* [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan
@ 2019-08-26 5:31 yufengmx
2019-08-26 5:31 ` yufengmx
0 siblings, 1 reply; 4+ messages in thread
From: yufengmx @ 2019-08-26 5:31 UTC (permalink / raw)
To: changqingx.wu, dts; +Cc: yufengmx
quote Nicolau, Radu
port start all action is not really a defect, but due to a race condition this
setup implies; the right way will be to start top level bond port only, and let it
propagate the start to the bonded ports and they will propagate it to
the real nics.
fix typo
yufengmx (1):
test_plans/pmd_stacked_bonded: update test plan
test_plans/pmd_stacked_bonded_test_plan.rst | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
--
1.9.3
^ permalink raw reply [flat|nested] 4+ messages in thread
* [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan
2019-08-26 5:31 [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan yufengmx
@ 2019-08-26 5:31 ` yufengmx
2019-08-26 7:10 ` Wu, ChangqingX
2019-09-04 5:04 ` Tu, Lijuan
0 siblings, 2 replies; 4+ messages in thread
From: yufengmx @ 2019-08-26 5:31 UTC (permalink / raw)
To: changqingx.wu, dts; +Cc: yufengmx
port start all action implies the race condition. The right way is to start
top level bond port only, and let it propagate the start action to slave bond ports
and its the real nics.
Signed-off-by: yufengmx <yufengx.mo@intel.com>
---
test_plans/pmd_stacked_bonded_test_plan.rst | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/test_plans/pmd_stacked_bonded_test_plan.rst b/test_plans/pmd_stacked_bonded_test_plan.rst
index be0da9a..a864f1f 100644
--- a/test_plans/pmd_stacked_bonded_test_plan.rst
+++ b/test_plans/pmd_stacked_bonded_test_plan.rst
@@ -41,7 +41,7 @@ based on RRC. The customer already uses Mellanox 100G NICs. Mellanox 100G NICs
support a proper x16 PCIe interface so the host sees a single netdev and that
netdev corresponds directly to the 100G Ethernet port. They indicated that in
their current system they bond multiple 100G NICs together, using DPDK bonding
-API in their application. They are interested in looking at an alternatve source
+API in their application. They are interested in looking at an alternative source
for the 100G NIC and are in conversation with Silicom who are shipping a 100G
RRC based NIC (something like Boulder Rapids). The issue they have with RRC NIC
is that the NIC presents as two PCIe interfaces (netdevs) instead of one. If the
@@ -140,9 +140,9 @@ steps
testpmd> show bonding config 3
testpmd> show bonding config 4
-#. start all ports to check ports start action::
+#. start top level bond port to check ports start action::
- testpmd> port start all
+ testpmd> port start 4
testpmd> start
#. close testpmd::
@@ -194,9 +194,9 @@ steps
testpmd> add bonding slave 3 4
testpmd> show bonding config 4
-#. start all bonded device ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 4
testpmd> start
#. send 100 tcp packets to portA 0 and portA 1::
@@ -260,9 +260,9 @@ steps
testpmd> add bonding slave 5 6
testpmd> show bonding config 6
-#. start all ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 6
testpmd> start
#. send 100 packets to portA 0/portA 0a/portA 1/portA 1a separately::
@@ -317,9 +317,9 @@ steps
testpmd> add bonding slave 3 4
testpmd> show bonding config 4
-#. start all ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 4
testpmd> start
#. send 100 packets to portA 0 and portA 1::
@@ -385,9 +385,9 @@ steps
testpmd> add bonding slave 5 6
testpmd> show bonding config 6
-#. start all ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 6
testpmd> start
#. send 100 packets to portA 0/portA 0a/portA 1/portA 1a separately::
--
1.9.3
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan
2019-08-26 5:31 ` yufengmx
@ 2019-08-26 7:10 ` Wu, ChangqingX
2019-09-04 5:04 ` Tu, Lijuan
1 sibling, 0 replies; 4+ messages in thread
From: Wu, ChangqingX @ 2019-08-26 7:10 UTC (permalink / raw)
To: Mo, YufengX, dts
Tested-by: Wu, ChangqingX <changqingx.wu@intel.com>
-----Original Message-----
From: Mo, YufengX
Sent: Monday, August 26, 2019 1:31 PM
To: Wu, ChangqingX <changqingx.wu@intel.com>; dts@dpdk.org
Cc: Mo, YufengX <yufengx.mo@intel.com>
Subject: [dts][PATCH V1]test_plans/pmd_stacked_bonded: update test plan
port start all action implies the race condition. The right way is to start top level bond port only, and let it propagate the start action to slave bond ports and its the real nics.
Signed-off-by: yufengmx <yufengx.mo@intel.com>
---
test_plans/pmd_stacked_bonded_test_plan.rst | 22 +++++++++++-----------
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/test_plans/pmd_stacked_bonded_test_plan.rst b/test_plans/pmd_stacked_bonded_test_plan.rst
index be0da9a..a864f1f 100644
--- a/test_plans/pmd_stacked_bonded_test_plan.rst
+++ b/test_plans/pmd_stacked_bonded_test_plan.rst
@@ -41,7 +41,7 @@ based on RRC. The customer already uses Mellanox 100G NICs. Mellanox 100G NICs support a proper x16 PCIe interface so the host sees a single netdev and that netdev corresponds directly to the 100G Ethernet port. They indicated that in their current system they bond multiple 100G NICs together, using DPDK bonding -API in their application. They are interested in looking at an alternatve source
+API in their application. They are interested in looking at an
+alternative source
for the 100G NIC and are in conversation with Silicom who are shipping a 100G RRC based NIC (something like Boulder Rapids). The issue they have with RRC NIC is that the NIC presents as two PCIe interfaces (netdevs) instead of one. If the @@ -140,9 +140,9 @@ steps
testpmd> show bonding config 3
testpmd> show bonding config 4
-#. start all ports to check ports start action::
+#. start top level bond port to check ports start action::
- testpmd> port start all
+ testpmd> port start 4
testpmd> start
#. close testpmd::
@@ -194,9 +194,9 @@ steps
testpmd> add bonding slave 3 4
testpmd> show bonding config 4
-#. start all bonded device ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 4
testpmd> start
#. send 100 tcp packets to portA 0 and portA 1::
@@ -260,9 +260,9 @@ steps
testpmd> add bonding slave 5 6
testpmd> show bonding config 6
-#. start all ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 6
testpmd> start
#. send 100 packets to portA 0/portA 0a/portA 1/portA 1a separately::
@@ -317,9 +317,9 @@ steps
testpmd> add bonding slave 3 4
testpmd> show bonding config 4
-#. start all ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 4
testpmd> start
#. send 100 packets to portA 0 and portA 1::
@@ -385,9 +385,9 @@ steps
testpmd> add bonding slave 5 6
testpmd> show bonding config 6
-#. start all ports::
+#. start top level bond port::
- testpmd> port start all
+ testpmd> port start 6
testpmd> start
#. send 100 packets to portA 0/portA 0a/portA 1/portA 1a separately::
--
1.9.3
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan
2019-08-26 5:31 ` yufengmx
2019-08-26 7:10 ` Wu, ChangqingX
@ 2019-09-04 5:04 ` Tu, Lijuan
1 sibling, 0 replies; 4+ messages in thread
From: Tu, Lijuan @ 2019-09-04 5:04 UTC (permalink / raw)
To: Mo, YufengX, Wu, ChangqingX, dts; +Cc: Mo, YufengX
Applied, thanks
> -----Original Message-----
> From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of yufengmx
> Sent: Monday, August 26, 2019 1:31 PM
> To: Wu, ChangqingX <changqingx.wu@intel.com>; dts@dpdk.org
> Cc: Mo, YufengX <yufengx.mo@intel.com>
> Subject: [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan
>
>
> port start all action implies the race condition. The right way is to start top
> level bond port only, and let it propagate the start action to slave bond
> ports and its the real nics.
>
> Signed-off-by: yufengmx <yufengx.mo@intel.com>
> ---
> test_plans/pmd_stacked_bonded_test_plan.rst | 22 +++++++++++-----------
> 1 file changed, 11 insertions(+), 11 deletions(-)
>
> diff --git a/test_plans/pmd_stacked_bonded_test_plan.rst
> b/test_plans/pmd_stacked_bonded_test_plan.rst
> index be0da9a..a864f1f 100644
> --- a/test_plans/pmd_stacked_bonded_test_plan.rst
> +++ b/test_plans/pmd_stacked_bonded_test_plan.rst
> @@ -41,7 +41,7 @@ based on RRC. The customer already uses Mellanox
> 100G NICs. Mellanox 100G NICs support a proper x16 PCIe interface so the
> host sees a single netdev and that netdev corresponds directly to the 100G
> Ethernet port. They indicated that in their current system they bond
> multiple 100G NICs together, using DPDK bonding -API in their application.
> They are interested in looking at an alternatve source
> +API in their application. They are interested in looking at an
> +alternative source
> for the 100G NIC and are in conversation with Silicom who are shipping a
> 100G RRC based NIC (something like Boulder Rapids). The issue they have
> with RRC NIC is that the NIC presents as two PCIe interfaces (netdevs)
> instead of one. If the @@ -140,9 +140,9 @@ steps
> testpmd> show bonding config 3
> testpmd> show bonding config 4
>
> -#. start all ports to check ports start action::
> +#. start top level bond port to check ports start action::
>
> - testpmd> port start all
> + testpmd> port start 4
> testpmd> start
>
> #. close testpmd::
> @@ -194,9 +194,9 @@ steps
> testpmd> add bonding slave 3 4
> testpmd> show bonding config 4
>
> -#. start all bonded device ports::
> +#. start top level bond port::
>
> - testpmd> port start all
> + testpmd> port start 4
> testpmd> start
>
> #. send 100 tcp packets to portA 0 and portA 1::
> @@ -260,9 +260,9 @@ steps
> testpmd> add bonding slave 5 6
> testpmd> show bonding config 6
>
> -#. start all ports::
> +#. start top level bond port::
>
> - testpmd> port start all
> + testpmd> port start 6
> testpmd> start
>
> #. send 100 packets to portA 0/portA 0a/portA 1/portA 1a separately::
> @@ -317,9 +317,9 @@ steps
> testpmd> add bonding slave 3 4
> testpmd> show bonding config 4
>
> -#. start all ports::
> +#. start top level bond port::
>
> - testpmd> port start all
> + testpmd> port start 4
> testpmd> start
>
> #. send 100 packets to portA 0 and portA 1::
> @@ -385,9 +385,9 @@ steps
> testpmd> add bonding slave 5 6
> testpmd> show bonding config 6
>
> -#. start all ports::
> +#. start top level bond port::
>
> - testpmd> port start all
> + testpmd> port start 6
> testpmd> start
>
> #. send 100 packets to portA 0/portA 0a/portA 1/portA 1a separately::
> --
> 1.9.3
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2019-09-04 5:04 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-08-26 5:31 [dts] [PATCH V1]test_plans/pmd_stacked_bonded: update test plan yufengmx
2019-08-26 5:31 ` yufengmx
2019-08-26 7:10 ` Wu, ChangqingX
2019-09-04 5:04 ` Tu, Lijuan
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).