Any suggestion is highly appreciated.
On Tue, Jun 2, 2020 at 3:40 PM Gowrishankar Muthukrishnan <
gmuth...@redhat.com> wrote:
> This patchset enables new feature "user defined events and notification".
> In short, by this feature, one can enable asynchronous and less int
Moved commands to set hugepages in every numa node into a macro
for tests to reuse.
Signed-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk.at | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/tests/system-dpdk.at b/tests/system-dpdk.at
index ab10f44
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v8:
- split patch into logical units of changes (as this series)
---
tests/system-dpdk.at | 280
OVS_DPDK_START could be split into two macros to start
ovsdb-server and ovs-vswitchd individually so that, user can
use ovs-vsctl to configure ovs tables for test requirements.
Signed-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk-macros.at | 17 -
tests/system-dpdk.at
This patch series enables system-dpdk test to validate TSO
negotiation.
Gowrishankar Muthukrishnan (4):
system-dpdk: filter non-available hugepages in ovs_vswitchd_stop
system-dpdk: split ovs_dpdk_start into db and daemon ops
system-dpdk: macro to set hugepages in numa nodes
system-dpdk
While checking log for stopping ovs-vswitchd, we can ignore
message about non-available hugepages as that is specific to
hugepages configuration in system and not required in the scope
log check.
Signed-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk.at | 5 +
1 file changed, 5
Thu, Jul 09, 2020 at 11:34:33AM +0530, Gowrishankar Muthukrishnan wrote:
> > This patch adds minimal check for userspace-tso in system-dpdk
> > tests, starting with verification on virtio negotiation.
> >
> > Signed-off-by: Gowrishankar Muthukrishnan
> > ---
> > v7:
Hi,
Any more ack(s) required for this patch to be merged ?.
Regards,
Gowrishankar
On Thu, May 14, 2020 at 8:06 PM William Tu wrote:
> On Mon, Apr 20, 2020 at 07:13:42PM +0530, Gowrishankar Muthukrishnan wrote:
> > Whenever the number of flows in the datapath crosses above
> >
Thanks Flavio, I did not realize my workaround patch in another commit
making the tests pass always :) .
Added corrections in ovs_stop accordingly now.
Thank you again.
--
Gowrishankar M
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v7:
- fixed ovs stop log check for added tests.
- test #8 is skipped using AT_ macro.
Sample test output
On Tue, Jul 7, 2020 at 7:19 PM Flavio Leitner wrote:
>
> Hi Gowrishankar,
>
> Thanks for following up with this patch.
>
> However, I think we need to reconsider the tests. It doesn't seem
> useful to have negotiation tests that will always fail because of
> a code that is marked as deprecated (v
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v6:
- refactored tests to drop vhostuser mode for ovs (depreciated feature)
and add tests to also validate tso bits from
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v5:
- fix killing testpmd with some sleep to complete (flavio).
---
tests/system-dpdk-macros.at | 17 +++-
tests/system
5|netdev_dpdk|WARN|dpdkvhostuser ports are
> considered deprecated; please migrate to dpdkvhostuserclient ports.
> +2020-07-03T00:11:35Z|00019|dpdk|WARN|EAL: Invalid NUMA socket, default
> to 0
> +2020-07-03T00:11:35Z|00021|dpdk|WARN|EAL: Invalid NUMA socket, default
> to 0
>
> M
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v4:
- configure offload settings in testpmd
---
tests/system-dpdk-macros.at | 17 +++-
tests/system-dpdk.at| 228
Hi,
Just curious to probe you again if you had a chance to check this feature.
Thanks in advance.
On Tue, Jun 2, 2020 at 3:40 PM Gowrishankar Muthukrishnan <
gmuth...@redhat.com> wrote:
> This patchset enables new feature "user defined events and notification".
> In short
Documentation notes on event library and its usage is included
in this patch.
Signed-off-by: Gowrishankar Muthukrishnan
---
Documentation/automake.mk| 1 +
Documentation/topics/index.rst | 1 +
Documentation/topics/user-defined-events.rst | 306
Add iperf3 performance testsuite for checking events.
Signed-off-by: Gowrishankar Muthukrishnan
---
This is not complete test suite as I would like to get first set of
feedbacks on implementation and write unit tests accordingly. This
test is to check if event API works expectedly while running
This patch injects timing some important functions as listed
below in datapath. This can be expanded but the list is short
for now.
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/dpif-netdev.c | 5 -
lib/netdev-dpdk.c | 3 +++
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a
This is sample tiny daemon which receives event notification
from the registered events and displays the message info in the
terminal.
Signed-off-by: Gowrishankar Muthukrishnan
---
utilities/automake.mk | 6 +-
utilities/ovs-testeventd.c | 156
This patch enables monitoring coverage counters through event API.
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/coverage.c | 61 ++
lib/coverage.h | 4
lib/timeval.c | 3 +++
3 files changed, 68 insertions(+)
diff --git a/lib
Timer notifications in event library require handling stopwatch
across one or more events, so share few required functions.
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/stopwatch.c | 20 +++-
lib/stopwatch.h | 9 +
2 files changed, 28 insertions(+), 1 deletion
is central library to define event based on
the input file from user (json format).
2. dedicated thread to monitor registered events and issue
notification.
3. event macros to hook at counters and functions for evaluating
events by above thread.
Signed-off-by: Gowrishankar Muthukrishnan
ved, I can work on enhancing this for optimations
whereever needed.
Gowrishankar Muthukrishnan (7):
event: add api to manage user defined events
stopwatch: expose required api for event library
dpif: hook timer event api in datapath functions
coverage: support conditional notification f
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v3:
- Removed other improvement(cleanup, optimization in general) for now
and sending only tso tests. Other improvements will
21, 2020 at 7:54 PM Ilya Maximets wrote:
> On 5/16/20 7:53 AM, Gowrishankar Muthukrishnan wrote:
> > After dpdk tests completes, cleaning up hugepage map files
> > created by tests is helpful to release used memory into
> > hugepage memory allocator.
> >
>
>
> > dnl Find number of sockets
> > AT_CHECK([lscpu], [], [stdout])
> > -AT_CHECK([cat stdout | grep "NUMA node(s)" | awk '{c=1; while
> (c++<$(3)) {printf "512,"}; print "512"}' > NUMA_NODE])
> > +AT_CHECK([cat stdout | grep "NUMA node(s)" | awk '{c=1; while
> (c++<$(3)) {printf "448,"}; print
>
> +dnl
> --
> +dnl validate tso negotiation (without userspace-tso)
> +AT_SETUP([OVS-DPDK - validate tso negotiation (with userspace-tso)])
>
Minor Typo here which I realized after sending the patch. AT_SETUP should
log "wi
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
v2:
- removed tso check in existing tests (added in v1)
- improved validations in host and guest end
- added tests for non-tso
-off-by: Gowrishankar Muthukrishnan
---
v2:
- fixed hunk to reflect intended changes.
---
tests/system-dpdk-macros.at | 19 +++
tests/system-dpdk.at| 10 ++
2 files changed, 25 insertions(+), 4 deletions(-)
diff --git a/tests/system-dpdk-macros.at b/tests/system
After dpdk tests completes, cleaning up hugepage map files
created by tests is helpful to release used memory into
hugepage memory allocator.
Signed-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk-macros.at | 13 +
tests/system-dpdk.at| 7 +++
2 files changed
This series adds test cases for TSO feature in OVS-DPDK,
along with few improvements and cleanup in testing
mechanism. Please find comments section in patch 3 for
the testing info.
Gowrishankar Muthukrishnan (3):
system-dpdk: cleanup stale hugepage files after tests
system-dpdk: use optimum
On Thu, May 14, 2020 at 8:06 PM William Tu wrote:
> On Mon, Apr 20, 2020 at 07:13:42PM +0530, Gowrishankar Muthukrishnan wrote:
> > Whenever the number of flows in the datapath crosses above
> > the flow limit set/autoconfigured, it is helpful to report
> > this event thr
I see this change fixing a commit b4c632526b pushed a few years back, but
still there is room for instrumentation like below. Could this patch be
reviewed ?.
Thanks,
Gowrishankar
On Mon, Apr 20, 2020 at 7:13 PM Gowrishankar Muthukrishnan <
gmuth...@redhat.com> wrote:
> Whenever the
After dpdk tests completes, cleaning up hugepage map files
created by tests is helpful to release used memory into
hugepage memory allocator.
Signed-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk-macros.at | 13 +
tests/system-dpdk.at| 7 +++
2 files changed
This patch adds minimal check for userspace-tso in system-dpdk
tests, starting with verification on virtio negotiation.
Signed-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk-macros.at | 16 +--
tests/system-dpdk.at| 102
2
-off-by: Gowrishankar Muthukrishnan
---
tests/system-dpdk-macros.at | 20 +++-
tests/system-dpdk.at| 10 ++
2 files changed, 25 insertions(+), 5 deletions(-)
diff --git a/tests/system-dpdk-macros.at b/tests/system-dpdk-macros.at
index b3a46d1..943cbdf 100644
--- a
This series adds test cases for TSO feature in OVS-DPDK,
along with few improvements and cleanup in testing
mechanism.
Gowrishankar Muthukrishnan (3):
system-dpdk: cleanup stale hugepage files after tests
system-dpdk: use optimum hugepages for dpdk tests
system-dpdk: add negotiation check
vswitch log when
a new flow can not be inserted in upcall processing in which
case ovs writes a warning, otherwise an auto correction
made by ovs to flush old flows without any intimation at all.
Signed-off-by: Gowrishankar Muthukrishnan
---
ofproto/ofproto-dpif-upcall.c | 6 ++
1 file changed
and numa_id.
Reported-at: https://bugzilla.redhat.com/1728616
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/dpif-netdev.c | 4
1 file changed, 4 insertions(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 4546b55e8..aeae66aea 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif
Could this patch be reviewed please.
Thanks,
Gowrishankar
On Tue, Oct 22, 2019 at 3:06 PM Gowrishankar Muthukrishnan <
gmuth...@redhat.com> wrote:
> There is no log about isolated rxq assignment in a pmd today, which
> sometimes could be useful to trace rxq/pmd pinning, when debu
, we need to handle such warning at the calling
place itself to debug quickly.
Thanks,
Gowrishankar
On Fri, Oct 25, 2019 at 2:47 AM Ben Pfaff wrote:
> On Mon, Oct 21, 2019 at 07:34:36PM +0530, Gowrishankar Muthukrishnan wrote:
> > It might be difficult to trace whether LACP PDU
.
Reported-at: https://bugzilla.redhat.com/1728616
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/dpif-netdev.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 4546b55e8..4d7c36787 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
/1728616
Signed-off-by: Gowrishankar Muthukrishnan
---
lib/dpif-netdev.c | 4
1 file changed, 4 insertions(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 4546b55e8..7a5afea00 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -4573,6 +4573,10 @@ rxq_scheduling(struct
It is helpful in reporting main thread that is yet to enable bond slave,
but link state was brought up by lacp thread and capture this desync
between ovs threads for debugging.
Fixes: a8448cb170 ("lacp: Avoid packet drop on LACP bond after link up")
Signed-off-by: Gowrishankar Mut
ats.
Adding a warning when there is tx failure could help user
debug at the root of this problem.
Signed-off-by: Gowrishankar Muthukrishnan
---
ofproto/ofproto-dpif.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c
in
Today (*), when a packet journey in the data path is disrupted and leading
towards its drop, we have OVS counters to auto-detect it and show at the
request of user space commands. Some category of drops are related to the
interfaces that can be queried from OVS DB table for that interface [2],
whil
Ignore below comment as I slightly overlooked it. invalid port is right
drop reason.
Thanks,
Gowrishankar
On Mon, Jan 7, 2019 at 6:12 PM Gowrishankar Muthukrishnan <
gmuth...@redhat.com> wrote:
> Hi Anju,
>
> On Mon, Jan 7, 2019 at 3:30 PM Anju Thomas
> wrote:
>
>
Hi Anju,
On Mon, Jan 7, 2019 at 3:30 PM Anju Thomas wrote:
>
> switch ((enum ovs_action_attr)type) {
> case OVS_ACTION_ATTR_OUTPUT:
> @@ -6548,6 +6625,8 @@ dp_execute_cb(void *aux_, struct dp_packet_batch
> *packets_,
> dp_packet_batch_add(&p->output_pkts, packet);
>
Hi,
> VLOG_DEFINE_THIS_MODULE(dpif_netdev);
>
>
Minor suggestions on naming variable/macros follow as below, as coding
itself is documentation IMO :).
+/* Auto Load Balancing Defaults */
> +#define ACCEPT_IMPROVE_DEFAULT (25)
>
Instead, how about LB_ACCEPTABLE_IMPROVEMENT ?
prefixing glo
Hi Anju,
Thank you reviewing my suggestions.
On 10/03/2018 04:54 PM, Anju Thomas wrote:
> Hi Gowrishankar,
>
> Thanks for your comments & suggestions
>
> Replies inline ..
>
> Regards & Thanks
> Anju
>
>
> -Original Message-
> From: Gowrishank
Hi Keshav,
Thank you for the patches. They certainly help us debugging packet drops
more meaningfully. Following your changes, some of the places where
additional reasons for drop could also be added as I hope helpful. Could
you please check below items as well.
__netdev_dpdk_vhost_send() AND netd
52 matches
Mail list logo