test suite reviews and discussions
 help / color / mirror / Atom feed
From: yufengmx <yufengx.mo@intel.com>
To: dts@dpdk.org, yinan.wang@intel.com
Cc: yufengmx <yufengx.mo@intel.com>
Subject: [dts] [PATCH V3 1/3] ethtool_stats: update test plan
Date: Fri,  1 Nov 2019 15:06:45 +0800	[thread overview]
Message-ID: <20191101070647.59192-2-yufengx.mo@intel.com> (raw)
In-Reply-To: <20191101070647.59192-1-yufengx.mo@intel.com>


use testpmd xstat data as reference data to check proc info tool xstat data.
update test plan content.

Signed-off-by: yufengmx <yufengx.mo@intel.com>
---
 test_plans/ethtool_stats_test_plan.rst | 66 ++++++++++++++++++--------
 1 file changed, 45 insertions(+), 21 deletions(-)

diff --git a/test_plans/ethtool_stats_test_plan.rst b/test_plans/ethtool_stats_test_plan.rst
index 13b5909..95f9e7a 100644
--- a/test_plans/ethtool_stats_test_plan.rst
+++ b/test_plans/ethtool_stats_test_plan.rst
@@ -47,13 +47,20 @@ that DPDK supports. The idea behind this epic is two fold as following.
     - the retrieval of aggregate stats upon request (Top level stats).
     - the retrieval of the extended NIC stats.
     - grouping of stats logically so they can be retrieved per logical grouping.
-    - the option to enable/disable the stats groups to retrieve similar to set private flags in ethtool.
+    - the option to enable/disable the stats groups to retrieve similar to set
+      private flags in ethtool.
 
 Prerequisites
 =============
 
 2xNICs (2 full duplex optical ports per NIC). One on dut, another one on tester,
-link them together. Update two nics' firmware to latest version.
+link them together. Update two nics' firmware to latest version::
+
+            Tester                          DUT
+          .-------.                      .-------.
+          | port0 | <------------------> | port0 |
+          | port1 | <------------------> | port1 |
+          '-------'                      '-------'
 
 Test cases
 ==========
@@ -64,8 +71,8 @@ bind two ports::
 
     ./usertools/dpdk-devbind.py --bind=igb_uio <pci address 1> <pci address 2>
 
-Test Case: xstat command set intergrity
----------------------------------------
+Test Case: xstat options
+------------------------
 
 check ``dpdk-procinfo`` tool support ``xstats`` command options.
 
@@ -80,7 +87,11 @@ steps:
 
 #. boot up ``testpmd``::
 
-    ./<target name>/app/testpmd -c 0x600006 -n 4  -- -i --port-topology=loop
+    ./<target name>/app/testpmd -c 0x3 -n 4  -- -i --port-topology=loop
+
+    testpmd> set fwd io
+    testpmd> clear port xstats all
+    testpmd> start
 
 #. run ``dpdk-procinfo`` tool::
 
@@ -88,25 +99,30 @@ steps:
 
 #. check ``dpdk-procinfo`` tool output should contain upper options.
 
-Test Case: xstat command
-------------------------
+Test Case: xstat statistic integrity
+------------------------------------
 
-check if port extended statistics can get right data.
+check if port extended statistics can access by xstat name or xstat id.
 
 steps:
 
 #. boot up ``testpmd``::
 
-    ./<target name>/app/testpmd -c 0x600006 -n 4  -- -i --port-topology=loop
+    ./<target name>/app/testpmd -c 0x3 -n 4  -- -i --port-topology=loop
+
+    testpmd> set fwd io
+    testpmd> clear port xstats all
+    testpmd> start
 
 #. send udp packet of 64/72/128/256/512/1024 size to port 0/1::
 
     sendp([Ether()/IP()/UDP()/Raw('\0'*60)], iface=<port 0 name>)
 
-#. run ``dpdk-procinfo`` tool with ``xstats`` option and check if all port have
-   a correct data of different size packet::
+#. run ``dpdk-procinfo`` tool with ``xstats`` option and check if all ports
+   extended statistics can access by xstat name or xstat id::
 
-    ./<target name>/app/dpdk-procinfo -- -p 3 --xstats
+    ./<target name>/app/dpdk-procinfo -- -p 3 --xstats-id <N>
+    ./<target name>/app/dpdk-procinfo -- -p 3 --xstats-name <statistic name>
 
 Test Case: xstat-reset command
 ------------------------------
@@ -117,7 +133,11 @@ steps:
 
 #. boot up ``testpmd``::
 
-    ./<target name>/app/testpmd -c 0x600006 -n 4  -- -i --port-topology=loop
+    ./<target name>/app/testpmd -c 0x3 -n 4  -- -i --port-topology=loop
+
+    testpmd> set fwd io
+    testpmd> clear port xstats all
+    testpmd> start
 
 #. send udp packet of 64/72/128/256/512/1024 size to port 0/1::
 
@@ -132,24 +152,26 @@ Test Case: xstat single statistic
 ---------------------------------
 
 check if port extended statistic name can be get by statistic id and check
-related data's correctness.
+related data's correctness with testpmd xstat data.
 
 steps:
 
 #. boot up ``testpmd``::
 
-    ./<target name>/app/testpmd -c 0x600006 -n 4  -- -i --port-topology=loop
-
-#. run ``dpdk-procinfo`` tool with ``xstats`` option to get all NIC extended
-   statistics parameters of port and use the output display sequence as parameter
-   index in the following search index::
+    ./<target name>/app/testpmd -c 0x3 -n 4  -- -i --port-topology=loop
 
-    ./<target name>/app/dpdk-procinfo -- -p 3 --xstats
+    testpmd> set fwd io
+    testpmd> clear port xstats all
+    testpmd> start
 
 #. send udp packet of 64/72/128/256/512/1024 size to port 0/1::
 
     sendp([Ether()/IP()/UDP()/Raw('\0'*60)], iface=<port 0 name>)
 
+#. run test pmd ``show port xstats all`` to get testpmd port xstat data:
+
+    testpmd> show port xstats all
+
 #. run ``dpdk-procinfo`` tool with ``xstats-id`` option to get the statistic
    name corresponding with the index id::
 
@@ -158,4 +180,6 @@ steps:
 #. run ``dpdk-procinfo`` tool with ``xstats-name`` option to get the statistic
    data corresponding with the statistic name::
 
-    ./<target name>/app/dpdk-procinfo -- -p 3 --xstats-name <statistic name>
\ No newline at end of file
+    ./<target name>/app/dpdk-procinfo -- -p 3 --xstats-name <statistic name>
+
+#. compare these proc info tool xstat values with testpmd xstat values.
\ No newline at end of file
-- 
2.21.0


  reply	other threads:[~2019-11-01  7:05 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-01  7:06 [dts] [PATCH V3 0/3] ethtool_stats: upload script and " yufengmx
2019-11-01  7:06 ` yufengmx [this message]
2019-11-12 15:34   ` [dts] [PATCH V3 1/3] ethtool_stats: " Wang, Yinan
2019-11-12 16:03     ` Wang, Yinan
2019-11-13  2:20   ` Wang, Yinan
2019-11-01  7:06 ` [dts] [PATCH V3 2/3] ethtool_stats: add ethtool_stats_test_plan index label yufengmx
2019-11-12 16:00   ` Wang, Yinan
2019-11-01  7:06 ` [dts] [PATCH V3 3/3] ethtool_stats: add automation script yufengmx
2019-11-12 15:38   ` Wang, Yinan
2019-11-12 16:02     ` Wang, Yinan
2019-11-13  1:10     ` Mo, YufengX
2019-11-13  2:19       ` Wang, Yinan
2019-11-13  2:20         ` Mo, YufengX
2019-11-12 16:00 ` [dts] [PATCH V3 0/3] ethtool_stats: upload script and update test plan Wang, Yinan
2019-11-22  3:33 ` Tu, Lijuan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191101070647.59192-2-yufengx.mo@intel.com \
    --to=yufengx.mo@intel.com \
    --cc=dts@dpdk.org \
    --cc=yinan.wang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).