Here is the the testing I just did:
1. Add log in function netdev_dpdk_policer_pkt_handle
static inline
bool
netdev_dpdk_policer_pkt_handle(struct rte_meter_srtcm
*meter,
struct rte_mbuf *pkt, uint64_t
time)
{
uint32_t pkt_len = rte_pktmbuf_pkt_len(pkt) -
> On Aug 30, 2017, at 3:57 PM, Russell Bryant wrote:
>
> On Wed, Aug 30, 2017 at 6:51 PM, Justin Pettit wrote:
>> This reverts commit 8294d78ab124ec6f4b51c621e9baf7f29b287bb9.
>>
>> This patch was mistakenly backported to branch-2.7. The change breaks
>> the
On Wed, Aug 30, 2017 at 2:09 PM, Flavio Leitner wrote:
> On Mon, 14 Aug 2017 16:18:14 -0400
> Aaron Conole wrote:
>
>> The upgrade from older Open vSwitch versions on RHEL will try, as much as
>> possible, to preserve the system. This means no new users or
Only curNbl in the forwarding context needs to be updated with
fragmented NBL.
Signed-off-by: Anand Kumar
---
datapath-windows/ovsext/Actions.c | 9 +
1 file changed, 1 insertion(+), 8 deletions(-)
diff --git a/datapath-windows/ovsext/Actions.c
On Wed, Aug 30, 2017 at 6:51 PM, Justin Pettit wrote:
> This reverts commit 8294d78ab124ec6f4b51c621e9baf7f29b287bb9.
>
> This patch was mistakenly backported to branch-2.7. The change breaks
> the OpenFlow API, which is not appropriate for a maintenance release.
>
>
This reverts commit 8294d78ab124ec6f4b51c621e9baf7f29b287bb9.
This patch was mistakenly backported to branch-2.7. The change breaks
the OpenFlow API, which is not appropriate for a maintenance release.
Signed-off-by: Justin Pettit
Reported-by: Russell Bryant
INVALID packets doesn't map to any ACL, but sometimes we need to
know when these packets are detected and dropped. This patch supports
logging for such packets and provides a global configuration in NB
to enable/disable this behavior. Default behavior is not logging,
so that it is backward
Signed-off-by: Han Zhou
---
tests/ovn.at | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/ovn.at b/tests/ovn.at
index fb9fc73..41ccc5a 100644
--- a/tests/ovn.at
+++ b/tests/ovn.at
@@ -5828,13 +5828,13 @@ packet="inport==\"lp1\" && eth.src==$lp1_mac
Hi,
Hope all is well!
Would you like to reach the Newly Updated Opt-In Email List which
includes complete contact details and tele-verified email addresses.
Every lead comes with Name, Company, Title, Website, Physical Address,
Employee Size, Revenue Size, Industry, SIC, Phone Number
On Wed, Aug 30, 2017 at 11:19:08PM +0300, Michael S. Tsirkin wrote:
> On Wed, Aug 30, 2017 at 04:25:10PM -0300, Eduardo Habkost wrote:
> > On Wed, Aug 30, 2017 at 05:23:39PM +0300, Michael S. Tsirkin wrote:
> > > On Wed, Aug 30, 2017 at 10:17:27AM -0300, Eduardo Habkost wrote:
> > > > I'm CCing
On Wed, Aug 30, 2017 at 9:33 AM, Ben Pfaff wrote:
> The current implementation of ovsdb-server caches only non-conditional
> monitors, that is, monitors for every table row, not those that monitor
> only rows that match some condition. To figure out which monitors are
>
Subject: [ovs-dev] [PATCH v2] datapath-windows: Add an upper limit to
conntrack entries
The current implementation lacked an upper bound of number of entries in
the system. Set the size to ~2M (2^21) for the time being.
Signed-off-by: Sairam Venugopal
---
The current implementation lacked an upper bound of number of entries in
the system. Set the size to ~2M (2^21) for the time being.
Signed-off-by: Sairam Venugopal
---
datapath-windows/ovsext/Conntrack.c | 10 ++
datapath-windows/ovsext/Conntrack.h | 1 +
2 files
On Wed, Aug 30, 2017 at 9:33 AM, Ben Pfaff wrote:
> This removes n_true_cnd from struct ovsdb_monitor_session_condition.
> It was an "optimization" that is not part of any inner loop, but
> make the code harder to reason about.
>
> Signed-off-by: Ben Pfaff
Thanks for
On Wed, Aug 30, 2017 at 04:25:10PM -0300, Eduardo Habkost wrote:
> On Wed, Aug 30, 2017 at 05:23:39PM +0300, Michael S. Tsirkin wrote:
> > On Wed, Aug 30, 2017 at 10:17:27AM -0300, Eduardo Habkost wrote:
> > > I'm CCing libvir-list and qemu-devel because I would like to get
> > > feedback from
On 8/29/17, 7:25 PM, "Yuanhan Liu" wrote:
On Wed, Aug 30, 2017 at 01:32:10AM +, Darrell Ball wrote:
>
> > Though that being said, this patchset still has issues
unresolved. The
> > major issue is that maybe most NIC (for
On 8/30/17, 12:19 AM, "Finn Christensen" wrote:
Hi,
Just wanted to add one comment to this correspondence. See below.
Thanks,
Finn Christensen
-Original Message-
From:
On Wed, Aug 30, 2017 at 05:23:39PM +0300, Michael S. Tsirkin wrote:
> On Wed, Aug 30, 2017 at 10:17:27AM -0300, Eduardo Habkost wrote:
> > I'm CCing libvir-list and qemu-devel because I would like to get
> > feedback from libvirt and QEMU developers too.
> >
> > On Tue, Aug 08, 2017 at 10:49:21PM
> -Original Message-
> From: Hannes Frederic Sowa [mailto:han...@stressinduktion.org]
> Sent: Wednesday, August 30, 2017 4:16 PM
> To: Mooney, Sean K
> Cc: Yang, Yi Y ; d...@openvswitch.org;
> net...@vger.kernel.org; jb...@redhat.com;
On Mon, 14 Aug 2017 16:18:14 -0400
Aaron Conole wrote:
> The upgrade from older Open vSwitch versions on RHEL will try, as much as
> possible, to preserve the system. This means no new users or groups are
> created. As an effect, it's possible for the chown to fail, because
Thanks for the patch Kevin
On 8/30/17, 10:45 AM, "Kevin Traynor" wrote:
rxq_cycle_sort summed the latest cycles from each queue for sorting.
While each comparison was correct with the latest cycles, the cycles
could change between calls to rxq_cycle_sort. In
Thanks Kevin
It is nice of you to fix a pre-existing coding style issue as part of this
patch.
Darrell
On 8/30/17, 10:45 AM, "Kevin Traynor" wrote:
A couple of trivial fixes for a ternery operator placement
and pointer declaration.
Fixes: 655856ef39b9
Thanks Kevin
Naming is hard.
The name looks a bit more intuitive and matches closely with the description
previously added.
Darrell
On 8/30/17, 10:45 AM, "Kevin Traynor" wrote:
rxq_interval was added before there was other #defines
and code related to rxq
rxq_cycle_sort summed the latest cycles from each queue for sorting.
While each comparison was correct with the latest cycles, the cycles
could change between calls to rxq_cycle_sort. In order to use
consistent values through each call to rxq_cycle_sort, sum the cycles
prior to rxq_cycle_sort
A couple of trivial fixes for a ternery operator placement
and pointer declaration.
Fixes: 655856ef39b9 ("dpif-netdev: Change rxq_scheduling to use rxq processing
cycles.")
Fixes: a2ac666d5265 ("dpif-netdev: Change definitions of 'idle' & 'processing'
cycles")
Cc: ciara.lof...@intel.com
rxq_interval was added before there was other #defines
and code related to rxq intervals.
Rename to rxq_next_cycles_store in order to make it more intuitive.
Reported-by: Ilya Maximets
Signed-off-by: Kevin Traynor
---
lib/dpif-netdev.c | 8
On 08/28/2017 02:31 PM, Ilya Maximets wrote:
>> Up to his point rxqs are sorted by processing cycles they
>> consumed and assigned to pmds in a round robin manner.
>>
>> Ian pointed out that on wrap around the most loaded pmd will be
>> the next one to be assigned an additional rxq and that it
On 08/28/2017 02:29 PM, Ilya Maximets wrote:
>> Previously rxqs were assigned to pmds by round robin in
>> port/queue order.
>>
>> Now that we have the processing cycles used for existing rxqs,
>> use that information to try and produced a better balanced
>> distribution of rxqs across pmds. i.e.
On 08/28/2017 02:28 PM, Ilya Maximets wrote:
>> Count the cycles used for processing an rxq during the
>> pmd rxq interval. As this is an in flight counter and
>> pmds run independently, also store the total cycles used
>> during the last full interval.
>>
>> Signed-off-by: Kevin Traynor
>> ---
The current implementation of ovsdb-server caches only non-conditional
monitors, that is, monitors for every table row, not those that monitor
only rows that match some condition. To figure out which monitors are
conditional, the code track the number of tables that have conditions that
are
>From: O Mahony, Billy
>Sent: Thursday, August 17, 2017 3:24 PM
>To: d...@openvswitch.org
>Cc: Kavanagh, Mark B ; O Mahony, Billy
>
>Subject: [PATCH v2 4/4] docs: Document ingress scheduling feature
>
>Signed-off-by: Billy O'Mahony
One other general comment about the series:
It is not a ‘big deal’ at this point since we are discussing high level
concepts, but
we would need to address these at some point anyways.
Could you run
./utilities/checkpatch.py
before the next iteration
Thanks Darrell
On 8/22/17, 11:24
>From: O Mahony, Billy
>Sent: Thursday, August 17, 2017 3:24 PM
>To: d...@openvswitch.org
>Cc: Kavanagh, Mark B ; O Mahony, Billy
>
>Subject: [PATCH v2 3/4] dpif-netdev: Add rxq prioritization
>
>If an rxq is marked as 'prioritized' then keep
Yes we do, sorry for the delay!
We are actively testing the performance of the patch series' with iperf as
simulation of kernel app and pktken as DPDK app in the guest:
a) OVS master
b) Ilya's patches w/o time-based batching
c) Ilya's time-based batching using CLOCK_MONOTONIC
d) Our TSC-based
>From: O Mahony, Billy
>Sent: Thursday, August 17, 2017 3:24 PM
>To: d...@openvswitch.org
>Cc: Kavanagh, Mark B ; O Mahony, Billy
>
>Subject: [PATCH v2 1/4] netdev: Add set_ingress_sched to netdev api
>
>Passes ingress_sched config item from
Aaron Conole writes:
> The upgrade from older Open vSwitch versions on RHEL will try, as much as
> possible, to preserve the system. This means no new users or groups are
> created. As an effect, it's possible for the chown to fail, because the
> hugetlbfs group may not
> In dpdk_do_tx_copy function, all packets were copied to mbuf first, but
> QoS checking may drop some of them.
> Move the QoS checking in front of copying data to mbuf, it helps to reduce
> useless copy.
Hi Zhenyu, thanks for the patch, I agree with the objective of the patch but
have some
OK, done, thanks for the suggestion.
On Wed, Aug 30, 2017 at 09:39:06AM -0400, Lance Richardson wrote:
> Travis-CI builds for branches 2.6 and 2.7 are also failing due to the
> 50 minute per-job limit, perhaps this should be back-ported to those
> branches as well.
>
> Regards,
>
>Lance
>
When building the openvswitch-custom.te file, it is important to have the
ability to filter out dpdk blocks depending on whether the system has been
configured with dpdk or not. This allows using all the standard .in file
blocks, as well as the dpdkstrip blocks, when constructing the selinux
This will be used by an upcoming commit to have @begin_ and @end_ dpdk
blocks to keep dpdk specific policy decisions only active when dpdk is
used.
Acked-by: Flavio Leitner
Signed-off-by: Aaron Conole
Tested-by: Jean Hsiao
---
This series brings about a policy update to openvswitch allowing it to
run on a RHEL / Fedora system, even as a non-root user, with selinux set
to Enforcing.
The first two patches make some changes to the way the selinux policy is
built to have a macro-like effect, allowing the dpdk policy to be
On 29.08.2017 19:27, Greg Rose wrote:
> On 08/28/2017 12:08 AM, Ilya Maximets wrote:
>> TIME_UPDATE_INTERVAL was removed long time ago.
>> Now each call leads to time update via syscall and it's
>> granularity is system dependent.
>>
>> Fixes: 31ef9f5178de ("timeval: Remove CACHE_TIME scheme.")
>>
Travis-CI builds for branches 2.6 and 2.7 are also failing due to the
50 minute per-job limit, perhaps this should be back-ported to those
branches as well.
Regards,
Lance
- Original Message -
> From: "Lance Richardson"
> To: d...@openvswitch.org
> Sent:
I'm CCing libvir-list and qemu-devel because I would like to get
feedback from libvirt and QEMU developers too.
On Tue, Aug 08, 2017 at 10:49:21PM +0300, Michael S. Tsirkin wrote:
> On Tue, Jul 18, 2017 at 03:42:08PM +0200, Maxime Coquelin wrote:
> > This is an revival from a thread I initiated
v6->v7
- Remove NSH GSO patches in v6 because Jiri Benc
reworked it as another patch series and they have
been merged.
- Change it to adapt to nsh kernel module added by NSH
GSO patch series
v5->v6
- Fix the rest comments for v4.
- Add NSH GSO support for VxLAN-gpe + NSH and
Eth +
[ View in browser ]( http://r.newsletter.bonescamail.nl/nru6rvq6oatrf.html
)
UNLOADED TODAY:
Yellow Croakers / Maluwa / Micropogonias Furnieri
Size: 1 kilo up
Packing: 10 kilo
Origin: Argentina
1 box: € 2,55
10 box: € 2,35
1 palet (60 box) € 2,25
3 palets (180 box) € 2,05
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Hannes Frederic Sowa
> Sent: Wednesday, August 30, 2017 10:53 AM
> To: Yang, Yi Y
> Cc: d...@openvswitch.org; net...@vger.kernel.org;
Hello,
Yi Yang writes:
[...]
> +struct ovs_key_nsh {
> + u8 flags;
> + u8 ttl;
> + u8 mdtype;
> + u8 np;
> + __be32 path_hdr;
> + __be32 context[NSH_MD1_CONTEXT_SIZE];
> +};
> +
> struct sw_flow_key {
> u8 tun_opts[IP_TUNNEL_OPTS_MAX];
>
Hello.
I'm currently trying to use Open vSwitch and OpenFlow to imitate the behavior
of a router in an Open vSwitch bridge.
However I've run into some problems when trying to route IPV6 traffic.
I'm trying to use OpenFlow to convert NDP requests into responses to simulate a
router interface.
It
Hi Darrell,
Yes, I'm still interested in this series. I had a lot of other work last
few weeks. I hope that I'll have enough time to reply to the questions
and concerns in a near future.
Best regards, Ilya Maximets.
On 30.08.2017 01:45, Darrell Ball wrote:
> Hi Jan/Ilya/Bhanu
>
> Just
Hi,
Just wanted to add one comment to this correspondence. See below.
Thanks,
Finn Christensen
-Original Message-
From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-boun...@openvswitch.org]
On Behalf Of Yuanhan Liu
Sent: 30. august 2017 04:25
To: Darrell Ball
Cc:
I applied the patch to dpdk_merge here
https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_darball_ovs_commits_dpdk-5Fmerge=DwIGaQ=uilaK90D4TOVoH58JNXRgQ=BVhFA09CGX7JQ5Ih-uZnsw=A2_FCacqbp2moAo3HGFlTuxsjONUGhlN42OBcAuQQ6w=b6btPKhgvOFr2GOUYvktND6kaC6jc3fXI-mXfvNgXOU=
This can be
52 matches
Mail list logo