On 08.12.2017 21:18, Flavio Leitner wrote:
> On Fri, Dec 08, 2017 at 06:38:10PM +0300, Ilya Maximets wrote:
>> netdev-dpdk appctl commands added to man pages.
>>
>> Signed-off-by: Ilya Maximets
>> Acked-by: Antonio Fischetti
>> ---
>> NEWS
New appctl 'netdev-dpdk/get-mempool-info' implemented to get result
of 'rte_mempool_list_dump()' function if no arguments passed and
'rte_mempool_dump()' if DPDK netdev passed as argument.
Could be used for debugging mbuf leaks and other mempool related
issues. Most useful in pair with `grep -v
netdev-dpdk appctl commands added to man pages.
Signed-off-by: Ilya Maximets
---
NEWS| 1 +
lib/automake.mk | 1 +
lib/netdev-dpdk-unixctl.man | 9 +
manpages.mk | 2 ++
vswitchd/ovs-vswitchd.8.in | 1 +
5
Second patch adds debug appctl to obtain mempool information from DPDK
including names, numbers of available mbufs, object sizes and memory
pointers. First patch introduces common place for 'netdev-dpdk' unixctl
commands documentation in man pages.
Version 4:
* Rebased on current
On 08.12.2017 19:36, Kavanagh, Mark B wrote:
>> From: Ilya Maximets [mailto:i.maxim...@samsung.com]
>> Sent: Friday, December 8, 2017 3:38 PM
>> To: ovs-dev@openvswitch.org
>> Cc: Heetae Ahn ; Fischetti, Antonio
>> ; Loftus, Ciara
Thanks Sugesh,
I'll rebase today and send an updated version that addresses your comments.
Best,
Mark
>-Original Message-
>From: Chandran, Sugesh
>Sent: Friday, December 8, 2017 6:10 PM
>To: Kavanagh, Mark B ; d...@openvswitch.org;
>qiud...@chinac.com
>Cc:
>From: Chandran, Sugesh
>Sent: Friday, December 8, 2017 6:05 PM
>To: Kavanagh, Mark B ; d...@openvswitch.org;
>qiud...@chinac.com
>Cc: Stokes, Ian ; Loftus, Ciara
>; santosh.shu...@caviumnetworks.com
>Subject: RE:
As per OpenFlow v1.3 specification, when an action list contains a group
action a copy of the packet is passed to the group for processing by the
group. This means that if there is an error encountered during group
processing, only the copy of packet should be dropped, but subsequent
actions in
Hi Billy,
My patches frequently get corrupted by our email system. When I submit the v4
of the series adapted to the reverted dp_netdev_pmd_thread struct changes I
will try out another method of sending them.
For now, could you please try the attached patch files? For me they apply
cleanly on
>From: Kavanagh, Mark B
>Sent: Monday, December 11, 2017 11:49 AM
>To: Chandran, Sugesh ; d...@openvswitch.org;
>qiud...@chinac.com
>Cc: Stokes, Ian ; Loftus, Ciara
>; santosh.shu...@caviumnetworks.com
>Subject: RE:
These options have existed for a while, but were not expressed in the
help information. Inform the user that these options exist, and give
some basic help.
Reported-by: Saravanan KR
Signed-off-by: Aaron Conole
---
lib/daemon-unix.c | 2 ++
1 file
On 11/12/17 15:07, Aaron Conole wrote:
> These options have existed for a while, but were not expressed in the
> help information. Inform the user that these options exist, and give
> some basic help.
>
> Reported-by: Saravanan KR
> Signed-off-by: Aaron Conole
Hi,
I've been working on the solution proposed by Jan. I faced a new issue with
patch ports. Let's have a config like in unit test 1025:
"1025: ofproto-dpif - balance-tcp bonding, different recirc flow "
+-+
| br-int| p5 (OF 5)
| o--<---
Fix up some instances where variable declarations were not close
enough to their use, as these were missed before. This is the
preferred art in OVS code and flagged heavily in code reviews.
This is highly desirable due to code clarity reasons.
There are also some cases where newlines were not
I was concerned a bit about binary compatibility between the future
userspace binaries and older kernel versions.
But Windows users usually just use the MSI (because of the signing issues).
Thanks both!
Alin.
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
Not needed anymore because 'may_steal' already handled on
dpif-netdev layer and always true.
Acked-by: Eelco Chaudron
Signed-off-by: Ilya Maximets
---
lib/dpif-netdev.c | 2 +-
lib/netdev-bsd.c | 4 ++--
lib/netdev-dpdk.c | 25
>From: Ilya Maximets [mailto:i.maxim...@samsung.com]
>Sent: Monday, December 11, 2017 1:19 PM
>To: ovs-dev@openvswitch.org
>Cc: Heetae Ahn ; Fischetti, Antonio
>; Loftus, Ciara ;
>Kavanagh, Mark B
In current implementation 'now' variable updated once on each
receive cycle and passed through the whole datapath via function
arguments. It'll be better to keep this variable inside PMD
thread structure to be able to get it at any time. Such solution
will save the stack memory and simplify
This patch-set inspired by [1] from Bhanuprakash Bodireddy.
Implementation of [1] looks very complex and introduces many pitfalls [2]
for later code modifications like possible packet stucks.
This version targeted to make simple and flexible output packet batching on
higher level without
Thanks Antonio for doing all this and pushing it forward.
Regarding patches 2-4:
I understand you want to save some code for various possible set and get
operations.
The prior art for these commands is however not generic set and get commands.
Sometimes, we have specific commands that can take
Cutlen already applied while processing OVS_ACTION_ATTR_OUTPUT.
Acked-by: Eelco Chaudron
Signed-off-by: Ilya Maximets
---
lib/netdev-bsd.c | 2 +-
lib/netdev-dpdk.c | 5 -
lib/netdev-dummy.c | 2 +-
lib/netdev-linux.c | 4 ++--
4 files
This allows to collect packets from more than one RX burst
and send them together with a configurable intervals.
'other_config:tx-flush-interval' can be used to configure
time that a packet can wait in output batch for sending.
dpif-netdev turned to microsecond resolution for time
measuring to
New statistics for 'pmd-stats-show' command:
average number of packets per output batch.
Acked-by: Eelco Chaudron
Signed-off-by: Ilya Maximets
---
lib/dpif-netdev.c | 27 ++-
1 file changed, 22 insertions(+), 5 deletions(-)
>From: Ilya Maximets [mailto:i.maxim...@samsung.com]
>Sent: Monday, December 11, 2017 1:19 PM
>To: ovs-dev@openvswitch.org
>Cc: Heetae Ahn ; Fischetti, Antonio
>; Loftus, Ciara ;
>Kavanagh, Mark B
While processing incoming batch of packets they are scattered
across many per-flow batches and sent separately.
This becomes an issue while using more than a few flows.
For example if we have balanced-tcp OvS bonding with 2 ports
there will be 256 datapath internal flows for each dp_hash
Hi Jan,
This is a very interesting patch with compelling results.
I had a few questions after reading the commit message for this patch:
How did you decide on 1M as the proposed size for the DFC?
Will the size of this DFC be configurable (i.e. map a different number of hash
bits)?
I was
Thanks Antonio for doing this.
1/ Given the comments on patches 2-4, I think the documentation would change in
dpctl.man to be attribute specific, if
we go that route.
I did not write it up yet, but most of it would be obvious.
One exception is how a case where setting a limit is
We use the command `netcfg` to install the Windows datapath.
Since we have both 32 and 64 bit installers available point it to the
platform specific binary.
Found while testing.
Signed-off-by: Alin Gabriel Serdean
---
windows/ovs-windows-installer/CustomActions.wxs | 9
Still LGTM, please add
Acked-by: Antonio Fischetti
> -Original Message-
> From: Kavanagh, Mark B
> Sent: Monday, December 11, 2017 4:37 PM
> To: Ilya Maximets ; ovs-dev@openvswitch.org
> Cc: Heetae Ahn ;
Hi Cian,
Thank you for your feedback and the interesting measurement results.
As first I was a bit surprised that the gain is smaller in your P2P test case.
I guess that is because your traffic passes the datapath (and hence EMC) only
once, while in our PVP test with VXLAN each packet causes
On 12/10/2017 4:39 PM, Yang, Yi wrote:
On Sat, Dec 09, 2017 at 12:42:32AM +0800, Gregory Rose wrote:
On 12/8/2017 6:04 AM, Yi Yang wrote:
v5->v6
- Rebase v5 to master
- Refactor netlink message format to align to NSH kernel
implementation
- Add dec_nsh_ttl unit test into
On Fri, Dec 08, 2017 at 10:14:49PM +, Stokes, Ian wrote:
> The following changes since commit 65d9759c4fc433dbda89ad3d7225c1f5eac274ca:
>
> ovsdb-data: Add OVS_WARN_UNUSED_RESULT annotations to function definitions.
> (2017-12-08 13:39:29 -0800)
>
> are available in the git repository at:
With this patch, kernel datapath testsuite can be run under valgrind by using
the "check-kernel-valgrind" target and the results can be found under directory
"tests/system-kmod-testsuite.dir/".
Signed-off-by: Yifeng Sun
---
Documentation/topics/testing.rst | 4
> On Fri, Dec 08, 2017 at 10:14:49PM +, Stokes, Ian wrote:
> > The following changes since commit
> 65d9759c4fc433dbda89ad3d7225c1f5eac274ca:
> >
> > ovsdb-data: Add OVS_WARN_UNUSED_RESULT annotations to function
> definitions. (2017-12-08 13:39:29 -0800)
> >
> > are available in the git
On 12/11/2017 10:55 AM, Ben Pfaff wrote:
On Mon, Dec 11, 2017 at 10:25:54AM -0800, Gregory Rose wrote:
On 12/10/2017 4:39 PM, Yang, Yi wrote:
On Sat, Dec 09, 2017 at 12:42:32AM +0800, Gregory Rose wrote:
On 12/8/2017 6:04 AM, Yi Yang wrote:
v5->v6
- Rebase v5 to master
- Refactor
Thank you for testing. I applied this to master and backported it as
far as OVS 2.6.
On Mon, Dec 11, 2017 at 11:04:37AM +0530, shivani dommeti wrote:
> Thank you Ben for suggesting the fix.
> OVS is working as expected after adding the changes.
>
>
> Regards,
> Shivani.
>
> On Fri, Dec 8,
One extra note inline
Thanks Darrell
On 12/11/17, 8:35 AM, "Darrell Ball" wrote:
Thanks Antonio for doing all this and pushing it forward.
Regarding patches 2-4:
I understand you want to save some code for various possible set and get
operations.
On Mon, Dec 11, 2017 at 03:11:25PM +, Markos Chandras wrote:
> On 11/12/17 15:07, Aaron Conole wrote:
> > These options have existed for a while, but were not expressed in the
> > help information. Inform the user that these options exist, and give
> > some basic help.
> >
> > Reported-by:
pr_op->pr_rule is pointing to memory in bond->hash. It shouldn't be written
if bond->hash is already freed.
This bug is reported by running kernel path testsuite under valgrind:
Invalid write of size 8
at 0x413D16: update_recirc_rules__ (bond.c:392)
by 0x414CA0: bond_unref (bond.c:290)
On Mon, Dec 11, 2017 at 10:25:54AM -0800, Gregory Rose wrote:
> On 12/10/2017 4:39 PM, Yang, Yi wrote:
> >On Sat, Dec 09, 2017 at 12:42:32AM +0800, Gregory Rose wrote:
> >>On 12/8/2017 6:04 AM, Yi Yang wrote:
> >>>v5->v6
> >>>- Rebase v5 to master
> >>>- Refactor netlink message format to
> "sparse" warns when odp_port_t is used directly in an inequality
> comparison. This avoids the warning.
>
> CC: Kevin Traynor
> Fixes: a130f1a89bd8 ("dpif-netdev: Add port/queue tiebreaker to
> rxq_cycle_sort.")
> Signed-off-by: Ben Pfaff
> ---
>
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Ben Pfaff
> Sent: Friday, December 8, 2017 11:24 PM
> To: d...@openvswitch.org
> Cc: Ben Pfaff
> Subject: [ovs-dev] [PATCH 3/5] ovsdb-idl: Fix assertion
On 12/11/2017 06:34 PM, Ben Pfaff wrote:
> "sparse" warns when odp_port_t is used directly in an inequality
> comparison. This avoids the warning.
>
> CC: Kevin Traynor
> Fixes: a130f1a89bd8 ("dpif-netdev: Add port/queue tiebreaker to
> rxq_cycle_sort.")
> Signed-off-by:
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Ben Pfaff
> Sent: Friday, December 8, 2017 11:24 PM
> To: d...@openvswitch.org
> Cc: Ben Pfaff
> Subject: [ovs-dev] [PATCH 2/5] ovsdb-idl: Fix indentation
LGTM. Just two nits.
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Ben Pfaff
> Sent: Friday, December 8, 2017 11:24 PM
> To: d...@openvswitch.org
> Cc: Ben Pfaff
> Subject: [ovs-dev] [PATCH 1/5]
Porque antes de arriesgar, hay que investigar
Auditoría para la adquisición de empresas: Due Diligence
18 de Diciembre - CP. y MAN. Fernando García Zárate9am-6pm
Due diligence es una herramienta que le ayudará a prevenir cualquier situación
que agravie sus intereses con la realización de una
On Mon, Dec 11, 2017 at 08:00:07PM +, Stokes, Ian wrote:
> > "sparse" warns when odp_port_t is used directly in an inequality
> > comparison. This avoids the warning.
> >
> > CC: Kevin Traynor
> > Fixes: a130f1a89bd8 ("dpif-netdev: Add port/queue tiebreaker to
> >
The following patches are available in the current Linux upstream
git repository:
183dea5 openvswitch: do not propagate headroom updates to internal port
311af51 openvswitch: use ktime_get_ts64() instead of ktime_get_ts()
67c8d22 openvswitch: fix the incorrect flow action alloc size
On Mon, Dec 11, 2017 at 08:18:35PM +, Kevin Traynor wrote:
> On 12/11/2017 06:34 PM, Ben Pfaff wrote:
> > "sparse" warns when odp_port_t is used directly in an inequality
> > comparison. This avoids the warning.
> >
> > CC: Kevin Traynor
> > Fixes: a130f1a89bd8
From: Christophe JAILLET
Upstream commit:
commit 5829e62ac17a40ab08c1b905565604a4b5fa7af6
Author: Christophe JAILLET
Date: Mon Sep 11 21:56:20 2017 +0200
openvswitch: Fix an error handling path in
From: Jiri Benc
Upstream commit:
commit 9354d452034273a50a4fd703bea31e5d6b1fc20b
Author: Jiri Benc
Date: Thu Nov 2 17:04:37 2017 -0200
openvswitch: reliable interface indentification in port dumps
This patch allows reliable
From: Andy Zhou
Upstream commit:
commit 9602c01e57f7b868d748c2ba2aef0efa64b71ffc
Author: Andy Zhou
Date: Fri Nov 10 12:09:41 2017 -0800
openvswitch: export get_dp() API.
Later patches will invoke get_dp() outside of datapath.c. Export it.
From: Andy Zhou
Upstream commit:
commit 5794040647de4011598a6d005fdad95d24fd385b
Author: Andy Zhou
Date: Fri Nov 10 12:09:40 2017 -0800
openvswitch: Add meter netlink definitions
Meter has its own netlink family. Define netlink messages and
From: Andy Zhou
Upstream commit:
commit 96fbc13d7e770b542d2d1fcf700d0baadc6e8063
Author: Andy Zhou
Date: Fri Nov 10 12:09:42 2017 -0800
openvswitch: Add meter infrastructure
OVS kernel datapath so far does not support Openflow meter action.
From: "Gustavo A. R. Silva"
Upstream commit:
commit b74912a2fdae9aadd20da502644aa8848c861954
Author: Gustavo A. R. Silva
Date: Tue Nov 14 14:26:16 2017 -0600
openvswitch: meter: fix NULL pointer dereference in
From: Wei Yongjun
Upstream commit:
commit 6dc14dc40a1d1dafd8491c349b5f3e15aabc4edb
Author: Wei Yongjun
Date: Tue Nov 14 06:27:12 2017 +
openvswitch: Using kfree_rcu() to simplify the code
The callback function of
From: Wei Yongjun
Upstream commit:
commit 8a860c2bcc84a8e4fbcabb928cd97e4c51b17d93
Author: Wei Yongjun
Date: Tue Nov 14 06:20:16 2017 +
openvswitch: Fix return value check in ovs_meter_cmd_features()
In case of error,
From: zhangliping
Upstream commit:
commit 67c8d22a73128ff910e2287567132530abcf5b71
Author: zhangliping
Date: Sat Nov 25 22:02:12 2017 +0800
openvswitch: fix the incorrect flow action alloc size
If we want to add a
From: "Gustavo A. R. Silva"
Upstream commit:
commit 2734166e89639c973c6e125ac8bcfc2d9db72b70
Author: Gustavo A. R. Silva
Date: Sat Nov 25 13:14:40 2017 -0600
net: openvswitch: datapath: fix data type in queue_gso_packets
Thanks, Darrell and Aaron. I applied this series to master.
On Wed, Dec 06, 2017 at 07:10:03PM +, Darrell Ball wrote:
> Thanks for the reviews Aaron.
>
> Darrell
>
>
>
> On 12/6/17, 11:04 AM, "ovs-dev-boun...@openvswitch.org on behalf of Aaron
> Conole"
Needs to go back to 2.6; at least the changes in lib/conntrack.c
Thanks Darrell
On 12/11/17, 2:20 PM, "ovs-dev-boun...@openvswitch.org on behalf of Ben Pfaff"
wrote:
On Wed, Dec 06, 2017 at 06:04:20PM -0800, Darrell Ball wrote:
> -Original Message-
> From: Ben Pfaff [mailto:b...@ovn.org]
> Sent: Tuesday, December 12, 2017 12:30 AM
> To: Alin Serdean
> Cc: d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH 1/5] ovsdb-idl: Improve comments.
>
> On Mon, Dec 11, 2017 at
OK, I made that change and applied it to branch-2.8. It didn't apply
cleanly to 2.7 or 2.6, can you look at that?
On Mon, Dec 11, 2017 at 10:28:32PM +, Darrell Ball wrote:
> Yes, it is Ben
>
> Thanks Darrell
>
> On 12/11/17, 2:27 PM, "Ben Pfaff" wrote:
>
> It fails to
On Mon, Dec 11, 2017 at 10:30:46PM +, Alin Serdean wrote:
>
>
> > -Original Message-
> > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> > boun...@openvswitch.org] On Behalf Of Ben Pfaff
> > Sent: Friday, December 8, 2017 11:25 PM
> > To: d...@openvswitch.org
> > Cc: Ben
I applied this on master.
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of aserd...@ovn.org
> Sent: Monday, December 11, 2017 4:23 PM
> To: 'Ilya Maximets' ; ovs-dev@openvswitch.org
> Cc: 'Lance
An address sanity check is done on icmp error packets to
check that the icmp error payload makes sense w.r.t. the
packet itself.
The sanity check was partially incorrect since it tried
to verify the source address of the error packet against the
original destination, which does not makes since
Ben
I sent a 2.7 patch here:
https://patchwork.ozlabs.org/patch/847308/
it should be applicable for 2.6 as well.
Thanks Darrell
On 12/11/17, 2:43 PM, "Ben Pfaff" wrote:
OK, I made that change and applied it to branch-2.8. It didn't apply
cleanly to 2.7 or 2.6, can you
On Tue, Dec 12, 2017 at 03:15:49AM +0800, Gregory Rose wrote:
> On 12/11/2017 10:55 AM, Ben Pfaff wrote:
> > On Mon, Dec 11, 2017 at 10:25:54AM -0800, Gregory Rose wrote:
> >> On 12/10/2017 4:39 PM, Yang, Yi wrote:
> >>> On Sat, Dec 09, 2017 at 12:42:32AM +0800, Gregory Rose wrote:
> On
> -Original Message-
> From: Ben Pfaff [mailto:b...@ovn.org]
> Sent: Tuesday, December 12, 2017 12:40 AM
> To: Alin Serdean
> Cc: d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH 1/5] ovsdb-idl: Improve comments.
>
> On Mon, Dec 11, 2017 at
From: Paolo Abeni
Upstream commit:
commit 183dea5818315c0a172d21ecbcd2554894bf01e3
Author: Paolo Abeni
Date: Thu Nov 30 15:35:33 2017 +0100
openvswitch: do not propagate headroom updates to internal port
After commit 3a927bc7cf9d
From: Arnd Bergmann
Upstream commit:
commit 311af51dcb5629f04976a8e451673f77e3301041
Author: Arnd Bergmann
Date: Mon Nov 27 12:41:38 2017 +0100
openvswitch: use ktime_get_ts64() instead of ktime_get_ts()
timespec is deprecated because of the
On Wed, Dec 06, 2017 at 06:04:20PM -0800, Darrell Ball wrote:
> An address sanity check is done on icmp error packets to
> check that the icmp error payload makes sense w.r.t. the
> packet itself.
>
> The sanity check was partially incorrect since it tried
> to verify the source address of the
On Mon, Dec 11, 2017 at 08:23:44PM +, Alin Serdean wrote:
> LGTM. Just two nits.
> > +/* Terminal error state that indicates that nothing useful can be done.
> > + * The most likely reason is that the database server doesn't actually
> > have
> [Alin Serdean] doesn't have, maybe?
On Mon, Dec 11, 2017 at 10:34:06PM +, Alin Serdean wrote:
>
>
> > -Original Message-
> > From: Ben Pfaff [mailto:b...@ovn.org]
> > Sent: Tuesday, December 12, 2017 12:30 AM
> > To: Alin Serdean
> > Cc: d...@openvswitch.org
> > Subject: Re: [ovs-dev]
A few suggestions Yi-hung
Thanks Darrell
On 11/21/17, 5:04 PM, "ovs-dev-boun...@openvswitch.org on behalf of Yi-Hung
Wei" wrote:
This patch adds support of flushing a conntrack entry specified by the
conntrack 5-tuple
Yes, it is Ben
Thanks Darrell
On 12/11/17, 2:27 PM, "Ben Pfaff" wrote:
It fails to apply due to conflicts in system-traffic.at. Is it safe to
drop that change and apply the rest?
On Mon, Dec 11, 2017 at 10:22:39PM +, Darrell Ball wrote:
> Needs to go
It fails to apply due to conflicts in system-traffic.at. Is it safe to
drop that change and apply the rest?
On Mon, Dec 11, 2017 at 10:22:39PM +, Darrell Ball wrote:
> Needs to go back to 2.6; at least the changes in lib/conntrack.c
>
> Thanks Darrell
>
> On 12/11/17, 2:20 PM,
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Ben Pfaff
> Sent: Friday, December 8, 2017 11:24 PM
> To: d...@openvswitch.org
> Cc: Ben Pfaff
> Subject: [ovs-dev] [PATCH 4/5] ovsdb-idl: Remove 'uuid'
An address sanity check is done on icmp error packets to
check that the icmp error payload makes sense w.r.t. the
packet itself.
The sanity check was partially incorrect since it tried
to verify the source address of the error packet against the
original destination, which does not makes since
I sent a V2 here
https://patchwork.ozlabs.org/patch/847315/
an extra signoff had snuck into v1
Thanks Darrell
On 12/11/17, 5:22 PM, "Darrell Ball" wrote:
Ben
I sent a 2.7 patch here:
https://patchwork.ozlabs.org/patch/847308/
it should be applicable
80 matches
Mail list logo