summaryrefslogtreecommitdiffstats
path: root/xlators/cluster/dht/src/dht-helper.c
diff options
context:
space:
mode:
authorKotresh HR <khiremat@redhat.com>2017-01-03 02:35:06 -0500
committerRaghavendra G <rgowdapp@redhat.com>2017-04-26 09:00:34 +0000
commit4076b73b2f4fb3cca0737974b124f33f76f9c9c1 (patch)
treecff52055113fd04c28d5a99719036d59522a51ff /xlators/cluster/dht/src/dht-helper.c
parent1538c98f5e33e0794830d5153f17a96ff28c9914 (diff)
feature/dht: Directory synchronization
Design doc: https://review.gluster.org/16876 Directory creation is now synchronized with blocking inodelk of the parent on the hashed subvolume followed by the entrylk on the hashed subvolume between dht_mkdir, dht_rmdir, dht_rename_dir and lookup selfheal mkdir. To maintain internal consistency of directories across all subvols of dht, we need locks. Specifically we are interested in: 1. Consistency of layout of a directory. Only one writer should modify the layout at a time. A writer (layout setting during directory heal as part of lookup) shouldn't modify the layout while there are readers (all other fops like create, mkdir etc., which consume layout) and readers shouldn't read the layout while a writer is in progress. Readers can read the layout simultaneously. Writer takes a WRITE inodelk on the directory (whose layout is being modified) across ALL subvols. Reader takes a READ inodelk on the directory (whose layout is being read) on ANY subvol. 2. Consistency of directory namespace across subvols. The path and associated gfid should be same on all subvols. A gfid should not be associated with more than one path on any subvol. All fops that can change directory names (mkdir, rmdir, renamedir, directory creation phase in lookup-heal) takes an entrylk on hashed subvol of the directory. NOTE1: In point 2 above, since dht takes entrylk on hashed subvol of a directory, the transaction itself is a consumer of layout on parent directory. So, the transaction is a reader of parent layout and does an inodelk on parent directory just like any other layout reader. So a mkdir (dir/subdir) would: > Acquire a READ inodelk on "dir" on any subvol. > Acquire an entrylk (dir, "subdir") on hashed subvol of "subdir". > creates directory on hashed subvol and possibly on non-hashed subvols. > UNLOCK (entrylk) > UNLOCK (inodelk) NOTE2: mkdir fop while setting the layout of the directory being created is considered as a reader, but NOT a writer. The reason is for a fop which can consume the layout of a directory to come either of the following conditions has to be true: > mkdir syscall from application has to complete. In this case no need of synchronization. > A lookup issued on the directory racing with mkdir has to complete. Since layout setting by a lookup is considered as a writer, only one of either mkdir or lookup will set the layout. Code re-organization: All the lock related routines are moved to "dht-lock.c" file. New wrapper function is introduced to take blocking inodelk followed by entrylk 'dht_protect_namespace' Updates #191 Change-Id: I01569094dfbe1852de6f586475be79c1ba965a31 Signed-off-by: Kotresh HR <khiremat@redhat.com> BUG: 1443373 Reviewed-on: https://review.gluster.org/15472 NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org>
Diffstat (limited to 'xlators/cluster/dht/src/dht-helper.c')
-rw-r--r--xlators/cluster/dht/src/dht-helper.c662
1 files changed, 14 insertions, 648 deletions
diff --git a/xlators/cluster/dht/src/dht-helper.c b/xlators/cluster/dht/src/dht-helper.c
index 0a2abfb697b..6f08f557730 100644
--- a/xlators/cluster/dht/src/dht-helper.c
+++ b/xlators/cluster/dht/src/dht-helper.c
@@ -12,8 +12,7 @@
#include "glusterfs.h"
#include "xlator.h"
#include "dht-common.h"
-#include "dht-helper.h"
-
+#include "dht-lock.h"
static void
dht_free_fd_ctx (dht_fd_ctx_t *fd_ctx)
@@ -400,170 +399,11 @@ dht_deitransform (xlator_t *this, uint64_t y, xlator_t **subvol_p)
return 0;
}
-char *
-dht_lock_asprintf (dht_lock_t *lock)
-{
- char *lk_buf = NULL;
- char gfid[GF_UUID_BUF_SIZE] = {0, };
-
- if (lock == NULL)
- goto out;
-
- uuid_utoa_r (lock->loc.gfid, gfid);
-
- gf_asprintf (&lk_buf, "%s:%s", lock->xl->name, gfid);
-
-out:
- return lk_buf;
-}
-
-void
-dht_log_lk_array (char *name, gf_loglevel_t log_level, dht_lock_t **lk_array,
- int count)
-{
- int i = 0;
- char *lk_buf = NULL;
-
- if ((lk_array == NULL) || (count == 0))
- goto out;
-
- for (i = 0; i < count; i++) {
- lk_buf = dht_lock_asprintf (lk_array[i]);
- gf_msg (name, log_level, 0, DHT_MSG_LK_ARRAY_INFO,
- "%d. %s", i, lk_buf);
- GF_FREE (lk_buf);
- }
-
-out:
- return;
-}
-
-void
-dht_lock_stack_destroy (call_frame_t *lock_frame)
-{
- dht_local_t *local = NULL;
-
- local = lock_frame->local;
-
- local->lock.locks = NULL;
- local->lock.lk_count = 0;
-
- DHT_STACK_DESTROY (lock_frame);
- return;
-}
-
-void
-dht_lock_free (dht_lock_t *lock)
-{
- if (lock == NULL)
- goto out;
-
- loc_wipe (&lock->loc);
- GF_FREE (lock->domain);
- mem_put (lock);
-
-out:
- return;
-}
-
-void
-dht_lock_array_free (dht_lock_t **lk_array, int count)
-{
- int i = 0;
- dht_lock_t *lock = NULL;
-
- if (lk_array == NULL)
- goto out;
-
- for (i = 0; i < count; i++) {
- lock = lk_array[i];
- lk_array[i] = NULL;
- dht_lock_free (lock);
- }
-
-out:
- return;
-}
-
-dht_lock_t *
-dht_lock_new (xlator_t *this, xlator_t *xl, loc_t *loc, short type,
- const char *domain)
-{
- dht_conf_t *conf = NULL;
- dht_lock_t *lock = NULL;
-
- conf = this->private;
-
- lock = mem_get0 (conf->lock_pool);
- if (lock == NULL)
- goto out;
-
- lock->xl = xl;
- lock->type = type;
-
- lock->domain = gf_strdup (domain);
- if (lock->domain == NULL) {
- dht_lock_free (lock);
- lock = NULL;
- goto out;
- }
-
- /* Fill only inode and gfid.
- posix and protocol/server give preference to pargfid/basename over
- gfid/inode for resolution if all the three parameters of loc_t are
- present. I want to avoid the following hypothetical situation:
-
- 1. rebalance did a lookup on a dentry and got a gfid.
- 2. rebalance acquires lock on loc_t which was filled with gfid and
- path (pargfid/bname) from step 1.
- 3. somebody deleted and recreated the same file
- 4. rename on the same path acquires lock on loc_t which now points
- to a different inode (and hence gets the lock).
- 5. rebalance continues to migrate file (note that not all fops done
- by rebalance during migration are inode/gfid based Eg., unlink)
- 6. rename continues.
- */
- lock->loc.inode = inode_ref (loc->inode);
- loc_gfid (loc, lock->loc.gfid);
-
-out:
- return lock;
-}
-
-int
-dht_local_lock_init (call_frame_t *frame, dht_lock_t **lk_array,
- int lk_count, fop_inodelk_cbk_t inodelk_cbk)
-{
- int ret = -1;
- dht_local_t *local = NULL;
-
- local = frame->local;
-
- if (local == NULL) {
- local = dht_local_init (frame, NULL, NULL, 0);
- }
-
- if (local == NULL) {
- goto out;
- }
-
- local->lock.inodelk_cbk = inodelk_cbk;
- local->lock.locks = lk_array;
- local->lock.lk_count = lk_count;
-
- ret = dht_lock_order_requests (local->lock.locks,
- local->lock.lk_count);
- if (ret < 0)
- goto out;
-
- ret = 0;
-out:
- return ret;
-}
-
void
dht_local_wipe (xlator_t *this, dht_local_t *local)
{
+ int i = 0;
+
if (!local)
return;
@@ -612,8 +452,16 @@ dht_local_wipe (xlator_t *this, dht_local_t *local)
local->selfheal.refreshed_layout = NULL;
}
- dht_lock_array_free (local->lock.locks, local->lock.lk_count);
- GF_FREE (local->lock.locks);
+ for (i = 0; i < 2; i++) {
+ dht_lock_array_free (local->lock[i].ns.parent_layout.locks,
+ local->lock[i].ns.parent_layout.lk_count);
+
+ GF_FREE (local->lock[i].ns.parent_layout.locks);
+
+ dht_lock_array_free (local->lock[i].ns.directory_ns.locks,
+ local->lock[i].ns.directory_ns.lk_count);
+ GF_FREE (local->lock[i].ns.directory_ns.locks);
+ }
GF_FREE (local->key);
@@ -657,6 +505,7 @@ dht_local_init (call_frame_t *frame, loc_t *loc, fd_t *fd, glusterfs_fop_t fop)
goto out;
inode = loc->inode;
+ local->hashed_subvol = dht_subvol_get_hashed (frame->this, loc);
}
if (fd) {
@@ -1727,22 +1576,6 @@ out:
return ret;
}
-void
-dht_set_lkowner (dht_lock_t **lk_array, int count, gf_lkowner_t *lkowner)
-{
- int i = 0;
-
- if (!lk_array || !lkowner)
- goto out;
-
- for (i = 0; i < count; i++) {
- lk_array[i]->lk_owner = *lkowner;
- }
-
-out:
- return;
-}
-
int
dht_subvol_status (dht_conf_t *conf, xlator_t *subvol)
{
@@ -1756,473 +1589,6 @@ dht_subvol_status (dht_conf_t *conf, xlator_t *subvol)
return 0;
}
-void
-dht_inodelk_done (call_frame_t *lock_frame)
-{
- fop_inodelk_cbk_t inodelk_cbk = NULL;
- call_frame_t *main_frame = NULL;
- dht_local_t *local = NULL;
-
- local = lock_frame->local;
- main_frame = local->main_frame;
-
- local->lock.locks = NULL;
- local->lock.lk_count = 0;
-
- inodelk_cbk = local->lock.inodelk_cbk;
- local->lock.inodelk_cbk = NULL;
-
- inodelk_cbk (main_frame, NULL, main_frame->this, local->lock.op_ret,
- local->lock.op_errno, NULL);
-
- dht_lock_stack_destroy (lock_frame);
- return;
-}
-
-int
-dht_inodelk_cleanup_cbk (call_frame_t *frame, void *cookie,
- xlator_t *this, int32_t op_ret, int32_t op_errno,
- dict_t *xdata)
-{
- dht_inodelk_done (frame);
- return 0;
-}
-
-int32_t
-dht_lock_count (dht_lock_t **lk_array, int lk_count)
-{
- int i = 0, locked = 0;
-
- if ((lk_array == NULL) || (lk_count == 0))
- goto out;
-
- for (i = 0; i < lk_count; i++) {
- if (lk_array[i]->locked)
- locked++;
- }
-out:
- return locked;
-}
-
-void
-dht_inodelk_cleanup (call_frame_t *lock_frame)
-{
- dht_lock_t **lk_array = NULL;
- int lk_count = 0, lk_acquired = 0;
- dht_local_t *local = NULL;
-
- local = lock_frame->local;
-
- lk_array = local->lock.locks;
- lk_count = local->lock.lk_count;
-
- lk_acquired = dht_lock_count (lk_array, lk_count);
- if (lk_acquired != 0) {
- dht_unlock_inodelk (lock_frame, lk_array, lk_count,
- dht_inodelk_cleanup_cbk);
- } else {
- dht_inodelk_done (lock_frame);
- }
-
- return;
-}
-
-int32_t
-dht_unlock_inodelk_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
- int32_t op_ret, int32_t op_errno, dict_t *xdata)
-{
- dht_local_t *local = NULL;
- int lk_index = 0, call_cnt = 0;
- char gfid[GF_UUID_BUF_SIZE] = {0};
-
- lk_index = (long) cookie;
-
- local = frame->local;
- if (op_ret < 0) {
- uuid_utoa_r (local->lock.locks[lk_index]->loc.gfid,
- gfid);
-
- gf_msg (this->name, GF_LOG_WARNING, op_errno,
- DHT_MSG_UNLOCKING_FAILED,
- "unlocking failed on %s:%s",
- local->lock.locks[lk_index]->xl->name,
- gfid);
- } else {
- local->lock.locks[lk_index]->locked = 0;
- }
-
- call_cnt = dht_frame_return (frame);
- if (is_last_call (call_cnt)) {
- dht_inodelk_done (frame);
- }
-
- return 0;
-}
-
-call_frame_t *
-dht_lock_frame (call_frame_t *parent_frame)
-{
- call_frame_t *lock_frame = NULL;
-
- lock_frame = copy_frame (parent_frame);
- if (lock_frame == NULL)
- goto out;
-
- set_lk_owner_from_ptr (&lock_frame->root->lk_owner, parent_frame->root);
-
-out:
- return lock_frame;
-}
-
-int32_t
-dht_unlock_inodelk (call_frame_t *frame, dht_lock_t **lk_array, int lk_count,
- fop_inodelk_cbk_t inodelk_cbk)
-{
- dht_local_t *local = NULL;
- struct gf_flock flock = {0,};
- int ret = -1 , i = 0;
- call_frame_t *lock_frame = NULL;
- int call_cnt = 0;
-
- GF_VALIDATE_OR_GOTO ("dht-locks", frame, done);
- GF_VALIDATE_OR_GOTO (frame->this->name, lk_array, done);
- GF_VALIDATE_OR_GOTO (frame->this->name, inodelk_cbk, done);
-
- call_cnt = dht_lock_count (lk_array, lk_count);
- if (call_cnt == 0) {
- ret = 0;
- goto done;
- }
-
- lock_frame = dht_lock_frame (frame);
- if (lock_frame == NULL) {
- gf_msg (frame->this->name, GF_LOG_WARNING, 0,
- DHT_MSG_UNLOCKING_FAILED,
- "cannot allocate a frame, not unlocking following "
- "locks:");
-
- dht_log_lk_array (frame->this->name, GF_LOG_WARNING, lk_array,
- lk_count);
- goto done;
- }
-
- ret = dht_local_lock_init (lock_frame, lk_array, lk_count, inodelk_cbk);
- if (ret < 0) {
- gf_msg (frame->this->name, GF_LOG_WARNING, 0,
- DHT_MSG_UNLOCKING_FAILED,
- "storing locks in local failed, not unlocking "
- "following locks:");
-
- dht_log_lk_array (frame->this->name, GF_LOG_WARNING, lk_array,
- lk_count);
-
- goto done;
- }
-
- local = lock_frame->local;
- local->main_frame = frame;
- local->call_cnt = call_cnt;
-
- flock.l_type = F_UNLCK;
-
- for (i = 0; i < local->lock.lk_count; i++) {
- if (!local->lock.locks[i]->locked)
- continue;
-
- lock_frame->root->lk_owner = local->lock.locks[i]->lk_owner;
- STACK_WIND_COOKIE (lock_frame, dht_unlock_inodelk_cbk,
- (void *)(long)i,
- local->lock.locks[i]->xl,
- local->lock.locks[i]->xl->fops->inodelk,
- local->lock.locks[i]->domain,
- &local->lock.locks[i]->loc, F_SETLK,
- &flock, NULL);
- if (!--call_cnt)
- break;
- }
-
- return 0;
-
-done:
- if (lock_frame)
- dht_lock_stack_destroy (lock_frame);
-
- /* no locks acquired, invoke inodelk_cbk */
- if (ret == 0)
- inodelk_cbk (frame, NULL, frame->this, 0, 0, NULL);
-
- return ret;
-}
-
-int32_t
-dht_nonblocking_inodelk_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
- int32_t op_ret, int32_t op_errno, dict_t *xdata)
-{
- dht_local_t *local = NULL;
- int lk_index = 0, call_cnt = 0;
- char gfid[GF_UUID_BUF_SIZE] = {0};
-
- local = frame->local;
- lk_index = (long) cookie;
-
- if (op_ret == -1) {
- local->lock.op_ret = -1;
- local->lock.op_errno = op_errno;
-
- if (local && local->lock.locks[lk_index]) {
- uuid_utoa_r (local->lock.locks[lk_index]->loc.inode->gfid,
- gfid);
-
- gf_msg_debug (this->name, op_errno,
- "inodelk failed on gfid: %s "
- "subvolume: %s", gfid,
- local->lock.locks[lk_index]->xl->name);
- }
-
- goto out;
- }
-
- local->lock.locks[lk_index]->locked = _gf_true;
-
-out:
- call_cnt = dht_frame_return (frame);
- if (is_last_call (call_cnt)) {
- if (local->lock.op_ret < 0) {
- dht_inodelk_cleanup (frame);
- return 0;
- }
-
- dht_inodelk_done (frame);
- }
-
- return 0;
-}
-
-int
-dht_nonblocking_inodelk (call_frame_t *frame, dht_lock_t **lk_array,
- int lk_count, fop_inodelk_cbk_t inodelk_cbk)
-{
- struct gf_flock flock = {0,};
- int i = 0, ret = 0;
- dht_local_t *local = NULL;
- call_frame_t *lock_frame = NULL;
-
- GF_VALIDATE_OR_GOTO ("dht-locks", frame, out);
- GF_VALIDATE_OR_GOTO (frame->this->name, lk_array, out);
- GF_VALIDATE_OR_GOTO (frame->this->name, inodelk_cbk, out);
-
- lock_frame = dht_lock_frame (frame);
- if (lock_frame == NULL)
- goto out;
-
- ret = dht_local_lock_init (lock_frame, lk_array, lk_count, inodelk_cbk);
- if (ret < 0) {
- goto out;
- }
-
- dht_set_lkowner (lk_array, lk_count, &lock_frame->root->lk_owner);
-
- local = lock_frame->local;
- local->main_frame = frame;
-
- local->call_cnt = lk_count;
-
- for (i = 0; i < lk_count; i++) {
- flock.l_type = local->lock.locks[i]->type;
-
- STACK_WIND_COOKIE (lock_frame, dht_nonblocking_inodelk_cbk,
- (void *) (long) i,
- local->lock.locks[i]->xl,
- local->lock.locks[i]->xl->fops->inodelk,
- local->lock.locks[i]->domain,
- &local->lock.locks[i]->loc, F_SETLK,
- &flock, NULL);
- }
-
- return 0;
-
-out:
- if (lock_frame)
- dht_lock_stack_destroy (lock_frame);
-
- return -1;
-}
-
-int32_t
-dht_blocking_inodelk_cbk (call_frame_t *frame, void *cookie, xlator_t *this,
- int32_t op_ret, int32_t op_errno, dict_t *xdata)
-{
- int lk_index = 0;
- int i = 0;
- dht_local_t *local = NULL;
- char gfid[GF_UUID_BUF_SIZE] = {0,};
-
- lk_index = (long) cookie;
-
- local = frame->local;
- if (op_ret == 0) {
- local->lock.locks[lk_index]->locked = _gf_true;
- } else {
- switch (op_errno) {
- case ESTALE:
- case ENOENT:
- if (local->lock.reaction != IGNORE_ENOENT_ESTALE) {
- gf_uuid_unparse (local->lock.locks[lk_index]->loc.gfid, gfid);
- local->lock.op_ret = -1;
- local->lock.op_errno = op_errno;
- gf_msg (this->name, GF_LOG_ERROR, op_errno,
- DHT_MSG_INODELK_FAILED,
- "inodelk failed on subvol %s. gfid:%s",
- local->lock.locks[lk_index]->xl->name,
- gfid);
- goto cleanup;
- }
- break;
- default:
- gf_uuid_unparse (local->lock.locks[lk_index]->loc.gfid, gfid);
- local->lock.op_ret = -1;
- local->lock.op_errno = op_errno;
- gf_msg (this->name, GF_LOG_ERROR, op_errno,
- DHT_MSG_INODELK_FAILED,
- "inodelk failed on subvol %s, gfid:%s",
- local->lock.locks[lk_index]->xl->name, gfid);
- goto cleanup;
- }
- }
-
- if (lk_index == (local->lock.lk_count - 1)) {
- for (i = 0; (i < local->lock.lk_count) &&
- (!local->lock.locks[i]->locked); i++)
- ;
-
- if (i == local->lock.lk_count) {
- local->lock.op_ret = -1;
- local->lock.op_errno = op_errno;
- }
-
- dht_inodelk_done (frame);
- } else {
- dht_blocking_inodelk_rec (frame, ++lk_index);
- }
-
- return 0;
-
-cleanup:
- dht_inodelk_cleanup (frame);
-
- return 0;
-}
-
-void
-dht_blocking_inodelk_rec (call_frame_t *frame, int i)
-{
- dht_local_t *local = NULL;
- struct gf_flock flock = {0,};
-
- local = frame->local;
-
- flock.l_type = local->lock.locks[i]->type;
-
- STACK_WIND_COOKIE (frame, dht_blocking_inodelk_cbk,
- (void *) (long) i,
- local->lock.locks[i]->xl,
- local->lock.locks[i]->xl->fops->inodelk,
- local->lock.locks[i]->domain,
- &local->lock.locks[i]->loc, F_SETLKW, &flock, NULL);
-
- return;
-}
-
-int
-dht_lock_request_cmp (const void *val1, const void *val2)
-{
- dht_lock_t *lock1 = NULL;
- dht_lock_t *lock2 = NULL;
- int ret = 0;
-
- lock1 = *(dht_lock_t **)val1;
- lock2 = *(dht_lock_t **)val2;
-
- GF_VALIDATE_OR_GOTO ("dht-locks", lock1, out);
- GF_VALIDATE_OR_GOTO ("dht-locks", lock2, out);
-
- ret = strcmp (lock1->xl->name, lock2->xl->name);
-
- if (ret == 0) {
- ret = gf_uuid_compare (lock1->loc.gfid, lock2->loc.gfid);
- }
-
-out:
- return ret;
-}
-
-int
-dht_lock_order_requests (dht_lock_t **locks, int count)
-{
- int ret = -1;
-
- if (!locks || !count)
- goto out;
-
- qsort (locks, count, sizeof (*locks), dht_lock_request_cmp);
- ret = 0;
-
-out:
- return ret;
-}
-
-int
-dht_blocking_inodelk (call_frame_t *frame, dht_lock_t **lk_array,
- int lk_count, dht_reaction_type_t reaction,
- fop_inodelk_cbk_t inodelk_cbk)
-{
- int ret = -1;
- call_frame_t *lock_frame = NULL;
- dht_local_t *local = NULL;
- dht_local_t *tmp_local = NULL;
- char gfid[GF_UUID_BUF_SIZE] = {0,};
-
- GF_VALIDATE_OR_GOTO ("dht-locks", frame, out);
- GF_VALIDATE_OR_GOTO (frame->this->name, lk_array, out);
- GF_VALIDATE_OR_GOTO (frame->this->name, inodelk_cbk, out);
-
- tmp_local = frame->local;
-
- lock_frame = dht_lock_frame (frame);
- if (lock_frame == NULL) {
- gf_uuid_unparse (tmp_local->loc.gfid, gfid);
- gf_msg ("dht", GF_LOG_ERROR, ENOMEM,
- DHT_MSG_LOCK_FRAME_FAILED,
- "memory allocation failed for lock_frame. gfid:%s"
- " path:%s", gfid, tmp_local->loc.path);
- goto out;
- }
-
- ret = dht_local_lock_init (lock_frame, lk_array, lk_count, inodelk_cbk);
- if (ret < 0) {
- gf_uuid_unparse (tmp_local->loc.gfid, gfid);
- gf_msg ("dht", GF_LOG_ERROR, ENOMEM,
- DHT_MSG_LOCAL_LOCK_INIT_FAILED,
- "dht_local_lock_init failed, gfid: %s path:%s", gfid,
- tmp_local->loc.path);
- goto out;
- }
-
- dht_set_lkowner (lk_array, lk_count, &lock_frame->root->lk_owner);
-
- local = lock_frame->local;
- local->lock.reaction = reaction;
- local->main_frame = frame;
-
- dht_blocking_inodelk_rec (lock_frame, 0);
-
- return 0;
-out:
- if (lock_frame)
- dht_lock_stack_destroy (lock_frame);
-
- return -1;
-}
inode_t*
dht_heal_path (xlator_t *this, char *path, inode_table_t *itable)
{