); + strncpy(ptr, gf->gf_u.gf_path, + strlen(gf->gf_u.gf_path)); + ptr += strlen(gf->gf_u.gf_path); *ptr = '/'; } - CDEBUG(D_INFO, "%s: get path %s "DFID" rec: "LPU64" ln: %u\n", + CDEBUG(D_INFO, "%s: get path %s "DFID" rec: %llu ln: %u\n", tgt->ltd_exp->exp_obd->obd_name, - gf->gf_path, PFID(&gf->gf_fid), gf->gf_recno, + gf->gf_u.gf_path, PFID(&gf->gf_fid), gf->gf_recno, gf->gf_linkno); if (rc == 0) @@ -754,7 +707,8 @@ repeat_fid2path: remote_gf->gf_fid = gf->gf_fid; remote_gf->gf_recno = -1; remote_gf->gf_linkno = -1; - memset(remote_gf->gf_path, 0, remote_gf->gf_pathlen); + memset(remote_gf->gf_u.gf_path, 0, remote_gf->gf_pathlen); + *remote_gf->gf_u.gf_root_fid = root_fid; gf = remote_gf; goto repeat_fid2path; @@ -775,13 +729,15 @@ static int lmv_hsm_req_count(struct lmv_obd *lmv, /* 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); + if (IS_ERR(curr_tgt)) + RETURN(PTR_ERR(curr_tgt)); if (obd_uuid_equals(&curr_tgt->ltd_uuid, &tgt_mds->ltd_uuid)) nr++; } return nr; } -static void lmv_hsm_req_build(struct lmv_obd *lmv, +static int lmv_hsm_req_build(struct lmv_obd *lmv, struct hsm_user_request *hur_in, const struct lmv_tgt_desc *tgt_mds, struct hsm_user_request *hur_out) @@ -795,6 +751,8 @@ static void lmv_hsm_req_build(struct lmv_obd *lmv, 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); + if (IS_ERR(curr_tgt)) + RETURN(PTR_ERR(curr_tgt)); if (obd_uuid_equals(&curr_tgt->ltd_uuid, &tgt_mds->ltd_uuid)) { hur_out->hur_user_item[nr_out] = hur_in->hur_user_item[i]; @@ -804,14 +762,17 @@ static void lmv_hsm_req_build(struct lmv_obd *lmv, hur_out->hur_request.hr_itemcount = nr_out; memcpy(hur_data(hur_out), hur_data(hur_in), hur_in->hur_request.hr_data_len); + + RETURN(0); } -static int lmv_hsm_ct_unregister(struct lmv_obd *lmv, unsigned int cmd, int len, - struct lustre_kernelcomm *lk, void *uarg) +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 kkuc_ct_data *kcd = NULL; + struct lmv_obd *lmv = &obd->u.lmv; + __u32 i; + int rc; ENTRY; /* unregister request (call from llapi_hsm_copytool_fini) */ @@ -829,23 +790,36 @@ static int lmv_hsm_ct_unregister(struct lmv_obd *lmv, unsigned int cmd, int len, * Unreached coordinators will get EPIPE on next requests * and will unregister automatically. */ - rc = libcfs_kkuc_group_rem(lk->lk_uid, lk->lk_group, (void **)&kcd); - if (kcd != NULL) - OBD_FREE_PTR(kcd); + 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, 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 lmv_obd *lmv = &obd->u.lmv; struct file *filp; __u32 i, j; - int err, rc; + int err; bool any_set = false; - struct kkuc_ct_data *kcd; + struct kkuc_ct_data kcd = { + .kcd_magic = KKUC_CT_DATA_MAGIC, + .kcd_archive = lk->lk_data, + }; + int rc = 0; ENTRY; + filp = fget(lk->lk_wfd); + if (!filp) + RETURN(-EBADF); + + rc = libcfs_kkuc_group_add(filp, &obd->obd_uuid, lk->lk_uid, + lk->lk_group, &kcd, sizeof(kcd)); + 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. */ @@ -854,13 +828,14 @@ static int lmv_hsm_ct_register(struct lmv_obd *lmv, unsigned int cmd, int len, if (tgt == NULL || tgt->ltd_exp == NULL) continue; + err = obd_iocontrol(cmd, tgt->ltd_exp, len, lk, uarg); if (err) { if (tgt->ltd_active) { /* permanent error */ CERROR("%s: iocontrol MDC %s on MDT" " idx %d cmd %x: err = %d\n", - class_exp2obd(lmv->exp)->obd_name, + lmv2obd_dev(lmv)->obd_name, tgt->ltd_uuid.uuid, i, cmd, err); rc = err; lk->lk_flags |= LK_FLG_STOP; @@ -872,7 +847,7 @@ static int lmv_hsm_ct_register(struct lmv_obd *lmv, unsigned int cmd, int len, 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 @@ -884,37 +859,23 @@ 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); - OBD_ALLOC_PTR(kcd); - if (kcd == NULL) { - fput(filp); - RETURN(-ENOMEM); - } - 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); - if (rc != 0) { - if (filp != NULL) - fput(filp); - OBD_FREE_PTR(kcd); - } +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 *uarg) + int len, void *karg, void __user *uarg) { struct obd_device *obddev = class_exp2obd(exp); struct lmv_obd *lmv = &obddev->u.lmv; @@ -954,7 +915,7 @@ 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); @@ -1011,19 +972,6 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, OBD_FREE_PTR(oqctl); break; } - case OBD_IOC_CHANGELOG_SEND: - case OBD_IOC_CHANGELOG_CLEAR: { - struct ioc_changelog *icc = karg; - - if (icc->icc_mdtindex >= count) - RETURN(-ENODEV); - - tgt = lmv->tgts[icc->icc_mdtindex]; - if (tgt == NULL || tgt->ltd_exp == NULL || !tgt->ltd_active) - RETURN(-ENODEV); - rc = obd_iocontrol(cmd, tgt->ltd_exp, sizeof(*icc), icc, NULL); - break; - } case LL_IOC_GET_CONNECT_FLAGS: { tgt = lmv->tgts[0]; if (tgt == NULL || tgt->ltd_exp == NULL) @@ -1091,8 +1039,8 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, } else { /* split fid list to their respective MDS */ for (i = 0; i < count; i++) { - unsigned int nr, reqlen; - int rc1; + int nr, rc1; + size_t reqlen; struct hsm_user_request *req; tgt = lmv->tgts[i]; @@ -1100,6 +1048,8 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, continue; nr = lmv_hsm_req_count(lmv, hur, tgt); + if (nr < 0) + RETURN(nr); if (nr == 0) /* nothing for this MDS */ continue; @@ -1110,11 +1060,12 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, OBD_ALLOC_LARGE(req, reqlen); if (req == NULL) RETURN(-ENOMEM); - - lmv_hsm_req_build(lmv, hur, tgt, req); - + rc1 = lmv_hsm_req_build(lmv, hur, tgt, req); + if (rc1 < 0) + GOTO(hsm_req_err, rc1); rc1 = obd_iocontrol(cmd, tgt->ltd_exp, reqlen, req, uarg); +hsm_req_err: if (rc1 != 0 && rc == 0) rc = rc1; OBD_FREE_LARGE(req, reqlen); @@ -1147,9 +1098,9 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, 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(obddev, cmd, len, lk, uarg); else - rc = lmv_hsm_ct_register(lmv, cmd, len, lk, uarg); + rc = lmv_hsm_ct_register(obddev, cmd, len, lk, uarg); break; } default: @@ -1182,58 +1133,15 @@ static int lmv_iocontrol(unsigned int cmd, struct obd_export *exp, RETURN(rc); } -#if 0 -static int lmv_all_chars_policy(int count, const char *name, - int len) -{ - unsigned int c = 0; - - while (len > 0) - c += name[--len]; - c = c % count; - return c; -} - -static int lmv_nid_policy(struct lmv_obd *lmv) -{ - struct obd_import *imp; - __u32 id; - - /* - * XXX: To get nid we assume that underlying obd device is mdc. - */ - imp = class_exp2cliimp(lmv->tgts[0].ltd_exp); - id = imp->imp_connection->c_self ^ (imp->imp_connection->c_self >> 32); - return id % lmv->desc.ld_tgt_count; -} - -static int lmv_choose_mds(struct lmv_obd *lmv, struct md_op_data *op_data, - placement_policy_t placement) -{ - switch (placement) { - case PLACEMENT_CHAR_POLICY: - return lmv_all_chars_policy(lmv->desc.ld_tgt_count, - op_data->op_name, - op_data->op_namelen); - case PLACEMENT_NID_POLICY: - return lmv_nid_policy(lmv); - - default: - break; - } - - CERROR("Unsupported placement policy %x\n", placement); - return -EINVAL; -} -#endif - /** * 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) { - struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_user_md *lum; + ENTRY; LASSERT(mds != NULL); @@ -1243,31 +1151,22 @@ static int lmv_placement_policy(struct obd_device *obd, RETURN(0); } - if (op_data->op_default_stripe_offset != -1) { + 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_tgt()). */ + 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; - RETURN(0); - } - - /** - * If stripe_offset is provided during setdirstripe - * (setdirstripe -i xx), xx MDS will be choosen. - */ - if (op_data->op_cli_flags & CLI_SET_MEA && op_data->op_data != NULL) { - struct lmv_user_md *lum; - - lum = op_data->op_data; - - if (le32_to_cpu(lum->lum_stripe_offset) != (__u32)-1) { - *mds = le32_to_cpu(lum->lum_stripe_offset); - } else { - /* -1 means default, which will be in the same MDT with - * the stripe */ - *mds = op_data->op_mds; - lum->lum_stripe_offset = cpu_to_le32(op_data->op_mds); - } + op_data->op_mds = *mds; + /* Correct the stripe offset in lum */ + if (lum != NULL) + lum->lum_stripe_offset = cpu_to_le32(*mds); } else { - /* Allocate new fid on target according to operation type and - * parent home mds. */ *mds = op_data->op_mds; } @@ -1365,21 +1264,15 @@ static int lmv_setup(struct obd_device *obd, struct lustre_cfg *lcfg) lmv->desc.ld_active_tgt_count = 0; lmv->max_def_easize = 0; lmv->max_easize = 0; - lmv->lmv_placement = PLACEMENT_CHAR_POLICY; 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); - lprocfs_alloc_md_stats(obd, 0); - rc = lprocfs_seq_create(obd->obd_proc_entry, "target_obd", - 0444, &lmv_proc_target_fops, obd); + rc = lmv_tunables_init(obd); if (rc) - CWARN("%s: error adding LMV target_obd file: rc = %d\n", + CWARN("%s: error adding LMV sysfs/debugfs files: 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) { @@ -1444,53 +1337,85 @@ 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->desc.ld_tgt_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_NETTYP(LNET_NIDNET(lnet_id.nid)) != LOLND) { + lmv->lmv_statfs_start = + lnet_id.nid % lmv->desc.ld_tgt_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; + __u32 i, idx; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - OBD_ALLOC(temp, sizeof(*temp)); if (temp == NULL) RETURN(-ENOMEM); - for (i = 0; i < lmv->desc.ld_tgt_count; i++) { - if (lmv->tgts[i] == NULL || lmv->tgts[i]->ltd_exp == NULL) + /* distribute statfs among MDTs */ + idx = lmv_select_statfs_mdt(lmv, flags); + + for (i = 0; i < lmv->desc.ld_tgt_count; i++, idx++) { + idx = idx % lmv->desc.ld_tgt_count; + if (lmv->tgts[idx] == NULL || lmv->tgts[idx]->ltd_exp == NULL) continue; - rc = obd_statfs(env, lmv->tgts[i]->ltd_exp, temp, + rc = obd_statfs(env, lmv->tgts[idx]->ltd_exp, temp, max_age, flags); if (rc) { CERROR("can't stat MDS #%d (%s), error %d\n", i, - lmv->tgts[i]->ltd_exp->exp_obd->obd_name, + lmv->tgts[idx]->ltd_exp->exp_obd->obd_name, rc); GOTO(out_free_temp, rc); } + if (temp->os_state & OS_STATE_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; @@ -1499,27 +1424,21 @@ out_free_temp: return rc; } -static int lmv_getstatus(struct obd_export *exp, - struct lu_fid *fid, - struct obd_capa **pc) +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; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - - rc = md_getstatus(lmv->tgts[0]->ltd_exp, fid, pc); + rc = md_get_root(lmv->tgts[0]->ltd_exp, fileset, fid); RETURN(rc); } static int lmv_getxattr(struct obd_export *exp, const struct lu_fid *fid, - struct obd_capa *oc, u64 valid, const char *name, - const char *input, int input_size, int output_size, - int flags, struct ptlrpc_request **request) + 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; @@ -1527,25 +1446,20 @@ static int lmv_getxattr(struct obd_export *exp, const struct lu_fid *fid, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - tgt = lmv_find_target(lmv, fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - rc = md_getxattr(tgt->ltd_exp, fid, oc, valid, name, input, - input_size, output_size, flags, request); + rc = md_getxattr(tgt->ltd_exp, fid, obd_md_valid, name, buf_size, req); - RETURN(rc); + RETURN(rc); } static int lmv_setxattr(struct obd_export *exp, const struct lu_fid *fid, - struct obd_capa *oc, 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; @@ -1553,19 +1467,14 @@ static int lmv_setxattr(struct obd_export *exp, const struct lu_fid *fid, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - tgt = lmv_find_target(lmv, fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - rc = md_setxattr(tgt->ltd_exp, fid, oc, 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); + RETURN(rc); } static int lmv_getattr(struct obd_export *exp, struct md_op_data *op_data, @@ -1577,10 +1486,6 @@ static int lmv_getattr(struct obd_export *exp, struct md_op_data *op_data, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - tgt = lmv_find_target(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -1600,13 +1505,8 @@ 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; - int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - CDEBUG(D_INODE, "CBDATA for "DFID"\n", PFID(fid)); /* @@ -1623,50 +1523,6 @@ static int lmv_null_inode(struct obd_export *exp, const struct lu_fid *fid) RETURN(0); } -static int lmv_find_cbdata(struct obd_export *exp, const struct lu_fid *fid, - ldlm_iterator_t it, void *data) -{ - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - int i; - int tgt; - int rc; - ENTRY; - - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - - 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. 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) { - CDEBUG(D_HA, "%s: "DFID" is inaccessible: rc = %d\n", - obd->obd_name, PFID(fid), tgt); - tgt = 0; - } - - if (lmv->tgts[tgt] == NULL || - lmv->tgts[tgt]->ltd_exp == NULL) - continue; - - rc = md_find_cbdata(lmv->tgts[tgt]->ltd_exp, fid, it, data); - if (rc) - RETURN(rc); - } - - RETURN(rc); -} - - static int lmv_close(struct obd_export *exp, struct md_op_data *op_data, struct md_open_data *mod, struct ptlrpc_request **request) { @@ -1676,10 +1532,6 @@ static int lmv_close(struct obd_export *exp, struct md_op_data *op_data, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - tgt = lmv_find_target(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -1689,28 +1541,33 @@ static int lmv_close(struct obd_export *exp, struct md_op_data *op_data, 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) +struct lmv_tgt_desc* +__lmv_locate_tgt(struct lmv_obd *lmv, struct lmv_stripe_md *lsm, + const char *name, int namelen, struct lu_fid *fid, u32 *mds, + bool post_migrate) { - struct lmv_tgt_desc *tgt; - const struct lmv_oinfo *oinfo; + struct lmv_tgt_desc *tgt; + const struct lmv_oinfo *oinfo; + + if (lsm == NULL || namelen == 0) { + tgt = lmv_find_target(lmv, fid); + if (IS_ERR(tgt)) + return tgt; + + LASSERT(mds); + *mds = tgt->ltd_idx; + 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, + post_migrate); if (IS_ERR(oinfo)) - RETURN(ERR_CAST(oinfo)); + return ERR_CAST(oinfo); } if (fid != NULL) @@ -1722,18 +1579,21 @@ lmv_locate_target_for_name(struct lmv_obd *lmv, struct lmv_stripe_md *lsm, CDEBUG(D_INFO, "locate on mds %u "DFID"\n", oinfo->lmo_mds, PFID(&oinfo->lmo_fid)); + return tgt; } + /** - * Locate mds by fid or name + * Locate mdt by fid or name * - * 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 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_post_migrate' + * indicates whether old or new layout is used to locate. * * 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. @@ -1743,53 +1603,53 @@ lmv_locate_target_for_name(struct lmv_obd *lmv, struct lmv_stripe_md *lsm, * ERR_PTR(errno) if failed. */ struct lmv_tgt_desc* -lmv_locate_mds(struct lmv_obd *lmv, struct md_op_data *op_data, +lmv_locate_tgt(struct lmv_obd *lmv, struct md_op_data *op_data, struct lu_fid *fid) { - 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; /* 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 && lsm != NULL) { - int i; + (int)op_data->op_mds != -1) { tgt = lmv_get_target(lmv, op_data->op_mds, NULL); if (IS_ERR(tgt)) return tgt; - /* refill the right parent fid */ - for (i = 0; i < lsm->lsm_md_stripe_count; i++) { - struct lmv_oinfo *oinfo; + if (lsm) { + int i; - oinfo = &lsm->lsm_md_oinfo[i]; - if (oinfo->lmo_mds == op_data->op_mds) { - *fid = oinfo->lmo_fid; - break; + /* refill the right parent fid */ + for (i = 0; i < lsm->lsm_md_stripe_count; i++) { + oinfo = &lsm->lsm_md_oinfo[i]; + if (oinfo->lmo_mds == op_data->op_mds) { + *fid = oinfo->lmo_fid; + break; + } } - } - /* Hmm, can not find the stripe by mdt_index(op_mds) */ - if (i == lsm->lsm_md_stripe_count) - tgt = ERR_PTR(-EINVAL); + if (i == lsm->lsm_md_stripe_count) + *fid = lsm->lsm_md_oinfo[0].lmo_fid; + } + } else if (lmv_is_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; + *fid = oinfo->lmo_fid; + op_data->op_mds = oinfo->lmo_mds; + tgt = lmv_get_target(lmv, oinfo->lmo_mds, NULL); + } else { + tgt = __lmv_locate_tgt(lmv, lsm, op_data->op_name, + op_data->op_namelen, fid, + &op_data->op_mds, + op_data->op_post_migrate); } - if (lsm == NULL || op_data->op_namelen == 0) { - tgt = lmv_find_target(lmv, fid); - if (IS_ERR(tgt)) - return tgt; - - op_data->op_mds = tgt->ltd_idx; - return tgt; - } - - return lmv_locate_target_for_name(lmv, lsm, op_data->op_name, - op_data->op_namelen, fid, - &op_data->op_mds); -} + return tgt; +} int lmv_create(struct obd_export *exp, struct md_op_data *op_data, const void *data, size_t datalen, umode_t mode, uid_t uid, @@ -1802,14 +1662,36 @@ int lmv_create(struct obd_export *exp, struct md_op_data *op_data, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - if (!lmv->desc.ld_active_tgt_count) RETURN(-EIO); - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); + if (lmv_is_dir_bad_hash(op_data->op_mea1)) + RETURN(-EBADF); + + if (lmv_is_dir_migrating(op_data->op_mea1)) { + /* + * if parent is migrating, create() needs to lookup existing + * name, to avoid creating new file under old layout of + * migrating directory, check old layout here. + */ + tgt = lmv_locate_tgt(lmv, op_data, &op_data->op_fid1); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + rc = md_getattr_name(tgt->ltd_exp, op_data, request); + if (!rc) { + ptlrpc_req_finished(*request); + *request = NULL; + RETURN(-EEXIST); + } + + if (rc != -ENOENT) + RETURN(rc); + + op_data->op_post_migrate = true; + } + + tgt = lmv_locate_tgt(lmv, op_data, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -1820,6 +1702,7 @@ int lmv_create(struct obd_export *exp, struct md_op_data *op_data, 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 */ @@ -1848,8 +1731,7 @@ int lmv_create(struct obd_export *exp, struct md_op_data *op_data, static int lmv_enqueue(struct obd_export *exp, struct ldlm_enqueue_info *einfo, - const union ldlm_policy_data *policy, - struct lookup_intent *it, struct md_op_data *op_data, + 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; @@ -1858,43 +1740,35 @@ lmv_enqueue(struct obd_export *exp, struct ldlm_enqueue_info *einfo, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - - CDEBUG(D_INODE, "ENQUEUE '%s' on "DFID"\n", - LL_IT2STR(it), PFID(&op_data->op_fid1)); + 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_find_target(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - CDEBUG(D_INODE, "ENQUEUE '%s' on "DFID" -> mds #%u\n", - LL_IT2STR(it), PFID(&op_data->op_fid1), tgt->ltd_idx); + CDEBUG(D_INODE, "ENQUEUE on "DFID" -> mds #%u\n", + PFID(&op_data->op_fid1), tgt->ltd_idx); - rc = md_enqueue(tgt->ltd_exp, einfo, policy, it, op_data, lockh, + rc = md_enqueue(tgt->ltd_exp, einfo, policy, op_data, lockh, extra_lock_flags); 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; - ENTRY; + 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; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); + ENTRY; - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); +retry: + tgt = lmv_locate_tgt(lmv, op_data, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -1903,31 +1777,28 @@ lmv_getattr_name(struct obd_export *exp,struct md_op_data *op_data, PFID(&op_data->op_fid1), tgt->ltd_idx); 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); @@ -1941,14 +1812,13 @@ lmv_getattr_name(struct obd_export *exp,struct md_op_data *op_data, NULL) static int lmv_early_cancel(struct obd_export *exp, struct lmv_tgt_desc *tgt, - struct md_op_data *op_data, - __u32 op_tgt, ldlm_mode_t mode, int bits, int flag) -{ - struct lu_fid *fid = md_op_data_fid(op_data, flag); - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - ldlm_policy_data_t policy = {{ 0 }}; - int rc = 0; + struct md_op_data *op_data, __u32 op_tgt, + enum ldlm_mode mode, int bits, int flag) +{ + struct lu_fid *fid = md_op_data_fid(op_data, flag); + struct lmv_obd *lmv = &exp->exp_obd->u.lmv; + union ldlm_policy_data policy = { { 0 } }; + int rc = 0; ENTRY; if (!fid_is_sane(fid)) @@ -1989,10 +1859,6 @@ static int lmv_link(struct obd_export *exp, struct md_op_data *op_data, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - LASSERT(op_data->op_namelen != 0); CDEBUG(D_INODE, "LINK "DFID":%*s to "DFID"\n", @@ -2002,19 +1868,40 @@ 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)); + if (lmv_is_dir_migrating(op_data->op_mea2)) { + struct lu_fid fid1 = op_data->op_fid1; + struct lmv_stripe_md *lsm1 = op_data->op_mea1; - op_data->op_fid2 = oinfo->lmo_fid; + /* + * avoid creating new file under old layout of migrating + * directory, check it here. + */ + tgt = __lmv_locate_tgt(lmv, op_data->op_mea2, op_data->op_name, + op_data->op_namelen, &op_data->op_fid2, + &op_data->op_mds, false); + tgt = lmv_locate_tgt(lmv, op_data, &op_data->op_fid1); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(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); + *request = NULL; + RETURN(-EEXIST); + } + + if (rc != -ENOENT) + RETURN(rc); } - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid2); + tgt = __lmv_locate_tgt(lmv, op_data->op_mea2, op_data->op_name, + op_data->op_namelen, &op_data->op_fid2, + &op_data->op_mds, true); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -2032,141 +1919,355 @@ 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; - int rc; - ENTRY; + 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; + int rc; - LASSERT(oldlen != 0); + ENTRY; - 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); + LASSERT(op_data->op_cli_flags & CLI_MIGRATE); - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); + 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); + parent_tgt = lmv_find_target(lmv, &op_data->op_fid1); + if (IS_ERR(parent_tgt)) + RETURN(PTR_ERR(parent_tgt)); - src_tgt = lmv_find_target(lmv, &op_data->op_fid3); - } 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); - if (IS_ERR(src_tgt)) - RETURN(PTR_ERR(src_tgt)); - } else { - src_tgt = lmv_find_target(lmv, &op_data->op_fid1); - if (IS_ERR(src_tgt)) - RETURN(PTR_ERR(src_tgt)); + if (lsm) { + __u32 hash_type = lsm->lsm_md_hash_type; + __u32 stripe_count = lsm->lsm_md_stripe_count; - op_data->op_mds = src_tgt->ltd_idx; + /* + * old stripes are appended after new stripes for migrating + * directory. + */ + if (lsm->lsm_md_hash_type & LMV_HASH_FLAG_MIGRATION) { + hash_type = lsm->lsm_md_migrate_hash; + stripe_count -= lsm->lsm_md_migrate_offset; } - if (op_data->op_mea2) { - struct lmv_stripe_md *lsm = op_data->op_mea2; - const struct lmv_oinfo *oinfo; + rc = lmv_name_to_stripe_index(hash_type, stripe_count, name, + namelen); + if (rc < 0) + RETURN(rc); - oinfo = lsm_name_to_stripe_info(lsm, new, newlen); - if (IS_ERR(oinfo)) - RETURN(PTR_ERR(oinfo)); + if (lsm->lsm_md_hash_type & LMV_HASH_FLAG_MIGRATION) + rc += lsm->lsm_md_migrate_offset; - op_data->op_fid2 = oinfo->lmo_fid; + /* save it in fid4 temporarily for early cancel */ + op_data->op_fid4 = lsm->lsm_md_oinfo[rc].lmo_fid; + sp_tgt = lmv_get_target(lmv, lsm->lsm_md_oinfo[rc].lmo_mds, + NULL); + if (IS_ERR(sp_tgt)) + RETURN(PTR_ERR(sp_tgt)); + + /* + * if parent is being migrated too, fill op_fid2 with target + * stripe fid, otherwise the target stripe is not created yet. + */ + if (lsm->lsm_md_hash_type & LMV_HASH_FLAG_MIGRATION) { + hash_type = lsm->lsm_md_hash_type & + ~LMV_HASH_FLAG_MIGRATION; + stripe_count = lsm->lsm_md_migrate_offset; + + rc = lmv_name_to_stripe_index(hash_type, stripe_count, + name, namelen); + if (rc < 0) + RETURN(rc); + + op_data->op_fid2 = lsm->lsm_md_oinfo[rc].lmo_fid; + tp_tgt = lmv_get_target(lmv, + lsm->lsm_md_oinfo[rc].lmo_mds, + NULL); + if (IS_ERR(tp_tgt)) + RETURN(PTR_ERR(tp_tgt)); } + } else { + sp_tgt = parent_tgt; } - if (IS_ERR(src_tgt)) - RETURN(PTR_ERR(src_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; - /* - * Cancel UPDATE locks on tgt parent (fid2), tgt_tgt is its - * own target. - */ - rc = lmv_early_cancel(exp, NULL, op_data, src_tgt->ltd_idx, - LCK_EX, MDS_INODELOCK_UPDATE, - MF_MDC_CANCEL_FID2); + child_tgt = lmv_find_target(lmv, &op_data->op_fid3); + if (IS_ERR(child_tgt)) + RETURN(PTR_ERR(child_tgt)); - if (rc != 0) + if (!S_ISDIR(op_data->op_mode) && tp_tgt) + rc = __lmv_fid_alloc(lmv, &target_fid, tp_tgt->ltd_idx); + else + 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_find_target(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, + /* cancel UPDATE lock of parent master object */ + rc = lmv_early_cancel(exp, parent_tgt, op_data, tgt->ltd_idx, 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); + + 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_idx, 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_idx, LCK_EX, MDS_INODELOCK_LOOKUP, MF_MDC_CANCEL_FID3); - if (rc != 0) + if (rc) RETURN(rc); + } + + /* cancel ELC locks of source */ + rc = lmv_early_cancel(exp, child_tgt, op_data, tgt->ltd_idx, 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(); + + if (lmv_is_dir_migrating(op_data->op_mea2)) { + struct lu_fid fid1 = op_data->op_fid1; + struct lmv_stripe_md *lsm1 = op_data->op_mea1; + + /* + * we avoid creating new file under old layout of migrating + * directory, if there is an existing file with new name under + * old layout, we can't unlink file in old layout and rename to + * new layout in one transaction, so return -EBUSY here.` + */ + tgt = __lmv_locate_tgt(lmv, op_data->op_mea2, new, newlen, + &op_data->op_fid2, &op_data->op_mds, + false); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + op_data->op_fid1 = op_data->op_fid2; + op_data->op_mea1 = op_data->op_mea2; + op_data->op_name = new; + op_data->op_namelen = newlen; + rc = md_getattr_name(tgt->ltd_exp, op_data, request); + op_data->op_fid1 = fid1; + op_data->op_mea1 = lsm1; + op_data->op_name = NULL; + op_data->op_namelen = 0; + if (!rc) { + ptlrpc_req_finished(*request); + *request = NULL; + RETURN(-EBUSY); + } + + if (rc != -ENOENT) + RETURN(rc); + } + + /* rename to new layout for migrating directory */ + tp_tgt = __lmv_locate_tgt(lmv, op_data->op_mea2, new, newlen, + &op_data->op_fid2, &op_data->op_mds, true); + 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)) { + tgt = lmv_find_target(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_idx, 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_idx, 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_find_target(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, + /* cancel ELC locks of source */ + rc = lmv_early_cancel(exp, src_tgt, op_data, tgt->ltd_idx, + LCK_EX, MDS_INODELOCK_ELC, MF_MDC_CANCEL_FID3); if (rc != 0) RETURN(rc); } - /* - * Cancel all the locks on tgt child (fid4). - */ - if (fid_is_sane(&op_data->op_fid4)) - rc = lmv_early_cancel(exp, NULL, op_data, src_tgt->ltd_idx, - LCK_EX, MDS_INODELOCK_FULL, - MF_MDC_CANCEL_FID4); +retry: + sp_tgt = __lmv_locate_tgt(lmv, op_data->op_mea1, old, oldlen, + &op_data->op_fid1, &op_data->op_mds, + op_data->op_post_migrate); + 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_idx, LCK_EX, + MDS_INODELOCK_UPDATE, MF_MDC_CANCEL_FID1); + if (rc != 0) + RETURN(rc); - CDEBUG(D_INODE, DFID":m%d to "DFID"\n", PFID(&op_data->op_fid1), - op_data->op_mds, PFID(&op_data->op_fid2)); + 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_idx, LCK_EX, + MDS_INODELOCK_LOOKUP, + MF_MDC_CANCEL_FID3); + if (rc != 0) + RETURN(rc); + } + } - rc = md_rename(src_tgt->ltd_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); - RETURN(rc); + 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 && rc != -EXDEV) + RETURN(rc); + + body = req_capsule_server_get(&(*request)->rq_pill, &RMF_MDT_BODY); + if (body == NULL) + RETURN(-EPROTO); + + /* Not cross-ref case, just get out of here. */ + if (likely(!(body->mbo_valid & OBD_MD_MDS))) + RETURN(rc); + + op_data->op_fid4 = body->mbo_fid1; + + ptlrpc_req_finished(*request); + *request = NULL; + + tgt = lmv_find_target(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_idx, 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, @@ -2178,12 +2279,9 @@ static int lmv_setattr(struct obd_export *exp, struct md_op_data *op_data, int rc = 0; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - - 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); @@ -2196,7 +2294,7 @@ 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 obd_capa *oc, struct ptlrpc_request **request) + struct ptlrpc_request **request) { struct obd_device *obd = exp->exp_obd; struct lmv_obd *lmv = &obd->u.lmv; @@ -2204,158 +2302,231 @@ static int lmv_fsync(struct obd_export *exp, const struct lu_fid *fid, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc != 0) - RETURN(rc); - tgt = lmv_find_target(lmv, fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - rc = md_fsync(tgt->ltd_exp, fid, oc, request); + rc = md_fsync(tgt->ltd_exp, fid, request); RETURN(rc); } -/** - * Get current minimum entry from striped directory - * - * This function will search the dir entry, whose hash value is the - * closest(>=) to @hash_offset, from all of sub-stripes, and it is - * only being called for striped directory. - * - * \param[in] exp export of LMV - * \param[in] op_data parameters transferred beween client MD stack - * stripe_information will be included in this - * parameter - * \param[in] cb_op ldlm callback being used in enqueue in - * mdc_read_page - * \param[in] hash_offset the hash value, which is used to locate - * minum(closet) dir entry - * \param[in|out] stripe_offset the caller use this to indicate the stripe - * index of last entry, so to avoid hash conflict - * between stripes. It will also be used to - * return the stripe index of current dir entry. - * \param[in|out] entp the minum entry and it also is being used - * to input the last dir entry to resolve the - * hash conflict - * - * \param[out] ppage the page which holds the minum entry - * - * \retval = 0 get the entry successfully - * negative errno (< 0) does not get the entry - */ -static int lmv_get_min_striped_entry(struct obd_export *exp, - struct md_op_data *op_data, - struct md_callback *cb_op, - __u64 hash_offset, int *stripe_offset, - struct lu_dirent **entp, - struct page **ppage) +struct stripe_dirent { + struct page *sd_page; + struct lu_dirpage *sd_dp; + struct lu_dirent *sd_ent; + bool sd_eof; +}; + +struct lmv_dir_ctxt { + struct lmv_obd *ldc_lmv; + struct md_op_data *ldc_op_data; + struct md_callback *ldc_cb_op; + __u64 ldc_hash; + int ldc_count; + struct stripe_dirent ldc_stripes[0]; +}; + +static inline void stripe_dirent_unload(struct stripe_dirent *stripe) { - 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 stripe_count; - struct lu_dirent *min_ent = NULL; - struct page *min_page = NULL; - int min_idx = 0; - int i; - int rc = 0; - ENTRY; + if (stripe->sd_page) { + kunmap(stripe->sd_page); + put_page(stripe->sd_page); + stripe->sd_page = NULL; + stripe->sd_ent = NULL; + } +} - stripe_count = lsm->lsm_md_stripe_count; - for (i = 0; i < stripe_count; i++) { - struct lu_dirent *ent = NULL; - struct page *page = NULL; - struct lu_dirpage *dp; - __u64 stripe_hash = hash_offset; +static inline void put_lmv_dir_ctxt(struct lmv_dir_ctxt *ctxt) +{ + int i; - tgt = lmv_get_target(lmv, lsm->lsm_md_oinfo[i].lmo_mds, NULL); - if (IS_ERR(tgt)) - GOTO(out, rc = PTR_ERR(tgt)); - - /* op_data will be shared by each stripe, so we need - * reset these value for each stripe */ - op_data->op_fid1 = lsm->lsm_md_oinfo[i].lmo_fid; - op_data->op_fid2 = lsm->lsm_md_oinfo[i].lmo_fid; - op_data->op_data = lsm->lsm_md_oinfo[i].lmo_root; -next: - rc = md_read_page(tgt->ltd_exp, op_data, cb_op, stripe_hash, - &page); - if (rc != 0) - GOTO(out, rc); + for (i = 0; i < ctxt->ldc_count; i++) + stripe_dirent_unload(&ctxt->ldc_stripes[i]); +} - dp = page_address(page); - for (ent = lu_dirent_start(dp); ent != NULL; - ent = lu_dirent_next(ent)) { - /* Skip dummy entry */ - if (le16_to_cpu(ent->lde_namelen) == 0) - continue; +/* 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; - if (le64_to_cpu(ent->lde_hash) < hash_offset) - 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_offset && - (*entp == ent || i < *stripe_offset)) - continue; + if (le64_to_cpu(ent->lde_hash) >= ctxt->ldc_hash) + break; + } - /* skip . and .. for other stripes */ - if (i != 0 && - (strncmp(ent->lde_name, ".", - le16_to_cpu(ent->lde_namelen)) == 0 || - strncmp(ent->lde_name, "..", - le16_to_cpu(ent->lde_namelen)) == 0)) - continue; + return ent; +} + +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; + int rc = 0; + + ENTRY; + + LASSERT(stripe == &ctxt->ldc_stripes[stripe_index]); + 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_eof = true; + break; + } + hash = end; + } + + 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)) { + rc = PTR_ERR(tgt); break; } - if (ent == NULL) { - stripe_hash = le64_to_cpu(dp->ldp_hash_end); + /* 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; + + rc = md_read_page(tgt->ltd_exp, op_data, ctxt->ldc_cb_op, hash, + &stripe->sd_page); + + op_data->op_fid1 = fid; + op_data->op_fid2 = fid; + op_data->op_data = inode; + + if (rc) + break; + + 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); + + stripe->sd_ent = ent; + if (rc) { + 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); +} + +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_find_target(lmv, &data->op_fid1); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); - kunmap(page); - page_cache_release(page); - page = NULL; + data->op_flags |= MF_MDC_CANCEL_FID1; + rc = md_file_resync(tgt->ltd_exp, data); + RETURN(rc); +} - /* reach the end of current stripe, go to next stripe */ - if (stripe_hash == MDS_DIR_END_OFF) +/** + * Get dirent with the closest hash for striped directory + * + * This function will search the dir entry, whose hash value is the + * closest(>=) to hash from all of sub-stripes, and it is only being called + * for striped directory. + * + * \param[in] ctxt dir read context + * + * \retval dirent get the entry successfully + * NULL does not get the entry, normally it means + * it reaches the end of the directory, while read + * stripe dirent error is ignored to allow partial + * access. + */ +static struct lu_dirent *lmv_dirent_next(struct lmv_dir_ctxt *ctxt) +{ + struct stripe_dirent *stripe; + struct lu_dirent *ent = NULL; + int i; + int min = -1; + + /* TODO: optimize with k-way merge sort */ + for (i = 0; i < ctxt->ldc_count; i++) { + stripe = &ctxt->ldc_stripes[i]; + if (stripe->sd_eof) + continue; + + if (!stripe->sd_ent) { + stripe_dirent_load(ctxt, stripe, i); + if (!stripe->sd_ent) { + LASSERT(stripe->sd_eof); continue; - else - goto next; + } } - if (min_ent != NULL) { - if (le64_to_cpu(min_ent->lde_hash) > - le64_to_cpu(ent->lde_hash)) { - min_ent = ent; - kunmap(min_page); - page_cache_release(min_page); - min_idx = i; - min_page = page; - } else { - kunmap(page); - page_cache_release(page); - page = NULL; - } - } else { - min_ent = ent; - min_page = page; - min_idx = i; + if (min == -1 || + le64_to_cpu(ctxt->ldc_stripes[min].sd_ent->lde_hash) > + le64_to_cpu(stripe->sd_ent->lde_hash)) { + min = i; + if (le64_to_cpu(stripe->sd_ent->lde_hash) == + ctxt->ldc_hash) + break; } } -out: - if (*ppage != NULL) { - kunmap(*ppage); - page_cache_release(*ppage); + if (min != -1) { + stripe = &ctxt->ldc_stripes[min]; + ent = stripe->sd_ent; + /* pop found dirent */ + stripe->sd_ent = stripe_dirent_get(ctxt, lu_dirent_next(ent), + min); } - *stripe_offset = min_idx; - *entp = min_ent; - *ppage = min_page; - RETURN(rc); + + return ent; } /** - * Build dir entry page from a striped directory + * Build dir entry page for striped directory * * This function gets one entry by @offset from a striped directory. It will * read entries from all of stripes, and choose one closest to the required @@ -2364,12 +2535,11 @@ out: * and .. in a directory. * 2. op_data will be shared by all of stripes, instead of allocating new * one, so need to restore before reusing. - * 3. release the entry page if that is not being chosen. * * \param[in] exp obd export refer to LMV * \param[in] op_data hold those MD parameters of read_entry * \param[in] cb_op ldlm callback being used in enqueue in mdc_read_entry - * \param[out] ldp the entry being read + * \param[in] offset starting hash offset * \param[out] ppage the page holding the entry. Note: because the entry * will be accessed in upper layer, so we need hold the * page until the usages of entry is finished, see @@ -2378,129 +2548,117 @@ out: * retval =0 if get entry successfully * <0 cannot get entry */ -static int lmv_read_striped_page(struct obd_export *exp, +static int lmv_striped_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 lu_fid master_fid = op_data->op_fid1; - struct inode *master_inode = op_data->op_data; - __u64 hash_offset = offset; - struct lu_dirpage *dp; - struct page *min_ent_page = NULL; - struct page *ent_page = NULL; - struct lu_dirent *ent; - void *area; - int ent_idx = 0; - struct lu_dirent *min_ent = NULL; - struct lu_dirent *last_ent; - size_t left_bytes; - int rc; + struct page *page = NULL; + struct lu_dirpage *dp; + void *start; + struct lu_dirent *ent; + struct lu_dirent *last_ent; + int stripe_count; + struct lmv_dir_ctxt *ctxt; + struct lu_dirent *next = NULL; + __u16 ent_size; + size_t left_bytes; + int rc = 0; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - /* Allocate a page and read entries from all of stripes and fill * the page by hash order */ - ent_page = alloc_page(GFP_KERNEL); - if (ent_page == NULL) + page = alloc_page(GFP_KERNEL); + if (!page) RETURN(-ENOMEM); /* Initialize the entry page */ - dp = kmap(ent_page); + dp = kmap(page); memset(dp, 0, sizeof(*dp)); dp->ldp_hash_start = cpu_to_le64(offset); - dp->ldp_flags |= LDF_COLLIDE; - area = dp + 1; - left_bytes = PAGE_CACHE_SIZE - sizeof(*dp); - ent = area; + start = dp + 1; + left_bytes = PAGE_SIZE - sizeof(*dp); + ent = start; last_ent = ent; - do { - __u16 ent_size; - /* Find the minum entry from all sub-stripes */ - rc = lmv_get_min_striped_entry(exp, op_data, cb_op, hash_offset, - &ent_idx, &min_ent, - &min_ent_page); - if (rc != 0) - GOTO(out, rc); - - /* If it can not get minum entry, it means it already reaches - * the end of this directory */ - if (min_ent == NULL) { - last_ent->lde_reclen = 0; - hash_offset = MDS_DIR_END_OFF; - GOTO(out, rc); + /* initalize dir read context */ + stripe_count = op_data->op_mea1->lsm_md_stripe_count; + OBD_ALLOC(ctxt, offsetof(typeof(*ctxt), ldc_stripes[stripe_count])); + if (!ctxt) + GOTO(free_page, rc = -ENOMEM); + ctxt->ldc_lmv = &exp->exp_obd->u.lmv; + ctxt->ldc_op_data = op_data; + ctxt->ldc_cb_op = cb_op; + ctxt->ldc_hash = offset; + ctxt->ldc_count = stripe_count; + + while (1) { + next = lmv_dirent_next(ctxt); + + /* end of directory */ + if (!next) { + ctxt->ldc_hash = MDS_DIR_END_OFF; + break; } + ctxt->ldc_hash = le64_to_cpu(next->lde_hash); - ent_size = le16_to_cpu(min_ent->lde_reclen); + ent_size = le16_to_cpu(next->lde_reclen); - /* the last entry lde_reclen is 0, but it might not - * the end of this entry of this temporay entry */ - if (ent_size == 0) + /* the last entry lde_reclen is 0, but it might not be the last + * one of this temporay dir page */ + if (!ent_size) ent_size = lu_dirent_calc_size( - le16_to_cpu(min_ent->lde_namelen), - le32_to_cpu(min_ent->lde_attrs)); - if (ent_size > left_bytes) { - last_ent->lde_reclen = cpu_to_le16(0); - hash_offset = le64_to_cpu(min_ent->lde_hash); - GOTO(out, rc); - } + le16_to_cpu(next->lde_namelen), + le32_to_cpu(next->lde_attrs)); + /* page full */ + if (ent_size > left_bytes) + break; - memcpy(ent, min_ent, ent_size); + memcpy(ent, next, ent_size); /* Replace . with master FID and Replace .. with the parent FID * of master object */ if (strncmp(ent->lde_name, ".", le16_to_cpu(ent->lde_namelen)) == 0 && le16_to_cpu(ent->lde_namelen) == 1) - fid_cpu_to_le(&ent->lde_fid, &master_fid); + fid_cpu_to_le(&ent->lde_fid, &op_data->op_fid1); else if (strncmp(ent->lde_name, "..", le16_to_cpu(ent->lde_namelen)) == 0 && le16_to_cpu(ent->lde_namelen) == 2) fid_cpu_to_le(&ent->lde_fid, &op_data->op_fid3); + CDEBUG(D_INODE, "entry %.*s hash %#llx\n", + le16_to_cpu(ent->lde_namelen), ent->lde_name, + le64_to_cpu(ent->lde_hash)); + left_bytes -= ent_size; ent->lde_reclen = cpu_to_le16(ent_size); last_ent = ent; ent = (void *)ent + ent_size; - hash_offset = le64_to_cpu(min_ent->lde_hash); - if (hash_offset == MDS_DIR_END_OFF) { - last_ent->lde_reclen = 0; - break; - } - } while (1); -out: - if (min_ent_page != NULL) { - kunmap(min_ent_page); - page_cache_release(min_ent_page); - } + }; - if (unlikely(rc != 0)) { - __free_page(ent_page); - ent_page = NULL; - } else { - if (ent == area) - dp->ldp_flags |= LDF_EMPTY; - dp->ldp_flags = cpu_to_le32(dp->ldp_flags); - dp->ldp_hash_end = cpu_to_le64(hash_offset); - } + last_ent->lde_reclen = 0; - /* We do not want to allocate md_op_data during each - * dir entry reading, so op_data will be shared by every stripe, - * then we need to restore it back to original value before - * return to the upper layer */ - op_data->op_fid1 = master_fid; - op_data->op_fid2 = master_fid; - op_data->op_data = master_inode; + if (ent == start) + dp->ldp_flags |= LDF_EMPTY; + else if (ctxt->ldc_hash == le64_to_cpu(last_ent->lde_hash)) + dp->ldp_flags |= LDF_COLLIDE; + dp->ldp_flags = cpu_to_le32(dp->ldp_flags); + dp->ldp_hash_end = cpu_to_le64(ctxt->ldc_hash); - *ppage = ent_page; + put_lmv_dir_ctxt(ctxt); + OBD_FREE(ctxt, offsetof(typeof(*ctxt), ldc_stripes[stripe_count])); - RETURN(rc); + *ppage = page; + + RETURN(0); + +free_page: + kunmap(page); + __free_page(page); + + return rc; } int lmv_read_page(struct obd_export *exp, struct md_op_data *op_data, @@ -2514,12 +2672,8 @@ int lmv_read_page(struct obd_export *exp, struct md_op_data *op_data, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc != 0) - RETURN(rc); - if (unlikely(lsm != NULL)) { - rc = lmv_read_striped_page(exp, op_data, cb_op, offset, ppage); + rc = lmv_striped_read_page(exp, op_data, cb_op, offset, ppage); RETURN(rc); } @@ -2557,71 +2711,34 @@ 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; - - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); -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, &op_data->op_fid1); + if (IS_ERR(parent_tgt)) + RETURN(PTR_ERR(parent_tgt)); + + if (likely(!fid_is_zero(&op_data->op_fid2))) { + tgt = lmv_find_target(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. @@ -2631,50 +2748,29 @@ 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) { + if (parent_tgt != tgt) rc = lmv_early_cancel(exp, parent_tgt, op_data, tgt->ltd_idx, 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) + 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); 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); @@ -2683,54 +2779,26 @@ 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_find_target(lmv, &op_data->op_fid2); + if (IS_ERR(tgt)) + RETURN(PTR_ERR(tgt)); + + goto retry; } -static int lmv_precleanup(struct obd_device *obd, enum obd_cleanup_stage stage) +static int lmv_precleanup(struct obd_device *obd) { - struct lmv_obd *lmv = &obd->u.lmv; - int rc = 0; - - switch (stage) { - case OBD_CLEANUP_EARLY: - /* XXX: here should be calling obd_precleanup() down to - * stack. */ - break; - case OBD_CLEANUP_EXPORTS: - fld_client_proc_fini(&lmv->lmv_fld); - lprocfs_obd_cleanup(obd); - lprocfs_free_md_stats(obd); - break; - default: - break; - } - RETURN(rc); + ENTRY; + 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); } /** @@ -2759,7 +2827,7 @@ static int lmv_get_info(const struct lu_env *env, struct obd_export *exp, obd = class_exp2obd(exp); if (obd == NULL) { - CDEBUG(D_IOCTL, "Invalid client cookie "LPX64"\n", + CDEBUG(D_IOCTL, "Invalid client cookie %#llx\n", exp->exp_handle.h_cookie); RETURN(-EINVAL); } @@ -2768,10 +2836,6 @@ static int lmv_get_info(const struct lu_env *env, struct obd_export *exp, if (keylen >= strlen("remote_flag") && !strcmp(key, "remote_flag")) { int i; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - LASSERT(*vallen == sizeof(__u32)); for (i = 0; i < lmv->desc.ld_tgt_count; i++) { struct lmv_tgt_desc *tgt = lmv->tgts[i]; @@ -2789,10 +2853,6 @@ static int lmv_get_info(const struct lu_env *env, struct obd_export *exp, } else if (KEY_IS(KEY_MAX_EASIZE) || KEY_IS(KEY_DEFAULT_EASIZE) || KEY_IS(KEY_CONN_DATA)) { - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - /* * Forwarding this request to first MDS, it should know LOV * desc. @@ -2839,7 +2899,7 @@ int lmv_set_info_async(const struct lu_env *env, struct obd_export *exp, obd = class_exp2obd(exp); if (obd == NULL) { - CDEBUG(D_IOCTL, "Invalid client cookie "LPX64"\n", + CDEBUG(D_IOCTL, "Invalid client cookie %#llx\n", exp->exp_handle.h_cookie); RETURN(-EINVAL); } @@ -2885,13 +2945,15 @@ 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, master %d hash_type %#x " "layout_version %d\n", lsm->lsm_md_stripe_count, lsm->lsm_md_master_mdt_index, lsm->lsm_md_hash_type, lsm->lsm_md_layout_version); @@ -2911,8 +2973,8 @@ static int lmv_unpack_md_v1(struct obd_export *exp, struct lmv_stripe_md *lsm, RETURN(rc); } -int lmv_unpack_md(struct obd_export *exp, struct lmv_stripe_md **lsmp, - const union lmv_mds_md *lmm, int stripe_count) +static int lmv_unpackmd(struct obd_export *exp, struct lmv_stripe_md **lsmp, + const union lmv_mds_md *lmm, size_t lmm_size) { struct lmv_stripe_md *lsm; int lsm_size; @@ -2926,31 +2988,15 @@ int lmv_unpack_md(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) - iput(lsm->lsm_md_oinfo[i].lmo_root); - } + + 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); OBD_FREE(lsm, lsm_size); *lsmp = NULL; RETURN(0); } - /* Alloc memmd */ - if (lsm == NULL && lmm == NULL) { - lsm_size = lmv_stripe_md_size(stripe_count); - OBD_ALLOC(lsm, lsm_size); - if (lsm == NULL) - RETURN(-ENOMEM); - lsm->lsm_md_stripe_count = stripe_count; - *lsmp = lsm; - RETURN(0); - } - if (le32_to_cpu(lmm->lmv_magic) == LMV_MAGIC_STRIPE) RETURN(-EPERM); @@ -3000,39 +3046,27 @@ int lmv_unpack_md(struct obd_export *exp, struct lmv_stripe_md **lsmp, RETURN(lsm_size); } -int lmv_alloc_memmd(struct lmv_stripe_md **lsmp, int stripes) -{ - return lmv_unpack_md(NULL, lsmp, NULL, stripes); -} - void lmv_free_memmd(struct lmv_stripe_md *lsm) { - lmv_unpack_md(NULL, &lsm, NULL, 0); + lmv_unpackmd(NULL, &lsm, NULL, 0); } EXPORT_SYMBOL(lmv_free_memmd); -int lmv_unpackmd(struct obd_export *exp, struct lov_stripe_md **lsmp, - struct lov_mds_md *lmm, int disk_len) -{ - return lmv_unpack_md(exp, (struct lmv_stripe_md **)lsmp, - (union lmv_mds_md *)lmm, disk_len); -} - static int lmv_cancel_unused(struct obd_export *exp, const struct lu_fid *fid, - ldlm_policy_data_t *policy, ldlm_mode_t mode, - ldlm_cancel_flags_t flags, void *opaque) + union ldlm_policy_data *policy, + enum ldlm_mode mode, enum ldlm_cancel_flags flags, + void *opaque) { - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - int rc = 0; - int err; - __u32 i; - ENTRY; + struct lmv_obd *lmv = &exp->exp_obd->u.lmv; + int rc = 0; + __u32 i; + ENTRY; - LASSERT(fid != NULL); + 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) continue; @@ -3045,8 +3079,9 @@ static int lmv_cancel_unused(struct obd_export *exp, const struct lu_fid *fid, RETURN(rc); } -int lmv_set_lock_data(struct obd_export *exp, __u64 *lockh, void *data, - __u64 *bits) +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]; @@ -3059,14 +3094,14 @@ int lmv_set_lock_data(struct obd_export *exp, __u64 *lockh, void *data, RETURN(rc); } -ldlm_mode_t lmv_lock_match(struct obd_export *exp, __u64 flags, - const struct lu_fid *fid, ldlm_type_t type, - ldlm_policy_data_t *policy, ldlm_mode_t mode, - struct lustre_handle *lockh) +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; - ldlm_mode_t rc; + enum ldlm_mode rc; int tgt; int i; ENTRY; @@ -3162,81 +3197,33 @@ int lmv_clear_open_replay_data(struct obd_export *exp, RETURN(md_clear_open_replay_data(tgt->ltd_exp, och)); } -static int lmv_get_remote_perm(struct obd_export *exp, - const struct lu_fid *fid, - struct obd_capa *oc, __u32 suppgid, - 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; - - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - - tgt = lmv_find_target(lmv, fid); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); - - rc = md_get_remote_perm(tgt->ltd_exp, fid, oc, suppgid, request); - RETURN(rc); -} - -static int lmv_renew_capa(struct obd_export *exp, struct obd_capa *oc, - renew_capa_cb_t cb) -{ - 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) - RETURN(rc); - - tgt = lmv_find_target(lmv, &oc->c_capa.lc_fid); - if (IS_ERR(tgt)) - RETURN(PTR_ERR(tgt)); - - rc = md_renew_capa(tgt->ltd_exp, oc, cb); - RETURN(rc); -} - -int lmv_unpack_capa(struct obd_export *exp, struct ptlrpc_request *req, - const struct req_msg_field *field, struct obd_capa **oc) -{ - struct lmv_obd *lmv = &exp->exp_obd->u.lmv; - struct lmv_tgt_desc *tgt = lmv->tgts[0]; - - if (tgt == NULL || tgt->ltd_exp == NULL) - RETURN(-EINVAL); - return md_unpack_capa(tgt->ltd_exp, req, field, oc); -} - int lmv_intent_getattr_async(struct obd_export *exp, - struct md_enqueue_info *minfo, - struct ldlm_enqueue_info *einfo) + struct md_enqueue_info *minfo) { - struct md_op_data *op_data = &minfo->mi_data; - struct obd_device *obd = exp->exp_obd; - struct lmv_obd *lmv = &obd->u.lmv; - struct lmv_tgt_desc *tgt = NULL; - int rc; + struct md_op_data *op_data = &minfo->mi_data; + struct obd_device *obd = exp->exp_obd; + struct lmv_obd *lmv = &obd->u.lmv; + struct lmv_tgt_desc *tgt = NULL; + int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); + if (!fid_is_sane(&op_data->op_fid2)) + RETURN(-EINVAL); - tgt = lmv_locate_mds(lmv, op_data, &op_data->op_fid1); + tgt = lmv_find_target(lmv, &op_data->op_fid1); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); - rc = md_intent_getattr_async(tgt->ltd_exp, minfo, einfo); + /* + * 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. + */ + + rc = md_intent_getattr_async(tgt->ltd_exp, minfo); RETURN(rc); } @@ -3249,10 +3236,6 @@ int lmv_revalidate_lock(struct obd_export *exp, struct lookup_intent *it, int rc; ENTRY; - rc = lmv_check_connect(obd); - if (rc) - RETURN(rc); - tgt = lmv_find_target(lmv, fid); if (IS_ERR(tgt)) RETURN(PTR_ERR(tgt)); @@ -3268,7 +3251,7 @@ int lmv_get_fid_from_lsm(struct obd_export *exp, const struct lmv_oinfo *oinfo; LASSERT(lsm != NULL); - oinfo = lsm_name_to_stripe_info(lsm, name, namelen); + oinfo = lsm_name_to_stripe_info(lsm, name, namelen, false); if (IS_ERR(oinfo)) return PTR_ERR(oinfo); @@ -3384,7 +3367,6 @@ struct obd_ops lmv_obd_ops = { .o_statfs = lmv_statfs, .o_get_info = lmv_get_info, .o_set_info_async = lmv_set_info_async, - .o_unpackmd = lmv_unpackmd, .o_notify = lmv_notify, .o_get_uuid = lmv_get_uuid, .o_iocontrol = lmv_iocontrol, @@ -3392,9 +3374,8 @@ struct obd_ops lmv_obd_ops = { }; struct md_ops lmv_md_ops = { - .m_getstatus = lmv_getstatus, + .m_get_root = lmv_get_root, .m_null_inode = lmv_null_inode, - .m_find_cbdata = lmv_find_cbdata, .m_close = lmv_close, .m_create = lmv_create, .m_enqueue = lmv_enqueue, @@ -3407,6 +3388,7 @@ 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, @@ -3418,27 +3400,26 @@ struct md_ops lmv_md_ops = { .m_merge_attr = lmv_merge_attr, .m_set_open_replay_data = lmv_set_open_replay_data, .m_clear_open_replay_data = lmv_clear_open_replay_data, - .m_renew_capa = lmv_renew_capa, - .m_unpack_capa = lmv_unpack_capa, - .m_get_remote_perm = lmv_get_remote_perm, .m_intent_getattr_async = lmv_intent_getattr_async, .m_revalidate_lock = lmv_revalidate_lock, .m_get_fid_from_lsm = lmv_get_fid_from_lsm, + .m_unpackmd = lmv_unpackmd, }; -int __init lmv_init(void) +static int __init lmv_init(void) { return class_register_type(&lmv_obd_ops, &lmv_md_ops, true, NULL, LUSTRE_LMV_NAME, NULL); } -static void lmv_exit(void) +static void __exit lmv_exit(void) { - class_unregister_type(LUSTRE_LMV_NAME); + class_unregister_type(LUSTRE_LMV_NAME); } -MODULE_AUTHOR("Sun Microsystems, Inc. "); -MODULE_DESCRIPTION("Lustre Logical Metadata Volume OBD driver"); +MODULE_AUTHOR("OpenSFS, Inc. "); +MODULE_DESCRIPTION("Lustre Logical Metadata Volume"); +MODULE_VERSION(LUSTRE_VERSION_STRING); MODULE_LICENSE("GPL"); module_init(lmv_init);