OpenZFS 2605, 6980, 6902

2605 want to resume interrupted zfs send
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Paul Dagnelie <pcd@delphix.com>
Reviewed by: Richard Elling <Richard.Elling@RichardElling.com>
Reviewed by: Xin Li <delphij@freebsd.org>
Reviewed by: Arne Jansen <sensille@gmx.net>
Approved by: Dan McDonald <danmcd@omniti.com>
Ported-by: kernelOfTruth <kerneloftruth@gmail.com>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>

OpenZFS-issue: https://www.illumos.org/issues/2605
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/9c3fd12

6980 6902 causes zfs send to break due to 32-bit/64-bit struct mismatch
Reviewed by: Paul Dagnelie <pcd@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Ported by: Brian Behlendorf <behlendorf1@llnl.gov>

OpenZFS-issue: https://www.illumos.org/issues/6980
OpenZFS-commit: https://github.com/openzfs/openzfs/commit/ea4a67f

Porting notes:
- All rsend and snapshop tests enabled and updated for Linux.
- Fix misuse of input argument in traverse_visitbp().
- Fix ISO C90 warnings and errors.
- Fix gcc 'missing braces around initializer' in
  'struct send_thread_arg to_arg =' warning.
- Replace 4 argument fletcher_4_native() with 3 argument version,
  this change was made in OpenZFS 4185 which has not been ported.
- Part of the sections for 'zfs receive' and 'zfs send' was
  rewritten and reordered to approximate upstream.
- Fix mktree xattr creation, 'user.' prefix required.
- Minor fixes to newly enabled test cases
- Long holds for volumes allowed during receive for minor registration.

tests/zfs-tests/tests/functional/cli_root/zfs_receive/Makefile.am

@@ -10,4 +10,7 @@ dist_pkgdata_SCRIPTS = \
 	zfs_receive_006_pos.ksh \
 	zfs_receive_007_neg.ksh \
 	zfs_receive_008_pos.ksh \
-	zfs_receive_009_neg.ksh
+	zfs_receive_009_neg.ksh \
+	zfs_receive_010_pos.ksh \
+	zfs_receive_011_pos.ksh \
+	zfs_receive_012_pos.ksh

tests/zfs-tests/tests/functional/cli_root/zfs_send/zfs_send_007_pos.ksh

@@ -61,7 +61,7 @@ test_pool ()
 	first_object=$(ls -i $mntpnt | awk '{print $1}')
 	log_must $ZFS snapshot $POOL/fs@a
 	while true; do
-		log_must $FIND $mntpnt -delete
+		log_must $FIND $mntpnt/* -delete
 		sync
 		log_must $MKFILES "$mntpnt/" 4000
 		FILE=$(ls -i $mntpnt | awk \

tests/zfs-tests/tests/functional/rsend/rsend.kshlib

@@ -74,9 +74,11 @@ function setup_test_model
 	if is_global_zone ; then
 		log_must $ZFS create -V 16M $pool/vol
 		log_must $ZFS create -V 16M $pool/$FS/vol
+		block_device_wait
 
 		log_must $ZFS snapshot $pool/$FS/vol@vsnap
 		log_must $ZFS clone $pool/$FS/vol@vsnap $pool/$FS/vclone
+		block_device_wait
 	fi
 
 	log_must snapshot_tree $pool/$FS/fs1/fs2@fsnap
@@ -199,10 +201,10 @@ function cmp_ds_prop
 	typeset dtst1=$1
 	typeset dtst2=$2
 
-	for item in "type" "origin" "volblocksize" "aclinherit" "aclmode" \
+	for item in "type" "origin" "volblocksize" "aclinherit" "acltype" \
 	    "atime" "canmount" "checksum" "compression" "copies" "devices" \
 	    "dnodesize" "exec" "quota" "readonly" "recordsize" "reservation" \
-	    "setuid" "sharenfs" "snapdir" "version" "volsize" "xattr" "zoned" \
+	    "setuid" "snapdir" "version" "volsize" "xattr" "zoned" \
 	    "mountpoint";
 	do
 		$ZFS get -H -o property,value,source $item $dtst1 >> \
@@ -393,7 +395,7 @@ function mk_files
 	for ((i=0; i<$nfiles; i=i+1)); do
 		$DD if=/dev/urandom \
 		    of=/$fs/file-$maxsize-$((i+$file_id_offset)) \
-		    bs=$(($RANDOM * $RANDOM % $maxsize)) \
+		    bs=$((($RANDOM * $RANDOM % ($maxsize - 1)) + 1)) \
 		    count=1 >/dev/null 2>&1 || log_fail \
 		    "Failed to create /$fs/file-$maxsize-$((i+$file_id_offset))"
 	done
@@ -438,7 +440,7 @@ function mess_file
 		# write the same value that's already there.
 		#
 		log_must eval "$DD if=/dev/urandom of=$file conv=notrunc " \
-		    "bs=1 count=2 oseek=$offset >/dev/null 2>&1"
+		    "bs=1 count=2 seek=$offset >/dev/null 2>&1"
 	else
 		log_must $TRUNCATE -s $offset $file
 	fi
@@ -523,20 +525,20 @@ function test_fs_setup
 		mk_files 100 1048576 0 $sendfs &
 		mk_files 10 10485760 0 $sendfs &
 		mk_files 1 104857600 0 $sendfs &
-		log_must $WAIT
+		wait
 		log_must $ZFS snapshot $sendfs@a
 
 		rm_files 200 256 0 $sendfs &
 		rm_files 200 131072 0 $sendfs &
 		rm_files 20 1048576 0 $sendfs &
 		rm_files 2 10485760 0 $sendfs &
-		log_must $WAIT
+		wait
 
 		mk_files 400 256 0 $sendfs &
 		mk_files 400 131072 0 $sendfs &
 		mk_files 40 1048576 0 $sendfs &
 		mk_files 4 10485760 0 $sendfs &
-		log_must $WAIT
+		wait
 
 		log_must $ZFS snapshot $sendfs@b
 		log_must eval "$ZFS send -v $sendfs@a >/$sendpool/initial.zsend"

tests/zfs-tests/tests/functional/rsend/rsend_012_pos.ksh

@@ -110,9 +110,6 @@ function cleanup
 		log_must $ZFS inherit $prop $POOL2
 	done
 
-	#if is_shared $POOL; then
-	#	log_must $ZFS set sharenfs=off $POOL
-	#fi
 	log_must setup_test_model $POOL
 
 	if [[ -d $TESTDIR ]]; then
@@ -131,7 +128,7 @@ for fs in "$POOL" "$POOL/pclone" "$POOL/$FS" "$POOL/$FS/fs1" \
 	"$POOL/$FS/fs1/fs2" "$POOL/$FS/fs1/fclone" ; do
 	rand_set_prop $fs aclinherit "discard" "noallow" "secure" "passthrough"
 	rand_set_prop $fs checksum "on" "off" "fletcher2" "fletcher4" "sha256"
-	rand_set_prop $fs aclmode "discard" "groupmask" "passthrough"
+	rand_set_prop $fs acltype "off" "noacl" "posixacl"
 	rand_set_prop $fs atime "on" "off"
 	rand_set_prop $fs checksum "on" "off" "fletcher2" "fletcher4" "sha256"
 	rand_set_prop $fs compression "on" "off" "lzjb" "gzip" \
@@ -161,7 +158,8 @@ done
 
 
 # Verify inherited property can be received
-rand_set_prop $POOL sharenfs "on" "off" "rw"
+rand_set_prop $POOL redundant_metadata "all" "most"
+rand_set_prop $POOL sync "standard" "always" "disabled"
 
 #
 # Duplicate POOL2 for testing

tests/zfs-tests/tests/functional/rsend/rsend_014_pos.ksh

@@ -46,6 +46,7 @@ log_must cleanup_pool $POOL2
 
 log_must eval "$ZFS send -R $POOL/$FS@final > $BACKDIR/fs-final-R"
 log_must eval "$ZFS receive -d $POOL2 < $BACKDIR/fs-final-R"
+block_device_wait
 log_must eval "$ZPOOL export $POOL"
 log_must eval "$ZPOOL import $POOL"