#include <linux/pagemap.h>
#include <linux/security.h>
#include <linux/user_namespace.h>
-#ifdef HAVE_UIDGID_HEADER
-# include <linux/uidgid.h>
-#endif
+#include <linux/uidgid.h>
#define DEBUG_SUBSYSTEM S_LLITE
struct dentry *dentry, *tmp_subdir;
DECLARE_LL_D_HLIST_NODE_PTR(p);
- ll_lock_dcache(dir);
+ spin_lock(&dir->i_lock);
ll_d_hlist_for_each_entry(dentry, p, &dir->i_dentry) {
spin_lock(&dentry->d_lock);
if (!list_empty(&dentry->d_subdirs)) {
}
spin_unlock(&dentry->d_lock);
}
- ll_unlock_dcache(dir);
+ spin_unlock(&dir->i_lock);
}
int ll_test_inode_by_fid(struct inode *inode, void *opaque)
return lu_fid_eq(&ll_i2info(inode)->lli_fid, opaque);
}
-int ll_dom_lock_cancel(struct inode *inode, struct ldlm_lock *lock)
+static int ll_dom_lock_cancel(struct inode *inode, struct ldlm_lock *lock)
{
struct lu_env *env;
struct ll_inode_info *lli = ll_i2info(inode);
- struct cl_layout clt = { .cl_layout_gen = 0, };
- int rc;
__u16 refcheck;
-
-
+ int rc;
ENTRY;
if (!lli->lli_clob) {
if (IS_ERR(env))
RETURN(PTR_ERR(env));
- rc = cl_object_layout_get(env, lli->lli_clob, &clt);
- if (rc) {
- CDEBUG(D_INODE, "Cannot get layout for "DFID"\n",
- PFID(ll_inode2fid(inode)));
- rc = -ENODATA;
- } else if (clt.cl_size == 0 || clt.cl_dom_comp_size == 0) {
- CDEBUG(D_INODE, "DOM lock without DOM layout for "DFID"\n",
+ /* reach MDC layer to flush data under the DoM ldlm lock */
+ rc = cl_object_flush(env, lli->lli_clob, lock);
+ if (rc == -ENODATA) {
+ CDEBUG(D_INODE, "inode "DFID" layout has no DoM stripe\n",
PFID(ll_inode2fid(inode)));
- } else {
- enum cl_fsync_mode mode;
- loff_t end = clt.cl_dom_comp_size - 1;
-
- mode = ldlm_is_discard_data(lock) ?
- CL_FSYNC_DISCARD : CL_FSYNC_LOCAL;
- rc = cl_sync_file_range(inode, 0, end, mode, 1);
- truncate_inode_pages_range(inode->i_mapping, 0, end);
+ /* most likely result of layout change, do nothing */
+ rc = 0;
}
+
cl_env_put(env, &refcheck);
RETURN(rc);
}
-void ll_lock_cancel_bits(struct ldlm_lock *lock, __u64 to_cancel)
+static void ll_lock_cancel_bits(struct ldlm_lock *lock, __u64 to_cancel)
{
struct inode *inode = ll_inode_from_resource_lock(lock);
struct ll_inode_info *lli;
CDEBUG(D_INODE, "cannot flush DoM data "
DFID": rc = %d\n",
PFID(ll_inode2fid(inode)), rc);
- lock_res_and_lock(lock);
- ldlm_set_kms_ignore(lock);
- unlock_res_and_lock(lock);
}
if (bits & MDS_INODELOCK_LAYOUT) {
lli = ll_i2info(inode);
if (bits & MDS_INODELOCK_UPDATE)
- lli->lli_update_atime = 1;
+ ll_file_set_flag(lli, LLIF_UPDATE_ATIME);
if ((bits & MDS_INODELOCK_UPDATE) && S_ISDIR(inode->i_mode)) {
CDEBUG(D_INODE, "invalidating inode "DFID" lli = %p, "
inode != inode->i_sb->s_root->d_inode)
ll_invalidate_aliases(inode);
+ if (bits & (MDS_INODELOCK_LOOKUP | MDS_INODELOCK_PERM))
+ forget_all_cached_acls(inode);
+
iput(inode);
RETURN_EXIT;
}
switch (lock->l_req_mode) {
case LCK_PR:
mode = LCK_PR;
+ /* fallthrough */
case LCK_PW:
mode |= LCK_CR;
break;
case LCK_CW:
mode = LCK_CW;
+ /* fallthrough */
case LCK_CR:
mode |= LCK_CR;
break;
return !!(bits);
}
-int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *desc,
+int ll_md_blocking_ast(struct ldlm_lock *lock, struct ldlm_lock_desc *ld,
void *data, int flag)
{
struct lustre_handle lockh;
- __u64 bits = lock->l_policy_data.l_inodebits.bits;
int rc;
ENTRY;
{
__u64 cancel_flags = LCF_ASYNC;
- if (ll_md_need_convert(lock)) {
- cancel_flags |= LCF_CONVERT;
- /* For lock convert some cancel actions may require
- * this lock with non-dropped canceled bits, e.g. page
- * flush for DOM lock. So call ll_lock_cancel_bits()
- * here while canceled bits are still set.
- */
- bits = lock->l_policy_data.l_inodebits.cancel_bits;
- if (bits & MDS_INODELOCK_DOM)
- ll_lock_cancel_bits(lock, MDS_INODELOCK_DOM);
+ /* if lock convert is not needed then still have to
+ * pass lock via ldlm_cli_convert() to keep all states
+ * correct, set cancel_bits to full lock bits to cause
+ * full cancel to happen.
+ */
+ if (!ll_md_need_convert(lock)) {
+ lock_res_and_lock(lock);
+ lock->l_policy_data.l_inodebits.cancel_bits =
+ lock->l_policy_data.l_inodebits.bits;
+ unlock_res_and_lock(lock);
}
+ rc = ldlm_cli_convert(lock, cancel_flags);
+ if (!rc)
+ RETURN(0);
+ /* continue with cancel otherwise */
ldlm_lock2handle(lock, &lockh);
rc = ldlm_cli_cancel(&lockh, cancel_flags);
if (rc < 0) {
break;
}
case LDLM_CB_CANCELING:
+ {
+ __u64 to_cancel = lock->l_policy_data.l_inodebits.bits;
+
/* Nothing to do for non-granted locks */
if (!ldlm_is_granted(lock))
break;
- if (ldlm_is_converting(lock)) {
- /* this is called on already converted lock, so
- * ibits has remained bits only and cancel_bits
- * are bits that were dropped.
- * Note that DOM lock is handled prior lock convert
- * and is excluded here.
+ /* If 'ld' is supplied then bits to be cancelled are passed
+ * implicitly by lock converting and cancel_bits from 'ld'
+ * should be used. Otherwise full cancel is being performed
+ * and lock inodebits are used.
+ *
+ * Note: we cannot rely on cancel_bits in lock itself at this
+ * moment because they can be changed by concurrent thread,
+ * so ldlm_cli_inodebits_convert() pass cancel bits implicitly
+ * in 'ld' parameter.
+ */
+ if (ld) {
+ /* partial bits cancel allowed only during convert */
+ LASSERT(ldlm_is_converting(lock));
+ /* mask cancel bits by lock bits so only no any unused
+ * bits are passed to ll_lock_cancel_bits()
*/
- bits = lock->l_policy_data.l_inodebits.cancel_bits &
- ~MDS_INODELOCK_DOM;
- } else {
- LASSERT(ldlm_is_canceling(lock));
+ to_cancel &= ld->l_policy_data.l_inodebits.cancel_bits;
}
- ll_lock_cancel_bits(lock, bits);
+ ll_lock_cancel_bits(lock, to_cancel);
break;
+ }
default:
LBUG();
}
discon_alias = invalid_alias = NULL;
- ll_lock_dcache(inode);
+ spin_lock(&inode->i_lock);
ll_d_hlist_for_each_entry(alias, p, &inode->i_dentry) {
LASSERT(alias != dentry);
dget_dlock(alias);
spin_unlock(&alias->d_lock);
}
- ll_unlock_dcache(inode);
+ spin_unlock(&inode->i_lock);
return alias;
}
static int ll_lookup_it_finish(struct ptlrpc_request *request,
struct lookup_intent *it,
struct inode *parent, struct dentry **de,
- void *secctx, __u32 secctxlen)
+ void *secctx, __u32 secctxlen,
+ ktime_t kstart)
{
struct inode *inode = NULL;
__u64 bits = 0;
}
}
+ if (it_disposition(it, DISP_OPEN_CREATE)) {
+ ll_stats_ops_tally(ll_i2sbi(parent), LPROC_LL_MKNOD,
+ ktime_us_delta(ktime_get(), kstart));
+ }
+
GOTO(out, rc = 0);
out:
- if (rc != 0 && it->it_op & IT_OPEN)
+ if (rc != 0 && it->it_op & IT_OPEN) {
+ ll_intent_drop_lock(it);
ll_open_cleanup((*de)->d_sb, request);
+ }
return rc;
}
-struct pcc_create_attach {
- struct pcc_dataset *pca_dataset;
- struct dentry *pca_dentry;
-};
-
static struct dentry *ll_lookup_it(struct inode *parent, struct dentry *dentry,
struct lookup_intent *it,
void **secctx, __u32 *secctxlen,
struct pcc_create_attach *pca)
{
+ ktime_t kstart = ktime_get();
struct lookup_intent lookup_it = { .it_op = IT_LOOKUP };
struct dentry *save = dentry, *retval;
struct ptlrpc_request *req = NULL;
}
if (it->it_op & IT_OPEN && it->it_flags & FMODE_WRITE &&
- dentry->d_sb->s_flags & MS_RDONLY)
+ dentry->d_sb->s_flags & SB_RDONLY)
RETURN(ERR_PTR(-EROFS));
if (it->it_op & IT_CREAT)
ll_unlock_md_op_lsm(op_data);
rc = ll_lookup_it_finish(req, it, parent, &dentry,
secctx != NULL ? *secctx : NULL,
- secctxlen != NULL ? *secctxlen : 0);
+ secctxlen != NULL ? *secctxlen : 0,
+ kstart);
if (rc != 0) {
ll_intent_release(it);
GOTO(out, retval = ERR_PTR(rc));
void *secctx = NULL;
__u32 secctxlen = 0;
struct ll_sb_info *sbi;
- struct pcc_create_attach pca = {NULL, NULL};
- struct pcc_dataset *dataset = NULL;
+ struct pcc_create_attach pca = { NULL, NULL };
int rc = 0;
ENTRY;
if (!filename_is_volatile(dentry->d_name.name,
dentry->d_name.len, NULL)) {
struct pcc_matcher item;
+ struct pcc_dataset *dataset;
item.pm_uid = from_kuid(&init_user_ns, current_uid());
item.pm_gid = from_kgid(&init_user_ns, current_gid());
dput(de);
goto out_release;
}
- if (dataset != NULL && dentry->d_inode) {
- rc = pcc_inode_create_fini(dataset,
- dentry->d_inode,
- pca.pca_dentry);
- if (rc) {
- if (de != NULL)
- dput(de);
- GOTO(out_release, rc);
- }
+
+ rc = pcc_inode_create_fini(dentry->d_inode, &pca);
+ if (rc) {
+ if (de != NULL)
+ dput(de);
+ GOTO(out_release, rc);
}
+
ll_set_created(opened, file);
+ } else {
+ /* Open the file with O_CREAT, but the file already
+ * existed on MDT. This may happend in the case that
+ * the LOOKUP ibits lock is revoked and the
+ * corresponding dentry cache is deleted.
+ * i.e. In the current Lustre, the truncate operation
+ * will revoke the LOOKUP ibits lock, and the file
+ * dentry cache will be invalidated. The following open
+ * with O_CREAT flag will call into ->atomic_open, the
+ * file was wrongly though as newly created file and
+ * try to auto cache the file. So after client knows it
+ * is not a DISP_OPEN_CREATE, it should cleanup the
+ * already created PCC copy.
+ */
+ pcc_create_attach_cleanup(dir->i_sb, &pca);
}
if (dentry->d_inode && it_disposition(it, DISP_OPEN_OPEN)) {
} else {
rc = finish_no_open(file, de);
}
+ } else {
+ pcc_create_attach_cleanup(dir->i_sb, &pca);
}
out_release:
- if (dataset != NULL)
- pcc_dataset_put(dataset);
ll_intent_release(it);
OBD_FREE(it, sizeof(*it));
it = ll_convert_intent(&nd->intent.open, nd->flags,
(nd->path.mnt->mnt_flags & MNT_READONLY) ||
- (nd->path.mnt->mnt_sb->s_flags & MS_RDONLY));
+ (nd->path.mnt->mnt_sb->s_flags & SB_RDONLY));
if (IS_ERR(it))
RETURN((struct dentry *)it);
}
dev_t rdev)
{
struct qstr *name = &dchild->d_name;
+ ktime_t kstart = ktime_get();
int err;
- ENTRY;
+ ENTRY;
CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p) mode %o dev %x\n",
name->len, name->name, PFID(ll_inode2fid(dir)), dir,
- mode, rdev);
+ mode, rdev);
if (!IS_POSIXACL(dir) || !exp_connect_umask(ll_i2mdexp(dir)))
mode &= ~current_umask();
- switch (mode & S_IFMT) {
- case 0:
- mode |= S_IFREG; /* for mode = 0 case, fallthrough */
- case S_IFREG:
- case S_IFCHR:
- case S_IFBLK:
- case S_IFIFO:
- case S_IFSOCK:
+ switch (mode & S_IFMT) {
+ case 0:
+ mode |= S_IFREG;
+ /* fallthrough */
+ case S_IFREG:
+ case S_IFCHR:
+ case S_IFBLK:
+ case S_IFIFO:
+ case S_IFSOCK:
err = ll_new_node(dir, dchild, NULL, mode, old_encode_dev(rdev),
LUSTRE_OPC_MKNOD);
- break;
- case S_IFDIR:
- err = -EPERM;
- break;
- default:
- err = -EINVAL;
- }
+ break;
+ case S_IFDIR:
+ err = -EPERM;
+ break;
+ default:
+ err = -EINVAL;
+ }
- if (!err)
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKNOD, 1);
+ if (!err)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKNOD,
+ ktime_us_delta(ktime_get(), kstart));
- RETURN(err);
+ RETURN(err);
}
#ifdef HAVE_IOP_ATOMIC_OPEN
static int ll_create_nd(struct inode *dir, struct dentry *dentry,
umode_t mode, bool want_excl)
{
+ ktime_t kstart = ktime_get();
int rc;
CFS_FAIL_TIMEOUT(OBD_FAIL_LLITE_CREATE_FILE_PAUSE, cfs_fail_val);
* volatile file name, so we use ll_mknod() here. */
rc = ll_mknod(dir, dentry, mode, 0);
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE, 1);
-
CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, unhashed %d\n",
dentry->d_name.len, dentry->d_name.name, d_unhashed(dentry));
+ if (!rc)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE,
+ ktime_us_delta(ktime_get(), kstart));
+
return rc;
}
#else /* !HAVE_IOP_ATOMIC_OPEN */
{
struct ll_dentry_data *lld = ll_d2d(dentry);
struct lookup_intent *it = NULL;
+ ktime_t kstart = ktime_get();
int rc;
CFS_FAIL_TIMEOUT(OBD_FAIL_LLITE_CREATE_FILE_PAUSE, cfs_fail_val);
filp = lookup_instantiate_filp(nd, dentry, NULL);
if (IS_ERR(filp))
rc = PTR_ERR(filp);
- }
+ }
out:
- ll_intent_release(it);
- OBD_FREE(it, sizeof(*it));
+ ll_intent_release(it);
+ OBD_FREE(it, sizeof(*it));
- if (!rc)
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE, 1);
+ if (!rc)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_CREATE,
+ ktime_us_delta(ktime_get(), kstart));
- return rc;
+ return rc;
}
#endif /* HAVE_IOP_ATOMIC_OPEN */
const char *oldpath)
{
struct qstr *name = &dchild->d_name;
+ ktime_t kstart = ktime_get();
int err;
ENTRY;
err = ll_new_node(dir, dchild, oldpath, S_IFLNK | S_IRWXUGO, 0,
LUSTRE_OPC_SYMLINK);
- if (!err)
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_SYMLINK, 1);
+ if (!err)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_SYMLINK,
+ ktime_us_delta(ktime_get(), kstart));
- RETURN(err);
+ RETURN(err);
}
static int ll_link(struct dentry *old_dentry, struct inode *dir,
struct ll_sb_info *sbi = ll_i2sbi(dir);
struct ptlrpc_request *request = NULL;
struct md_op_data *op_data;
+ ktime_t kstart = ktime_get();
int err;
ENTRY;
"target=%.*s\n", PFID(ll_inode2fid(src)), src,
PFID(ll_inode2fid(dir)), dir, name->len, name->name);
- op_data = ll_prep_md_op_data(NULL, src, dir, name->name, name->len,
- 0, LUSTRE_OPC_ANY, NULL);
- if (IS_ERR(op_data))
- RETURN(PTR_ERR(op_data));
+ op_data = ll_prep_md_op_data(NULL, src, dir, name->name, name->len,
+ 0, LUSTRE_OPC_ANY, NULL);
+ if (IS_ERR(op_data))
+ RETURN(PTR_ERR(op_data));
- err = md_link(sbi->ll_md_exp, op_data, &request);
- ll_finish_md_op_data(op_data);
- if (err)
- GOTO(out, err);
+ err = md_link(sbi->ll_md_exp, op_data, &request);
+ ll_finish_md_op_data(op_data);
+ if (err)
+ GOTO(out, err);
- ll_update_times(request, dir);
- ll_stats_ops_tally(sbi, LPROC_LL_LINK, 1);
- EXIT;
+ ll_update_times(request, dir);
+ ll_stats_ops_tally(sbi, LPROC_LL_LINK,
+ ktime_us_delta(ktime_get(), kstart));
+ EXIT;
out:
- ptlrpc_req_finished(request);
- RETURN(err);
+ ptlrpc_req_finished(request);
+ RETURN(err);
}
static int ll_mkdir(struct inode *dir, struct dentry *dchild, ll_umode_t mode)
{
struct qstr *name = &dchild->d_name;
- int err;
- ENTRY;
+ ktime_t kstart = ktime_get();
+ int err;
+ ENTRY;
CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
name->len, name->name, PFID(ll_inode2fid(dir)), dir);
err = ll_new_node(dir, dchild, NULL, mode, 0, LUSTRE_OPC_MKDIR);
if (err == 0)
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKDIR, 1);
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_MKDIR,
+ ktime_us_delta(ktime_get(), kstart));
RETURN(err);
}
static int ll_rmdir(struct inode *dir, struct dentry *dchild)
{
struct qstr *name = &dchild->d_name;
- struct ptlrpc_request *request = NULL;
- struct md_op_data *op_data;
- int rc;
- ENTRY;
+ struct ptlrpc_request *request = NULL;
+ struct md_op_data *op_data;
+ ktime_t kstart = ktime_get();
+ int rc;
+
+ ENTRY;
CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
name->len, name->name, PFID(ll_inode2fid(dir)), dir);
op_data->op_fid3 = *ll_inode2fid(dchild->d_inode);
op_data->op_fid2 = op_data->op_fid3;
- rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
- ll_finish_md_op_data(op_data);
- if (rc == 0) {
- ll_update_times(request, dir);
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, 1);
- }
+ rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
+ ll_finish_md_op_data(op_data);
+ if (!rc)
+ ll_update_times(request, dir);
- ptlrpc_req_finished(request);
- RETURN(rc);
+ ptlrpc_req_finished(request);
+ if (!rc)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR,
+ ktime_us_delta(ktime_get(), kstart));
+ RETURN(rc);
}
/**
{
struct ptlrpc_request *request = NULL;
struct md_op_data *op_data;
+ ktime_t kstart = ktime_get();
int rc;
ENTRY;
op_data->op_cli_flags |= CLI_RM_ENTRY;
rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
ll_finish_md_op_data(op_data);
- if (rc == 0) {
+ if (!rc)
ll_update_times(request, dir);
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR, 1);
- }
ptlrpc_req_finished(request);
+ if (!rc)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_RMDIR,
+ ktime_us_delta(ktime_get(), kstart));
RETURN(rc);
}
struct ptlrpc_request *request = NULL;
struct md_op_data *op_data;
struct mdt_body *body;
+ ktime_t kstart = ktime_get();
int rc;
+
ENTRY;
+
CDEBUG(D_VFSTRACE, "VFS Op:name=%.*s, dir="DFID"(%p)\n",
name->len, name->name, PFID(ll_inode2fid(dir)), dir);
RETURN(PTR_ERR(op_data));
op_data->op_fid3 = *ll_inode2fid(dchild->d_inode);
-
+ /* notify lower layer if inode has dirty pages */
+ if (S_ISREG(dchild->d_inode->i_mode) &&
+ ll_i2info(dchild->d_inode)->lli_clob &&
+ dirty_cnt(dchild->d_inode))
+ op_data->op_cli_flags |= CLI_DIRTY_DATA;
op_data->op_fid2 = op_data->op_fid3;
rc = md_unlink(ll_i2sbi(dir)->ll_md_exp, op_data, &request);
ll_finish_md_op_data(op_data);
set_nlink(dchild->d_inode, body->mbo_nlink);
ll_update_times(request, dir);
- ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_UNLINK, 1);
out:
ptlrpc_req_finished(request);
+ if (!rc)
+ ll_stats_ops_tally(ll_i2sbi(dir), LPROC_LL_UNLINK,
+ ktime_us_delta(ktime_get(), kstart));
RETURN(rc);
}
struct ptlrpc_request *request = NULL;
struct ll_sb_info *sbi = ll_i2sbi(src);
struct md_op_data *op_data;
+ ktime_t kstart = ktime_get();
int err;
ENTRY;
if (tgt_dchild->d_inode != NULL)
op_data->op_fid4 = *ll_inode2fid(tgt_dchild->d_inode);
- err = md_rename(sbi->ll_md_exp, op_data,
- src_name->name, src_name->len,
- tgt_name->name, tgt_name->len, &request);
- ll_finish_md_op_data(op_data);
- if (!err) {
- ll_update_times(request, src);
- ll_update_times(request, tgt);
- ll_stats_ops_tally(sbi, LPROC_LL_RENAME, 1);
- }
+ err = md_rename(sbi->ll_md_exp, op_data,
+ src_name->name, src_name->len,
+ tgt_name->name, tgt_name->len, &request);
+ ll_finish_md_op_data(op_data);
+ if (!err) {
+ ll_update_times(request, src);
+ ll_update_times(request, tgt);
+ }
- ptlrpc_req_finished(request);
+ ptlrpc_req_finished(request);
- if (err == 0)
+ if (!err) {
d_move(src_dchild, tgt_dchild);
+ ll_stats_ops_tally(sbi, LPROC_LL_RENAME,
+ ktime_us_delta(ktime_get(), kstart));
+ }
RETURN(err);
}