Tests to port DTS

	A	B	C	D	E	F	G
1	ETHDEV TESTS	Status	Difficulty	Priority (1-5) (5 = highest)	old test plan	notes	Candidate to write this suite
2	Scatter	DONE	Medium	4
3	MTU_Update	SELECTED	Easy	4			Alex
4	Jumboframes	SELECTED	Easy	4			Nick
5	RxTx Offload	SELECTED		4		Majority of verification in old DTS verifies based on how many packets received and which queue received traffic. Heavy expectation of a quiet wire	Jeremy
6	Queue start/stop	SELECTED		3		linux specific network interface management	Dean
7	VLAN Offload			3		need to verify not Intel NIC specific	Dean
8	Dual VLAN	SELECTED		3		need to verify not Intel NIC specific. Has a lot of overlap with the VLAN test suite that isn't really related to dual VLANs which can likely get cut out. Some features are Intel specific such as tpid and extended offload.	Jeremy
9	Checksum Offload	SELECTED		3	https://git.dpdk.org/tools/dts/tree/test_plans/checksum_offload_test_plan.rst	Partially complete, removed the tx test cases due to lack of tx flag support in verbose output	Dean
10	tso_suite	SELECTED				Needs to validate failure of checksums which can be awkward, but this functionality could potentially be taken from the Rx/Tx offload suie	Dean
11	ddp_gtp					All DDP tests might be Intel specific. The test plan specifically uses a profile that is made for Intel, and after a quick search, it looks like MLX might support DDP, but I haven't found any available package files for it.
12	ddp_gtp_qregion
13	ddp_l2tvp3
14	ddp_mpls
15	ddp_ppp_m2ls
16	stats_checks_test_plan				https://git.dpdk.org/tools/dts/tree/test_plans/stats_checks_test_plan.rst	Testcase checking number of packets recieved requires quiet wire, which we are not providing (at least yet) in new DTS. Because we aren't supporting quiet wires, a potential strategy for this is "did the stats increase by at least x amount" rather than strictly x amount. Need to check if this is still valuable information.	Jeremy
17	mac_filter	SELECTED					Nick
18	sriov_kvm
19	inline_ipsec
20	Interrupt_pmd			3		Need to look into how this event effects the interactive shell buffer. Other link state change events have been known to completely break the output buffer. Maybe the improving interactive shell output gathering series will help with this.
21	pf_smoke	SELECTED					Jeremy
22	vf_smoke						Jeremy
23	queue_region	SELECTED
24	speed_capabilities	SELECTED		3			Alex
25	pmd					Will be relatively straightforward once we have a better process for distinguishing noise from framework packets in verbose output.
26	pmd_power					Requires l3fwd app, turbostat package, and waitpkg package. Generally looking to see how ports operate at different frequencies and ensuring that they stay in certain frequency ranges when handling packets.
27	blocklist	SELECTED	Easy		https://git.dpdk.org/tools/dts/tree/test_plans/blocklist_test_plan.rst	Start testpmd with -b {device}, and check if device is brought up by testpmd or not. Might be a good one for Luca to write as he already has a patch in motion for show port stats/info in testpmd.	Luca
28	link_status_interrupt			3		uses example app other than testpmd. Uses lots of linux commands for managing network interfaces.
29	vf_rss		Medium	4	https://git.dpdk.org/tools/dts/tree/test_plans/vf_rss_test_plan.rst
30	RSSreta (redirection table update)	SELECTED	Medium	4			Alex
31	rss_key_update	SELECTED	Medium	4			Alex
32	l2fwd sample app test	SELECTED	Easy		https://git.dpdk.org/tools/dts/tree/test_plans/l2fwd_test_plan.rst		Luca
33	pmdrss_hash_test_plan		Easy
34	external_memory				https://git.dpdk.org/tools/dts/tree/test_plans/external_memory_test_plan.rst
35
36
37	port_control	SELECTED	easy		https://git.dpdk.org/tools/dts/tree/test_plans/port_control_test_plan.rst		Jeremy
38	tx_preparation	SELECTED	easy		https://git.dpdk.org/tools/dts/tree/test_plans/tx_preparation_test_plan.rst		Patrick
39	ieee1588				https://git.dpdk.org/tools/dts/tree/test_plans/ieee1588_test_plan.rst	requires ieee1588 fwd mode, which doesn't seem to be recognized by testpmd. 1588 dc flag seems to be removed from dpdk meson.
40	ipgre	SELECTED	Medium		https://git.dpdk.org/tools/dts/tree/test_plans/ipgre_test_plan.rst		Nick
41	nvgre	SELECTED	Medium		https://git.dpdk.org/tools/dts/tree/test_plans/nvgre_test_plan.rst	nvgre is almost identical to the above 'ipgre' test suite, but both suites technically tests different functionality. nvgre leverages gre to provides is service, so two test suites are needed. However, both tests are identical in a lot of ways, so it makes sense to write this suite alongside ipgre and perhaps publish both in a series	Nick
42	vlan_ethertype_config				https://git.dpdk.org/tools/dts/tree/test_plans/vlan_ethertype_config_test_plan.rst		Nick
43
44	multicast				https://git.dpdk.org/tools/dts/tree/test_plans/multicast_test_plan.rst	Uses a sample app, but testing could potentially be replicated by making similar routing tables using flow rules. I tried looking into the sample app some more and couldn't get it to extract the metrics from testpmd.
45	packet_capture/pmdpcap				https://git.dpdk.org/tools/dts/tree/test_plans/packet_capture_test_plan.rst \| https://git.dpdk.org/tools/dts/tree/test_plans/pmdpcap_test_plan.rst	This suite uses a sample app, but you could try and do similar verification on just the .pcap files that are being created from tcpdump/scapy instead of the ones created by the sample application
46	softnic			1 (req test case conf files)	https://git.dpdk.org/tools/dts/tree/test_plans/softnic_test_plan.rst	This suite uses configuration files and then basically starts testpmd with a vdev and verifies that received packets match an expected structure.
47	uni_pkt				https://git.dpdk.org/tools/dts/tree/test_plans/uni_pkt_test_plan.rst	Seems basically like it is verifying SW and HW types in the testpmd verbose output. Says that it requires ABI? Unsure why this is explicitly called out	Dean
48	promisc_support		easy		none?	We should have a test which clearly is testing promiscuous support. This should include validating correct behavior for both promisc enabled and promisc disabled, and also re-validating these work after a port start/stop (which should not affect promisc mode). So, we could have something like testcase 1 involves disabling promisc, sending 2 packets (one with correct mac, one with incorrect mac) and checking whether they were recieved. Then, enable promisc, query testpmd for promisc mode, and then send the two packets again, checking that the are recieved in both cases. testcase 2 could be the same thing as testcase 1 but include a port restart(s) in order to validate that the promisc mode is unaffected by port restarts, as is stated in rte__ethdev. This can all be modified, or broken out into additional testcases etc. but this is my rough idea.	alex
49	port restart configuration persistency				none?	David M indicated in his email about needed ethdev coverage that port config persistency between restarts should be tested. According to the ethdev description, these should be persistent: - MTU - flow control settings - receive mode configuration (promiscuous mode, all-multicast mode, hardware checksum mode, RSS/VMDq settings etc.) - VLAN filtering configuration - default MAC address - MAC addresses supplied to MAC address array - flow director filtering mode (but not filtering rules) - NIC queue statistics mappings and these can be persistent, based on a capability which can be queried - flow rules - flow-related shared objects, e.g. indirect actions I think for this series, I propose we only set the values, query testpmd for port config, restart, and query testpmd again, but not actually send packets.
50	vf_port_start_stop				https://git.dpdk.org/tools/dts/tree/test_plans/vf_port_start_stop_test_plan.rst	This is kind of an extension of the previous item, but obviously A. it has a dependency on vf support, which Jeremy is working on currently, and B. it involves actually sending packets after the restart. I think this suite could actually be extended to validate much more port configuration than the limited scope covered in this case.
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100