age on vhost socket disconnect,
---
Documentation/topics/userspace-tso.rst | 26 ++-
lib/netdev-dpdk.c | 100 -
2 files changed, 120 insertions(+), 6 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
On 6/23/23 23:52, Ilya Maximets wrote:
On 6/23/23 19:33, David Marchand wrote:
On Fri, Jun 23, 2023 at 5:14 PM Ilya Maximets wrote:
On 6/23/23 15:16, David Marchand wrote:
On Wed, Jun 21, 2023 at 12:13 AM Ilya Maximets wrote:
On 6/14/23 21:03, Mike Pattrick wrote:
The netdev receiving
On 2/28/23 16:37, Ilya Maximets wrote:
On 2/28/23 16:16, Maxime Coquelin wrote:
Hi,
On 2/28/23 10:33, Wan Junjie wrote:
Hi,
On Tue, Feb 28, 2023 at 4:03 PM David Marchand
wrote:
Hello,
On Fri, Feb 24, 2023 at 2:55 AM Han Ding wrote:
On 2/23/23 10:35, Han Ding wrote:
When use ovs
Hi,
On 2/28/23 10:33, Wan Junjie wrote:
Hi,
On Tue, Feb 28, 2023 at 4:03 PM David Marchand
wrote:
Hello,
On Fri, Feb 24, 2023 at 2:55 AM Han Ding wrote:
On 2/23/23 10:35, Han Ding wrote:
When use ovs-vsctl to delete vhostuserclient port while qemu destroy virtio,
there is a deadlock
if a fair number of updates is pending,
---
lib/netdev-dpdk.c | 122 ++
1 file changed, 102 insertions(+), 20 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openv
Hi David,
On 1/12/23 21:55, David Marchand wrote:
As Ilya reported, we have a ABBA deadlock between DPDK vq->access_lock
and OVS dev->mutex when OVS main thread refreshes statistics, while a
vring state change event is being processed for a same vhost port.
To break from this situation, move
---
lib/netdev-dpdk.c | 9 -
1 file changed, 9 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
=33
tx_q1_size_256_511_packets=15
tx_q1_size_65_127_packets=21
tx_q1_undersize_packets=12
Reviewed-by: Maxime Coquelin
Signed-off-by: David Marchand
---
Changes since v6:
- fixed tx_retries,
- dropped netdev_dpdk_vhost_update_[rt]x_counters helpers. This makes
it clearer that a lock can be taken in the vhost rx
=33
tx_q1_size_256_511_packets=15
tx_q1_size_65_127_packets=21
tx_q1_undersize_packets=12
Reviewed-by: Maxime Coquelin
Signed-off-by: David Marchand
---
Changes since v5:
- added missing dev->stats_lock acquire in netdev_dpdk_vhost_get_stats,
- changed netdev_dpdk_vhost_update_[rt]x_counters to take dev->s
=48
tx_q1_multicast_packets=33
tx_q1_size_256_511_packets=15
tx_q1_size_65_127_packets=21
tx_q1_undersize_packets=12
Reviewed-by: Maxime Coquelin
Signed-off-by: David Marchand
---
Changes since v5:
- added missing dev->stats_lock acquire in netdev_dpdk_vhost_get_stats,
- changed netdev_dpdk_vhost_upda
On 12/2/22 15:59, Mike Pattrick wrote:
On Fri, Dec 2, 2022 at 5:37 AM Maxime Coquelin
wrote:
On 12/2/22 11:09, David Marchand wrote:
On Wed, Nov 30, 2022 at 9:30 PM Ilya Maximets wrote:
Shouldn't this be 0x7f instead?
0x3f doesn't enable bit #6, which is responsible for dumping
shared
On 12/2/22 11:09, David Marchand wrote:
On Wed, Nov 30, 2022 at 9:30 PM Ilya Maximets wrote:
Shouldn't this be 0x7f instead?
0x3f doesn't enable bit #6, which is responsible for dumping
shared huge pages. Or am I missing something?
That's a good point, the hugepage may or may not be
good to me, thanks for working on this!
Reviewed-by: Maxime Coquelin
Regards,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
Hi David,
On 10/7/22 13:16, David Marchand wrote:
Request per virtqueue statistics from the vhost library and expose them
as per port OVS custom stats.
Thanks for the patch!
Note:
- the vhost stats API is experimental at the moment, this patch is
sent as a demonstrator,
I'm going to
Hi Ilya,
On 1/17/22 19:01, Ilya Maximets wrote:
On 1/5/22 09:19, Maxime Coquelin wrote:
Hash-based Tx steering feature will enable steering Tx
packets on transmit queues based on their hashes. In order
to test the feature, it is needed to be able to get the
per-queue statistics for Vhost-user
Hi David,
On 9/20/22 17:50, David Marchand wrote:
I did not finish reviewing the rest, but I have an additional comment
on the virtio/vhost features negotiations.
On Wed, Jul 13, 2022 at 8:45 PM Mike Pattrick wrote:
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index
On 9/20/22 11:28, Phelan, Michael wrote:
-Original Message-
From: Maxime Coquelin
Sent: Monday 19 September 2022 12:39
To: Phelan, Michael ; d...@openvswitch.org
Cc: i.maxim...@ovn.org
Subject: Re: [ovs-dev] [v2] dpdk: Use DPDK 21.11.2 release.
On 9/16/22 16:23, Michael Phelan
K 21.11.3.
+ It should be noted that DPDK 21.11.0 does not benefit from the numerous
+ bug and CVE fixes addressed since its release.
+ If a user wishes to benefit from these fixes it is recommended to use
+ DPDK 21.11.2.
v3.0.0 - 15 Aug
al Message-
From: Maxime Coquelin
Sent: Thursday 15 September 2022 09:35
To: Kevin Traynor ; i.maxim...@ovn.org; Phelan,
Michael ; d...@openvswitch.org
Cc: David Marchand
Subject: Re: [ovs-dev] [PATCH] dpdk: Use DPDK 21.11.2 release.
On 9/15/22 10:27, Kevin Traynor wrote:
Hi Michael/Maxime,
On 9/15/22 10:27, Kevin Traynor wrote:
Hi Michael/Maxime,
On 14/09/2022 16:53, Kevin Traynor wrote:
On 14/09/2022 16:18, Maxime Coquelin wrote:
Hi Kevin,
On 9/14/22 16:14, Kevin Traynor wrote:
Hi Maxime,
On 14/09/2022 15:45, Maxime Coquelin wrote:
Hi,
On 9/8/22 13:15, Michael Phelan
Hi Kevin,
On 9/14/22 16:14, Kevin Traynor wrote:
Hi Maxime,
On 14/09/2022 15:45, Maxime Coquelin wrote:
Hi,
On 9/8/22 13:15, Michael Phelan wrote:
CI Linux build script updated to use DPDK version 21.11.2.
Documentation updated to use DPDK version 21.11.2.
NEWS entry added for DPDK 21.11.2
On 9/14/22 15:59, Phelan, Michael wrote:
Hi Maxime,
Thanks for the heads up, is there a timeline on when 21.11.3 is expected to be
released?
v21.11.3 is planned to be released in December.
Regards,
Maxime
Kind regards,
Michael.
-Original Message-
From: Maxime Coquelin
Sent
Hi,
On 9/8/22 13:15, Michael Phelan wrote:
CI Linux build script updated to use DPDK version 21.11.2.
Documentation updated to use DPDK version 21.11.2.
NEWS entry added for DPDK 21.11.2 validation.
Updated FAQ to recommend use of DPDK version 21.11.2.
For your information, we just discovered
On 7/11/22 23:06, David Marchand wrote:
On Wed, Jul 6, 2022 at 11:00 PM David Marchand
wrote:
On Fri, Jul 1, 2022 at 5:58 AM Mike Pattrick wrote:
From: Flavio Leitner
Now that there is a segmentation in software as a fall back in
case a netdev doesn't support TCP segmentation
On 7/11/22 23:02, David Marchand wrote:
Hello,
Maxime, Ilya, your opinion will probably help.
On Wed, Jul 6, 2022 at 10:59 PM David Marchand
wrote:
@@ -5088,19 +5117,22 @@ netdev_dpdk_vhost_client_reconfigure(struct netdev
*netdev)
goto unlock;
}
+
(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
| 6 +++---
lib/netdev-linux.c | 14 +++---
lib/netdev.c | 16 +++-
5 files changed, 26 insertions(+), 28 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https
-linux.c | 2 +-
lib/netdev.c | 4 ++--
3 files changed, 4 insertions(+), 4 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
++---
lib/netdev-dummy.c | 8 +-
lib/netdev-linux.c | 56 ++---
4 files changed, 457 insertions(+), 456 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo
++--
lib/ipf.c | 6 +++---
lib/netdev-dpdk.c | 20 ++--
lib/netdev-linux.c | 24
lib/netdev.c | 14 +++---
6 files changed, 50 insertions(+), 50 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
| 20 ++--
lib/netdev-linux.c| 10 +-
lib/netdev-provider.h | 10 +-
lib/netdev.c | 8
4 files changed, 24 insertions(+), 24 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
___
dev
| 16 +-
lib/netdev-linux.c| 8 ++---
lib/netdev-provider.h | 8 ++---
lib/netdev.c | 6 ++--
5 files changed, 55 insertions(+), 55 deletions(-)
Acked-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d
Sunil Pai G
---
tests/system-dpdk.at | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
Thanks for taking care of the adaptation:
Acked-by: Maxime Coquelin
Maxime
___
dev mailing list
d...@openvswitch
On 4/7/22 17:01, Ilya Maximets wrote:
On 4/7/22 16:42, Van Haaren, Harry wrote:
-Original Message-
From: Ilya Maximets
Sent: Thursday, April 7, 2022 3:40 PM
To: Maxime Coquelin ; Van Haaren, Harry
; Morten Brørup ;
Richardson, Bruce
Cc: i.maxim...@ovn.org; Pai G, Sunil ; Stokes, Ian
0:02 AM
To: Morten Brørup ; Richardson, Bruce
Cc: Maxime Coquelin ; Pai G, Sunil
; Stokes, Ian ; Hu, Jiayu
; Ferriter, Cian ; Ilya Maximets
; ovs-dev@openvswitch.org; d...@dpdk.org; Mcnamara,
John ; O'Driscoll, Tim ;
Finn, Emma
Subject: RE: OVS DPDK DMA-Dev library/Design Discussion
-Original
On 3/31/22 17:21, David Marchand wrote:
On Thu, Mar 31, 2022 at 4:41 PM Eelco Chaudron wrote:
The dp_netdev_get_pmd() function is using only the hash of the core_id
to get the pmd structure. So in case of hash collisions, the wrong pmd
is returned.
This patch is fixing this by checking
On 3/30/22 04:02, Hu, Jiayu wrote:
-Original Message-
From: Ilya Maximets
Sent: Wednesday, March 30, 2022 1:45 AM
To: Morten Brørup ; Richardson, Bruce
Cc: i.maxim...@ovn.org; Maxime Coquelin ;
Van Haaren, Harry ; Pai G, Sunil
; Stokes, Ian ; Hu, Jiayu
; Ferriter, Cian ; ovs-
d
Hi Morten,
On 3/29/22 16:44, Morten Brørup wrote:
From: Van Haaren, Harry [mailto:harry.van.haa...@intel.com]
Sent: Tuesday, 29 March 2022 15.02
From: Morten Brørup
Sent: Tuesday, March 29, 2022 1:51 PM
Having thought more about it, I think that a completely different architectural
Hi Sunil,
On 2/1/22 15:23, Pai G, Sunil wrote:
Hi ,
This version of the patch seems to have negative impact on performance for
burst traffic profile[1].
Benefits seen with the previous version (v2) was up to ~1.6x for 1568 byte
packets compared to ~1.2x seen with the current design (v3) as
etdev: Introduce hash-based Tx packet steering
mode.")
Reported-by: Kevin Traynor
Reported-by: Ilya Maximets
Acked-by: Kevin Traynor
Signed-off-by: Maxime Coquelin
---
Documentation/topics/userspace-tx-steering.rst | 13 +++--
1 file changed, 7 insertions(+), 6 deletions(-)
This patch documents PMD's other_config:tx-steering option.
Acked-by: Kevin Traynor
Signed-off-by: Maxime Coquelin
---
vswitchd/vswitch.xml | 23 +++
1 file changed, 23 insertions(+)
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index 064e0facf..0c6632617 100644
hread mode documentation (Ilya)
Maxime Coquelin (2):
vswitchd.xml: Add missing tx-steering PMD option.
Documentation: Fix userspace Tx steering section.
.../topics/userspace-tx-steering.rst | 13 ++-
vswitchd/vswitch.xml | 23 +++
2 files c
On 1/20/22 14:20, Kevin Traynor wrote:
On 20/01/2022 12:16, Ilya Maximets wrote:
On 1/18/22 11:51, Kevin Traynor wrote:
On 17/01/2022 22:25, Maxime Coquelin wrote:
This patch documents PMD's other_config:tx-steering option.
Signed-off-by: Maxime Coquelin
---
vswitchd/vswitch.xml | 22
Hi, Ilya, Kevin,
On 1/20/22 13:16, Ilya Maximets wrote:
On 1/18/22 11:51, Kevin Traynor wrote:
On 17/01/2022 22:25, Maxime Coquelin wrote:
This patch documents PMD's other_config:tx-steering option.
Signed-off-by: Maxime Coquelin
---
vswitchd/vswitch.xml | 22 ++
1
On 1/17/22 18:44, Ilya Maximets wrote:
On 1/5/22 09:19, Maxime Coquelin wrote:
This series introduces a new hash-based Tx packets steering
mode alognside existing XPS and static modes. The goal is
to provide a mode where all the transmit queues are used,
whatever the number of PMD threads
This patch documents PMD's other_config:tx-steering option.
Signed-off-by: Maxime Coquelin
---
vswitchd/vswitch.xml | 22 ++
1 file changed, 22 insertions(+)
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index 026b5e2ca..ef7f8f2c8 100644
--- a/vswitchd
11.0.0.1,icmp_type=8,icmp_code=0
actions=" > flow.txt
+for i in `seq 1 1022`; do printf
"set_field:0x399->reg13,set_field:0x$i->reg15,resubmit(,39),"; done >> flow.txt
+echo "resubmit(,39)" >> flow.txt
Minor comment, but maybe using printf everywhere would
Hi Ilya,
On 1/5/22 09:19, Maxime Coquelin wrote:
This patch adds a new hash Tx steering mode that
distributes the traffic on all the Tx queues, whatever the
number of PMD threads. It would be useful for guests
expecting traffic to be distributed on all the vCPUs.
The idea here is to re-use
in dp_netdev_pmd_flush_output_on_port(),
only for this hash mode.
Signed-off-by: Maxime Coquelin
Reviewed-by: David Marchand
---
Documentation/automake.mk | 1 +
Documentation/topics/index.rst| 1 +
.../topics/userspace-tx-steering.rst | 73 ++
lib/dpif-netdev.c
.
Signed-off-by: Maxime Coquelin
Reviewed-by: David Marchand
---
tests/dpif-netdev.at | 67
1 file changed, 67 insertions(+)
diff --git a/tests/dpif-netdev.at b/tests/dpif-netdev.at
index 53eee185a..b7f1c3f8b 100644
--- a/tests/dpif-netdev.at
+++ b
t;
per-queue custom statistics for Vhost-user ports.
Suggested-by David Marchand
Signed-off-by: Maxime Coquelin
Reviewed-by: David Marchand
---
lib/netdev-dpdk.c | 147 +++---
1 file changed, 138 insertions(+), 9 deletions(-)
diff --git a/lib/netdev-dp
A boolean is currently used to differenciate between the
static and XPS Tx queue modes.
Since we are going to introduce a new steering mode, replace
this boolean with an enum.
This patch does not introduce functionnal changes.
Signed-off-by: Maxime Coquelin
Reviewed-by: David Marchand
This patch adds Rx and Tx per-queue statistics. It will be
used to test hash-based Tx packet steering. Only "bytes",
and "packets" per-queue custom statistics are added, as
there are no global "errors" counters in netdev-dummy.
Signed-off-by: Maxime Coquelin
Reviewe
hash|default (Ilya)
- Fix use-after-free of tx_port->txq_pkts (Kevin)
- Fix out-of-bound writes on Vhost-user rxq stats (David)
- Implement Tx steering test in dpif-netdev without fuzzing (Ilya)
- Remove Vhost-user tx-steering tests (Ilya)
- Fix various indents & other cosmetic issues (David)
- Mov
Hi David,
On 1/3/22 19:42, David Marchand wrote:
On Fri, Dec 17, 2021 at 4:10 PM Maxime Coquelin
wrote:
Hash-based Tx steering feature will enable steering Tx
packets on transmit queues based on their hashes. In order
to test the feature, it is needed to be able to get the
per-queue
7 files changed, 61 insertions(+), 7 deletions(-)
The path looks good to me. Multiprocess was not supported before with
OVS, so disabling it is not a problem.
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
ht
.
Signed-off-by: Maxime Coquelin
---
tests/dpif-netdev.at | 67
1 file changed, 67 insertions(+)
diff --git a/tests/dpif-netdev.at b/tests/dpif-netdev.at
index 53eee185a..b7f1c3f8b 100644
--- a/tests/dpif-netdev.at
+++ b/tests/dpif-netdev.at
@@ -635,3
in dp_netdev_pmd_flush_output_on_port(),
only for this hash mode.
Signed-off-by: Maxime Coquelin
---
Documentation/automake.mk | 1 +
Documentation/topics/index.rst| 1 +
.../topics/userspace-tx-steering.rst | 75 ++
lib/dpif-netdev.c
A boolean is currently used to differenciate between the
static and XPS Tx queue modes.
Since we are going to introduce a new steering mode, replace
this boolean with an enum.
This patch does not introduce functionnal changes.
Signed-off-by: Maxime Coquelin
---
lib/dpif-netdev.c | 25
This patch adds Rx and Tx per-queue statistics. It will be
used to test hash-based Tx packet steering. Only "bytes",
and "packets" per-queue custom statistics are added, as
there are no global "errors" counters in netdev-dummy.
Signed-off-by: Maxime Coquelin
t;
per-queue custom statistics for Vhost-user ports.
Suggested-by David Marchand
Signed-off-by: Maxime Coquelin
---
lib/netdev-dpdk.c | 149 +++---
1 file changed, 140 insertions(+), 9 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
ind
dpif-netdev without fuzzing (Ilya)
- Remove Vhost-user tx-steering tests (Ilya)
- Fix various indents & other cosmetic issues (David)
- Move documenation out of dpdk/
Maxime Coquelin (5):
netdev-dpdk: Introduce per rxq/txq Vhost-user statistics.
netdev-dummy: Introduce per rxq/txq statistics.
On 12/17/21 14:20, Maxime Coquelin wrote:
On 12/17/21 13:52, Ilya Maximets wrote:
On 12/17/21 13:23, Ilya Maximets wrote:
On 12/16/21 21:04, Maxime Coquelin wrote:
Hi Kevin,
On 12/16/21 19:00, Kevin Traynor wrote:
On 15/12/2021 16:26, Maxime Coquelin wrote:
This patch adds a new hash
On 12/17/21 14:03, Kevin Traynor wrote:
On 17/12/2021 12:23, Ilya Maximets wrote:
On 12/16/21 21:04, Maxime Coquelin wrote:
Hi Kevin,
On 12/16/21 19:00, Kevin Traynor wrote:
On 15/12/2021 16:26, Maxime Coquelin wrote:
This patch adds a new hash Tx steering mode that
distributes
On 12/17/21 13:52, Ilya Maximets wrote:
On 12/17/21 13:23, Ilya Maximets wrote:
On 12/16/21 21:04, Maxime Coquelin wrote:
Hi Kevin,
On 12/16/21 19:00, Kevin Traynor wrote:
On 15/12/2021 16:26, Maxime Coquelin wrote:
This patch adds a new hash Tx steering mode that
distributes the traffic
On 12/17/21 10:45, Kevin Traynor wrote:
On 16/12/2021 20:04, Maxime Coquelin wrote:
Hi Kevin,
On 12/16/21 19:00, Kevin Traynor wrote:
On 15/12/2021 16:26, Maxime Coquelin wrote:
This patch adds a new hash Tx steering mode that
distributes the traffic on all the Tx queues, whatever
Hi Kevin,
On 12/16/21 19:00, Kevin Traynor wrote:
On 15/12/2021 16:26, Maxime Coquelin wrote:
This patch adds a new hash Tx steering mode that
distributes the traffic on all the Tx queues, whatever the
number of PMD threads. It would be useful for guests
expecting traffic to be distributed
On 12/16/21 18:58, Kevin Traynor wrote:
On 15/12/2021 16:26, Maxime Coquelin wrote:
Hash-based Tx steering feature will enable steering Tx
packets on transmit queues based on their hashes. In order
to test the feature, it is needed to be able to get the
per-queue statistics for Vhost-user
in dp_netdev_pmd_flush_output_on_port(),
only for this hash mode.
Signed-off-by: Maxime Coquelin
---
Documentation/automake.mk | 1 +
Documentation/topics/index.rst| 1 +
.../topics/userspace-tx-steering.rst | 78 +++
lib/dpif-netdev.c
A boolean is currently used to differenciate between the
static and XPS Tx queue modes.
Since we are going to introduce a new steering mode, replace
this boolean with an enum.
This patch does not introduce functionnal changes.
Signed-off-by: Maxime Coquelin
---
lib/dpif-netdev.c | 25
.
Signed-off-by: Maxime Coquelin
---
tests/dpif-netdev.at | 44
1 file changed, 44 insertions(+)
diff --git a/tests/dpif-netdev.at b/tests/dpif-netdev.at
index 53eee185a..482e5029f 100644
--- a/tests/dpif-netdev.at
+++ b/tests/dpif-netdev.at
@@ -635,3
This patch adds Rx and Tx per-queue statistics. It will be
used to test hash-based Tx packet steering. Only "bytes",
and "packets" per-queue custom statistics are added, as
there are no global "errors" counters in netdev-dummy.
Signed-off-by: Maxime Coquelin
t;
per-queue custom statistics for Vhost-user ports.
Suggested-by David Marchand
Signed-off-by: Maxime Coquelin
---
lib/netdev-dpdk.c | 149 +++---
1 file changed, 140 insertions(+), 9 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
ind
tests (Ilya)
- Fix various indents & other cosmetic issues (David)
- Move documenation out of dpdk/
Maxime Coquelin (5):
netdev-dpdk: Introduce per rxq/txq Vhost-user statistics.
netdev-dummy: Introduce per rxq/txq statistics.
dpif-netdev: Introduce Tx queue mode.
dpif-netdev: Intro
On 12/15/21 15:34, Maxime Coquelin wrote:
This patch adds Rx and Tx per-queue statistics. It will be
used to test hash-based Tx packet steering. Only "bytes",
and "packets" per-queue custom statistics are added, as
there are no global "errors" counters in netdev
.
Signed-off-by: Maxime Coquelin
---
tests/dpif-netdev.at | 44
1 file changed, 44 insertions(+)
diff --git a/tests/dpif-netdev.at b/tests/dpif-netdev.at
index 53eee185a..482e5029f 100644
--- a/tests/dpif-netdev.at
+++ b/tests/dpif-netdev.at
@@ -635,3
t;
per-queue custom statistics for Vhost-user ports.
Suggested-by David Marchand
Signed-off-by: Maxime Coquelin
---
lib/netdev-dpdk.c | 149 +++---
1 file changed, 140 insertions(+), 9 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
ind
in dp_netdev_pmd_flush_output_on_port(),
only for this hash mode.
Signed-off-by: Maxime Coquelin
---
Documentation/automake.mk | 1 +
Documentation/topics/index.rst| 1 +
.../topics/userspace-tx-steering.rst | 74 +++
lib/dpif-netdev.c
A boolean is currently used to differenciate between the
static and XPS Tx queue modes.
Since we are going to introduce a new steering mode, replace
this boolean with an enum.
This patch does not introduce functionnal changes.
Signed-off-by: Maxime Coquelin
---
lib/dpif-netdev.c | 25
This patch adds Rx and Tx per-queue statistics. It will be
used to test hash-based Tx packet steering. Only "bytes",
and "packets" per-queue custom statistics are added, as
there are no global "errors" counters in netdev-dummy.
Signed-off-by: Maxime Coquelin
ous indents & other cosmetic issues (David)
- Move documenation out of dpdk/
Maxime Coquelin (5):
netdev-dpdk: Introduce per rxq/txq Vhost-user statistics.
netdev-dummy: Introduce per rxq/txq statistics.
dpif-netdev: Introduce Tx queue mode.
dpif-netdev: Introduce hash-based Tx packet steeri
-a >$OVS_RUNDIR/testpmd-dpdkvhostuserclient0.log
2>&1 &
OVS_WAIT_UNTIL([grep "virtio is now ready for processing" ovs-vswitchd.log])
+OVS_WAIT_UNTIL([ip link show dev tap0 | grep -qw LOWER_UP])
dnl Move the tap devices to the namespaces
AT_CHECK([ps aux | grep
On 12/9/21 17:29, Ilya Maximets wrote:
On 12/9/21 16:04, David Marchand wrote:
On Thu, Dec 9, 2021 at 3:51 PM Maxime Coquelin
wrote:
On 12/7/21 21:37, David Marchand wrote:
+struct netdev_dpdk_vhost_q_stats {
+uint64_t bytes;
+uint64_t packets;
+uint64_t errors
On 12/7/21 21:37, David Marchand wrote:
+struct netdev_dpdk_vhost_q_stats {
+uint64_t bytes;
+uint64_t packets;
+uint64_t errors;
+};
+
/* Custom software stats for dpdk ports */
struct netdev_dpdk_sw_stats {
/* No. of retries when unable to transmit. */
@@ -206,6
On 12/8/21 21:25, David Marchand wrote:
On Wed, Nov 24, 2021 at 10:24 PM Maxime Coquelin
wrote:
@@ -2385,6 +2408,7 @@ netdev_dpdk_vhost_update_rx_counters(struct netdev_dpdk
*dev,
}
stats->rx_bytes += packet_size;
+q_stats->bytes += packe
Hi Ilya,
On 12/8/21 00:23, Ilya Maximets wrote:
On 11/24/21 22:23, Maxime Coquelin wrote:
This series introduces a new HXPS Tx mode alognside existing
XPS and static modes. The goal is to provide a mode where all
the transmit queues are used, whatever the number of PMD
threads. This may
Hi David,
On 12/7/21 21:37, David Marchand wrote:
Hey Maxime,
On Wed, Nov 24, 2021 at 10:24 PM Maxime Coquelin
wrote:
HXPS feature will enable steering Tx packets on transmist
transmit*
queues based on their hashes. In order to test the feature,
"their hashes" is ambi
| 115 ---
11 files changed, 98 insertions(+), 86 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
---
Documentation/topics/dpdk/bridge.rst | 22 +--
tests/dpif-netdev.at | 167 +++
tests/system-dpdk.at | 194 ---
3 files changed, 178 insertions(+), 205 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
space after DPDK_PCI_ADDR?
Other than that, it looks good to me, feel free to add my:
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
100755 tests/genpkts.py
delete mode 100755 tests/mfex_fuzzy.py
delete mode 100644 tests/pcap/mfex_test.pcap
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
++---
2 files changed, 31 insertions(+), 28 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
Hi,
On 11/24/21 22:23, Maxime Coquelin wrote:
This series introduces a new HXPS Tx mode alognside existing
XPS and static modes. The goal is to provide a mode where all
the transmit queues are used, whatever the number of PMD
threads. This may be used with Vhost-user ports, where the
guest
Instead of waiting 10 seconds for testpmd to start, this
patch makes use of OVS_WAIT_UNTIL() macro to wait for
the Virtio device readiness notification in ovs-vswitchd
logs.
Signed-off-by: Maxime Coquelin
---
tests/system-dpdk.at | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff
packets injected on the dummy port are
are spread on the two vhsot-user queues.
Signed-off-by: Maxime Coquelin
---
tests/system-dpdk.at | 135 +++
1 file changed, 135 insertions(+)
diff --git a/tests/system-dpdk.at b/tests/system-dpdk.at
index 894e01316
(),
only for this HXPS mode.
Signed-off-by: Maxime Coquelin
---
Documentation/automake.mk | 1 +
Documentation/topics/dpdk/hxps.rst | 51 +++
Documentation/topics/dpdk/index.rst | 1 +
lib/dpif-netdev.c | 78 +++--
4 files
A boolean is currently used to differenciate between the
static and XPS Tx queue modes.
Since we are going to introduce a new RSS mode, replace
this boolean with an enum.
This patch does not introduce functionnal changes.
Signed-off-by: Maxime Coquelin
---
lib/dpif-netdev.c | 25
tatistics for Vhost-user ports.
Suggested-by David Marchand
Signed-off-by: Maxime Coquelin
---
lib/netdev-dpdk.c | 143 +++---
1 file changed, 135 insertions(+), 8 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index ca92c947a..e80d5b4ab
.
The series also introduces two DPDK tests for Vhost-user
multiqueue, with and without HXPS enabled.
Maxime Coquelin (4):
netdev-dpdk: Introduce per rxq/txq Vhost-user statistics
dpif-netdev: Introduce Tx queue mode
dpif-netdev: Add HXPS Tx queue mode
system-dpdk: Add tests for HXPS
On 11/19/21 13:08, Kevin Traynor wrote:
On 19/11/2021 10:31, David Marchand wrote:
On Fri, Nov 19, 2021 at 10:40 AM Maxime Coquelin
wrote:
Hi Kevin, David,
On 11/18/21 16:31, Kevin Traynor wrote:
On 15/10/2021 16:04, David Marchand wrote:
When troubleshooting multiqueue setups, having
1 - 100 of 195 matches
Mail list logo