gNFS: Subdir mount does not work on UDP proto

After enabling nfs.mount-udp, mounting a subdir on a volume over
NFS fails. Because mountudpproc3_mnt_3_svc() invokes nfs3_rootfh()
which internally calls mnt3_mntpath_to_export() to resolve the
mount path. mnt3_mntpath_to_export() just works if the mount path
requested is volume itself. It is not able to resolve, if the path
is a subdir inside the volume.

MOUNT over TCP uses mnt3_find_export() to resolve subdir path but
UDP can't use this routine because mnt3_find_export() needs the
req data (of type rpcsvc_request_t) and it's available only for
TCP version of RPC.

FIX:
(1) Use syncop_lookup() framework to resolve the MOUNT PATH by
    breaking it into components and resolve component-by-component.
    i.e. glfs_resolve_at () API from libgfapi shared object.
(2) If MOUNT PATH is subdir, then make sure subdir export is not
    disabled.
(3) Add auth mechanism to respect nfs.rpc-auth-allow/reject and
    subdir auth i.e. nfs.export-dir
(4) Enhanced error handling for MOUNT over UDP

Change-Id: I42ee69415d064b98af4f49773026562824f684d1
BUG: 1118311
Signed-off-by: Santosh Kumar Pradhan <spradhan@redhat.com>
Reviewed-on: http://review.gluster.org/8346
Reviewed-by: soumya k <skoduri@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Niels de Vos <ndevos@redhat.com>

2 files changed, 88 insertions, 0 deletions

diff --git a/api/src/glfs-internal.h b/api/src/glfs-internal.h
index 6fdec961a1c..ff875280ce6 100644
--- a/api/src/glfs-internal.h
+++ b/api/src/glfs-internal.h
@@ -226,4 +226,55 @@ int glfs_getxattr_process (void *value, size_t size, dict_t *xattr,
 /* Sends RPC call to glusterd to fetch required volume info */
 int glfs_get_volume_info (struct glfs *fs);
 
+/*
+  SYNOPSIS
+
+       glfs_new_from_ctx: Creates a virtual mount object by taking a
+       glusterfs_ctx_t object.
+
+  DESCRIPTION
+
+       glfs_new_from_ctx() is not same as glfs_new(). It takes the
+       glusterfs_ctx_t object instead of creating one by glusterfs_ctx_new().
+       Again the usage is restricted to NFS MOUNT over UDP i.e. in
+       glfs_resolve_at() which would take fs object as input but never use
+       (purpose is not to change the ABI of glfs_resolve_at()).
+
+  PARAMETERS
+
+       @ctx: glusterfs_ctx_t object
+
+  RETURN VALUES
+
+       fs     : Pointer to the newly created glfs_t object.
+       NULL   : Otherwise.
+*/
+
+struct glfs *glfs_new_from_ctx (glusterfs_ctx_t *ctx);
+
+/*
+  SYNOPSIS
+
+       glfs_free_from_ctx: Free up the memory occupied by glfs_t object
+       created by glfs_new_from_ctx().
+
+  DESCRIPTION
+
+       The glfs_t object allocated by glfs_new_from_ctx() must be released
+       by the caller using this routine. The usage is restricted to NFS
+       MOUNT over UDP i.e.
+       __mnt3udp_get_export_subdir_inode ()
+                                => glfs_resolve_at().
+
+  PARAMETERS
+
+       @fs: The glfs_t object to be deallocated.
+
+  RETURN VALUES
+
+       void
+*/
+
+void glfs_free_from_ctx (struct glfs *fs);
+
 #endif /* !_GLFS_INTERNAL_H */
diff --git a/api/src/glfs.c b/api/src/glfs.c
index bcabb29f857..c7c4c00450a 100644
--- a/api/src/glfs.c
+++ b/api/src/glfs.c
@@ -554,6 +554,43 @@ glfs_new (const char *volname)
 }
 
 
+struct glfs *
+glfs_new_from_ctx (glusterfs_ctx_t *ctx)
+{
+        struct glfs    *fs = NULL;
+
+        if (!ctx)
+                return NULL;
+
+        fs = GF_CALLOC (1, sizeof (*fs), glfs_mt_glfs_t);
+        if (!fs)
+                return NULL;
+        fs->ctx = ctx;
+
+        (void) pthread_cond_init (&fs->cond, NULL);
+
+        (void) pthread_mutex_init (&fs->mutex, NULL);
+
+        INIT_LIST_HEAD (&fs->openfds);
+
+        return fs;
+}
+
+
+void
+glfs_free_from_ctx (struct glfs *fs)
+{
+        if (!fs)
+                return;
+
+        (void) pthread_cond_destroy (&fs->cond);
+
+        (void) pthread_mutex_destroy (&fs->mutex);
+
+        GF_FREE (fs);
+}
+
+
 int
 glfs_set_volfile (struct glfs *fs, const char *volfile)
 {