Dear user of openvswitch.org,
We have received reports that your email account was used to send a huge amount
of junk e-mail messages during the recent week.
Most likely your computer was compromised and now contains a trojaned proxy
server.
We recommend you to follow the instructions in the at
On Wed, Aug 26, 2015 at 04:29:50PM -0700, Andy Zhou wrote:
> This is not a review, since I am not that familiar with this code yet.
> Just a few questions
>
> Should compare_cmps_3way() be of type 'int' instead of bool?
Yes, thanks, fixed.
> Should crash_or() also make use of disjunction_matches
On Wed, Aug 26, 2015 at 04:21:42PM -0700, Andy Zhou wrote:
> A few questions in line:
>
> On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
> > Signed-off-by: Ben Pfaff
> > ---
> > ovn/lib/expr.c | 19 +++
> > 1 file changed, 15 insertions(+), 4 deletions(-)
> >
> > diff --git a
On Wed, Aug 26, 2015 at 01:58:31PM -0700, Justin Pettit wrote:
>
> > On Aug 25, 2015, at 9:37 PM, Ben Pfaff wrote:
> >
> > +/* Implementation of crush_cmps() for expr->type == EXPR_T_OR. */
> > static struct expr *
> > crush_or(struct expr *expr, const struct expr_symbol *symbol)
>
> Since you
A lot of tests need to initialize the OVS_RUNDIR, OVS_LOGDIR, etc.
variables to point to the directory in which the tests run. Until now,
each of them has had to do this individually, which is redundant. This
commit starts to do this automatically.
Signed-off-by: Ben Pfaff
---
tests/jsonrpc-py
A shell function doesn't need quoted and unquoted variants and it
integrates naturally with other shell code.
Signed-off-by: Ben Pfaff
---
tests/ofproto-dpif.at | 61 ---
tests/ofproto-macros.at | 4 +--
tests/ofproto.at | 6 ++
On Thu, Aug 27, 2015 at 10:05:26AM +0530, mallika jain wrote:
> Is it possible to port OVS on a platform which is not Linux based?
Yes, OVS already supports FreeBSD, NetBSD, and Windows.
> How will it affect development ?
It takes more effort.
___
dev
Hi,
Is it possible to port OVS on a platform which is not Linux based?
How will it affect development ?
Regards,
Mallika Jain
Sr Engineer
Tata Elxsi Bangalore, India
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/
In test runs, I've occasionally seen mysterious failures in which
parse_listening_port failed to find the listening port even though an
examination of the log file showed that it was there. I spent some time
trying to figure out what was going wrong. It seemed like everything was
lined up properl
AT_CHECK runs its commands in a subshell. That means that (among other
effects), any variable assignments within its commands will disappear after
the commands' completion. That doesn't matter for any of the existing
users, which don't do the sorts of things that affect an outer shell
environment
Sometimes OVS_WAIT_WHILE and OVS_WAIT_UNTIL are used for conditions that
tend to be true immediately but sometimes take a moment or two. This
change makes such cases 0.1 seconds faster (which could add up over
hundreds of tests).
Signed-off-by: Ben Pfaff
---
tests/ovs-macros.at | 5 -
1 fil
On Wed, Aug 26, 2015 at 05:42:46PM -0700, Jarno Rajahalme wrote:
> Without an explicit bounds check GCC 4.9 issues an array out of bounds
> error. This patch adds explicit checks which will however be
> optimized away as the relevant parameters are compile-time constants.
>
> Signed-off-by: Jarno
On Wed, Aug 26, 2015 at 05:42:45PM -0700, Jarno Rajahalme wrote:
> Initialize the whole 'exact_match_mask' in
> mf_mask_field_and_prereqs(), as geneve options can be bigger than 16
> bytes.
>
> Requested-by: Jesse Gross
> Signed-off-by: Jarno Rajahalme
I wish *so much* that the GCC extension fo
On Wed, Aug 26, 2015 at 5:42 PM, Jarno Rajahalme wrote:
> Initialize the whole 'exact_match_mask' in
> mf_mask_field_and_prereqs(), as geneve options can be bigger than 16
> bytes.
>
> Requested-by: Jesse Gross
> Signed-off-by: Jarno Rajahalme
Acked-by: Jesse Gross
I think we can actually do
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
> On Aug 26, 2015, at 4:57 PM, Jesse Gross wrote:
>
> On Fri, Aug 21, 2015 at 3:25 PM, Jarno Rajahalme
> wrote:
>> mf_mask_field_and_prereqs() used to memset a static variable again and
>> again. Now that mf_value is larger (due to tun_metadata field), this
>> is more expensive. Avoid this b
I sent a patch that adds the checks.
Jarno
> On Aug 26, 2015, at 5:12 PM, Ben Pfaff wrote:
>
> I think the compiler is wrong here. Not sure what to do about it yet
> though.
>
> On Wed, Aug 26, 2015 at 04:50:21PM -0700, Ben Pfaff wrote:
>> GCC 4.9 doesn't like flowmap_clear() and flowmap_se
Without an explicit bounds check GCC 4.9 issues an array out of bounds
error. This patch adds explicit checks which will however be
optimized away as the relevant parameters are compile-time constants.
Signed-off-by: Jarno Rajahalme
---
lib/flow.h | 12 +---
1 file changed, 9 insertions
Initialize the whole 'exact_match_mask' in
mf_mask_field_and_prereqs(), as geneve options can be bigger than 16
bytes.
Requested-by: Jesse Gross
Signed-off-by: Jarno Rajahalme
---
lib/meta-flow.c | 5 ++---
lib/meta-flow.h | 7 +++
2 files changed, 9 insertions(+), 3 deletions(-)
diff --gi
On Wed, Aug 26, 2015 at 4:58 PM, Ben Pfaff wrote:
> On Wed, Aug 26, 2015 at 04:56:20PM -0700, Andy Zhou wrote:
>> On Wed, Aug 26, 2015 at 4:46 PM, Ben Pfaff wrote:
>> > On Wed, Aug 26, 2015 at 12:34:07PM -0700, Andy Zhou wrote:
>> >> On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
>> >> > +
A couple of folks have pointed out that the way we attach ACLs to either
physical ports or to pairs leaves some room for ambiguity. My
proposal is that we discourage the use of both types of ACLs on the same
physical port. We can’t enforce this (AFAIK) in the database itself, but we can
recomm
> On Aug 26, 2015, at 4:50 PM, Ben Pfaff wrote:
>
> GCC 4.9 doesn't like flowmap_clear() and flowmap_set(). Maybe that's
> why you had those seemingly redundant tests in there? I'm taking a
> closer look to see whether I believe its claims though.
>
> CC lib/classifier.lo
>In f
I think the compiler is wrong here. Not sure what to do about it yet
though.
On Wed, Aug 26, 2015 at 04:50:21PM -0700, Ben Pfaff wrote:
> GCC 4.9 doesn't like flowmap_clear() and flowmap_set(). Maybe that's
> why you had those seemingly redundant tests in there? I'm taking a
> closer look to se
On Fri, Aug 21, 2015 at 3:25 PM, Jarno Rajahalme wrote:
> mf_mask_field_and_prereqs() used to memset a static variable again and
> again. Now that mf_value is larger (due to tun_metadata field), this
> is more expensive. Avoid this by using static initialization.
>
> mf_mask_field_and_prereqs()
On Wed, Aug 26, 2015 at 04:56:20PM -0700, Andy Zhou wrote:
> On Wed, Aug 26, 2015 at 4:46 PM, Ben Pfaff wrote:
> > On Wed, Aug 26, 2015 at 12:34:07PM -0700, Andy Zhou wrote:
> >> On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
> >> > +SSET_FOR_EACH_SAFE (name, next, a) {
> >> > +if
On Wed, Aug 26, 2015 at 4:46 PM, Ben Pfaff wrote:
> On Wed, Aug 26, 2015 at 12:34:07PM -0700, Andy Zhou wrote:
>> Acked-by: Andy Zhou
>
> Thanks, I applied this to master.
>
>> A question in line.
>>
>> On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
>> > This will acquire its first user in an
GCC 4.9 doesn't like flowmap_clear() and flowmap_set(). Maybe that's
why you had those seemingly redundant tests in there? I'm taking a
closer look to see whether I believe its claims though.
CC lib/classifier.lo
In file included from ../lib/match.h:20:0,
fro
On Wed, Aug 26, 2015 at 12:34:07PM -0700, Andy Zhou wrote:
> Acked-by: Andy Zhou
Thanks, I applied this to master.
> A question in line.
>
> On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
> > This will acquire its first user in an upcoming commit.
> >
> > Signed-off-by: Ben Pfaff
> > ---
>
On Wed, Aug 26, 2015 at 10:42:34PM +, Nithin Raju wrote:
> > On Aug 26, 2015, at 2:37 PM, Sairam Venugopal wrote:
> >
> > Signed-off-by: Sairam Venugopal
>
> Thanks for the fix.
>
> Acked-by: Nithin Raju
Thanks, applied to master and branch-2.4.
__
This is not a review, since I am not that familiar with this code yet.
Just a few questions
Should compare_cmps_3way() be of type 'int' instead of bool?
Should crash_or() also make use of disjunction_matches_string() in
case of string?
On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
> The exp
A few questions in line:
On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> ovn/lib/expr.c | 19 +++
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/ovn/lib/expr.c b/ovn/lib/expr.c
> index 510a15e..20dd9c5 100644
> --- a/ovn/
> On Aug 26, 2015, at 1:08 PM, Ben Pfaff wrote:
>
> On Fri, Aug 21, 2015 at 03:25:23PM -0700, Jarno Rajahalme wrote:
>> mf_mask_field_and_prereqs() used to memset a static variable again and
>> again. Now that mf_value is larger (due to tun_metadata field), this
>> is more expensive. Avoid thi
> On Aug 26, 2015, at 1:09 PM, Ben Pfaff wrote:
>
> On Wed, Aug 26, 2015 at 01:04:24PM -0700, Jarno Rajahalme wrote:
>>
>>> On Aug 26, 2015, at 12:46 PM, Ben Pfaff wrote:
>>>
>>> On Fri, Aug 21, 2015 at 03:25:22PM -0700, Jarno Rajahalme wrote:
Tunnel metadata has grown large since the ad
> On Aug 26, 2015, at 2:40 PM, Ben Pfaff wrote:
>
> On Fri, Aug 21, 2015 at 03:25:21PM -0700, Jarno Rajahalme wrote:
>> Classifier partitions allowed skipping subtables when if was known
>> from the flow's metadata field that the subtable cannot possibly
>> match. This functionality was later i
> On Aug 26, 2015, at 10:06 AM, Ben Pfaff wrote:
>
> On Fri, Aug 21, 2015 at 03:25:20PM -0700, Jarno Rajahalme wrote:
>> Struct miniflow is now sometimes used just as a map. Define a new
>> struct flowmap for that purpose. The flowmap is defined as an array of
>> maps, and it is automatically
> On Aug 26, 2015, at 9:06 AM, Ben Pfaff wrote:
>
> On Fri, Aug 21, 2015 at 03:25:19PM -0700, Jarno Rajahalme wrote:
>> This makes stage mask computation happen only when a subtable is
>> inserted and allows simplification of the main lookup function.
>>
>> Classifier benchmark shows that this
Well, if so, that makes life easy ;-)
On Wed, Aug 26, 2015 at 03:07:34PM -0700, Andy Zhou wrote:
> Can we assume ovsdb-server will be launched by the process that has
> the right owner and group?
>
> On Wed, Aug 26, 2015 at 2:19 PM, Ben Pfaff wrote:
> > I guess the question is from what users an
Thanks. I tested this and realized that it introduced a bug, and that
there was another place to fix too, so I sent a v2:
http://openvswitch.org/pipermail/dev/2015-August/059256.html
On Wed, Aug 26, 2015 at 02:49:11PM -0700, Alex Wang wrote:
> Thx for informing the portability issue.
>
>
"sed -i" isn't entirely portable, and we can avoid it by using the
argument to check_logs as intended.
Signed-off-by: Ben Pfaff
Acked-by: Alex Wang
---
v1->v2: Retain $1 in check_logs call from OVN_CONTROLLER_VTEP_STOP.
Also drop other, now-unneeded, call to "sed -i".
tests/ovn-controller-vt
> On Aug 26, 2015, at 2:37 PM, Sairam Venugopal wrote:
>
> Signed-off-by: Sairam Venugopal
Thanks for the fix.
Acked-by: Nithin Raju
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Can we assume ovsdb-server will be launched by the process that has
the right owner and group?
On Wed, Aug 26, 2015 at 2:19 PM, Ben Pfaff wrote:
> I guess the question is from what users and groups should ovsdb-server
> accept connections. If it's only supposed to accept connections from
> root
Thx for informing the portability issue.
I used 'sed -i' since I only saw this in ovsdb-server.log, so originally,
did not
want to miss same error logs in other *.log file,
But, since 'sed -i' is not portable, I think it is good to do it in
check_log()
Acked-by: Alex Wang
On Wed, Aug 26, 2015
The previous systemd units for OVN had systemd create a rundir for
each daemon (ovn-northd and ovn-controller). This "worked", in that
the services did start successfully. However, they didn't actually
work. In practice, both services make use of files in the main ovs
rundir, so just run everyth
Signed-off-by: Russell Bryant
---
rhel/automake.mk| 1 +
rhel/openvswitch-fedora.spec.in | 10 +-
rhel/usr_lib_systemd_system_ovn-controller-vtep.service | 12
3 files changed, 22 insertions(+), 1 deletion(-)
c
Most real deployments will need to customize the database locations
for ovn-controller and ovn-controller-vtep. Instead of making them
override the entire command used to start the daemons, provide and
document some environment variables that can be overridden in a custom
config file.
Signed-off-
These patches make the OVN packages for Fedora actually work (oops) and make it
easier to override the db locations for the OVN and VTEP databases.
Russell Bryant (3):
rhel: Use same rundir for ovs and ovn.
rhel: Add systemd suport for ovn-controller-vtep.
rhel: Add variables for OVN and VTE
"sed -i" isn't entirely portable, and we can avoid it by using the
argument to check_logs as intended.
CC: Alex Wang
Signed-off-by: Ben Pfaff
---
tests/ovn-controller-vtep.at | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tests/ovn-controller-vtep.at b/tests/ovn-controlle
On Fri, Aug 21, 2015 at 03:25:21PM -0700, Jarno Rajahalme wrote:
> Classifier partitions allowed skipping subtables when if was known
> from the flow's metadata field that the subtable cannot possibly
> match. This functionality was later implemented in a more general
> fashion by staged lookup, w
On Wed, Aug 26, 2015 at 01:59:39PM -0700, Jarno Rajahalme wrote:
>
> > On Aug 26, 2015, at 1:41 PM, Ben Pfaff wrote:
> >
> > On Wed, Aug 26, 2015 at 01:31:25PM -0700, Jarno Rajahalme wrote:
> >>
> >>> On Aug 26, 2015, at 10:15 AM, Ben Pfaff wrote:
> >>>
> >>> On Fri, Aug 21, 2015 at 03:25:21P
Signed-off-by: Sairam Venugopal
---
datapath-windows/ovsext/Datapath.c| 10 +++---
datapath-windows/ovsext/Flow.c| 17 -
datapath-windows/ovsext/Netlink/Netlink.c | 18 ++
datapath-windows/ovsext/Netlink/Netlink.h | 6 --
datapath-windo
I guess the question is from what users and groups should ovsdb-server
accept connections. If it's only supposed to accept connections from
root or a specified group, then it needs to have the correct privileges
to at least create a root-owned socket or a socket with that group.
On Fri, Aug 21, 2
On Fri, Aug 21, 2015 at 11:10:32PM -0700, Alex Wang wrote:
> This commit adds a new key-value pair, 'punix_file_group=',
> to the 'other_config' column in the 'Controller' table. This new config
> allows user to change the punix socket file's group ownership, so that
> non-root process can also co
> On Aug 26, 2015, at 1:41 PM, Ben Pfaff wrote:
>
> On Wed, Aug 26, 2015 at 01:31:25PM -0700, Jarno Rajahalme wrote:
>>
>>> On Aug 26, 2015, at 10:15 AM, Ben Pfaff wrote:
>>>
>>> On Fri, Aug 21, 2015 at 03:25:21PM -0700, Jarno Rajahalme wrote:
Classifier partitions allowed skipping subta
> On Aug 25, 2015, at 9:37 PM, Ben Pfaff wrote:
>
> +/* Implementation of crush_cmps() for expr->type == EXPR_T_OR. */
> static struct expr *
> crush_or(struct expr *expr, const struct expr_symbol *symbol)
Since you added this for crush_or(), do you want to do the same for crush_and().
> -/* C
On Fri, Aug 21, 2015 at 11:10:31PM -0700, Alex Wang wrote:
> This commit relaxes the whitelist format for punix path for
> service controller. Instead of only allowing
> punix:/.controller, the new format
> allows any suffix, like punix:/.*.
>
> Signed-off-by: Alex Wang
I think there's still a
On Wed, Aug 26, 2015 at 01:31:25PM -0700, Jarno Rajahalme wrote:
>
> > On Aug 26, 2015, at 10:15 AM, Ben Pfaff wrote:
> >
> > On Fri, Aug 21, 2015 at 03:25:21PM -0700, Jarno Rajahalme wrote:
> >> Classifier partitions allowed skipping subtables when if was known
> >> from the flow's metadata fie
> On Aug 26, 2015, at 10:15 AM, Ben Pfaff wrote:
>
> On Fri, Aug 21, 2015 at 03:25:21PM -0700, Jarno Rajahalme wrote:
>> Classifier partitions allowed skipping subtables when if was known
>> from the flow's metadata field that the subtable cannot possibly
>> match. This functionality was later
lQÉ,´?±(ý«"t˼alÄØ¥
f,;ÆQ¨1ô¬¢$Èð]HN<» Çt¥#ÓÁ4M|¥Ñ¨*³k®Ëk&¦êñ0¨cè
ÎàÙá-|'À.¡gX3ÄÈ
jKùÊUP¡m¿Üô̱ten·
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Aug 26, 2015 at 01:04:24PM -0700, Jarno Rajahalme wrote:
>
> > On Aug 26, 2015, at 12:46 PM, Ben Pfaff wrote:
> >
> > On Fri, Aug 21, 2015 at 03:25:22PM -0700, Jarno Rajahalme wrote:
> >> Tunnel metadata has grown large since the addition of Geneve options.
> >> Copying tunnel metadata f
On Fri, Aug 21, 2015 at 03:25:23PM -0700, Jarno Rajahalme wrote:
> mf_mask_field_and_prereqs() used to memset a static variable again and
> again. Now that mf_value is larger (due to tun_metadata field), this
> is more expensive. Avoid this by using static initialization.
>
> mf_mask_field_and_p
> On Aug 26, 2015, at 12:46 PM, Ben Pfaff wrote:
>
> On Fri, Aug 21, 2015 at 03:25:22PM -0700, Jarno Rajahalme wrote:
>> Tunnel metadata has grown large since the addition of Geneve options.
>> Copying tunnel metadata for performing a lookup is not necessary.
>> Change recirc_metadata to use a p
Hi,
Currently for normal rule when destination address is not found in mac table
packet is flooded to all ports connected to bridge except incoming port. So if
incoming traffic is coming with high rate then flooding rate is maximum which
reduced performance of the sysyem. So is there any func
On Fri, Aug 21, 2015 at 03:25:22PM -0700, Jarno Rajahalme wrote:
> Tunnel metadata has grown large since the addition of Geneve options.
> Copying tunnel metadata for performing a lookup is not necessary.
> Change recirc_metadata to use a pointer to struct flow_tnl, and only
> copy the tunnel metad
Acked-by: Andy Zhou
A question in line.
On Tue, Aug 25, 2015 at 9:37 PM, Ben Pfaff wrote:
> This will acquire its first user in an upcoming commit.
>
> Signed-off-by: Ben Pfaff
> ---
> lib/sset.c | 14 ++
> lib/sset.h | 3 +++
> 2 files changed, 17 insertions(+)
>
> diff --git a/
I agree this should move to the OpenStack mailing list as this related to
it.
I agree with Ben and Russell here and don't think any of this needs to be
in OVN,
I am not that familiar with that effort, but there is an effort in Nova to
provide generic VIF binding
library (https://github.com/jaypipe
On 08/26/2015 11:04 AM, Chandran, Sugesh wrote:
> HI Russel,
>
> Please find the work flow as below.
> 1.A user requests nova to boot a vm
> 2.Nova schedules the vm based on the flavor and image properties to a host
> 3.Nova asks neutron to bind the neutron port basing the selected hos
Hi,
I've checked it further, based on Stephen's suggestion I've tried perf
top as well. The results were the same, it spends a lot of time in that
part of the code, and there are high number of branch load misses
(BR_MISS_PRED_RETIRED) around there too.
I've also started to strip down miniflow
On Fri, Aug 21, 2015 at 03:25:21PM -0700, Jarno Rajahalme wrote:
> Classifier partitions allowed skipping subtables when if was known
> from the flow's metadata field that the subtable cannot possibly
> match. This functionality was later implemented in a more general
> fashion by staged lookup, w
Hi,
On 24/08/15 12:43, Traynor, Kevin wrote:
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Zoltan Kiss
Sent: Friday, August 21, 2015 7:05 PM
To: d...@dpdk.org; dev@openvswitch.org
Cc: Richardson, Bruce; Ananyev, Konstantin
Subject: [ovs-dev] OVS-DPDK pe
On Fri, Aug 21, 2015 at 03:25:20PM -0700, Jarno Rajahalme wrote:
> Struct miniflow is now sometimes used just as a map. Define a new
> struct flowmap for that purpose. The flowmap is defined as an array of
> maps, and it is automatically sized according to the size of struct
> flow, so it will be
Hi, Ben.
It sounds good.
Thank you for kind reply.
I will try.
Thanks,
Junguk
2015-08-26 12:54 GMT-04:00 Ben Pfaff :
> OK.
>
> OVS using DPDK can connect physical machines without using the special
> drivers for VM support. I don't know whether the OVS build requires
> them to be present.
>
>
OK.
OVS using DPDK can connect physical machines without using the special
drivers for VM support. I don't know whether the OVS build requires
them to be present.
OpenFlow controllers should be able to control OVS with DPDK in the same
way as OVS without DPDK.
On Wed, Aug 26, 2015 at 12:52:30PM
I am sorry that I am not clear about term.
My question is about OVS.
Thanks,
Junguk
2015-08-26 12:49 GMT-04:00 Ben Pfaff :
> On Wed, Aug 26, 2015 at 12:47:39PM -0400, Junguk Cho wrote:
> > In OVS 2.4 version, it includes DPDK as mainstream since there is
> > INSTALL.DPDK.md.
> > Is it different
On Wed, Aug 26, 2015 at 12:47:39PM -0400, Junguk Cho wrote:
> In OVS 2.4 version, it includes DPDK as mainstream since there is
> INSTALL.DPDK.md.
> Is it different from OVDK?
> In other words, OVS + DPDK in version OVS 2.4 is not OVDK?
OVDK is Intel's fork. Open vSwitch is just Open vSwitch or "
Hi.
In OVS 2.4 version, it includes DPDK as mainstream since there is
INSTALL.DPDK.md.
Is it different from OVDK?
In other words, OVS + DPDK in version OVS 2.4 is not OVDK?
Thanks,
Junguk
2015-08-26 12:43 GMT-04:00 Ben Pfaff :
> On Wed, Aug 26, 2015 at 12:32:37PM -0400, Junguk Cho wrote:
> > Do
On 26 August 2015 at 09:04, Ben Pfaff wrote:
> On Fri, Aug 21, 2015 at 03:25:18PM -0700, Jarno Rajahalme wrote:
>> Reserve the use of size_t for sizes of objects in bytes.
>>
>> Suggested-by: Joe Stringer
>> Signed-off-by: Jarno Rajahalme
>
> I think that size_t is a perfectly appropriate type f
On Wed, Aug 26, 2015 at 12:32:37PM -0400, Junguk Cho wrote:
> Does OVDK works as a switch to connect physical machines?
> In this case, we do not include compile option to use IVSHEM and vHost
> function.
> Is it right?
>
> Can we use SDN controller like ryu, pox for controlling OVDK?
I think tha
Hi,
Does OVDK works as a switch to connect physical machines?
In this case, we do not include compile option to use IVSHEM and vHost
function.
Is it right?
Can we use SDN controller like ryu, pox for controlling OVDK?
Thanks,
Junguk
2015-08-25 18:20 GMT-04:00 Ben Pfaff :
> On Tue, Aug 25, 201
On Fri, Aug 21, 2015 at 03:25:19PM -0700, Jarno Rajahalme wrote:
> This makes stage mask computation happen only when a subtable is
> inserted and allows simplification of the main lookup function.
>
> Classifier benchmark shows that this speeds up the classification
> (with wildcards) about 5%.
>
On Fri, Aug 21, 2015 at 03:25:18PM -0700, Jarno Rajahalme wrote:
> Reserve the use of size_t for sizes of objects in bytes.
>
> Suggested-by: Joe Stringer
> Signed-off-by: Jarno Rajahalme
I think that size_t is a perfectly appropriate type for the number of
elements in an array or an array inde
When dpdk configuration changes, all pmd threads are recreated
and rx queues of each port are reloaded. After this process,
rx queue could be mapped to a different pmd thread other than
the one before reconfiguration. However, this is totally
transparent to ofproto layer modules. So, if the ofpr
On Wed, Aug 26, 2015 at 11:35:30AM -0400, Russell Bryant wrote:
> On 08/26/2015 11:11 AM, Ben Pfaff wrote:
> > Reported-by: Russell Bryant
> > Signed-off-by: Ben Pfaff
> > ---
> > python/build/nroff.py | 11 ---
> > 1 file changed, 8 insertions(+), 3 deletions(-)
>
> Tested against my c
Sorry for the delayed reply,
I tested this patch yesterday, found that this is not right fix,
Turns out, dp_netdev_reset_pmd_threads() just creates (or restores) pmd
threads... and we call dp_netdev_destroy_all_pmds() at the beginning of the
function. So, when purge_cb() is called, the flows in
On 08/26/2015 11:11 AM, Ben Pfaff wrote:
> Reported-by: Russell Bryant
> Signed-off-by: Ben Pfaff
> ---
> python/build/nroff.py | 11 ---
> 1 file changed, 8 insertions(+), 3 deletions(-)
Tested against my comment that broke the ovn-architecture doc and it
works now.
Acked-by: Russell
Reported-by: Russell Bryant
Signed-off-by: Ben Pfaff
---
python/build/nroff.py | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/python/build/nroff.py b/python/build/nroff.py
index 1611c49..537bfd5 100644
--- a/python/build/nroff.py
+++ b/python/build/nroff.py
@@ -9
On Wed, Aug 26, 2015 at 09:10:32AM -0400, Russell Bryant wrote:
> On 08/25/2015 02:11 PM, Russell Bryant wrote:
> > ovn-controller previously required the integration bridge to be
> > created before running ovn-controller. This patch makes
> > ovn-controller automatically create it if it doesn't a
Introduce a new logical port type called "localnet". A logical port
with this type also has an option called "network_name". A "localnet"
logical port represents a connection to a network that is locally
accessible from each chassis running ovn-controller. ovn-controller
will use the ovn-bridge-
v6->v7
- Fix xml2nroff error on doc addition in patch 2.
- Fix possible use of uninitialized variable pointed out by clang in patch 1.
v5->v6
- Rebase on OVN tunnel key changes
- Add example usage for "localnet" logical port type to commit message
Russell Bryant (3):
ovn: Add bridge mapping
ovn-controller previously required the integration bridge to be
created before running ovn-controller. This patch makes
ovn-controller automatically create it if it doesn't already exist.
Signed-off-by: Russell Bryant
---
ovn/controller/ovn-controller.c | 65 ++--
Add a new OVN configuration entry in the Open_vSwitch database called
"ovn-bridge-mappings". This allows the configuration of mappings
between a physical network name and an OVS bridge that provides
connectivity to that network.
For example, if you wanted to configure "physnet1" to map to "br-eth
HI Russel,
Please find the work flow as below.
1. A user requests nova to boot a vm
2. Nova schedules the vm based on the flavor and image properties to a host
3. Nova asks neutron to bind the neutron port basing the selected host
name and some limited information in binding profil
On 08/26/2015 10:58 AM, Ben Pfaff wrote:
> On Tue, Aug 25, 2015 at 02:11:44PM -0400, Russell Bryant wrote:
>> ovn-controller previously required the integration bridge to be
>> created before running ovn-controller. This patch makes
>> ovn-controller automatically create it if it doesn't already e
On Tue, Aug 25, 2015 at 02:11:44PM -0400, Russell Bryant wrote:
> ovn-controller previously required the integration bridge to be
> created before running ovn-controller. This patch makes
> ovn-controller automatically create it if it doesn't already exist.
>
> Signed-off-by: Russell Bryant
Cla
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On 08/25/2015 02:11 PM, Russell Bryant wrote:
> ovn-controller previously required the integration bridge to be
> created before running ovn-controller. This patch makes
> ovn-controller automatically create it if it doesn't already exist.
>
> Signed-off-by: Russell Bryant
> ---
> ovn/controlle
Update relevant artifacts to add support for DPDK v2.1.0
- INSTALL.DPDK.md
- acinclude.m4: Change DPDK library name
- netdev-dpdk: Add 16 bytes extra padding to mbuf size to adapt to DPDK bug
fix that changes the treatment of the requested mbuf size
- build.sh: Change DPDK version number
No
The message was undeliverable due to the following reason:
Your message was not delivered because the destination server was
not reachable within the allowed queue period. The amount of time
a message is queued before it is returned depends on local configura-
tion parameters.
Most likely there i
,M}Û±lU*]q×ÖLó/%ÖïWvƬ0#O
7jà ~¹Ôòí`3xkWÍéÓr²?.ÞÎI§»Ò§ôÀNñI:ºn
õq{[«õ[foFúî§óXí5FPaÒ°æõjñ,ãVÑ¥óñ«ÄpôrËÛdUXÍw 9ÀBEÃÚ¯\\C¢w?uk-dkõòÁå
¯4×Ê¡¯ï}·xÎuX¼iéñkͲT±NDt«£ÒÝÂy.¨i.óõVÏ°ûÀ4
ù
J6hó7y0¬ùÃU>ÐèÚÑ)Çu¶R¢ü-Õt.x[ôËz¤ÜÄ×!åÄCÞ
d¹Öz:;òxÍæqä?Ù'iѯâ¨-Lºé¦ ¹"³
7ò4]uëf½
98 matches
Mail list logo
