I don't fully get what's going on here, but the removal looks correct.
Assuming it allows builds to complete, it's better than what was there. :-)
--Justin
On Jun 28, 2013, at 8:56 PM, Ben Pfaff wrote:
> XenServer builds are failing because of link errors reporting that
> __sync_fetch_and__
Thank you for the reviews. I applied patches 1, 2, and 4 to master.
On Sat, Jun 29, 2013 at 11:25:01AM +0900, Joe Stringer wrote:
> Looks good.
>
> On Sat, Jun 29, 2013 at 8:27 AM, Ben Pfaff wrote:
> > Signed-off-by: Ben Pfaff
> > ---
> > OPENFLOW-1.1+ | 12
> > 1 files changed
On Sat, Jun 29, 2013 at 10:14:16AM +0800, ZhengLingyun wrote:
> /* Ensure that the last "unsigned long" in the bitmap only has as many
> * 1-bits as there actually should be. */
> bitmap[n_longs - 1] = (1UL << (n_bits % BITMAP_ULONG_BITS)) - 1;
> from bitmap_allocate1(),
> It will fail when n_
XenServer builds are failing because of link errors reporting that
__sync_fetch_and__ were not found, for in add/sub/and/xor/or
and in 1/2/4/8. We're not actually using these RMW operations yet,
so as a stopgap measure just drop the tests.
The correct long-term fix is probably to do Autoconf li
Looks good.
On Sat, Jun 29, 2013 at 8:27 AM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> OPENFLOW-1.1+ | 12
> 1 files changed, 0 insertions(+), 12 deletions(-)
>
> diff --git a/OPENFLOW-1.1+ b/OPENFLOW-1.1+
> index fb39873..d08a46e 100644
> --- a/OPENFLOW-1.1+
> +++ b/OP
Looks good to me.
On Sat, Jun 29, 2013 at 8:27 AM, Ben Pfaff wrote:
> Other test names use a space here, so this one should also.
>
> Signed-off-by: Ben Pfaff
> ---
> tests/ofproto.at |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/tests/ofproto.at b/tests/ofproto
Looks good to me.
On Sat, Jun 29, 2013 at 8:27 AM, Ben Pfaff wrote:
> This corrects "version 0x03and earlier" to "version 0x03 and earlier".
>
> Signed-off-by: Ben Pfaff
> ---
> lib/vconn.c |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/lib/vconn.c b/lib/vconn.c
/* Ensure that the last "unsigned long" in the bitmap only has as many
* 1-bits as there actually should be. */
bitmap[n_longs - 1] = (1UL << (n_bits % BITMAP_ULONG_BITS)) - 1;
from bitmap_allocate1(),
It will fail when n_bits % BITMAP_ULONG_BITS == 0, does it matter?
unsigned long *
bitmap_a
Added appctl commands to override the bfd forwarding status. Values of
true/false/normal are allowed. When set to normal, the bfd forwarding status
is left unchanged. Else, the forwarding status is set to the specified
value - true/false.
Signed-off-by: Pavithra Ramesh
---
lib/bfd.c | 47 +
Thanks for the reviews. I pushed both patches to the relevant branches.
--Justin
On Jun 28, 2013, at 6:31 PM, Ethan Jackson wrote:
> Looks good to me thanks.
>
> Acked-by: Ethan Jackson
>
>
> On Fri, Jun 28, 2013 at 6:26 PM, Justin Pettit wrote:
>> Add new "dpif/disable-megaflows" and "d
Looks good to me thanks.
Acked-by: Ethan Jackson
On Fri, Jun 28, 2013 at 6:26 PM, Justin Pettit wrote:
> Add new "dpif/disable-megaflows" and "dpif/enable-megaflows" commands to
> ovs-appctl to disable and enable megaflows, respectively. By default,
> megaflows are enabled, and these commands
On Jun 28, 2013, at 5:37 PM, Ethan Jackson wrote:
> Please put a period after the Feature number. Not sure why, but
> that's been the convention.
Okay.
> Instead of calling the variable "disable_megaflows", I think the code
> would be a bit easier to read if we called it "megaflows" and
> ini
Add new "dpif/disable-megaflows" and "dpif/enable-megaflows" commands to
ovs-appctl to disable and enable megaflows, respectively. By default,
megaflows are enabled, and these commands should only be used for
debugging.
Feature #18265.
Requested-by: Ronghua Zhang
Signed-off-by: Justin Pettit
-
On Fri, Jun 28, 2013 at 4:07 PM, Pravin B Shelar wrote:
> Openvswitch uses function from NET_IPGRE_DEMUX module.
> Add Kconfig dependency to fix following compilation errors:
> http://marc.info/?l=linux-netdev&m=137244035226634
>
> CC: Jesse Gross
> Reported-by: Randy Dunlap
> Signed-off-by: Pra
Please put a period after the Feature number. Not sure why, but
that's been the convention.
Instead of calling the variable "disable_megaflows", I think the code
would be a bit easier to read if we called it "megaflows" and
initialize it to true. Negative flags are a bit confusing.
I don't thin
Acked-by: Ethan Jackson
On Fri, Jun 28, 2013 at 5:21 PM, Justin Pettit wrote:
> When determinining what fields to un-wildcard for the symmetric L4 hash,
> don't include the IPv6 address fields if the packet isn't IPv6.
>
> Reported-by: Jarno Rajahalme
> Signed-off-by: Justin Pettit
> ---
> li
This patch doesn't apply to master. Please be sure to rebase it
before sending the next version.
> static bool bfd_in_poll(const struct bfd *);
> @@ -194,6 +195,8 @@ static uint32_t generate_discriminator(void);
> static void bfd_put_details(struct ds *, const struct bfd *);
> static void bfd_
When determinining what fields to un-wildcard for the symmetric L4 hash,
don't include the IPv6 address fields if the packet isn't IPv6.
Reported-by: Jarno Rajahalme
Signed-off-by: Justin Pettit
---
lib/flow.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lib/flow.c
Add new "dpif/disable-megaflows" and "dpif/enable-megaflows" commands to
ovs-appctl to disable and enable megaflows, respectively. By default,
megaflows are enabled, and these commands should only be used for
debugging.
Feature #18265
Requested-by: Ronghua Zhang
Signed-off-by: Justin Pettit
--
Jesse,
On 06/29/2013 12:44 AM, Jesse Gross wrote:
Pravin pointed out that the RHEL6 openvswitch_handle_frame_hook
unregister is happening from the RCU callback, which means that we
immediately free the vport after removing the handler. This seems
racy. I also noticed that we have asymmetric use
Signed-off-by: Ben Pfaff
---
FAQ| 16 +++---
NEWS |2 +
lib/ofp-util.c | 125 +++
lib/ofp-util.h | 19 ++-
tests/learn.at |2 +-
tests/ofp-print.at |
Signed-off-by: Ben Pfaff
---
DESIGN | 120 +--
1 files changed, 116 insertions(+), 4 deletions(-)
diff --git a/DESIGN b/DESIGN
index f3e9309..5dd6b89 100644
--- a/DESIGN
+++ b/DESIGN
@@ -84,8 +84,8 @@ OFPP_LOCAL as a physical port and
Signed-off-by: Ben Pfaff
---
OPENFLOW-1.1+ | 12
1 files changed, 0 insertions(+), 12 deletions(-)
diff --git a/OPENFLOW-1.1+ b/OPENFLOW-1.1+
index fb39873..d08a46e 100644
--- a/OPENFLOW-1.1+
+++ b/OPENFLOW-1.1+
@@ -62,10 +62,6 @@ probably incomplete.
so the value is not ne
Signed-off-by: Ben Pfaff
---
tests/ofproto.at | 22 --
1 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/tests/ofproto.at b/tests/ofproto.at
index 53d5cad..5fc33d6 100644
--- a/tests/ofproto.at
+++ b/tests/ofproto.at
@@ -423,7 +423,7 @@ NXST_FLOW reply:
OVS_VS
Other test names use a space here, so this one should also.
Signed-off-by: Ben Pfaff
---
tests/ofproto.at |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/tests/ofproto.at b/tests/ofproto.at
index b3823a3..53d5cad 100644
--- a/tests/ofproto.at
+++ b/tests/ofproto.at
@@ -
Signed-off-by: Ben Pfaff
---
lib/learn.c|3 ++-
lib/ofp-parse.c|2 ++
lib/ofp-util.c |3 +++
lib/ofp-util.h | 41 +
ofproto/ofproto-dpif.c |1 +
ofproto/ofproto.c |2 +-
utilities/ovs-ofctl.c |
This patch series adds basic support for the OpenFlow 1.1 protocol.
(Previously, 1.2 and 1.3 have been supported, but not 1.1.) This
is the second or third time I've sent this series; it has never
attracted a review.
Ben Pfaff (7):
vconn: Fix formatting of user message.
tests: Fix typo in tes
This corrects "version 0x03and earlier" to "version 0x03 and earlier".
Signed-off-by: Ben Pfaff
---
lib/vconn.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/lib/vconn.c b/lib/vconn.c
index 449a36e..15ac119 100644
--- a/lib/vconn.c
+++ b/lib/vconn.c
@@ -461,7 +461,7 @
On 06/28/13 16:07, Pravin B Shelar wrote:
> Openvswitch uses function from NET_IPGRE_DEMUX module.
> Add Kconfig dependency to fix following compilation errors:
> http://marc.info/?l=linux-netdev&m=137244035226634
>
> CC: Jesse Gross
> Reported-by: Randy Dunlap
> Signed-off-by: Pravin Shelar
A
> What I expected to see was that each compose_output_action() would add
> the mirrors for that output to a set of mirrors to ultimately use, and
> that somewhere else (just one place) we would add the mirrors for the
> input to this same set, and that at the end we would use the set of
> mirrors t
Added appctl commands to override the bfd forwarding status. Values of
true/false/normal are allowed. When set to normal, the bfd forwarding status
is left unchanged. Else, the forwarding status is set to the specified
value - true/false.
---
lib/bfd.c | 46 ++
Openvswitch uses function from NET_IPGRE_DEMUX module.
Add Kconfig dependency to fix following compilation errors:
http://marc.info/?l=linux-netdev&m=137244035226634
CC: Jesse Gross
Reported-by: Randy Dunlap
Signed-off-by: Pravin Shelar
---
v2-v3:
- Added openvswitch gre config option.
- add
On Fri, Jun 28, 2013 at 03:34:26PM -0700, Ethan Jackson wrote:
> Should the multithreaded bool be atomic? I doubt it matters, but
> perhaps if pthread_create()'s implementation is odd, we could have a
> problem.
>
> Acked-by: Ethan Jackson
Thanks for the review.
I don't think it's necessary fo
Thanks for the review.
On Fri, Jun 28, 2013 at 03:23:24PM -0700, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
> On Wed, Jun 19, 2013 at 1:17 PM, Ben Pfaff wrote:
> > pthread_once() is portable but it does not allow passing any parameters to
> > the initialization function, which is often in
> I'm not really happy with adding another hash lookup along this path,
> but we can deal with that in profiling later.
It's just a pointer hash, so I doubt it will matter, but if it does we
can always pass it as an argument to tnl_port_send(). It's caller
could cash it trivially.
Thanks for the
On Fri, Jun 28, 2013 at 03:20:49PM -0700, Andy Zhou wrote:
> ofproto/trace command was not updated to take mega flow input. Only
> information provided in the key fields are used for tracing.
>
> Signed-off-by: Andy Zhou
Applied to master and branch-1.11.
It wasn't obvious at first why this wou
Thomas,
Pravin pointed out that the RHEL6 openvswitch_handle_frame_hook
unregister is happening from the RCU callback, which means that we
immediately free the vport after removing the handler. This seems
racy. I also noticed that we have asymmetric use of
netdev_rx_handler_register/unregister (I
> Oh, that was a note to try to implement the C11 "kill_dependency()"
> function. I don't think I can, but I don't think it matters. Mind if
> I just delete the comment?
Sounds good to me.
>> In ovs-atomic.h is it fair to assume that sparse supports c11
>> features? If not, does it make ense t
Should the multithreaded bool be atomic? I doubt it matters, but
perhaps if pthread_create()'s implementation is odd, we could have a
problem.
Acked-by: Ethan Jackson
On Wed, Jun 19, 2013 at 1:17 PM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> lib/command-line.c|4 ++-
> lib
On Fri, Jun 28, 2013 at 03:14:36PM -0700, Ethan Jackson wrote:
> Thanks for writing up something so thorough. Better to get it done
> now then jerry rig stuff ad hoc as we need it.
>
> There are a couple of "XXX kill_dependency" comments in the code.
> Could you expand the comment a bit to explai
Acked-by: Ethan Jackson
On Wed, Jun 19, 2013 at 1:17 PM, Ben Pfaff wrote:
> pthread_once() is portable but it does not allow passing any parameters to
> the initialization function, which is often inconvenient, because it means
> that the function can only access data declared at file scope. Th
ofproto/trace command was not updated to take mega flow input. Only
information provided in the key fields are used for tracing.
Signed-off-by: Andy Zhou
---
ofproto/ofproto-dpif.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofpro
Thanks for writing up something so thorough. Better to get it done
now then jerry rig stuff ad hoc as we need it.
There are a couple of "XXX kill_dependency" comments in the code.
Could you expand the comment a bit to explain what it's referring to?
In ovs-atomic.h is it fair to assume that spar
I think that there are some memory leaks in the new meters code. I
don't, for example, see any "free" in ofproto_destroy() to match up
with this xzalloc() in ofproto_create():
ofproto->meters = xzalloc((ofproto->meter_features.max_meters + 1)
* sizeof(struct mete
On Tue, Jun 25, 2013 at 1:08 AM, Rajahalme, Jarno (NSN - FI/Espoo)
wrote:
> I also tried to figure out how the group table should be implemented. I
> haven't done anything for it yet, but I'm rather convinced that the datapath
> should maintain the group table so that there would be no need for
On Fri, Jun 28, 2013 at 10:50 AM, Ben Pfaff wrote:
> On Thu, Jun 27, 2013 at 01:39:52AM +0300, Jarno Rajahalme wrote:
>>
>> Signed-off-by: Jarno Rajahalme
>> ---
>> v2: Remove resetting of provider meter id on failure in dpif_meter_set().
>
> Changes to need Jesse's approval, so I'm handing
> th
On Fri, Jun 28, 2013 at 2:00 PM, Randy Dunlap wrote:
> On 06/28/13 12:39, Pravin B Shelar wrote:
>> Openvswitch uses function from NET_IPGRE_DEMUX module.
>> Add Kconfig dependency to fix following compilation errors:
>> http://marc.info/?l=linux-netdev&m=137244035226634
>>
>> CC: Jesse Gross
>>
This code has been kindly contributed by Centec Networks. I have
spent only a few minutes going over it, so it needs futher review,
but I thought I'd put it out to avoid possibly duplicating work
with others in the community (as happened with the "meters" code
a little while back).
Ben Pfaff (1):
This is the convention that I've meant to maintain, but missed a number of
messages. This corrects them.
Signed-off-by: Ben Pfaff
---
lib/learning-switch.c | 28 ++--
lib/ofp-msgs.h| 28 ++--
lib/ofp-print.c | 28 ++
On 06/28/13 12:39, Pravin B Shelar wrote:
> Openvswitch uses function from NET_IPGRE_DEMUX module.
> Add Kconfig dependency to fix following compilation errors:
> http://marc.info/?l=linux-netdev&m=137244035226634
>
> CC: Jesse Gross
> Reported-by: Randy Dunlap
> Signed-off-by: Pravin Shelar
W
Thanks, I applied this.
On Fri, Jun 28, 2013 at 1:41 AM, Rajahalme, Jarno (NSN - FI/Espoo)
wrote:
> Now "make modules_install" works for me too, thanks.
>
> Jarno
>
> On Jun 28, 2013, at 3:26 , ext Jesse Gross wrote:
>
>> The Open vSwitch kernel module now has dependencies on symbols
>> exporte
This doesn't constitute a full review, but I have a couple of comments.
Can you describe a realistic situation in which a user configured
their hypervisor to run LACP, the negotiation failed, and the correct
behavior is to forward traffic despite a clear problem in their
network config? At best,
On Fri, Jun 28, 2013 at 10:31:01AM -0700, Ben Pfaff wrote:
> On Fri, Jun 28, 2013 at 07:44:05PM +0300, Jarno Rajahalme wrote:
> > Keep track of usable protocols while parsing, rather than checking for them
> > afterwards. Check the usable protocols for actions as well as matches.
> > This fixes si
Thanks Alex, applied to master.
On Thu, Jun 27, 2013 at 12:58:42PM -0700, Alex Wang wrote:
> Looks good to me,
>
>
> On Wed, Jun 19, 2013 at 1:55 PM, Ben Pfaff wrote:
>
> > OFPGMFC is so nice these constants said it twice.
> >
> > Signed-off-by: Ben Pfaff
> > ---
> > lib/ofp-errors.h |8
Openvswitch uses function from NET_IPGRE_DEMUX module.
Add Kconfig dependency to fix following compilation errors:
http://marc.info/?l=linux-netdev&m=137244035226634
CC: Jesse Gross
Reported-by: Randy Dunlap
Signed-off-by: Pravin Shelar
---
v1-v2:
- use select to resolve dependency.
---
net/op
On Fri, Jun 28, 2013 at 11:54 AM, Ben Hutchings
wrote:
> On Fri, 2013-06-28 at 11:30 -0700, Pravin B Shelar wrote:
>> Openvswitch uses function from NET_IPGRE_DEMUX module.
>> Add Kconfig dependency to fix compilation errors.
>
> That's not an obvious dependency, and this may result in OPENVSWITCH
On Fri, 2013-06-28 at 11:30 -0700, Pravin B Shelar wrote:
> Openvswitch uses function from NET_IPGRE_DEMUX module.
> Add Kconfig dependency to fix compilation errors.
That's not an obvious dependency, and this may result in OPENVSWITCH
being inexplicably disabled when people update their custom ke
On Jun 28, 2013, at 1:00 AM, "Rajahalme, Jarno (NSN - FI/Espoo)"
wrote:
>> +if (flow->dl_type == htons(ETH_TYPE_IP)) {
>> +memset(&wc->masks.nw_src, 0xff, sizeof wc->masks.nw_src);
>> +memset(&wc->masks.nw_dst, 0xff, sizeof wc->masks.nw_dst);
>> +} else {
Openvswitch uses function from NET_IPGRE_DEMUX module.
Add Kconfig dependency to fix compilation errors.
CC: Jesse Gross
Reported-by: Randy Dunlap
Signed-off-by: Pravin Shelar
---
net/openvswitch/Kconfig |1 +
net/openvswitch/vport-gre.c |2 --
net/openvswitch/vport.c |3 --
On Thu, Jun 27, 2013 at 01:39:52AM +0300, Jarno Rajahalme wrote:
>
> Signed-off-by: Jarno Rajahalme
> ---
> v2: Remove resetting of provider meter id on failure in dpif_meter_set().
Changes to need Jesse's approval, so I'm handing
this over to him.
Thanks,
Ben.
___
On Thu, Jun 27, 2013 at 12:52:47PM -0700, Gurucharan Shetty wrote:
> On Thu, Jun 27, 2013 at 10:27 AM, Ben Pfaff wrote:
>
> > On Wed, Jun 26, 2013 at 11:29:36AM -0700, Gurucharan Shetty wrote:
> > > On Wed, Jun 26, 2013 at 10:51 AM, Gurucharan Shetty > >wrote:
> > >
> > > >
> > > >
> > > > On Tu
Otherwise, input with invalid trailing data was accepted, such as input
that had 7 colon-separated segments instead of 6.
Signed-off-by: Ben Pfaff
---
I first sent out this patch on May 8.
lib/meta-flow.c | 21 +
1 files changed, 13 insertions(+), 8 deletions(-)
diff --gi
Signed-off-by: Ben Pfaff
---
I first sent out this patch on May 8.
DESIGN |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/DESIGN b/DESIGN
index f3e9309..0037eea 100644
--- a/DESIGN
+++ b/DESIGN
@@ -185,7 +185,7 @@ Each column is interpreted as follows.
- OF1.0 an
On Fri, Jun 28, 2013 at 07:44:03PM +0300, Jarno Rajahalme wrote:
> Usually the table id in flow mods is 255, which means that goto table
> instruction cannot be checked before the table is picked (for flow add),
> or the rules to be modified are found (flow mod).
>
> Move goto table checking from
On Fri, Jun 28, 2013 at 04:55:19PM +, Rajahalme, Jarno (NSN - FI/Espoo)
wrote:
>
> On Jun 28, 2013, at 19:46 , ext Ben Pfaff wrote:
>
> > On Fri, Jun 28, 2013 at 07:44:04PM +0300, Jarno Rajahalme wrote:
> >> Any protocol version can match dl_type=0x86dd, but only NXM and OXM can
> >> match
Commit bdebeece5 (lacp: Require successful LACP negotiations when
configured.) makes successful LACP negotiation mandatory.
This change makes it configurable. The user has the option to
set other-config:lacp-fallback-slb to true if pre 1.5.0 behavior
is needed. The switch will fallback to balance-s
On Jun 28, 2013, at 19:46 , ext Ben Pfaff wrote:
> On Fri, Jun 28, 2013 at 07:44:04PM +0300, Jarno Rajahalme wrote:
>> Any protocol version can match dl_type=0x86dd, but only NXM and OXM can match
>> IPv6 addresses, or nd_target.
>> Update unit tests to check for this.
>>
>> Signed-off-by: Jarno
On Fri, Jun 28, 2013 at 07:44:04PM +0300, Jarno Rajahalme wrote:
> Any protocol version can match dl_type=0x86dd, but only NXM and OXM can match
> IPv6 addresses, or nd_target.
> Update unit tests to check for this.
>
> Signed-off-by: Jarno Rajahalme
I sent out a similar patch on May 8, will you
Usually the table id in flow mods is 255, which means that goto table
instruction cannot be checked before the table is picked (for flow add),
or the rules to be modified are found (flow mod).
Move goto table checking from decode (ofp-util) to actions checking
(ofp-actions), and postpone the actio
Any protocol version can match dl_type=0x86dd, but only NXM and OXM can match
IPv6 addresses, or nd_target.
Update unit tests to check for this.
Signed-off-by: Jarno Rajahalme
---
lib/ofp-util.c | 13 +
tests/ovs-ofctl.at |6 --
2 files changed, 9 insertions(+), 10 dele
On Fri, Jun 28, 2013 at 09:33:00AM -0700, Ben Pfaff wrote:
> On Fri, Jun 28, 2013 at 04:28:40PM +, Rajahalme, Jarno (NSN - FI/Espoo)
> wrote:
> >
> > On Jun 28, 2013, at 19:09 , ext Ben Pfaff wrote:
> >
> > > On Fri, Jun 28, 2013 at 07:17:09AM +, Rajahalme, Jarno (NSN -
> > > FI/Espoo)
On Jun 28, 2013, at 19:33 , ext Ben Pfaff wrote:
> On Fri, Jun 28, 2013 at 04:28:40PM +, Rajahalme, Jarno (NSN - FI/Espoo)
> wrote:
>>
>> On Jun 28, 2013, at 19:09 , ext Ben Pfaff wrote:
>>
>>> On Fri, Jun 28, 2013 at 07:17:09AM +, Rajahalme, Jarno (NSN - FI/Espoo)
>>> wrote:
>>
On Fri, Jun 28, 2013 at 04:28:40PM +, Rajahalme, Jarno (NSN - FI/Espoo)
wrote:
>
> On Jun 28, 2013, at 19:09 , ext Ben Pfaff wrote:
>
> > On Fri, Jun 28, 2013 at 07:17:09AM +, Rajahalme, Jarno (NSN - FI/Espoo)
> > wrote:
> >>
> >> On Jun 28, 2013, at 1:19 , ext Ben Pfaff wrote:
> >>> o
On Jun 28, 2013, at 19:09 , ext Ben Pfaff wrote:
> On Fri, Jun 28, 2013 at 07:17:09AM +, Rajahalme, Jarno (NSN - FI/Espoo)
> wrote:
>>
>> On Jun 28, 2013, at 1:19 , ext Ben Pfaff wrote:
>>> ofproto: Use another method to find meter_id.
>>
>> OK, with the rationale that actions lists are sh
On Fri, Jun 28, 2013 at 2:22 AM, Jarno Rajahalme
wrote:
> Taken the version checks in datapath/linux/compat/gre.c this seems correct.
>
> Signed-off-by: Jarno Rajahalme
Looks good.
I will apply it soon.
> ---
> datapath/linux/compat/include/net/gre.h |4 ++--
> 1 file changed, 2 insertions(
On Fri, Jun 28, 2013 at 07:17:09AM +, Rajahalme, Jarno (NSN - FI/Espoo)
wrote:
>
> On Jun 28, 2013, at 1:19 , ext Ben Pfaff wrote:
> > ofproto: Use another method to find meter_id.
>
> OK, with the rationale that actions lists are short on average, so
> it does not matter scanning them twic
Thanks Alex, applied.
On Thu, Jun 27, 2013 at 01:02:10PM -0700, Alex Wang wrote:
> Looks good to me,
>
>
> On Wed, Jun 26, 2013 at 8:35 AM, Ben Pfaff wrote:
>
> > Signed-off-by: Ben Pfaff
> > ---
> > vswitchd/bridge.c |5 +
> > 1 file changed, 1 insertion(+), 4 deletions(-)
> >
> > d
Taken the version checks in datapath/linux/compat/gre.c this seems correct.
Signed-off-by: Jarno Rajahalme
---
datapath/linux/compat/include/net/gre.h |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/datapath/linux/compat/include/net/gre.h
b/datapath/linux/compat/inclu
Now "make modules_install" works for me too, thanks.
Jarno
On Jun 28, 2013, at 3:26 , ext Jesse Gross wrote:
> The Open vSwitch kernel module now has dependencies on symbols
> exported by other kernel modules (currently just for GRE). In
> order for it to load, the dependencies must be correct
On Jun 28, 2013, at 4:43 , ext Jesse Gross wrote:
> It's a validation limitation: if you have a vlan packet then you need
> to have an exact match on a corresponding EtherType. If you only want
> to match on the Ethernet header then it's fully maskable.
>
On OpenFlow (& meta-flow etc.) ethertyp
On Jun 28, 2013, at 4:16 , ext Justin Pettit wrote:
...
> diff --git a/lib/flow.c b/lib/flow.c
> index a42fea1..1a5084b 100644
> --- a/lib/flow.c
> +++ b/lib/flow.c
> @@ -781,7 +781,8 @@ flow_hash_symmetric_l4(const struct flow *flow, uint32_t
> basis)
>
> /* Masks the fields in 'wc' that are us
I sent some comments (mostly about comments it seems) earlier you may want to
revisit (or not):
http://openvswitch.org/pipermail/dev/2013-May/027297.html
Jarno
On Jun 28, 2013, at 1:30 , ext Ben Pfaff wrote:
> We have a controller that puts many rules with different metadata values
> into the
On Jun 28, 2013, at 1:19 , ext Ben Pfaff wrote:
> I spent some time looking at and playing with patch 4/5 "v2: ofproto:
> Implement OpenFlow 1.3 meter table." and decided that there was a fair bit
> that I wanted to tweak. This series represent what I came up with:
>
> IMPORTANT: In this series
Thanks, Andy and Ben. I tested and pushed this to all affected branches.
--Justin
On Jun 27, 2013, at 10:02 PM, Andy Zhou wrote:
> ovs-dpctl sometimes displays wildcarded fields as exact match. This
> patch fixes those cases.
>
> This patch implements the following logic. When OVS_FLOW_ATTR_
84 matches
Mail list logo
