On 14.12.2017 19:26, Kevin Traynor wrote:
> On 12/14/2017 03:30 PM, Ilya Maximets wrote:
>> Hello Ciara,
>> Thanks for the patches.
>>
>> I did not review the code. But I have few general concerns about number
>> of tx descriptors in HW NICs inline.
>>
>> Best regards, Ilya Maximets.
>>
>> On 08.12
> Hello Ciara,
> Thanks for the patches.
>
> I did not review the code. But I have few general concerns about number of
> tx descriptors in HW NICs inline.
>
> Best regards, Ilya Maximets.
>
> On 08.12.2017 18:26, Ciara Loftus wrote:
> > Enabled per port like so:
> > ovs-vsctl set Interface dpdk
Hi,
> >> What if we have 2, 4, 10 HW NICs and a few different destinations for
> >> traffic from VM?
> >> What if we have 2, 4, 10 HW NICs added to balanced OVS bonding and just a
> >> few different packet flows?
> >> What if we have HW NICs with small number of TX queues (like VFs) and XPS
> >
One general concern:
Why we're adding new (experimental) feature to deprecated port type?
If we'll not do that, this patch will not be needed.
IMHO, it's better to keep vhostuser ports intact to avoid possible
regressive changes. Also, this patch increases code size.
Some comments about patch itse
This patch enables optional dequeue zero copy for vHost ports.
This gives a performance increase for some use cases. I'm using
the cover letter to report my results.
vhost (vm1) -> vhost (vm2)
Using testpmd to source (txonly) in vm1 and sink (rxonly) in vm2.
4C1Q 64B packets: 5.05Mpps -> 5.52Mpps
dpdkvhostuser and dpdkvhostuserclient ports share a lot of the same
setup & configuration code. Create a common function they can share in
order to remove some duplication of code.
Signed-off-by: Ciara Loftus
---
lib/netdev-dpdk.c | 122 +-
1 f
Enabled per port like so:
ovs-vsctl set Interface dpdkvhostuserclient0 options:dq-zero-copy=true
The feature is disabled by default and can only be enabled/disabled when
a vHost port is down.
When packets from a vHost device with zero copy enabled are destined for
a 'dpdk' port, the number of tx
Not a full review.
General thoughts:
If following conditions are true:
1. We don't need to add new feature to deprecated vhostuser port.
2. We actually don't need to have ability to change ZC config if
vhost-server-path
already configured.
Let me explain this condition:
To change 'dq
Hi Bhanu,
I've tested v6 and LGTM, works as expected.
To check the behavior I blocked one or more of the PMD with the trick of
attaching cgdb. So with
utilities/ovs-appctl keepalive/pmd-health-show
I could see for example 2 of the 3 PMDs - those I intentionally blocked -
correctly reported
LGTM,
Tested-by: Antonio Fischetti
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [o
LGTM,
Tested-by: Antonio Fischetti
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [o
LGTM,
Tested-by: Antonio Fischetti
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [o
LGTM,
Tested-by: Antonio Fischetti
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [
LGTM,
Tested-by: Antonio Fischetti
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [o
> From: ovs-dev-boun...@openvswitch.org
> [mailto:ovs-dev-boun...@openvswitch.org] On Behalf Of Ciara Loftus
> Sent: Friday, 15 December, 2017 14:14
> To: d...@openvswitch.org
> Subject: [ovs-dev] [PATCH v6 2/2] netdev-dpdk: Enable optional dequeue zero
> copy for vHost User
>
> Enabled per port
LGTM,
Tested-by: Antonio Fischetti
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [o
I think this needs a trivial rebase, other than that it's ok.
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@ope
LGTM,
Acked-by: Antonio Fischetti
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Bhanuprakash Bodireddy
> Sent: Friday, December 8, 2017 12:04 PM
> To: d...@openvswitch.org
> Subject: [ovs-dev] [PATCH v6 7/8] Documen
Currently, "make rpm-fedora" doesn't work correctly since the new ovsdb
manpages (ovsdb.5, ovsdb.7 and ovsdb-server.7) were added.
This patch adds the new ovsdb manpages in the %files list in the spec
file.
CC: Ben Pfaff
Fixes: 12b84d50e032 ("ovsdb: Improve documentation.")
Signed-off-by: Timoth
On Fri, Dec 15, 2017 at 05:09:40PM +0100, Timothy Redaelli wrote:
> Currently, "make rpm-fedora" doesn't work correctly since the new ovsdb
> manpages (ovsdb.5, ovsdb.7 and ovsdb-server.7) were added.
>
> This patch adds the new ovsdb manpages in the %files list in the spec
> file.
>
> CC: Ben Pf
On Thu, Dec 14, 2017 at 05:55:36PM -0800, Justin Pettit wrote:
>
>
> > On Dec 13, 2017, at 3:10 PM, Ben Pfaff wrote:
> >
> > diff --git a/ovsdb/ovsdb-client.1.in b/ovsdb/ovsdb-client.1.in
> > index 694a7abed46d..2e2df5e5aa7f 100644
> > --- a/ovsdb/ovsdb-client.1.in
> > +++ b/ovsdb/ovsdb-client.
On Thu, Dec 14, 2017 at 04:01:55PM -0800, Yifeng Sun wrote:
> Thanks. It is a good reading and learning. I found a small issue in the
> line below.
>
> +A boolean value, true if authorized insertions and authorized are
> allowed,
>
> For "insertions and authorized", maybe you are saying "inse
On 12/15/2017 12:32 PM, Jan Scheurich wrote:
> Hi,
>
What if we have 2, 4, 10 HW NICs and a few different destinations for
traffic from VM?
What if we have 2, 4, 10 HW NICs added to balanced OVS bonding and just a
few different packet flows?
What if we have HW NICs with
On 12/15/2017 09:39 AM, Ilya Maximets wrote:
>>> What if we have 2, 4, 10 HW NICs and a few different destinations for
>>> traffic from VM?
>>> What if we have 2, 4, 10 HW NICs added to balanced OVS bonding and just a
>>> few different packet flows?
>>> What if we have HW NICs with small number o
> On 12/15/2017 12:32 PM, Jan Scheurich wrote:
> > Hi,
> >
> What if we have 2, 4, 10 HW NICs and a few different destinations for
> traffic from VM?
> What if we have 2, 4, 10 HW NICs added to balanced OVS bonding and
> just a few different packet flows?
> What if we have HW NICs wi
This member was only used in one particular code path, so this commit
adds code to pass it around as a function parameter instead.
Signed-off-by: Ben Pfaff
---
ovsdb/ovsdb-server.c | 40
1 file changed, 16 insertions(+), 24 deletions(-)
diff --git a/ovsd
Thanks Darrell for your review.
I agree with your comments, if we have to consider R/W commands with
more than one parameter is much better to have specific functions
as you explained and showed.
I'll rework accordingly and post a v4.
Regards,
-Antonio
> -Original Message-
> From: Darrel
Thanks Darrell and Stephen for your suggestions.
I'll rework accordingly in v4.
Antonio
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Monday, December 11, 2017 6:02 PM
> To: Fischetti, Antonio ;
> d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v3 5/5]
Hi Antonio
Alternatively, given that most of code is rewritten, I was planning on just
submitting the patches, which are in process, and adding you as co-author.
Will that work for you?
Thanks Darrell
On 12/15/17, 11:05 AM, "Fischetti, Antonio" wrote:
Thanks Darrell for your review.
Sure, that's perfect.
Thanks, Antonio
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Friday, December 15, 2017 7:23 PM
> To: Fischetti, Antonio ;
> d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v3 2/5] conntrack: add commands to r/w CT
> parameters.
>
Added a unit test case for testing the condition when a veth interface
is added to br0 and then the veth interface is deleted from the
system and added back with the same name.
Signed-off-by: Ashish Varma
---
v1-v2
Incorporated the comments by Yi-Hung Wei:
Added system-interface.at test to syste
Mida y minimice el impacto económico por desastres
Blindaje económico empresarial contra desastres
17 de Enero - Mtro. David Lozada Martínez - 9am-8pm
Inscríbase hoy a este seminario, en el cual nuestro experto le enseñará no sólo
a identificar los riesgos y sus impactos, sino a establecer medi
MSVC does not allow to re-define unnamed structure in union.
Thus, this fix defines the struct outside of the anonymous union
in order to calculate the padded size.
Signed-off-by: Shireesh Kumar Singh
---
include/openvswitch/util.h | 14 +++---
1 file changed, 7 insertions(+), 7 deletion
C++ does not allow implicit conversion from void pointer to a specific
pointer type. This change adds explicit typecasts to appropriate types
wherever needed.
---
lib/netlink.h | 3 ++-
lib/ovs-thread.h | 4 ++--
lib/packets.h | 15 ++-
lib/socket-util.h | 4 ++--
4 files c
C++ compiler cannot find matching function calls due to
unsupported implicit conversions. This change adds appropriate
explicit typecasts to match the existing function prototypes.
---
lib/ovs-atomic-msvc.h | 59 +++
1 file changed, 36 insertions(+),
This change fixes compilation issues related to designated
initializers, which are not supported by C++ compier.
Signed-off-by: Shireesh Kumar Singh
---
lib/unaligned.h | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/lib/unaligned.h b/lib/unaligned.h
index a150
v4:
Rebased
Modified commit message
v3:
Rebased
Split patch into subparts
v2:
Rebased
Fixed formatting erros
Removed Windows specific macro
Shireesh Singh (4):
Windows: Fixing Windows C++ compilation issues with unnamed structure
lib: Adding explicit typecasts to fix C++ com
MSVC does not allow to re-define unnamed structure in union.
Thus, this fix defines the struct outside of the anonymous union
in order to calculate the padded size.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Shireesh Kumar Singh
Co-authored-by:Sairam Venugopal
---
include/openvswitch/u
C++ does not allow implicit conversion from void pointer to a specific
pointer type. This change adds explicit typecasts to appropriate types
wherever needed.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Sairam Venugopal
Co-authored-by: Sairam Venugopal
---
lib/netlink.h | 3 ++-
l
C++ compiler cannot find matching function calls due to
unsupported implicit conversions. This change adds appropriate
explicit typecasts to match the existing function prototypes.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Sairam Venugopal
Co-authored-by: Sairam Venugopal
---
lib/ovs-
This change fixes compilation issues related to designated
initializers, which are not supported by C++ compier.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Sairam Venugopal
Co-authored-by: Sairam Venugopal
---
lib/unaligned.h | 13 ++---
1 file changed, 10 insertions(+), 3 dele
v5:
Rebased
Added version tag in commit message
v4:
Rebased
Modified commit message
v3:
Rebased
Split patch into subparts
v2:
Rebased
Fixed formatting erros
Removed Windows specific macro
Shireesh Singh (4):
Windows: Fixing Windows C++ compilation issues with unnamed structure
MSVC does not allow to redefine unnamed structure in union.
Thus, this fix defines the struct outside of the anonymous union
in order to calculate the padded size.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Shireesh Kumar Singh
Co-authored-by:Sairam Venugopal
---
include/openvswitch/ut
C++ does not allow implicit conversion from void pointer to a specific
pointer type. This change adds explicit typecasts to appropriate types
wherever needed.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Sairam Venugopal
Co-authored-by: Sairam Venugopal
---
lib/netlink.h | 3 ++-
l
C++ compiler cannot find matching function calls due to
unsupported implicit conversions. This change adds appropriate
explicit typecasts to match the existing function prototypes.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Sairam Venugopal
Co-authored-by: Sairam Venugopal
---
lib/ovs-
This change fixes compilation issues related to designated
initializers, which are not supported by C++ compiler.
Signed-off-by: Shireesh Kumar Singh
Signed-off-by: Sairam Venugopal
Co-authored-by: Sairam Venugopal
---
lib/unaligned.h | 13 ++---
1 file changed, 10 insertions(+), 3 del
Presently, alg expectations are removed by being time expired.
This was intended to happen before the control connections and
was intended to minimize the extra work involved for tracking and
removing the expectations. This is not the best option and
conceptually an expectation should not exist wi
47 matches
Mail list logo