[RFC blktests v1 1/1] nvme/050: test DNR handling on reconnect

[Daniel Wagner]

When the host gets disconnected and tries to reconnect,
it should honor the DNR bit and do not retry to connect
with the same parameters.

Signed-off-by: Daniel Wagner <dwagner at suse.de>
---
 tests/nvme/050     | 126 +++++++++++++++++++++++++++++++++++++++++++++
 tests/nvme/050.out |   2 +
 2 files changed, 128 insertions(+)
 create mode 100644 tests/nvme/050
 create mode 100644 tests/nvme/050.out

diff --git a/tests/nvme/050 b/tests/nvme/050
new file mode 100644
index 000000000000..d33eb24e2f13
--- /dev/null
+++ b/tests/nvme/050
@@ -0,0 +1,126 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-3.0+
+# Copyright (C) 2023 SUSE LLC
+#
+# Test DNR is handled on connnect attempt with invalid arguments.
+
+. tests/nvme/rc
+
+DESCRIPTION="test DNR is handled on connect attempt with invalid arguments"
+
+requires() {
+	_nvme_requires
+	_require_nvme_trtype tcp rdma fc
+	_require_min_cpus 2
+}
+
+nvmf_wait_for_state() {
+	local def_state_timeout=5
+	local subsys_name="$1"
+	local state="$2"
+	local timeout="${3:-$def_state_timeout}"
+	local nvmedev
+	local state_file
+	local start_time
+	local end_time
+
+	nvmedev=$(_find_nvme_dev "${subsys_name}")
+	state_file="/sys/class/nvme-fabrics/ctl/${nvmedev}/state"
+
+	start_time=$(date +%s)
+	while ! grep -q "${state}" "${state_file}"; do
+		sleep 1
+		end_time=$(date +%s)
+		if (( end_time - start_time > timeout )); then
+			echo "expected state \"${state}\" not " \
+				"reached within ${timeout} seconds"
+			return 1
+		fi
+	done
+
+	return 0
+}
+
+nvmf_wait_for_ctrl_delete() {
+	local def_state_timeout=5
+	local nvmedev="$1"
+	local timeout="${2:-$def_state_timeout}"
+	local ctrl="/sys/class/nvme-fabrics/ctl/${nvmedev}/state"
+	local start_time
+	local end_time
+
+	start_time=$(date +%s)
+	while [ -f "${ctrl}" ]; do
+		sleep 1
+		end_time=$(date +%s)
+		if (( end_time - start_time > timeout )); then
+			echo "controller \"${nvmedev}\" not deleted" \
+				"within ${timeout} seconds"
+			return 1
+		fi
+	done
+
+	return 0
+}
+
+set_nvmet_attr_qid_max() {
+	local nvmet_subsystem="$1"
+	local qid_max="$2"
+	local cfs_path="${NVMET_CFS}/subsystems/${nvmet_subsystem}"
+
+	echo "${qid_max}" > "${cfs_path}/attr_qid_max"
+}
+
+test() {
+	echo "Running ${TEST_NAME}"
+
+	_setup_nvmet
+
+	local port
+	local loop_dev
+	local file_path="$TMPDIR/img"
+	local subsys_name="blktests-subsystem-1"
+	local hostid="77b49aba-06b4-431a-9af8-75e318740f1a"
+	local hostnqn="nqn.2014-08.org.nvmexpress:uuid:${hostid}"
+	local cfs_path="${NVMET_CFS}/subsystems/${subsys_name}"
+	local nvmedev
+
+	truncate -s "${nvme_img_size}" "${file_path}"
+
+	loop_dev="$(losetup -f --show "${file_path}")"
+
+	_create_nvmet_subsystem "${subsys_name}" "${loop_dev}" \
+		"91fdba0d-f87b-4c25-b80f-db7be1418b9e"
+	port="$(_create_nvmet_port "${nvme_trtype}")"
+	_add_nvmet_subsys_to_port "${port}" "${subsys_name}"
+
+	_nvme_connect_subsys "${nvme_trtype}" "${subsys_name}" \
+		--hostnqn "${hostnqn}" \
+		--reconnect-delay 1 \
+		--ctrl-loss-tmo 10
+
+	nvmf_wait_for_state "${subsys_name}" "live"
+	nvmedev=$(_find_nvme_dev "${subsys_name}")
+
+	# Only allow connects from ${def_hostnqn}
+	echo 0 > "${cfs_path}/attr_allow_any_host"
+
+	# Force a reconnect
+	set_nvmet_attr_qid_max "${subsys_name}" 1
+
+	# The reconnect fails with the DNR bit set
+	# Thus the host should remove the controller
+	nvmf_wait_for_ctrl_delete "${nvmedev}"
+
+	_nvme_disconnect_subsys "${subsys_name}" >> "$FULL" 2>&1
+
+	_remove_nvmet_subsystem_from_port "${port}" "${subsys_name}"
+	_remove_nvmet_subsystem "${subsys_name}"
+	_remove_nvmet_port "${port}"
+
+	losetup -d "${loop_dev}"
+
+	rm "${file_path}"
+
+	echo "Test complete"
+}
diff --git a/tests/nvme/050.out b/tests/nvme/050.out
new file mode 100644
index 000000000000..b78b05f78424
--- /dev/null
+++ b/tests/nvme/050.out
@@ -0,0 +1,2 @@
+Running nvme/050
+Test complete
-- 
2.41.0