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>

1 files changed, 37 insertions, 62 deletions

diff --git a/xlators/cluster/dht/src/dht-selfheal.c b/xlators/cluster/dht/src/dht-selfheal.c
index 0838a627521..de9d30c047f 100644
--- a/xlators/cluster/dht/src/dht-selfheal.c
+++ b/xlators/cluster/dht/src/dht-selfheal.c
@@ -13,6 +13,7 @@
 #include "xlator.h"
 #include "dht-common.h"
 #include "dht-messages.h"
+#include "dht-lock.h"
 #include "glusterfs-acl.h"
 
 #define DHT_SET_LAYOUT_RANGE(layout,i,srt,chunk,path)    do {           \
@@ -85,7 +86,13 @@ dht_selfheal_dir_finish (call_frame_t *frame, xlator_t *this, int ret,
         int           lock_count = 0;
 
         local = frame->local;
-        lock_count = dht_lock_count (local->lock.locks, local->lock.lk_count);
+
+        /* Unlock entrylk */
+        dht_unlock_entrylk_wrapper (frame, &local->lock[0].ns.directory_ns);
+
+        /* Unlock inodelk */
+        lock_count = dht_lock_count (local->lock[0].ns.parent_layout.locks,
+                                     local->lock[0].ns.parent_layout.lk_count);
         if (lock_count == 0)
                 goto done;
 
@@ -100,14 +107,15 @@ dht_selfheal_dir_finish (call_frame_t *frame, xlator_t *this, int ret,
                 goto done;
         }
 
-        lock_local->lock.locks = local->lock.locks;
-        lock_local->lock.lk_count = local->lock.lk_count;
+        lock_local->lock[0].ns.parent_layout.locks = local->lock[0].ns.parent_layout.locks;
+        lock_local->lock[0].ns.parent_layout.lk_count = local->lock[0].ns.parent_layout.lk_count;
 
-        local->lock.locks = NULL;
-        local->lock.lk_count = 0;
+        local->lock[0].ns.parent_layout.locks = NULL;
+        local->lock[0].ns.parent_layout.lk_count = 0;
 
-        dht_unlock_inodelk (lock_frame, lock_local->lock.locks,
-                            lock_local->lock.lk_count,
+        dht_unlock_inodelk (lock_frame,
+                            lock_local->lock[0].ns.parent_layout.locks,
+                            lock_local->lock[0].ns.parent_layout.lk_count,
                             dht_selfheal_unlock_cbk);
         lock_frame = NULL;
 
@@ -579,7 +587,8 @@ dht_selfheal_layout_lock (call_frame_t *frame, dht_layout_t *layout,
                         lk_array[i] = dht_lock_new (frame->this,
                                                     conf->subvolumes[i],
                                                     &local->loc, F_WRLCK,
-                                                    DHT_LAYOUT_HEAL_DOMAIN);
+                                                    DHT_LAYOUT_HEAL_DOMAIN,
+                                                    NULL);
                         if (lk_array[i] == NULL) {
                                 gf_uuid_unparse (local->stbuf.ia_gfid, gfid);
                                 gf_msg (THIS->name, GF_LOG_ERROR, ENOMEM,
@@ -604,7 +613,7 @@ dht_selfheal_layout_lock (call_frame_t *frame, dht_layout_t *layout,
 
                 lk_array[0] = dht_lock_new (frame->this, local->hashed_subvol,
                                             &local->loc, F_WRLCK,
-                                            DHT_LAYOUT_HEAL_DOMAIN);
+                                            DHT_LAYOUT_HEAL_DOMAIN, NULL);
                 if (lk_array[0] == NULL) {
                         gf_uuid_unparse (local->stbuf.ia_gfid, gfid);
                         gf_msg (THIS->name, GF_LOG_ERROR, ENOMEM,
@@ -615,14 +624,14 @@ dht_selfheal_layout_lock (call_frame_t *frame, dht_layout_t *layout,
                 }
         }
 
-        local->lock.locks = lk_array;
-        local->lock.lk_count = count;
+        local->lock[0].layout.my_layout.locks = lk_array;
+        local->lock[0].layout.my_layout.lk_count = count;
 
         ret = dht_blocking_inodelk (frame, lk_array, count, FAIL_ON_ANY_ERROR,
                                     dht_selfheal_layout_lock_cbk);
         if (ret < 0) {
-                local->lock.locks = NULL;
-                local->lock.lk_count = 0;
+                local->lock[0].layout.my_layout.locks = NULL;
+                local->lock[0].layout.my_layout.lk_count = 0;
                 goto err;
         }
 
@@ -1454,8 +1463,8 @@ dht_selfheal_dir_mkdir_lock_cbk (call_frame_t *frame, void *cookie,
                 }
 
                 gf_msg (this->name, GF_LOG_WARNING, op_errno,
-                        DHT_MSG_INODE_LK_ERROR,
-                        "acquiring inodelk failed for %s",
+                        DHT_MSG_ENTRYLK_ERROR,
+                        "acquiring entrylk after inodelk failed for %s",
                         local->loc.path);
 
                 local->op_errno = op_errno;
@@ -1487,15 +1496,9 @@ dht_selfheal_dir_mkdir (call_frame_t *frame, loc_t *loc,
         int           missing_dirs = 0;
         int           i     = 0;
         int           ret   = -1;
-        int           count = 1;
         dht_local_t  *local = NULL;
-        dht_conf_t   *conf  = NULL;
-        xlator_t     *this = NULL;
-        dht_lock_t   **lk_array = NULL;
 
         local = frame->local;
-        this = frame->this;
-        conf = this->private;
 
         local->selfheal.force_mkdir = force;
         local->selfheal.hole_cnt = 0;
@@ -1511,44 +1514,16 @@ dht_selfheal_dir_mkdir (call_frame_t *frame, loc_t *loc,
                 return 0;
         }
 
-        count = conf->subvolume_cnt;
-
-        /* Locking on all subvols in the mkdir phase of lookup selfheal is
-           is done to synchronize with rmdir/rename.
-        */
-        lk_array = GF_CALLOC (count, sizeof (*lk_array), gf_common_mt_char);
-        if (lk_array == NULL)
-                goto err;
-
-        for (i = 0; i < count; i++) {
-                lk_array[i] = dht_lock_new (frame->this,
-                                            conf->subvolumes[i],
-                                            &local->loc, F_WRLCK,
-                                            DHT_LAYOUT_HEAL_DOMAIN);
-                if (lk_array[i] == NULL)
-                        goto err;
-        }
-
-        local->lock.locks = lk_array;
-        local->lock.lk_count = count;
-
-        ret = dht_blocking_inodelk (frame, lk_array, count,
-                                    IGNORE_ENOENT_ESTALE,
-                                    dht_selfheal_dir_mkdir_lock_cbk);
+        local->current = &local->lock[0];
+        ret = dht_protect_namespace (frame, loc, local->hashed_subvol,
+                                     &local->current->ns,
+                                     dht_selfheal_dir_mkdir_lock_cbk);
 
-        if (ret < 0) {
-                local->lock.locks = NULL;
-                local->lock.lk_count = 0;
+        if (ret < 0)
                 goto err;
-        }
 
         return 0;
 err:
-        if (lk_array != NULL) {
-                dht_lock_array_free (lk_array, count);
-                GF_FREE (lk_array);
-        }
-
         return -1;
 }
 
@@ -2379,9 +2354,9 @@ dht_update_commit_hash_for_layout_unlock (call_frame_t *frame, xlator_t *this)
 
         local = frame->local;
 
-        ret = dht_unlock_inodelk (frame, local->lock.locks,
-                                  local->lock.lk_count,
-                                  dht_update_commit_hash_for_layout_done);
+        ret = dht_unlock_inodelk (frame, local->lock[0].layout.my_layout.locks,
+                                   local->lock[0].layout.my_layout.lk_count,
+                                   dht_update_commit_hash_for_layout_done);
         if (ret < 0) {
                 /* preserve oldest error, just ... */
                 if (!local->op_ret) {
@@ -2614,19 +2589,19 @@ dht_update_commit_hash_for_layout (call_frame_t *frame)
                 lk_array[i] = dht_lock_new (frame->this,
                                             conf->local_subvols[i],
                                             &local->loc, F_WRLCK,
-                                            DHT_LAYOUT_HEAL_DOMAIN);
+                                            DHT_LAYOUT_HEAL_DOMAIN, NULL);
                 if (lk_array[i] == NULL)
                         goto err;
         }
 
-        local->lock.locks = lk_array;
-        local->lock.lk_count = count;
+        local->lock[0].layout.my_layout.locks = lk_array;
+        local->lock[0].layout.my_layout.lk_count = count;
 
         ret = dht_blocking_inodelk (frame, lk_array, count, FAIL_ON_ANY_ERROR,
                                     dht_update_commit_hash_for_layout_resume);
         if (ret < 0) {
-                local->lock.locks = NULL;
-                local->lock.lk_count = 0;
+                local->lock[0].layout.my_layout.locks = NULL;
+                local->lock[0].layout.my_layout.lk_count = 0;
                 goto err;
         }