> Signed-off-by: Joe Stringer
> ---
> utilities/ovs-dev.py |2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/utilities/ovs-dev.py b/utilities/ovs-dev.py
> index 97fab6c..d979afd 100755
> --- a/utilities/ovs-dev.py
> +++ b/utilities/ovs-dev.py
> @@ -144,7 +144,7 @@ comma
> Make treats tabs very differently from spaces at the beginning of a line,
> so this test must use a tab instead of a space. This partially reverts
> commit a0903134d2d60 (acinclude.m4: Expand tabs).
>
> Without this commit, the build system never enables checking with sparse
> because it never
On 5/19/14, 10:13 AM, YAMAMOTO Takashi wrote:
Signed-off-by: Joe Stringer
---
utilities/ovs-dev.py |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/utilities/ovs-dev.py b/utilities/ovs-dev.py
index 97fab6c..d979afd 100755
--- a/utilities/ovs-dev.py
+++ b/utilities/ovs-de
On 19 May 2014 19:53, Lori Jakab wrote:
> On 5/19/14, 10:13 AM, YAMAMOTO Takashi wrote:
>
>> Signed-off-by: Joe Stringer
>>> ---
>>> utilities/ovs-dev.py |2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/utilities/ovs-dev.py b/utilities/ovs-dev.py
>>> index 97f
> On 19 May 2014 19:53, Lori Jakab wrote:
>
>> On 5/19/14, 10:13 AM, YAMAMOTO Takashi wrote:
>>
>>> Signed-off-by: Joe Stringer
---
utilities/ovs-dev.py |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/utilities/ovs-dev.py b/utilities/ovs-dev.py
>
On Fri, May 16, 2014 at 02:44:40PM -0700, Jarno Rajahalme wrote:
> Do not cache the 'tag' and 'max_priority' in the subtable array. This
> makes later changes to classifier easier.
>
> Also makes the 'cls_subtables*' functions to always leave the
> subtables array in a consistent state. This inc
On Fri, May 16, 2014 at 02:44:41PM -0700, Jarno Rajahalme wrote:
> When, during a classifier lookup, we narrow down to a single potential
> rule, it is enough to match on ("unwildcard") one bit that differs
> between the packet and the rule.
>
> This is a special case of the more general algorithm
On Mon, May 19, 2014 at 04:17:32PM +0900, YAMAMOTO Takashi wrote:
> > Make treats tabs very differently from spaces at the beginning of a line,
> > so this test must use a tab instead of a space. This partially reverts
> > commit a0903134d2d60 (acinclude.m4: Expand tabs).
> >
> > Without this com
On Fri, May 16, 2014 at 03:23:11PM -0700, Alex Wang wrote:
> Acked-by: Alex Wang
>
> Hey Ben, Ethan,
>
> Could either of you help review this?
>
> I think it is ready.
I am taking a look. Alex, thanks for doing multiple rounds of review
already! We're getting to be much more of a team in ter
Looks good but needs a Signed-off-by.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Signed-off-by: Ryan Wilson
Acked-by: Alex Wang
---
lib/ovs-rcu.h |2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/ovs-rcu.h b/lib/ovs-rcu.h
index 4b451b2..8327517 100644
--- a/lib/ovs-rcu.h
+++ b/lib/ovs-rcu.h
@@ -129,6 +129,7 @@
*/
#if __GNUC__
#define OVSRCU_TYPE(TYPE) struct {
Shoot, total brain fart on my part. Just sent out another version of the patch
with the signed off by.
Ryan Wilson
Member of Technical Staff
wr...@vmware.com
3401 Hillview Avenue, Palo Alto, CA
650.427.1511 Office
916.588.7783 Mobile
On May 19, 2014, at 7:59 AM, Ben Pfaff wrote:
> Looks good b
On Mon, May 19, 2014 at 12:09:24AM -0700, Ryan Wilson wrote:
> Signed-off-by: Ryan Wilson
> Acked-by: Alex Wang
Applied, thanks!
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Fri, May 16, 2014 at 06:59:02AM -0700, Ryan Wilson wrote:
> Before, a global read-write lock protected the ofproto-dpif /
> ofproto-dpif-xlate
> interface. Handler and revalidator threads had to wait while configuration was
> being changed. This patch implements RCU locking which allows handler
On Mon, May 19, 2014 at 8:56 AM, Ben Pfaff wrote:
> On Fri, May 16, 2014 at 06:59:02AM -0700, Ryan Wilson wrote:
>> Before, a global read-write lock protected the ofproto-dpif /
>> ofproto-dpif-xlate
>> interface. Handler and revalidator threads had to wait while configuration
>> was
>> being ch
Thanks, applied.
On Fri, May 09, 2014 at 09:11:18AM -0700, Jarno Rajahalme wrote:
> Acked-by: Jarno Rajahalme
>
> On May 8, 2014, at 4:47 PM, Ben Pfaff wrote:
>
> > Signed-off-by: Ben Pfaff
> > ---
> > lib/ovs-rcu.h |2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --g
On May 19, 2014, at 7:48 AM, Ben Pfaff wrote:
> On Fri, May 16, 2014 at 02:44:40PM -0700, Jarno Rajahalme wrote:
>> Do not cache the 'tag' and 'max_priority' in the subtable array. This
>> makes later changes to classifier easier.
>>
>> Also makes the 'cls_subtables*' functions to always leave
On Mon, May 19, 2014 at 10:38:38AM -0700, Jarno Rajahalme wrote:
>
> On May 19, 2014, at 7:48 AM, Ben Pfaff wrote:
>
> > On Fri, May 16, 2014 at 02:44:40PM -0700, Jarno Rajahalme wrote:
> >> Do not cache the 'tag' and 'max_priority' in the subtable array. This
> >> makes later changes to classi
Do not cache the 'tag' and 'max_priority' in the subtable array. This
makes later changes to classifier easier.
Change CLS_SUBTABLES_FOR_EACH to iterate with a regular array index.
Also makes the 'cls_subtables*' functions to always leave the
subtables array in a consistent state. This includes
When, during a classifier lookup, we narrow down to a single potential
rule, it is enough to match on ("unwildcard") one bit that differs
between the packet and the rule.
This is a special case of the more general algorithm, where it is
sufficient to match on enough bits that separates the packet
Hey Ryan,
Still high level comments first,
Could you break the patch into three? (your change about crediting the
group stats, Andy's test, and use RCU for group configuration)
The group stats crediting part of code looks good to me.
The test makes sense.
I haven't reviewed the "rcu for group c
On Fri, May 09, 2014 at 07:35:21PM +0300, Alexandru Copot wrote:
> On Fri, May 9, 2014 at 6:42 PM, Ben Pfaff wrote:
> > On Tue, May 06, 2014 at 09:28:29PM +0300, Alexandru Copot wrote:
> >> Allow port_mod messages in a bundle and apply them to copies
> >> of struct ofport. The final state is commi
ovsdb-client's 'monitor' command works with --detach such
that the parent detaches after printing initial transactions in
the database. This is a little tricky to implement
in windows. So for windows, send the process to background with
'&' and then sleep for a second to let the intial transactions
Otherwise, on windows unit tests data sometimes is
not seen in output files.
Signed-off-by: Gurucharan Shetty
---
utilities/ovs-ofctl.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/utilities/ovs-ofctl.c b/utilities/ovs-ofctl.c
index d62d328..1034e8b 100644
--- a/utilities/ovs-ofctl.c
This is mostly for unit tests that use ovs-ofctl monitor.
service_start() creates a new process and waits till
daemonize_complete() is called by the child.
Signed-off-by: Gurucharan Shetty
---
utilities/ovs-ofctl.c |1 +
1 file changed, 1 insertion(+)
diff --git a/utilities/ovs-ofctl.c b/ut
So far, we log the kernel assigned port number when the port number is
not specified. On Windows, this happens multiple times because "unix"
sockets are implemented internally via TCP ports. This means that many tests,
specially the ovs-ofctl monitor tests, need to filter out the
additional message
Without it, garbage values make it to set_dscp function
in Windows.
Signed-off-by: Gurucharan Shetty
---
vswitchd/bridge.c |1 +
1 file changed, 1 insertion(+)
diff --git a/vswitchd/bridge.c b/vswitchd/bridge.c
index 618a705..6038cf0 100644
--- a/vswitchd/bridge.c
+++ b/vswitchd/bridge.c
@@
We should use Andy's patch for the tests since I didn't write the tests:
http://patchwork.openvswitch.org/patch/3626/
I'll split the group RCU and the crediting stats parts of this commit on my
next version.
Ryan
- Original Message -
From: "Alex Wang"
To: "Ryan Wilson"
Cc: dev
Thanks Ryan, this is a great refactoring.
Looks good to me,
Minor issues:
1. Could you rebase the patch against master? Need to fix some new calls,
added after you posted the patch.
On Wed, May 7, 2014 at 3:14 PM, Ryan Wilson wrote:
> The upcall hander keeps a hash table which hashes flow
On Fri, May 16, 2014 at 8:08 AM, wrote:
> 2014-05-16, Jesse Gross:
>> On Thu, May 15, 2014 at 1:28 AM, wrote:
>>> Hi Jesse,
>>>
>>> (below)
>>>
>>> 2014-05-09, Jesse Gross:
On Fri, Apr 25, 2014 at 4:56 AM, wrote:
> Hi Jesse,
>
> (inlined below)
>
> Jesse Gross :
>
Rather than setting and resetting the 'mark' field in the ukey, this
patch introduces a seq to track whether a flow has been seen during the
most recent dump. This tidies the code and simplifies the logic for
detecting when flows are duplicated from the datapath.
Signed-off-by: Joe Stringer
---
Signed-off-by: Ryan Wilson
---
v2: Fixed bug with group stats all buckets, cleaned up ofgroup unref code,
added Andy Zhou's test for more complete test coverage
v3: Split group reference count, support for group and bucket stats into 2
patches. Commit Andy Zhou's test patch separately.
---
o
Signed-off-by: Ryan Wilson
---
v2: Fixed bug with group stats all buckets, cleaned up ofgroup unref code,
added Andy Zhou's test for more complete test coverage
v3: Split group ref/unref, support for group and bucket stats, and Andy Zhou's
tests into 3 patches.
---
lib/ofp-util.h
From: Andy Zhou
Signed-off-by: Andy Zhou
Signed-off-by: Ryan Wilson
---
tests/ofproto-dpif.at | 44
1 file changed, 44 insertions(+)
diff --git a/tests/ofproto-dpif.at b/tests/ofproto-dpif.at
index fad8f2c..b87f735 100644
--- a/tests/ofproto-dpif
---
debian/openvswitch-vtep.default |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/debian/openvswitch-vtep.default b/debian/openvswitch-vtep.default
index 2e888e7..5049aeb 100644
--- a/debian/openvswitch-vtep.default
+++ b/debian/openvswitch-vtep.default
@@ -1,4 +1,4 @@
#
Signed-off-by: David Yang
---
debian/openvswitch-vtep.default |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/debian/openvswitch-vtep.default b/debian/openvswitch-vtep.default
index 2e888e7..5049aeb 100644
--- a/debian/openvswitch-vtep.default
+++ b/debian/openvswitch-vtep.
Hey Andy,
Could you review the first two packets?
The second patch looks good to me,
For this patch, I'm concerned about the use of ovsrcu_postpone(), since
there is no
rcu protected variable in 'struct ofgroup'. I suggest, we could move the
code in
group_destroy_cb() to ofproto_group_unref().
Looks good to me. If no one else have any comments, I will apply this
in an hour..
On Mon, May 19, 2014 at 2:52 PM, David Yang wrote:
> Signed-off-by: David Yang
> ---
> debian/openvswitch-vtep.default |2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/debian/openvswitc
Acked-by: Alex Wang
On Mon, May 19, 2014 at 6:20 AM, Ryan Wilson wrote:
> From: Andy Zhou
>
> Signed-off-by: Andy Zhou
> Signed-off-by: Ryan Wilson
> ---
> tests/ofproto-dpif.at | 44
> 1 file changed, 44 insertions(+)
>
> diff --git a/tests/o
On Mon, May 19, 2014 at 02:52:35PM -0700, David Yang wrote:
> Signed-off-by: David Yang
Why?
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Sat, May 17, 2014 at 10:32:24AM +0900, Simon Horman wrote:
> On Fri, May 16, 2014 at 03:17:18PM -0700, Jesse Gross wrote:
> > On Fri, May 16, 2014 at 2:48 PM, Thomas Graf wrote:
> > > On 05/16/14 at 02:29pm, Jesse Gross wrote:
> > >> On Fri, May 16, 2014 at 1:48 AM, Simon Horman wrote:
> > >>
If someone were to install this package, we'd expect that they'd want
this enabled since the default behavior after package install should
be to start the daemons. Right now, the VTEP start script doesn't do
anything unless this flag is set to true.
On Mon, May 19, 2014 at 3:00 PM, Ben Pfaff wrot
On Mon, Mar 31, 2014 at 08:37:48AM -0700, Ben Pfaff wrote:
> When openvswitch stops before libvirt shuts down VMs, it makes it hard for
> libvirt to remove virtual network interfaces (ovs-vsctl cannot access the
> database socket, which has been removed). This commit should ensure that
> the VMs g
On Thu, Apr 3, 2014 at 12:45 PM, Karol Mroz wrote:
> On 2014-04-03, Ben Pfaff wrote:
>> On Thu, Apr 03, 2014 at 06:45:47PM +, Karol Mroz wrote:
>>> On 2014-04-03, Ben Pfaff wrote:
>>> > On Tue, Apr 01, 2014 at 11:15:41AM -0700, Karol Mroz wrote:
>>> >> I'm experiencing a problem starting ope
David, thanks. Please add that information to the commit message.
Justin and David, I see that you two co-authored the Debian packaging
for the VTEP code. What was the original reason not to enable VTEP by
default on install? Is the reason no longer valid?
On Mon, May 19, 2014 at 03:14:40PM -0
Signed-off-by: Joe Stringer
---
v2: Make it configurable.
---
utilities/ovs-dev.py |4
1 file changed, 4 insertions(+)
diff --git a/utilities/ovs-dev.py b/utilities/ovs-dev.py
index 97fab6c..d08ef39 100755
--- a/utilities/ovs-dev.py
+++ b/utilities/ovs-dev.py
@@ -145,6 +145,8 @@ command
On Thu, Apr 10, 2014 at 12:50:11PM +0200, Thomas Graf wrote:
> a425a102-c317-4743-b0ba-79d59ff04a74
> Bridge "br0"
> [...]
> Port test
> Interface test
> type: vxlan
> options: {unknown="1"}
> error: "test: could not se
On Fri, Apr 11, 2014 at 06:34:07PM -0300, Flavio Leitner wrote:
> From: Daniel Borkmann
>
> Add basic header structure and definitions into packet.h.
>
> Acked-by: Thomas Graf
> Signed-off-by: Daniel Borkmann
> Signed-off-by: Flavio Leitner
Applied, thanks.
__
On Fri, Apr 11, 2014 at 06:34:06PM -0300, Flavio Leitner wrote:
> From: Daniel Borkmann
>
> Add the definition of Internet Group Management Protocol.
>
> Signed-off-by: Cong Wang
> Signed-off-by: Daniel Borkmann
> Acked-by: Thomas Graf
> Signed-off-by: Flavio Leitner
Applied, thanks.
__
On Fri, Apr 11, 2014 at 06:34:08PM -0300, Flavio Leitner wrote:
> RFC4541- IGMP and MLD Snooping Switches Considerations
> recommends to have different actions for local and
> non-local multicast traffic.
>
> Acked-by: Daniel Borkmann
> Signed-off-by: Thomas Graf
> Signed-off-by: Flavio Leitner
If someone were to install this package, we'd expect that they'd want this
enabled since the default behavior after package install should be to start the
daemons. Right now, the VTEP start script doesn't do anything unless this flag
is set to true.
Bug #1246486
Signed-off-by: David Yang
---
de
On Fri, Apr 11, 2014 at 06:34:13PM -0300, Flavio Leitner wrote:
> No functionality is changed. This is needed on
> a further commit adding REV_MCAST_SNOOPING.
>
> Acked-by: Thomas Graf
> Acked-by: Daniel Borkmann
> Signed-off-by: Flavio Leitner
Applied, thanks.
On Fri, Apr 11, 2014 at 06:34:14PM -0300, Flavio Leitner wrote:
> No functional changes. This is just for better readability
> when the multicast snooping learning and sending functions
> are added to the code.
>
> Acked-by: Thomas Graf
> Acked-by: Daniel Borkmann
> Signed-off-by: Flavio Leitne
On Fri, Apr 11, 2014 at 06:34:14PM -0300, Flavio Leitner wrote:
> No functional changes. This is just for better readability
> when the multicast snooping learning and sending functions
> are added to the code.
>
> Acked-by: Thomas Graf
> Acked-by: Daniel Borkmann
> Signed-off-by: Flavio Leitne
On Fri, Apr 11, 2014 at 06:34:15PM -0300, Flavio Leitner wrote:
> This is also needed for multicast snooping.
>
> Acked-by: Thomas Graf
> Acked-by: Daniel Borkmann
> Signed-off-by: Flavio Leitner
Applied, thanks!
___
dev mailing list
dev@openvswitch.
Ok, after long last, I was able to get my perf environment to work. Here are
the results for the TCP_CRR (300 flows on server209/210 to be exact) test with
master with and without the flow hash table in up call.
The mean and median transmissions/second are 2-3 lower without the hash table;
I ra
This was previously logging that there was no such function, but passing
the test anyway.
Signed-off-by: Joe Stringer
---
tests/ofproto.at |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/tests/ofproto.at b/tests/ofproto.at
index dde715e..2f1df2d 100644
--- a/tests/ofpr
> On Mon, Mar 31, 2014 at 08:37:48AM -0700, Ben Pfaff wrote:
>> When openvswitch stops before libvirt shuts down VMs, it makes it hard for
>> libvirt to remove virtual network interfaces (ovs-vsctl cannot access the
>> database socket, which has been removed). This commit should ensure that
>> the
I have some miscellaneous comments on things that I noticed, all of
which are pretty small. I will probably have a few more tomorrow but
my hope is that we can get this in soon.
On Thu, May 15, 2014 at 4:07 PM, Simon Horman wrote:
> diff --git a/datapath/actions.c b/datapath/actions.c
> index 603
So I did an experiment where I added 500 and 1000 ports and then deleted 500
and 1000 ports with and without this patch on both machines with 8 GB and 62 GB
memory. Weirdly enough, adding / deleting ports with the RCU patch turned out
to actually be faster than without. My only explanation here
Sorry Gurucharan, totally forgot to answer your question!
After interspersing these tests with random calls to reload the kernel module,
it doesn't appear to affect time in any significant way.
Ryan Wilson
Member of Technical Staff
wr...@vmware.com
3401 Hillview Avenue, Palo Alto, CA
650.427.151
And sorry for the email spam but here's a test I did earlier with the RCU patch
when I first finished it. I sent the email to Alex, but unfortunately not to
the rest of the list. This may give some insight into the case where this patch
is the most helpful
So I performance tested this and
Ok turns out my Openflow rules weren't totally correct (they were flooding all
ports like a hub instead of forwarding properly). After adjusting them, I
achieved equivalent performance with and without my upcall patch (both achieved
161-162 trans/second). I'll submit my other version of the patc
The upcall handler keeps a hash table which hashes flow to a list of
corresponding packets. This used to be necessary as packets with the same flow
had similar actions and calculating actions used to be a performance bottleneck.
Now that userspace action calculation performance has improved, there
On Mon, May 19, 2014 at 10:25 PM, Ryan Wilson wrote:
> Ok turns out my Openflow rules weren't totally correct (they were flooding
> all ports like a hub instead of forwarding properly). After adjusting them,
> I achieved equivalent performance with and without my upcall patch (both
> achieved 161
65 matches
Mail list logo
