When executing conntrack actions on skbuffs with checksum mode
CHECKSUM_COMPLETE, the checksum must be updated to account for
header pushes and pulls. Otherwise we get "hw csum failure"
logs similar to this (ICMP packet received on geneve tunnel
via ixgbe NIC):
[ 405.740065] genev_sys_6081: hw
[ View in browser ]( http://r.newsletter.bonescamail.nl/nru6qw66oatrf.html
)
NEW PRODUCTION PANREADY DORADE DENTON
ORIGIN: MAROCCO
PACKING: 5 KILO
SPECIFICATION: GUTTED AND SCALED
SIZE: +/- 4 PIECES PER KILO
PRICES:
1 BOX € 2,89
10 BOX € 2,69
PALET(145 BOXES) €2,59 = € 12,95 PER
On Thu, Jan 12, 2017 at 12:25:27PM -0800, Andy Zhou wrote:
> On Thu, Jan 12, 2017 at 9:16 AM, Ben Pfaff wrote:
>
> > The lists these days prefer an ovs- prefix. Currently all of the build
> > emails are being dropped because it is missing.
> >
> > Signed-off-by: Ben Pfaff
This patch adds support for Ipv4 fragments in conntrack module.
Individual fragments are not tracked, the reassembled Ipv4 datagram is treated
as a single ct entry.
Added MRU field in OvsForwardingContext, to keep track of Maximum recieved unit
from all the recieved IPv4 fragments.
---
This patch retains the MRU value for the reassembled IP datagram in the
OvsForwardingContext when the packet is forwarded to userspace and/or
retrived from userspace.
Also retain the MRU value when there are any deferred actions for the
current NBL.
---
datapath-windows/ovsext/Actions.c| 48
This patch adds functionalities to handle IPv4 fragments, which will be
used by Conntrack module.
Added a new structure to hold the Ipv4 fragments and a hash table to
hold Ipv4 datagram entries. Also added a clean up thread that runs
every minute to delete the expired IPv4 datagram entries.
The
MRU value is updated only for the Ipv4 fragments. If it is non zero,
then fragment the NBL based on MRU value and send out the new NBL to
the vnic.
---
datapath-windows/ovsext/Actions.c | 26 ++
1 file changed, 26 insertions(+)
diff --git
With this patch, OvsTcpSegmentNBL not only supports tcp segments but also the
IP fragments.
To reflect the new changes, renamed function name from OvsTcpSegmentNBL
to OvsFragmentNBL and created a wrapper for OvsTcpSegmentNBL.
---
datapath-windows/ovsext/BufferMgmt.c | 189
Hi,
It seems that shelling out to ovs-ofctl very quickly can lead to bug where
it reports an OFPT_ERROR.
We were able to constantly reproduce within minutes of running the above
flow modifications on Unbutu.
Any help, hints or guidance would be appreciated. I'd be happy to pursue
some
On Thu, Jan 12, 2017 at 9:16 AM, Ben Pfaff wrote:
> The lists these days prefer an ovs- prefix. Currently all of the build
> emails are being dropped because it is missing.
>
> Signed-off-by: Ben Pfaff
>
Acked-by: Andy Zhou
>
>
Hi Daniele,
I've checked that without this patch I was getting
ERROR: 2316 tests were run,
75 failed unexpectedly.
2 tests were skipped.
Instead after applying this patch I get
ERROR: 2316 tests were run,
42 failed unexpectedly.
2 tests were skipped.
In particular, after I apply this patch the
set_tunnel_config() always logs a warning, even on success. This
shouldn't happen.
Without this, some unit tests fail.
Fixes: 9fff138ec3a6("netdev: Add 'errp' to set_config().")
Signed-off-by: Daniele Di Proietto
---
lib/netdev-vport.c | 10 ++
1 file changed, 6
Instead of counting all polling cycles as processing cycles, only count
the cycles where packets were received from the polling.
Signed-off-by: Georg Schmuecking
Signed-off-by: Ciara Loftus
Co-authored-by: Ciara Loftus
'pmd-stats-show' now reports the number of entries in the EMC as well as
the percentage full it is.
Eg. For 2048 entries the EMC is reported as 25% full as the maximum
capacity is 8192 entries.
Signed-off-by: Ciara Loftus
Signed-off-by: Georg Schmuecking
On Thu, Jan 12, 2017 at 4:05 AM, Joe Stringer wrote:
> On 10 January 2017 at 16:00, Jarno Rajahalme wrote:
> > Acked-by: Jarno Rajahalme
>
> Thanks, applied to master.
>
> Numan, I believe that this is an issue regardless of whether it fixes
> your
Hi, Daniele.
Thanks for v3.
Acked-by: Ilya Maximets
On 09.01.2017 06:14, Daniele Di Proietto wrote:
> The first two commits of the series are trivial bugfixes for dpif-netdev.
>
> Then the series fixes a long standing bug that caused a crash when the
> admin link state
Hi,
Thanks for the patch.
Is this part of a series or is it a standalone patch?
I don't see the added value to add the wrapper just from this patch.
Thanks,
Alin.
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of
Hi,
Thanks a lot for the series, it will be a great addition to the Windows
datapath!
One small nit it does not compile under release because of `ovspool` is defined
only in debug mode.
I also tried to set up an environment and for some reason the fragments were
stalling. I tried icmp and
On Wed, Jan 11, 2017 at 4:39 PM, Jarno Rajahalme wrote:
> Looks good to me:
>
> Acked-by: Jarno Rajahalme
>
Thanks, pushed.
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
> On Jan 12, 2017, at 9:23 AM, Ben Pfaff wrote:
>
> On Mon, Jan 09, 2017 at 08:46:28PM -0800, Justin Pettit wrote:
>> I think it would be helpful to have a comment describing this
>> function. Also mentioning that the caller maintains ownership of
>> 'text'.
>
> There was a lot
On Mon, Jan 09, 2017 at 08:46:28PM -0800, Justin Pettit wrote:
> I think it would be helpful to have a comment describing this
> function. Also mentioning that the caller maintains ownership of
> 'text'.
There was a lot of missing comments and documentation. I added a bunch.
> I may be missing
Hi Alin, I plan to replace all occurrences of "portId ==
switchContext->virtualExternalPortId" by calling into the wrapper. Will send
out a subsequent patch soon.
Thanks,
Shashank
From: Alin Serdean
Sent: Thursday, January
On 12/01/2017 09:33, "Ben Pfaff" wrote:
>On Thu, Jan 12, 2017 at 12:23:55AM -0800, Daniele Di Proietto wrote:
>> set_tunnel_config() always logs a warning, even on success. This
>> shouldn't happen.
>>
>> Without this, some unit tests fail.
>>
>> Fixes: 9fff138ec3a6("netdev:
Use memory_order_release when updating the tables version number to
make sure no memory accesses before the atomic_store (possibly
relating to setting up the new version) are reordered to take place
after the atomic_store, which makes the new version available to other
threads.
Correspondingly,
This patch is part of the OVS-DPDK performance optimizations presented
on the OVS fall conference
(http://openvswitch.org/support/ovscon2016/8/1400-gray.pdf)
The Exact Match Cache does not perform well in use cases with a high
numbers of parallel packet flows. When the flow count exceeds 8k,
Hi Ilya,
thanks for your detailed feedback. I've added a couple of replies
inline about when the instant send happens and the may_steal param.
Regards,
Antonio
> -Original Message-
> From: Ilya Maximets [mailto:i.maxim...@samsung.com]
> Sent: Tuesday, December 20, 2016 12:47 PM
> To:
Commit 9fff138ec3a6dbe75073d16cba7fbe86ac273c36 "netdev: Add 'errp' to
set_config()." breaks the unit tests because netdev-vport now logs lots
of blank lines. I am unsure of the right fix--is it to just drop the
new VLOG_WARN call?
Thanks,
Ben.
___
On 12/01/2017 09:24, "Ben Pfaff" wrote:
>Commit 9fff138ec3a6dbe75073d16cba7fbe86ac273c36 "netdev: Add 'errp' to
>set_config()." breaks the unit tests because netdev-vport now logs lots
>of blank lines. I am unsure of the right fix--is it to just drop the
>new VLOG_WARN call?
28 matches
Mail list logo