cluster/dht: Change the subvolume encoding in d_off to be a "global"

position in the graph rather than relative (local) to a particular
translator.

Encoding the volume in this way allows a single translator to manage
which brick is currently being scanned for directory entries. Using a
single translator minimizes allocated bits in the d_off. It also allows
multiple DHT translators in the same graph to have a common frame of
reference (the graph position) for which brick is being read. Multiple
DHT translators are needed for the Tiering feature.

The fix builds off a previous change (9332) which removed subvolume
encoding from AFR. The fix makes an equivalent change to the EC
translator.

More background can be found in fix 9332 and gluster-dev discussions [1].

DHT and AFR/EC are responsibile (as before) for choosing which brick to
enumerate directory entries in over the readdir lifecycle.

The client translator receiving the readdir fop encodes the dht_t. It
is referred to as the "leaf node" in the graph and corresponds to the
brick being scanned.

When DHT decodes the d_off, it translates the leaf node to a local
subvolume, which represents the next node in the graph leading to
the brick.

Tracking of leaf nodes is done in common utility functions. Leaf nodes
counts and positional information are updated on a graph switch.

[1] www.gluster.org/pipermail/gluster-devel/2015-January/043592.html

Change-Id: Iaf0ea86d7046b1ceadbad69d88707b243077ebc8
BUG: 1190734
Signed-off-by: Dan Lambright <dlambrig@redhat.com>
Reviewed-on: http://review.gluster.org/9688
Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com>
Reviewed-by: Vijay Bellur <vbellur@redhat.com>
Tested-by: Vijay Bellur <vbellur@redhat.com>

5 files changed, 30 insertions, 6 deletions

diff --git a/xlators/protocol/client/src/client-handshake.c b/xlators/protocol/client/src/client-handshake.c
index 531b38eaf83..90192ad2ac8 100644
--- a/xlators/protocol/client/src/client-handshake.c
+++ b/xlators/protocol/client/src/client-handshake.c
@@ -1194,6 +1194,8 @@ client_setvolume_cbk (struct rpc_req *req, struct iovec *iov, int count, void *m
         }
         */
 
+        conf->client_id = glusterfs_leaf_position(this);
+
         gf_log (this->name, GF_LOG_INFO,
                 "Connected to %s, attached to remote volume '%s'.",
                 conf->rpc->conn.name,
diff --git a/xlators/protocol/client/src/client-helpers.c b/xlators/protocol/client/src/client-helpers.c
index 5d9f00fdc70..be5e7b57739 100644
--- a/xlators/protocol/client/src/client-helpers.c
+++ b/xlators/protocol/client/src/client-helpers.c
@@ -141,12 +141,16 @@ client_local_wipe (clnt_local_t *local)
 }
 
 int
-unserialize_rsp_dirent (struct gfs3_readdir_rsp *rsp, gf_dirent_t *entries)
+unserialize_rsp_dirent (xlator_t *this, struct gfs3_readdir_rsp *rsp,
+                        gf_dirent_t *entries)
 {
         struct gfs3_dirlist  *trav      = NULL;
 	gf_dirent_t          *entry     = NULL;
         int                   entry_len = 0;
         int                   ret       = -1;
+        clnt_conf_t          *conf = NULL;
+
+        conf = this->private;
 
         trav = rsp->reply;
         while (trav) {
@@ -156,7 +160,8 @@ unserialize_rsp_dirent (struct gfs3_readdir_rsp *rsp, gf_dirent_t *entries)
                         goto out;
 
                 entry->d_ino  = trav->d_ino;
-                entry->d_off  = trav->d_off;
+                gf_itransform (this, trav->d_off, &entry->d_off,
+                                      conf->client_id);
                 entry->d_len  = trav->d_len;
                 entry->d_type = trav->d_type;
 
@@ -182,12 +187,17 @@ unserialize_rsp_direntp (xlator_t *this, fd_t *fd,
         inode_table_t        *itable    = NULL;
         int                   entry_len = 0;
         int                   ret       = -1;
+        clnt_conf_t          *conf      = NULL;
 
         trav = rsp->reply;
 
         if (fd)
                 itable = fd->inode->table;
 
+        conf = this->private;
+        if (!conf)
+                goto out;
+
         while (trav) {
                 entry_len = gf_dirent_size (trav->name);
                 entry = GF_CALLOC (1, entry_len, gf_common_mt_gf_dirent_t);
@@ -195,7 +205,8 @@ unserialize_rsp_direntp (xlator_t *this, fd_t *fd,
                         goto out;
 
                 entry->d_ino  = trav->d_ino;
-                entry->d_off  = trav->d_off;
+                gf_itransform (this, trav->d_off, &entry->d_off,
+                                      conf->client_id);
                 entry->d_len  = trav->d_len;
                 entry->d_type = trav->d_type;
 
diff --git a/xlators/protocol/client/src/client-rpc-fops.c b/xlators/protocol/client/src/client-rpc-fops.c
index 7ca91e9880c..10abe845c0c 100644
--- a/xlators/protocol/client/src/client-rpc-fops.c
+++ b/xlators/protocol/client/src/client-rpc-fops.c
@@ -2450,7 +2450,7 @@ client3_3_readdir_cbk (struct rpc_req *req, struct iovec *iov, int count,
 
         INIT_LIST_HEAD (&entries.list);
         if (rsp.op_ret > 0) {
-                unserialize_rsp_dirent (&rsp, &entries);
+                unserialize_rsp_dirent (this, &rsp, &entries);
         }
 
         GF_PROTOCOL_DICT_UNSERIALIZE (frame->this, xdata,
diff --git a/xlators/protocol/client/src/client.c b/xlators/protocol/client/src/client.c
index d9e7ccd0c4f..7fca01c5b6a 100644
--- a/xlators/protocol/client/src/client.c
+++ b/xlators/protocol/client/src/client.c
@@ -24,6 +24,7 @@
 
 #include "xdr-rpc.h"
 #include "glusterfs3.h"
+#include "gf-dirent.h"
 
 extern rpc_clnt_prog_t clnt_handshake_prog;
 extern rpc_clnt_prog_t clnt_dump_prog;
@@ -1913,6 +1914,9 @@ client_readdir (call_frame_t *frame, xlator_t *this, fd_t *fd,
         if (!conf || !conf->fops)
                 goto out;
 
+        if (off != 0)
+                off = gf_dirent_orig_offset(this, off);
+
         args.fd = fd;
         args.size = size;
         args.offset = off;
@@ -1948,6 +1952,9 @@ client_readdirp (call_frame_t *frame, xlator_t *this, fd_t *fd,
         if (!conf || !conf->fops)
                 goto out;
 
+        if (off != 0)
+                off = gf_dirent_orig_offset(this, off);
+
         args.fd = fd;
         args.size = size;
         args.offset = off;
@@ -2447,7 +2454,7 @@ build_client_config (xlator_t *this, clnt_conf_t *conf)
 {
         int                     ret = -1;
 
-        if (!conf)
+       if (!conf)
                 goto out;
 
         GF_OPTION_INIT ("frame-timeout", conf->rpc_conf.rpc_timeout,
@@ -2470,6 +2477,8 @@ build_client_config (xlator_t *this, clnt_conf_t *conf)
 
         GF_OPTION_INIT ("send-gids", conf->send_gids, bool, out);
 
+        conf->client_id = glusterfs_leaf_position(this);
+
         ret = client_check_remote_host (this, this->options);
         if (ret)
                 goto out;
diff --git a/xlators/protocol/client/src/client.h b/xlators/protocol/client/src/client.h
index 7157e120dda..40200b6afc4 100644
--- a/xlators/protocol/client/src/client.h
+++ b/xlators/protocol/client/src/client.h
@@ -85,6 +85,7 @@ typedef struct clnt_conf {
         rpc_clnt_prog_t       *handshake;
         rpc_clnt_prog_t       *dump;
 
+        int                    client_id;
         uint64_t               reopen_fd_count; /* Count of fds reopened after a
                                                    connection is established */
         gf_lock_t              rec_lock;
@@ -228,7 +229,8 @@ int client_submit_request (xlator_t *this, void *req,
                            struct iovec *rsp_payload, int rsp_count,
                            struct iobref *rsp_iobref, xdrproc_t xdrproc);
 
-int unserialize_rsp_dirent (struct gfs3_readdir_rsp *rsp, gf_dirent_t *entries);
+int unserialize_rsp_dirent (xlator_t *this, struct gfs3_readdir_rsp *rsp,
+                            gf_dirent_t *entries);
 int unserialize_rsp_direntp (xlator_t *this, fd_t *fd,
                              struct gfs3_readdirp_rsp *rsp, gf_dirent_t *entries);