si_meminfo(&si);
pages = si.totalram - si.totalhigh;
- if (pages >> (20 - PAGE_CACHE_SHIFT) < 512) {
- lru_page_max = pages / 2;
- } else {
- lru_page_max = (pages / 4) * 3;
- }
+ lru_page_max = pages / 2;
- /* initialize lru data */
+ /* initialize ll_cache data */
cfs_atomic_set(&sbi->ll_cache.ccc_users, 0);
sbi->ll_cache.ccc_lru_max = lru_page_max;
cfs_atomic_set(&sbi->ll_cache.ccc_lru_left, lru_page_max);
spin_lock_init(&sbi->ll_cache.ccc_lru_lock);
CFS_INIT_LIST_HEAD(&sbi->ll_cache.ccc_lru);
+ cfs_atomic_set(&sbi->ll_cache.ccc_unstable_nr, 0);
+ init_waitqueue_head(&sbi->ll_cache.ccc_unstable_waitq);
+
sbi->ll_ra_info.ra_max_pages_per_file = min(pages / 32,
SBI_DEFAULT_READAHEAD_MAX);
sbi->ll_ra_info.ra_max_pages = sbi->ll_ra_info.ra_max_pages_per_file;
OBD_CONNECT_JOBSTATS | OBD_CONNECT_LVB_TYPE |
OBD_CONNECT_LAYOUTLOCK | OBD_CONNECT_PINGLESS |
OBD_CONNECT_MAX_EASIZE |
- OBD_CONNECT_FLOCK_DEAD;
+ OBD_CONNECT_FLOCK_DEAD |
+ OBD_CONNECT_DISP_STRIPE;
if (sbi->ll_flags & LL_SBI_SOM_PREVIEW)
data->ocd_connect_flags |= OBD_CONNECT_SOM;
else
sbi->ll_md_brw_size = PAGE_CACHE_SIZE;
- if (data->ocd_connect_flags & OBD_CONNECT_LAYOUTLOCK) {
- LCONSOLE_INFO("Layout lock feature supported.\n");
+ if (data->ocd_connect_flags & OBD_CONNECT_LAYOUTLOCK)
sbi->ll_flags |= LL_SBI_LAYOUT_LOCK;
- }
if (data->ocd_ibits_known & MDS_INODELOCK_XATTR) {
if (!(data->ocd_connect_flags & OBD_CONNECT_MAX_EASIZE)) {
ll_d_hlist_for_each(tmp, &inode->i_dentry)
dentry_count++;
- CERROR("inode %p dump: dev=%s ino=%lu mode=%o count=%u, %d dentries\n",
- inode, ll_i2mdexp(inode)->exp_obd->obd_name, inode->i_ino,
+ CERROR("%s: inode %p dump: dev=%s fid="DFID
+ " mode=%o count=%u, %d dentries\n",
+ ll_get_fsname(inode->i_sb, NULL, 0), inode,
+ ll_i2mdexp(inode)->exp_obd->obd_name, PFID(ll_inode2fid(inode)),
inode->i_mode, atomic_read(&inode->i_count), dentry_count);
}
" flags=0x%x, fsdata=%p, %d subdirs\n", dentry,
dentry->d_name.len, dentry->d_name.name,
dentry->d_parent->d_name.len, dentry->d_parent->d_name.name,
- dentry->d_parent, dentry->d_inode, d_refcount(dentry),
+ dentry->d_parent, dentry->d_inode, d_count(dentry),
dentry->d_flags, dentry->d_fsdata, subdirs);
if (dentry->d_inode != NULL)
ll_dump_inode(dentry->d_inode);
*flags &= ~tmp;
goto next;
}
-#if LUSTRE_VERSION_CODE < OBD_OCD_VERSION(2, 5, 50, 0)
- tmp = ll_set_opt("acl", s1, LL_SBI_ACL);
- if (tmp) {
- /* Ignore deprecated mount option. The client will
- * always try to mount with ACL support, whether this
- * is used depends on whether server supports it. */
- LCONSOLE_ERROR_MSG(0x152, "Ignoring deprecated "
- "mount option 'acl'.\n");
- goto next;
- }
- tmp = ll_set_opt("noacl", s1, LL_SBI_ACL);
- if (tmp) {
- LCONSOLE_ERROR_MSG(0x152, "Ignoring deprecated "
- "mount option 'noacl'.\n");
- goto next;
- }
+#if LUSTRE_VERSION_CODE < OBD_OCD_VERSION(2, 6, 51, 0)
+ tmp = ll_set_opt("acl", s1, LL_SBI_ACL);
+ if (tmp) {
+ /* Ignore deprecated mount option. The client will
+ * always try to mount with ACL support, whether this
+ * is used depends on whether server supports it. */
+ LCONSOLE_ERROR_MSG(0x152, "Ignoring deprecated "
+ "mount option 'acl'.\n");
+ goto next;
+ }
+ tmp = ll_set_opt("noacl", s1, LL_SBI_ACL);
+ if (tmp) {
+ LCONSOLE_ERROR_MSG(0x152, "Ignoring deprecated "
+ "mount option 'noacl'.\n");
+ goto next;
+ }
#else
#warning "{no}acl options have been deprecated since 1.8, please remove them"
#endif
if (cfg == NULL)
RETURN(-ENOMEM);
- cfs_module_get();
+ try_module_get(THIS_MODULE);
- /* client additional sb info */
- lsi->lsi_llsbi = sbi = ll_init_sbi();
- if (!sbi) {
- cfs_module_put(THIS_MODULE);
- OBD_FREE_PTR(cfg);
- RETURN(-ENOMEM);
- }
+ /* client additional sb info */
+ lsi->lsi_llsbi = sbi = ll_init_sbi();
+ if (!sbi) {
+ module_put(THIS_MODULE);
+ OBD_FREE_PTR(cfg);
+ RETURN(-ENOMEM);
+ }
err = ll_options(lsi->lsi_lmd->lmd_opts, &sbi->ll_flags);
if (err)
struct lustre_sb_info *lsi = s2lsi(sb);
struct ll_sb_info *sbi = ll_s2sbi(sb);
char *profilenm = get_profile_name(sb);
- int force = 1, next;
+ int ccc_count, next, force = 1, rc = 0;
ENTRY;
CDEBUG(D_VFSTRACE, "VFS Op: sb %p - %s\n", sb, profilenm);
force = obd->obd_force;
}
+ /* Wait for unstable pages to be committed to stable storage */
+ if (force == 0) {
+ struct l_wait_info lwi = LWI_INTR(LWI_ON_SIGNAL_NOOP, NULL);
+ rc = l_wait_event(sbi->ll_cache.ccc_unstable_waitq,
+ cfs_atomic_read(&sbi->ll_cache.ccc_unstable_nr) == 0,
+ &lwi);
+ }
+
+ ccc_count = cfs_atomic_read(&sbi->ll_cache.ccc_unstable_nr);
+ if (force == 0 && rc != -EINTR)
+ LASSERTF(ccc_count == 0, "count: %i\n", ccc_count);
+
+
/* We need to set force before the lov_disconnect in
lustre_common_put_super, since l_d cleans up osc's as well. */
if (force) {
ll_free_sbi(sb);
lsi->lsi_llsbi = NULL;
- lustre_common_put_super(sb);
+ lustre_common_put_super(sb);
- cfs_module_put(THIS_MODULE);
+ cl_env_cache_purge(~0);
- EXIT;
+ module_put(THIS_MODULE);
+
+ EXIT;
} /* client_put_super */
struct inode *ll_inode_from_resource_lock(struct ldlm_lock *lock)
struct ll_sb_info *sbi = ll_i2sbi(inode);
ENTRY;
- CDEBUG(D_VFSTRACE, "VFS Op:inode=%lu/%u(%p)\n", inode->i_ino,
- inode->i_generation, inode);
+ CDEBUG(D_VFSTRACE, "VFS Op:inode="DFID"(%p)\n",
+ PFID(ll_inode2fid(inode)), inode);
if (S_ISDIR(inode->i_mode)) {
/* these should have been cleared in ll_file_release */
* from OSTs and send setattr to back to MDS. */
rc = ll_som_update(inode, op_data);
} else if (rc) {
- CERROR("inode %lu mdc truncate failed: rc = %d\n",
- inode->i_ino, rc);
+ CERROR("%s: inode "DFID" mdc truncate failed: rc = %d\n",
+ ll_i2sbi(inode)->ll_md_exp->exp_obd->obd_name,
+ PFID(ll_inode2fid(inode)), rc);
}
RETURN(rc);
}
struct md_op_data *op_data = NULL;
struct md_open_data *mod = NULL;
bool file_is_released = false;
- int rc = 0, rc1 = 0;
- ENTRY;
+ int rc = 0, rc1 = 0;
+ ENTRY;
- CDEBUG(D_VFSTRACE, "%s: setattr inode %p/fid:"DFID" from %llu to %llu, "
- "valid %x, hsm_import %d\n",
- ll_get_fsname(inode->i_sb, NULL, 0), inode,
- PFID(&lli->lli_fid), i_size_read(inode), attr->ia_size,
- attr->ia_valid, hsm_import);
+ CDEBUG(D_VFSTRACE, "%s: setattr inode "DFID"(%p) from %llu to %llu, "
+ "valid %x, hsm_import %d\n",
+ ll_get_fsname(inode->i_sb, NULL, 0), PFID(&lli->lli_fid),
+ inode, i_size_read(inode), attr->ia_size, attr->ia_valid,
+ hsm_import);
if (attr->ia_valid & ATTR_SIZE) {
/* Check new size against VFS/VM file size limit and rlimit */
LTIME_S(attr->ia_mtime), LTIME_S(attr->ia_ctime),
cfs_time_current_sec());
- /* If we are changing file size, file content is modified, flag it. */
- if (attr->ia_valid & ATTR_SIZE) {
- attr->ia_valid |= MDS_OPEN_OWNEROVERRIDE;
- spin_lock(&lli->lli_lock);
- lli->lli_flags |= LLIF_DATA_MODIFIED;
- spin_unlock(&lli->lli_lock);
- }
-
/* We always do an MDS RPC, even if we're only changing the size;
* only the MDS knows whether truncate() should fail with -ETXTBUSY */
down_write(&lli->lli_trunc_sem);
}
- memcpy(&op_data->op_attr, attr, sizeof(*attr));
-
- /* Open epoch for truncate. */
- if (exp_connect_som(ll_i2mdexp(inode)) &&
- (attr->ia_valid & (ATTR_SIZE | ATTR_MTIME | ATTR_MTIME_SET)))
- op_data->op_flags = MF_EPOCH_OPEN;
-
/* truncate on a released file must failed with -ENODATA,
* so size must not be set on MDS for released file
* but other attributes must be set
if (lsm && lsm->lsm_pattern & LOV_PATTERN_F_RELEASED)
file_is_released = true;
ccc_inode_lsm_put(inode, lsm);
- }
- /* if not in HSM import mode, clear size attr for released file
- * we clear the attribute send to MDT in op_data, not the original
- * received from caller in attr which is used later to
- * decide return code */
- if (file_is_released && (attr->ia_valid & ATTR_SIZE) && !hsm_import)
- op_data->op_attr.ia_valid &= ~ATTR_SIZE;
+ if (!hsm_import && attr->ia_valid & ATTR_SIZE) {
+ if (file_is_released) {
+ rc = ll_layout_restore(inode, 0, attr->ia_size);
+ if (rc < 0)
+ GOTO(out, rc);
- rc = ll_md_setattr(dentry, op_data, &mod);
- if (rc)
- GOTO(out, rc);
+ file_is_released = false;
+ ll_layout_refresh(inode, &gen);
+ }
- /* truncate failed (only when non HSM import), others succeed */
- if (file_is_released) {
- if ((attr->ia_valid & ATTR_SIZE) && !hsm_import)
- GOTO(out, rc = -ENODATA);
- else
- GOTO(out, rc = 0);
+ /* If we are changing file size, file content is
+ * modified, flag it. */
+ attr->ia_valid |= MDS_OPEN_OWNEROVERRIDE;
+ spin_lock(&lli->lli_lock);
+ lli->lli_flags |= LLIF_DATA_MODIFIED;
+ spin_unlock(&lli->lli_lock);
+ op_data->op_bias |= MDS_DATA_MODIFIED;
+ }
}
+ memcpy(&op_data->op_attr, attr, sizeof(*attr));
+
+ /* Open epoch for truncate. */
+ if (exp_connect_som(ll_i2mdexp(inode)) && !hsm_import &&
+ (attr->ia_valid & (ATTR_SIZE | ATTR_MTIME | ATTR_MTIME_SET)))
+ op_data->op_flags = MF_EPOCH_OPEN;
+
+ rc = ll_md_setattr(dentry, op_data, &mod);
+ if (rc)
+ GOTO(out, rc);
+
/* RPC to MDT is sent, cancel data modification flag */
if (rc == 0 && (op_data->op_bias & MDS_DATA_MODIFIED)) {
spin_lock(&lli->lli_lock);
spin_unlock(&lli->lli_lock);
}
- ll_ioepoch_open(lli, op_data->op_ioepoch);
- if (!S_ISREG(inode->i_mode))
- GOTO(out, rc = 0);
+ ll_ioepoch_open(lli, op_data->op_ioepoch);
+ if (!S_ISREG(inode->i_mode) || file_is_released)
+ GOTO(out, rc = 0);
if (attr->ia_valid & (ATTR_SIZE |
ATTR_ATIME | ATTR_ATIME_SET |
ATTR_MTIME | ATTR_MTIME_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 = ll_setattr_ost(inode, attr);
+ /* 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 = ll_setattr_ost(inode, attr);
}
- EXIT;
+ EXIT;
out:
- if (op_data) {
- if (op_data->op_ioepoch) {
- rc1 = ll_setattr_done_writing(inode, op_data, mod);
- if (!rc)
- rc = rc1;
- }
- ll_finish_md_op_data(op_data);
- }
+ if (op_data) {
+ if (op_data->op_ioepoch) {
+ rc1 = ll_setattr_done_writing(inode, op_data, mod);
+ if (!rc)
+ rc = rc1;
+ }
+ ll_finish_md_op_data(op_data);
+ }
if (!S_ISDIR(inode->i_mode)) {
up_write(&lli->lli_trunc_sem);
mutex_lock(&inode->i_mutex);
if (fid_is_sane(&lli->lli_fid)) {
LASSERTF(lu_fid_eq(&lli->lli_fid, &body->fid1),
"Trying to change FID "DFID
- " to the "DFID", inode %lu/%u(%p)\n",
- PFID(&lli->lli_fid), PFID(&body->fid1),
- inode->i_ino, inode->i_generation, inode);
+ " to the "DFID", inode "DFID"(%p)\n",
+ PFID(&lli->lli_fid), PFID(&body->fid1),
+ PFID(ll_inode2fid(inode)), inode);
} else
lli->lli_fid = body->fid1;
}
if (lli->lli_flags & (LLIF_DONE_WRITING |
LLIF_EPOCH_PENDING |
LLIF_SOM_DIRTY)) {
- CERROR("ino %lu flags %u still has "
- "size authority! do not trust "
- "the size got from MDS\n",
- inode->i_ino, lli->lli_flags);
+ CERROR("%s: inode "DFID" flags %u still"
+ " has size authority! do not "
+ "trust the size from MDS\n",
+ sbi->ll_md_exp->exp_obd->obd_name,
+ PFID(ll_inode2fid(inode)),
+ lli->lli_flags);
} else {
/* Use old size assignment to avoid
* deadlock bz14138 & bz14326 */
* deadlock bz14138 & bz14326 */
i_size_write(inode, body->size);
- CDEBUG(D_VFSTRACE, "inode=%lu, updating i_size %llu\n",
- inode->i_ino, (unsigned long long)body->size);
+ CDEBUG(D_VFSTRACE,
+ "inode="DFID", updating i_size %llu\n",
+ PFID(ll_inode2fid(inode)),
+ (unsigned long long)body->size);
}
if (body->valid & OBD_MD_FLBLOCKS)
/* Workaround for LU-118 */
if (inode->i_data.nrpages) {
- TREE_READ_LOCK_IRQ(&inode->i_data);
- TREE_READ_UNLOCK_IRQ(&inode->i_data);
+ spin_lock_irq(&inode->i_data.tree_lock);
+ spin_unlock_irq(&inode->i_data.tree_lock);
LASSERTF(inode->i_data.nrpages == 0,
- "inode=%lu/%u(%p) nrpages=%lu, see "
- "http://jira.whamcloud.com/browse/LU-118\n",
- inode->i_ino, inode->i_generation, inode,
+ "inode="DFID"(%p) nrpages=%lu, see "
+ "http://jira.whamcloud.com/browse/LU-118\n",
+ PFID(ll_inode2fid(inode)), inode,
inode->i_data.nrpages);
}
/* Workaround end */
rc = md_getattr(sbi->ll_md_exp, op_data, &req);
ll_finish_md_op_data(op_data);
if (rc) {
- CERROR("failure %d inode %lu\n", rc, inode->i_ino);
+ CERROR("%s: failure inode "DFID": rc = %d\n",
+ sbi->ll_md_exp->exp_obd->obd_name,
+ PFID(ll_inode2fid(inode)), rc);
RETURN(-abs(rc));
}
OBD_FREE_PTR(ioc_data);
}
- /* Really, we'd like to wait until there are no requests outstanding,
- * and then continue. For now, we just invalidate the requests,
- * schedule() and sleep one second if needed, and hope.
- */
- cfs_schedule();
- EXIT;
+ /* Really, we'd like to wait until there are no requests outstanding,
+ * and then continue. For now, we just invalidate the requests,
+ * schedule() and sleep one second if needed, and hope.
+ */
+ schedule();
+ EXIT;
}
int ll_remount_fs(struct super_block *sb, int *flags, char *data)