}
/* metadata statahead is enabled by default */
+ sbi->ll_sa_running_max = LL_SA_RUNNING_DEF;
sbi->ll_sa_max = LL_SA_RPC_DEF;
atomic_set(&sbi->ll_sa_total, 0);
atomic_set(&sbi->ll_sa_wrong, 0);
EXIT;
}
-static inline int obd_connect_has_secctx(struct obd_connect_data *data)
-{
- return data->ocd_connect_flags & OBD_CONNECT_FLAGS2 &&
- data->ocd_connect_flags2 & OBD_CONNECT2_FILE_SECCTX;
-}
-
static int client_common_fill_super(struct super_block *sb, char *md, char *dt,
struct vfsmount *mnt)
{
struct inode *root = NULL;
- struct ll_sb_info *sbi = ll_s2sbi(sb);
- struct obd_device *obd;
- struct obd_statfs *osfs = NULL;
- struct ptlrpc_request *request = NULL;
- struct obd_connect_data *data = NULL;
- struct obd_uuid *uuid;
- struct md_op_data *op_data;
- struct lustre_md lmd;
+ struct ll_sb_info *sbi = ll_s2sbi(sb);
+ struct obd_statfs *osfs = NULL;
+ struct ptlrpc_request *request = NULL;
+ struct obd_connect_data *data = NULL;
+ struct obd_uuid *uuid;
+ struct md_op_data *op_data;
+ struct lustre_md lmd;
u64 valid;
- int size, err, checksum;
- ENTRY;
+ int size, err, checksum;
- obd = class_name2obd(md);
- if (!obd) {
+ ENTRY;
+ sbi->ll_md_obd = class_name2obd(md);
+ if (!sbi->ll_md_obd) {
CERROR("MD %s: not setup or attached\n", md);
RETURN(-EINVAL);
}
OBD_CONNECT_GRANT_PARAM |
OBD_CONNECT_SHORTIO | OBD_CONNECT_FLAGS2;
- data->ocd_connect_flags2 = OBD_CONNECT2_FLR;
+ data->ocd_connect_flags2 = OBD_CONNECT2_FLR |
+ OBD_CONNECT2_LOCK_CONVERT |
+ OBD_CONNECT2_DIR_MIGRATE |
+ OBD_CONNECT2_SUM_STATFS;
#ifdef HAVE_LRU_RESIZE_SUPPORT
if (sbi->ll_flags & LL_SBI_LRU_RESIZE)
OBD_CONNECT_LARGE_ACL;
#endif
- data->ocd_cksum_types = cksum_types_supported_client();
+ data->ocd_cksum_types = obd_cksum_types_supported_client();
if (OBD_FAIL_CHECK(OBD_FAIL_MDC_LIGHTWEIGHT))
/* flag mdc connection as lightweight, only used for test
if (sbi->ll_flags & LL_SBI_ALWAYS_PING)
data->ocd_connect_flags &= ~OBD_CONNECT_PINGLESS;
-#if defined(HAVE_SECURITY_DENTRY_INIT_SECURITY) && defined(CONFIG_SECURITY)
- data->ocd_connect_flags2 |= OBD_CONNECT2_FILE_SECCTX;
-#endif /* HAVE_SECURITY_DENTRY_INIT_SECURITY */
+ obd_connect_set_secctx(data);
data->ocd_brw_size = MD_MAX_BRW_SIZE;
- err = obd_connect(NULL, &sbi->ll_md_exp, obd, &sbi->ll_sb_uuid, data, NULL);
+ err = obd_connect(NULL, &sbi->ll_md_exp, sbi->ll_md_obd,
+ &sbi->ll_sb_uuid, data, NULL);
if (err == -EBUSY) {
LCONSOLE_ERROR_MSG(0x14f, "An MDT (md %s) is performing "
"recovery, of which this client is not a "
* can make sure the client can be mounted as long as MDT0 is
* avaible */
err = obd_statfs(NULL, sbi->ll_md_exp, osfs,
- ktime_get_seconds() -OBD_STATFS_CACHE_SECONDS,
+ ktime_get_seconds() - OBD_STATFS_CACHE_SECONDS,
OBD_STATFS_FOR_MDT0);
if (err)
GOTO(out_md_fid, err);
}
}
- obd = class_name2obd(dt);
- if (!obd) {
+ sbi->ll_dt_obd = class_name2obd(dt);
+ if (!sbi->ll_dt_obd) {
CERROR("DT %s: not setup or attached\n", dt);
GOTO(out_md_fid, err = -ENODEV);
}
OBD_CONNECT_LAYOUTLOCK |
OBD_CONNECT_PINGLESS | OBD_CONNECT_LFSCK |
OBD_CONNECT_BULK_MBITS | OBD_CONNECT_SHORTIO |
- OBD_CONNECT_FLAGS2;
+ OBD_CONNECT_FLAGS2 | OBD_CONNECT_GRANT_SHRINK;
/* The client currently advertises support for OBD_CONNECT_LOCKAHEAD_OLD so it
* can interoperate with an older version of lockahead which was released prior
/* OBD_CONNECT_CKSUM should always be set, even if checksums are
* disabled by default, because it can still be enabled on the
- * fly via /proc. As a consequence, we still need to come to an
- * agreement on the supported algorithms at connect time */
+ * fly via /sys. As a consequence, we still need to come to an
+ * agreement on the supported algorithms at connect time
+ */
data->ocd_connect_flags |= OBD_CONNECT_CKSUM;
if (OBD_FAIL_CHECK(OBD_FAIL_OSC_CKSUM_ADLER_ONLY))
data->ocd_cksum_types = OBD_CKSUM_ADLER;
else
- data->ocd_cksum_types = cksum_types_supported_client();
+ data->ocd_cksum_types = obd_cksum_types_supported_client();
#ifdef HAVE_LRU_RESIZE_SUPPORT
data->ocd_connect_flags |= OBD_CONNECT_LRU_RESIZE;
"ocd_grant: %d\n", data->ocd_connect_flags,
data->ocd_version, data->ocd_grant);
- obd->obd_upcall.onu_owner = &sbi->ll_lco;
- obd->obd_upcall.onu_upcall = cl_ocd_update;
+ sbi->ll_dt_obd->obd_upcall.onu_owner = &sbi->ll_lco;
+ sbi->ll_dt_obd->obd_upcall.onu_upcall = cl_ocd_update;
data->ocd_brw_size = DT_MAX_BRW_SIZE;
- err = obd_connect(NULL, &sbi->ll_dt_exp, obd, &sbi->ll_sb_uuid, data,
- NULL);
+ err = obd_connect(NULL, &sbi->ll_dt_exp, sbi->ll_dt_obd,
+ &sbi->ll_sb_uuid, data, NULL);
if (err == -EBUSY) {
LCONSOLE_ERROR_MSG(0x150, "An OST (dt %s) is performing "
"recovery, of which this client is not a "
}
checksum = sbi->ll_flags & LL_SBI_CHECKSUM;
- err = obd_set_info_async(NULL, sbi->ll_dt_exp, sizeof(KEY_CHECKSUM),
- KEY_CHECKSUM, sizeof(checksum), &checksum,
- NULL);
- if (err) {
- CERROR("%s: Set checksum failed: rc = %d\n",
- sbi->ll_dt_exp->exp_obd->obd_name, err);
- GOTO(out_root, err);
+ if (sbi->ll_checksum_set) {
+ err = obd_set_info_async(NULL, sbi->ll_dt_exp,
+ sizeof(KEY_CHECKSUM), KEY_CHECKSUM,
+ sizeof(checksum), &checksum, NULL);
+ if (err) {
+ CERROR("%s: Set checksum failed: rc = %d\n",
+ sbi->ll_dt_exp->exp_obd->obd_name, err);
+ GOTO(out_root, err);
+ }
}
cl_sb_init(sb);
if (osfs != NULL)
OBD_FREE_PTR(osfs);
- if (sbi->ll_proc_root != NULL) {
- err = lprocfs_ll_register_obd(sb, dt);
+ if (sbi->ll_dt_obd) {
+ err = sysfs_create_link(&sbi->ll_kset.kobj,
+ &sbi->ll_dt_obd->obd_kset.kobj,
+ sbi->ll_dt_obd->obd_type->typ_name);
if (err < 0) {
CERROR("%s: could not register %s in llite: rc = %d\n",
dt, ll_get_fsname(sb, NULL, 0), err);
err = 0;
}
- err = lprocfs_ll_register_obd(sb, md);
+ }
+
+ if (sbi->ll_md_obd) {
+ err = sysfs_create_link(&sbi->ll_kset.kobj,
+ &sbi->ll_md_obd->obd_kset.kobj,
+ sbi->ll_md_obd->obd_type->typ_name);
if (err < 0) {
CERROR("%s: could not register %s in llite: rc = %d\n",
md, ll_get_fsname(sb, NULL, 0), err);
out_dt:
obd_disconnect(sbi->ll_dt_exp);
sbi->ll_dt_exp = NULL;
+ sbi->ll_dt_obd = NULL;
out_md_fid:
obd_fid_fini(sbi->ll_md_exp->exp_obd);
out_md:
obd_disconnect(sbi->ll_md_exp);
sbi->ll_md_exp = NULL;
+ sbi->ll_md_obd = NULL;
out:
if (data != NULL)
OBD_FREE_PTR(data);
obd_disconnect(sbi->ll_dt_exp);
sbi->ll_dt_exp = NULL;
- lprocfs_ll_unregister_mountpoint(sbi);
+ ll_debugfs_unregister_super(sb);
obd_fid_fini(sbi->ll_md_exp->exp_obd);
obd_disconnect(sbi->ll_md_exp);
static inline int ll_set_opt(const char *opt, char *data, int fl)
{
- if (strncmp(opt, data, strlen(opt)) != 0)
- return(0);
- else
- return(fl);
+ if (strncmp(opt, data, strlen(opt)) != 0)
+ return 0;
+ else
+ return fl;
}
/* non-client-specific mount options are parsed in lmd_parse */
-static int ll_options(char *options, int *flags)
+static int ll_options(char *options, struct ll_sb_info *sbi)
{
- int tmp;
- char *s1 = options, *s2;
- ENTRY;
+ int tmp;
+ char *s1 = options, *s2;
+ int *flags = &sbi->ll_flags;
+ ENTRY;
- if (!options)
- RETURN(0);
+ if (!options)
+ RETURN(0);
- CDEBUG(D_CONFIG, "Parsing opts %s\n", options);
+ CDEBUG(D_CONFIG, "Parsing opts %s\n", options);
while (*s1) {
CDEBUG(D_SUPER, "next opt=%s\n", s1);
goto next;
}
- tmp = ll_set_opt("checksum", s1, LL_SBI_CHECKSUM);
- if (tmp) {
- *flags |= tmp;
- goto next;
- }
- tmp = ll_set_opt("nochecksum", s1, LL_SBI_CHECKSUM);
- if (tmp) {
- *flags &= ~tmp;
- goto next;
- }
+ tmp = ll_set_opt("checksum", s1, LL_SBI_CHECKSUM);
+ if (tmp) {
+ *flags |= tmp;
+ sbi->ll_checksum_set = 1;
+ goto next;
+ }
+ tmp = ll_set_opt("nochecksum", s1, LL_SBI_CHECKSUM);
+ if (tmp) {
+ *flags &= ~tmp;
+ sbi->ll_checksum_set = 1;
+ goto next;
+ }
tmp = ll_set_opt("lruresize", s1, LL_SBI_LRU_RESIZE);
if (tmp) {
*flags |= tmp;
lli->lli_opendir_pid = 0;
lli->lli_sa_enabled = 0;
lli->lli_def_stripe_offset = -1;
+ init_rwsem(&lli->lli_lsm_sem);
} else {
mutex_init(&lli->lli_size_mutex);
lli->lli_symlink_name = NULL;
memset(lli->lli_jobid, 0, sizeof(lli->lli_jobid));
}
+#define MAX_STRING_SIZE 128
+
#ifndef HAVE_SUPER_SETUP_BDI_NAME
#define LSI_BDI_INITIALIZED 0x00400000
# define BDI_CAP_MAP_COPY 0
#endif
-#define MAX_STRING_SIZE 128
-
static int super_setup_bdi_name(struct super_block *sb, char *fmt, ...)
{
struct lustre_sb_info *lsi = s2lsi(sb);
struct config_llog_instance *cfg;
/* %p for void* in printf needs 16+2 characters: 0xffffffffffffffff */
const int instlen = sizeof(cfg->cfg_instance) * 2 + 2;
- int md_len = 0;
- int dt_len = 0;
- int err;
- ENTRY;
+ char name[MAX_STRING_SIZE];
+ int md_len = 0;
+ int dt_len = 0;
+ char *ptr;
+ int len;
+ int err;
+ ENTRY;
CDEBUG(D_VFSTRACE, "VFS Op: sb %p\n", sb);
try_module_get(THIS_MODULE);
OBD_ALLOC_PTR(cfg);
if (cfg == NULL)
- GOTO(out_free, err = -ENOMEM);
+ GOTO(out_free_cfg, err = -ENOMEM);
/* client additional sb info */
lsi->lsi_llsbi = sbi = ll_init_sbi();
if (!sbi)
- GOTO(out_free, err = -ENOMEM);
+ GOTO(out_free_cfg, err = -ENOMEM);
- err = ll_options(lsi->lsi_lmd->lmd_opts, &sbi->ll_flags);
+ err = ll_options(lsi->lsi_lmd->lmd_opts, sbi);
if (err)
- GOTO(out_free, err);
+ GOTO(out_free_cfg, err);
err = super_setup_bdi_name(sb, "lustre-%p", sb);
if (err)
- GOTO(out_free, err);
+ GOTO(out_free_cfg, err);
#ifndef HAVE_DCACHE_LOCK
/* kernel >= 2.6.38 store dentry operations in sb->s_d_op. */
sb->s_d_op = &ll_d_ops;
#endif
+ /* Get fsname */
+ len = strlen(lsi->lsi_lmd->lmd_profile);
+ ptr = strrchr(lsi->lsi_lmd->lmd_profile, '-');
+ if (ptr && (strcmp(ptr, "-client") == 0))
+ len -= 7;
- /* Call lprocfs_ll_register_mountpoint() before lustre_process_log()
- * so that "llite.*.*" params can be processed correctly. */
- if (proc_lustre_fs_root != NULL) {
- err = lprocfs_ll_register_mountpoint(proc_lustre_fs_root, sb);
- if (err < 0) {
- CERROR("%s: could not register mountpoint in llite: "
- "rc = %d\n", ll_get_fsname(sb, NULL, 0), err);
- err = 0;
- }
+ /* Mount info */
+ snprintf(name, MAX_STRING_SIZE, "%.*s-%p", len,
+ lsi->lsi_lmd->lmd_profile, sb);
+
+ /* Call ll_debugfs_register_super() before lustre_process_log()
+ * so that "llite.*.*" params can be processed correctly.
+ */
+ err = ll_debugfs_register_super(sb, name);
+ if (err < 0) {
+ CERROR("%s: could not register mountpoint in llite: rc = %d\n",
+ ll_get_fsname(sb, NULL, 0), err);
+ err = 0;
}
/* Generate a string unique to this super, in case some joker tries
- to mount the same fs at two mount points.
- Use the address of the super itself.*/
+ * to mount the same fs at two mount points.
+ * Use the address of the super itself.
+ */
cfg->cfg_instance = sb;
cfg->cfg_uuid = lsi->lsi_llsbi->ll_sb_uuid;
cfg->cfg_callback = class_config_llog_handler;
/* set up client obds */
err = lustre_process_log(sb, profilenm, cfg);
if (err < 0)
- GOTO(out_proc, err);
+ GOTO(out_debugfs, err);
/* Profile set with LCFG_MOUNTOPT so we can find our mdc and osc obds */
lprof = class_get_profile(profilenm);
LCONSOLE_ERROR_MSG(0x156, "The client profile '%s' could not be"
" read from the MGS. Does that filesystem "
"exist?\n", profilenm);
- GOTO(out_proc, err = -EINVAL);
+ GOTO(out_debugfs, err = -EINVAL);
}
CDEBUG(D_CONFIG, "Found profile %s: mdc=%s osc=%s\n", profilenm,
lprof->lp_md, lprof->lp_dt);
dt_len = strlen(lprof->lp_dt) + instlen + 2;
OBD_ALLOC(dt, dt_len);
if (!dt)
- GOTO(out_proc, err = -ENOMEM);
+ GOTO(out_profile, err = -ENOMEM);
snprintf(dt, dt_len - 1, "%s-%p", lprof->lp_dt, cfg->cfg_instance);
md_len = strlen(lprof->lp_md) + instlen + 2;
OBD_ALLOC(md, md_len);
if (!md)
- GOTO(out_proc, err = -ENOMEM);
+ GOTO(out_free_dt, err = -ENOMEM);
snprintf(md, md_len - 1, "%s-%p", lprof->lp_md, cfg->cfg_instance);
/* connections, registrations, sb setup */
err = client_common_fill_super(sb, md, dt, mnt);
if (err < 0)
- GOTO(out_proc, err);
+ GOTO(out_free_md, err);
sbi->ll_client_common_fill_super_succeeded = 1;
-out_proc:
- if (err < 0)
- lprocfs_ll_unregister_mountpoint(sbi);
-out_free:
+out_free_md:
if (md)
OBD_FREE(md, md_len);
+out_free_dt:
if (dt)
OBD_FREE(dt, dt_len);
- if (lprof != NULL)
+out_profile:
+ if (lprof)
class_put_profile(lprof);
+out_debugfs:
+ if (err < 0)
+ ll_debugfs_unregister_super(sb);
+out_free_cfg:
if (cfg)
OBD_FREE_PTR(cfg);
+
if (err)
ll_put_super(sb);
else if (sbi->ll_flags & LL_SBI_VERBOSE)
{
struct lu_fid *fid;
struct lmv_stripe_md *lsm = md->lmv;
+ struct ll_inode_info *lli = ll_i2info(inode);
int i;
LASSERT(lsm != NULL);
+
+ CDEBUG(D_INODE, "%s: "DFID" set dir layout:\n",
+ ll_get_fsname(inode->i_sb, NULL, 0),
+ PFID(&lli->lli_fid));
+ lsm_md_dump(D_INODE, lsm);
+
/* XXX sigh, this lsm_root initialization should be in
* LMV layer, but it needs ll_iget right now, so we
* put this here right now. */
* where the initialization of slave inode is slightly
* different, so it reset lsm_md to NULL to avoid
* initializing lsm for slave inode. */
- /* For migrating inode, master stripe and master object will
- * be same, so we only need assign this inode */
- if (lsm->lsm_md_hash_type & LMV_HASH_FLAG_MIGRATION && i == 0)
- lsm->lsm_md_oinfo[i].lmo_root = inode;
- else
- lsm->lsm_md_oinfo[i].lmo_root =
+ lsm->lsm_md_oinfo[i].lmo_root =
ll_iget_anon_dir(inode->i_sb, fid, md);
-
if (IS_ERR(lsm->lsm_md_oinfo[i].lmo_root)) {
int rc = PTR_ERR(lsm->lsm_md_oinfo[i].lmo_root);
lsm->lsm_md_oinfo[i].lmo_root = NULL;
+ while (i-- > 0) {
+ iput(lsm->lsm_md_oinfo[i].lmo_root);
+ lsm->lsm_md_oinfo[i].lmo_root = NULL;
+ }
return rc;
}
}
- return 0;
-}
+ lli->lli_lsm_md = lsm;
-static inline int lli_lsm_md_eq(const struct lmv_stripe_md *lsm_md1,
- const struct lmv_stripe_md *lsm_md2)
-{
- return lsm_md1->lsm_md_magic == lsm_md2->lsm_md_magic &&
- lsm_md1->lsm_md_stripe_count == lsm_md2->lsm_md_stripe_count &&
- lsm_md1->lsm_md_master_mdt_index ==
- lsm_md2->lsm_md_master_mdt_index &&
- lsm_md1->lsm_md_hash_type == lsm_md2->lsm_md_hash_type &&
- lsm_md1->lsm_md_layout_version ==
- lsm_md2->lsm_md_layout_version &&
- strcmp(lsm_md1->lsm_md_pool_name,
- lsm_md2->lsm_md_pool_name) == 0;
+ return 0;
}
static int ll_update_lsm_md(struct inode *inode, struct lustre_md *md)
{
struct ll_inode_info *lli = ll_i2info(inode);
struct lmv_stripe_md *lsm = md->lmv;
- int rc;
+ int rc = 0;
+
ENTRY;
LASSERT(S_ISDIR(inode->i_mode));
CDEBUG(D_INODE, "update lsm %p of "DFID"\n", lli->lli_lsm_md,
PFID(ll_inode2fid(inode)));
- /* no striped information from request. */
- if (lsm == NULL) {
- if (lli->lli_lsm_md == NULL) {
- RETURN(0);
- } else if (lli->lli_lsm_md->lsm_md_hash_type &
- LMV_HASH_FLAG_MIGRATION) {
- /* migration is done, the temporay MIGRATE layout has
- * been removed */
- CDEBUG(D_INODE, DFID" finish migration.\n",
- PFID(ll_inode2fid(inode)));
- lmv_free_memmd(lli->lli_lsm_md);
- lli->lli_lsm_md = NULL;
- RETURN(0);
- } else {
- /* The lustre_md from req does not include stripeEA,
- * see ll_md_setattr */
- RETURN(0);
+ /*
+ * no striped information from request, lustre_md from req does not
+ * include stripeEA, see ll_md_setattr()
+ */
+ if (!lsm)
+ RETURN(0);
+
+ /*
+ * normally dir layout doesn't change, only take read lock to check
+ * that to avoid blocking other MD operations.
+ */
+ if (lli->lli_lsm_md)
+ down_read(&lli->lli_lsm_sem);
+ else
+ down_write(&lli->lli_lsm_sem);
+
+ /*
+ * if dir layout mismatch, check whether version is increased, which
+ * means layout is changed, this happens in dir migration and lfsck.
+ */
+ if (lli->lli_lsm_md && !lsm_md_eq(lli->lli_lsm_md, lsm)) {
+ if (lsm->lsm_md_layout_version <=
+ lli->lli_lsm_md->lsm_md_layout_version) {
+ CERROR("%s: "DFID" dir layout mismatch:\n",
+ ll_get_fsname(inode->i_sb, NULL, 0),
+ PFID(&lli->lli_fid));
+ lsm_md_dump(D_ERROR, lli->lli_lsm_md);
+ lsm_md_dump(D_ERROR, lsm);
+ GOTO(unlock, rc = -EINVAL);
}
+
+ /* layout changed, switch to write lock */
+ up_read(&lli->lli_lsm_sem);
+ down_write(&lli->lli_lsm_sem);
+ ll_dir_clear_lsm_md(inode);
}
- /* set the directory layout */
- if (lli->lli_lsm_md == NULL) {
+ /* set directory layout */
+ if (!lli->lli_lsm_md) {
struct cl_attr *attr;
rc = ll_init_lsm_md(inode, md);
+ up_write(&lli->lli_lsm_sem);
if (rc != 0)
RETURN(rc);
/* set md->lmv to NULL, so the following free lustre_md
* will not free this lsm */
md->lmv = NULL;
- lli->lli_lsm_md = lsm;
+
+ /*
+ * md_merge_attr() may take long, since lsm is already set,
+ * switch to read lock.
+ */
+ down_read(&lli->lli_lsm_sem);
OBD_ALLOC_PTR(attr);
if (attr == NULL)
- RETURN(-ENOMEM);
+ GOTO(unlock, rc = -ENOMEM);
/* validate the lsm */
rc = md_merge_attr(ll_i2mdexp(inode), lsm, attr,
ll_md_blocking_ast);
if (rc != 0) {
OBD_FREE_PTR(attr);
- RETURN(rc);
+ GOTO(unlock, rc);
}
if (md->body->mbo_valid & OBD_MD_FLNLINK)
md->body->mbo_mtime = attr->cat_mtime;
OBD_FREE_PTR(attr);
-
- CDEBUG(D_INODE, "Set lsm %p magic %x to "DFID"\n", lsm,
- lsm->lsm_md_magic, PFID(ll_inode2fid(inode)));
- RETURN(0);
- }
-
- /* Compare the old and new stripe information */
- if (!lsm_md_eq(lli->lli_lsm_md, lsm)) {
- struct lmv_stripe_md *old_lsm = lli->lli_lsm_md;
- int idx;
-
- CERROR("%s: inode "DFID"(%p)'s lmv layout mismatch (%p)/(%p)"
- "magic:0x%x/0x%x stripe count: %d/%d master_mdt: %d/%d"
- "hash_type:0x%x/0x%x layout: 0x%x/0x%x pool:%s/%s\n",
- ll_get_fsname(inode->i_sb, NULL, 0), PFID(&lli->lli_fid),
- inode, lsm, old_lsm,
- lsm->lsm_md_magic, old_lsm->lsm_md_magic,
- lsm->lsm_md_stripe_count,
- old_lsm->lsm_md_stripe_count,
- lsm->lsm_md_master_mdt_index,
- old_lsm->lsm_md_master_mdt_index,
- lsm->lsm_md_hash_type, old_lsm->lsm_md_hash_type,
- lsm->lsm_md_layout_version,
- old_lsm->lsm_md_layout_version,
- lsm->lsm_md_pool_name,
- old_lsm->lsm_md_pool_name);
-
- for (idx = 0; idx < old_lsm->lsm_md_stripe_count; idx++) {
- CERROR("%s: sub FIDs in old lsm idx %d, old: "DFID"\n",
- ll_get_fsname(inode->i_sb, NULL, 0), idx,
- PFID(&old_lsm->lsm_md_oinfo[idx].lmo_fid));
- }
-
- for (idx = 0; idx < lsm->lsm_md_stripe_count; idx++) {
- CERROR("%s: sub FIDs in new lsm idx %d, new: "DFID"\n",
- ll_get_fsname(inode->i_sb, NULL, 0), idx,
- PFID(&lsm->lsm_md_oinfo[idx].lmo_fid));
- }
-
- RETURN(-EIO);
}
+unlock:
+ up_read(&lli->lli_lsm_sem);
- RETURN(0);
+ RETURN(rc);
}
void ll_clear_inode(struct inode *inode)
*
* In case of HSMimport, we only set attr on MDS.
*/
-int ll_setattr_raw(struct dentry *dentry, struct iattr *attr, bool hsm_import)
+int ll_setattr_raw(struct dentry *dentry, struct iattr *attr,
+ enum op_xvalid xvalid, bool hsm_import)
{
struct inode *inode = dentry->d_inode;
struct ll_inode_info *lli = ll_i2info(inode);
RETURN(-EPERM);
}
- /* We mark all of the fields "set" so MDS/OST does not re-set them */
- if (!(attr->ia_valid & ATTR_CTIME_SET) &&
- (attr->ia_valid & ATTR_CTIME)) {
+ /* We mark all of the fields "set" so MDS/OST does not re-set them */
+ if (!(xvalid & OP_XVALID_CTIME_SET) &&
+ (attr->ia_valid & ATTR_CTIME)) {
attr->ia_ctime = current_time(inode);
- attr->ia_valid |= ATTR_CTIME_SET;
- }
+ xvalid |= OP_XVALID_CTIME_SET;
+ }
if (!(attr->ia_valid & ATTR_ATIME_SET) &&
(attr->ia_valid & ATTR_ATIME)) {
attr->ia_atime = current_time(inode);
if (!hsm_import && attr->ia_valid & ATTR_SIZE) {
/* If we are changing file size, file content is
- * modified, flag it. */
- attr->ia_valid |= MDS_OPEN_OWNEROVERRIDE;
+ * modified, flag it.
+ */
+ xvalid |= OP_XVALID_OWNEROVERRIDE;
op_data->op_bias |= MDS_DATA_MODIFIED;
ll_file_clear_flag(lli, LLIF_DATA_MODIFIED);
}
+ if (attr->ia_valid & ATTR_FILE) {
+ struct ll_file_data *fd = LUSTRE_FPRIVATE(attr->ia_file);
+
+ if (fd->fd_lease_och)
+ op_data->op_bias |= MDS_TRUNC_KEEP_LEASE;
+ }
+
op_data->op_attr = *attr;
+ op_data->op_xvalid = xvalid;
rc = ll_md_setattr(dentry, op_data);
if (rc)
if (!S_ISREG(inode->i_mode) || hsm_import)
GOTO(out, rc = 0);
- if (attr->ia_valid & (ATTR_SIZE |
- ATTR_ATIME | ATTR_ATIME_SET |
- ATTR_MTIME | ATTR_MTIME_SET |
- ATTR_CTIME | ATTR_CTIME_SET)) {
+ if (attr->ia_valid & (ATTR_SIZE | ATTR_ATIME | ATTR_ATIME_SET |
+ ATTR_MTIME | ATTR_MTIME_SET | ATTR_CTIME) ||
+ xvalid & OP_XVALID_CTIME_SET) {
/* For truncate and utimes sending attributes to OSTs, setting
* mtime/atime to the past will be performed under PW [0:EOF]
* extent lock (new_size:EOF for truncate). It may seem
* excessive to send mtime/atime updates to OSTs when not
* setting times to past, but it is necessary due to possible
- * time de-synchronization between MDT inode and OST objects */
- rc = cl_setattr_ost(lli->lli_clob, attr, 0);
+ * time de-synchronization between MDT inode and OST objects
+ */
+ rc = cl_setattr_ost(lli->lli_clob, attr, xvalid, 0);
}
/* If the file was restored, it needs to set dirty flag.
int ll_setattr(struct dentry *de, struct iattr *attr)
{
int mode = de->d_inode->i_mode;
+ enum op_xvalid xvalid = 0;
if ((attr->ia_valid & (ATTR_CTIME|ATTR_SIZE|ATTR_MODE)) ==
(ATTR_CTIME|ATTR_SIZE|ATTR_MODE))
- attr->ia_valid |= MDS_OPEN_OWNEROVERRIDE;
+ xvalid |= OP_XVALID_OWNEROVERRIDE;
if (((attr->ia_valid & (ATTR_MODE|ATTR_FORCE|ATTR_SIZE)) ==
(ATTR_SIZE|ATTR_MODE)) &&
!(attr->ia_valid & ATTR_KILL_SGID))
attr->ia_valid |= ATTR_KILL_SGID;
- /* avoid polluted from ATTR_TIMES_SET,
- * projid is not expected to be set here */
- attr->ia_valid &= ~MDS_ATTR_PROJID;
-
- return ll_setattr_raw(de, attr, false);
+ return ll_setattr_raw(de, attr, xvalid, false);
}
-int ll_statfs_internal(struct super_block *sb, struct obd_statfs *osfs,
+int ll_statfs_internal(struct ll_sb_info *sbi, struct obd_statfs *osfs,
u32 flags)
{
- struct ll_sb_info *sbi = ll_s2sbi(sb);
struct obd_statfs obd_osfs;
time64_t max_age;
int rc;
RETURN(rc);
}
- osfs->os_type = sb->s_magic;
+ osfs->os_type = LL_SUPER_MAGIC;
CDEBUG(D_SUPER, "MDC blocks %llu/%llu objects %llu/%llu\n",
osfs->os_bavail, osfs->os_blocks, osfs->os_ffree,osfs->os_files);
+ if (osfs->os_state & OS_STATE_SUM)
+ GOTO(out, rc);
+
if (sbi->ll_flags & LL_SBI_LAZYSTATFS)
flags |= OBD_STATFS_NODELAY;
osfs->os_ffree = obd_osfs.os_ffree;
}
+out:
RETURN(rc);
}
int ll_statfs(struct dentry *de, struct kstatfs *sfs)
ll_stats_ops_tally(ll_s2sbi(sb), LPROC_LL_STAFS, 1);
/* Some amount of caching on the client is allowed */
- rc = ll_statfs_internal(sb, &osfs, 0);
+ rc = ll_statfs_internal(ll_s2sbi(sb), &osfs, OBD_STATFS_SUM);
if (rc)
return rc;
mutex_unlock(&lli->lli_size_mutex);
}
+void ll_update_inode_flags(struct inode *inode, int ext_flags)
+{
+ inode->i_flags = ll_ext_to_inode_flags(ext_flags);
+ if (ext_flags & LUSTRE_PROJINHERIT_FL)
+ ll_file_set_flag(ll_i2info(inode), LLIF_PROJECT_INHERIT);
+ else
+ ll_file_clear_flag(ll_i2info(inode), LLIF_PROJECT_INHERIT);
+}
+
int ll_update_inode(struct inode *inode, struct lustre_md *md)
{
struct ll_inode_info *lli = ll_i2info(inode);
/* Clear i_flags to remove S_NOSEC before permissions are updated */
if (body->mbo_valid & OBD_MD_FLFLAGS)
- inode->i_flags = ll_ext_to_inode_flags(body->mbo_flags);
+ ll_update_inode_flags(inode, body->mbo_flags);
if (body->mbo_valid & OBD_MD_FLMODE)
inode->i_mode = (inode->i_mode & S_IFMT) |
(body->mbo_mode & ~S_IFMT);
void ll_delete_inode(struct inode *inode)
{
struct ll_inode_info *lli = ll_i2info(inode);
+ struct address_space *mapping = &inode->i_data;
+ unsigned long nrpages;
ENTRY;
if (S_ISREG(inode->i_mode) && lli->lli_clob != NULL)
* otherwise we may lose data while umount */
cl_sync_file_range(inode, 0, OBD_OBJECT_EOF, CL_FSYNC_LOCAL, 1);
- truncate_inode_pages_final(&inode->i_data);
+ truncate_inode_pages_final(mapping);
- LASSERTF(inode->i_data.nrpages == 0, "inode="DFID"(%p) nrpages=%lu, "
- "see https://jira.hpdd.intel.com/browse/LU-118\n",
- PFID(ll_inode2fid(inode)), inode, inode->i_data.nrpages);
+ /* Workaround for LU-118: Note nrpages may not be totally updated when
+ * truncate_inode_pages() returns, as there can be a page in the process
+ * of deletion (inside __delete_from_page_cache()) in the specified
+ * range. Thus mapping->nrpages can be non-zero when this function
+ * returns even after truncation of the whole mapping. Only do this if
+ * npages isn't already zero.
+ */
+ nrpages = mapping->nrpages;
+ if (nrpages) {
+ spin_lock_irq(&mapping->tree_lock);
+ nrpages = mapping->nrpages;
+ spin_unlock_irq(&mapping->tree_lock);
+ } /* Workaround end */
+
+ LASSERTF(nrpages == 0, "%s: inode="DFID"(%p) nrpages=%lu, "
+ "see https://jira.whamcloud.com/browse/LU-118\n",
+ ll_get_fsname(inode->i_sb, NULL, 0),
+ PFID(ll_inode2fid(inode)), inode, nrpages);
#ifdef HAVE_SBOPS_EVICT_INODE
ll_clear_inode(inode);
int ll_iocontrol(struct inode *inode, struct file *file,
unsigned int cmd, unsigned long arg)
{
- struct ll_sb_info *sbi = ll_i2sbi(inode);
- struct ptlrpc_request *req = NULL;
- int rc, flags = 0;
- ENTRY;
+ struct ll_sb_info *sbi = ll_i2sbi(inode);
+ struct ptlrpc_request *req = NULL;
+ int rc, flags = 0;
+ ENTRY;
- switch(cmd) {
- case FSFILT_IOC_GETFLAGS: {
+ switch (cmd) {
+ case FS_IOC_GETFLAGS: {
struct mdt_body *body;
struct md_op_data *op_data;
flags = body->mbo_flags;
- ptlrpc_req_finished(req);
+ ptlrpc_req_finished(req);
RETURN(put_user(flags, (int __user *)arg));
- }
- case FSFILT_IOC_SETFLAGS: {
+ }
+ case FS_IOC_SETFLAGS: {
struct iattr *attr;
struct md_op_data *op_data;
struct cl_object *obj;
+ struct fsxattr fa = { 0 };
if (get_user(flags, (int __user *)arg))
RETURN(-EFAULT);
- op_data = ll_prep_md_op_data(NULL, inode, NULL, NULL, 0, 0,
- LUSTRE_OPC_ANY, NULL);
- if (IS_ERR(op_data))
- RETURN(PTR_ERR(op_data));
+ fa.fsx_projid = ll_i2info(inode)->lli_projid;
+ if (flags & LUSTRE_PROJINHERIT_FL)
+ fa.fsx_xflags = FS_XFLAG_PROJINHERIT;
+
+ rc = ll_ioctl_check_project(inode, &fa);
+ if (rc)
+ RETURN(rc);
+
+ op_data = ll_prep_md_op_data(NULL, inode, NULL, NULL, 0, 0,
+ LUSTRE_OPC_ANY, NULL);
+ if (IS_ERR(op_data))
+ RETURN(PTR_ERR(op_data));
op_data->op_attr_flags = flags;
- op_data->op_attr.ia_valid |= ATTR_ATTR_FLAG;
+ op_data->op_xvalid |= OP_XVALID_FLAGS;
rc = md_setattr(sbi->ll_md_exp, op_data, NULL, 0, &req);
ll_finish_md_op_data(op_data);
ptlrpc_req_finished(req);
if (rc)
RETURN(rc);
- inode->i_flags = ll_ext_to_inode_flags(flags);
+ ll_update_inode_flags(inode, flags);
obj = ll_i2info(inode)->lli_clob;
if (obj == NULL)
if (attr == NULL)
RETURN(-ENOMEM);
- attr->ia_valid = ATTR_ATTR_FLAG;
- rc = cl_setattr_ost(obj, attr, flags);
+ rc = cl_setattr_ost(obj, attr, OP_XVALID_FLAGS, flags);
OBD_FREE_PTR(attr);
RETURN(rc);
}
op_data->op_fid1 = body->mbo_fid1;
- op_data->op_handle = body->mbo_handle;
+ op_data->op_open_handle = body->mbo_open_handle;
op_data->op_mod_time = ktime_get_real_seconds();
md_close(exp, op_data, NULL, &close_req);
ptlrpc_req_finished(close_req);
return rc;
}
-int ll_process_config(struct lustre_cfg *lcfg)
+/*
+ * this is normally called in ll_fini_md_op_data(), but sometimes it needs to
+ * be called early to avoid deadlock.
+ */
+void ll_unlock_md_op_lsm(struct md_op_data *op_data)
{
- struct super_block *sb;
- unsigned long x;
- int rc = 0;
- char *ptr;
+ if (op_data->op_mea2_sem) {
+ up_read(op_data->op_mea2_sem);
+ op_data->op_mea2_sem = NULL;
+ }
- /* The instance name contains the sb: lustre-client-aacfe000 */
- ptr = strrchr(lustre_cfg_string(lcfg, 0), '-');
- if (!ptr || !*(++ptr))
- return -EINVAL;
- if (sscanf(ptr, "%lx", &x) != 1)
- return -EINVAL;
- sb = (struct super_block *)x;
- /* This better be a real Lustre superblock! */
- LASSERT(s2lsi(sb)->lsi_lmd->lmd_magic == LMD_MAGIC);
-
- /* Note we have not called client_common_fill_super yet, so
- proc fns must be able to handle that! */
- rc = class_process_proc_param(PARAM_LLITE, lprocfs_llite_obd_vars,
- lcfg, sb);
- if (rc > 0)
- rc = 0;
- return rc;
+ if (op_data->op_mea1_sem) {
+ up_read(op_data->op_mea1_sem);
+ op_data->op_mea1_sem = NULL;
+ }
}
/* this function prepares md_op_data hint for passing it down to MD stack. */
ll_i2gids(op_data->op_suppgids, i1, i2);
op_data->op_fid1 = *ll_inode2fid(i1);
op_data->op_default_stripe_offset = -1;
+
if (S_ISDIR(i1->i_mode)) {
+ down_read(&ll_i2info(i1)->lli_lsm_sem);
+ op_data->op_mea1_sem = &ll_i2info(i1)->lli_lsm_sem;
op_data->op_mea1 = ll_i2info(i1)->lli_lsm_md;
if (opc == LUSTRE_OPC_MKDIR)
op_data->op_default_stripe_offset =
if (i2) {
op_data->op_fid2 = *ll_inode2fid(i2);
- if (S_ISDIR(i2->i_mode))
+ if (S_ISDIR(i2->i_mode)) {
+ if (i2 != i1) {
+ down_read(&ll_i2info(i2)->lli_lsm_sem);
+ op_data->op_mea2_sem =
+ &ll_i2info(i2)->lli_lsm_sem;
+ }
op_data->op_mea2 = ll_i2info(i2)->lli_lsm_md;
+ }
} else {
fid_zero(&op_data->op_fid2);
}
void ll_finish_md_op_data(struct md_op_data *op_data)
{
+ ll_unlock_md_op_lsm(op_data);
security_release_secctx(op_data->op_file_secctx,
op_data->op_file_secctx_size);
OBD_FREE_PTR(op_data);