On Thu, Sep 27, 2012 at 11:16 AM, Jesse Gross wrote:
> On Thu, Sep 27, 2012 at 9:26 AM, Ben Pfaff wrote:
> > On Wed, Sep 26, 2012 at 04:13:41PM +0900, Simon Horman wrote:
> >> This patch provides an implementation of the non-datapath portions
> >> of MPLS matches and actions.
> >>
> >> This patc
On Thu, Sep 27, 2012 at 5:44 PM, Simon Horman wrote:
> On Thu, Sep 27, 2012 at 04:45:22PM -0700, Jesse Gross wrote:
> > On Thu, Sep 27, 2012 at 12:05 PM, ravi kerur wrote:
> > > On Thu, Sep 27, 2012 at 11:16 AM, Jesse Gross
> wrote:
> > >> On Thu, Sep 27,
This patch looks very similar to the ones I had posted earlier albeit one
key difference being a variable added in ovs_skb_cb and obvious difference
in offset calculation code which is done based on this field. Some questions
1. Functions in this patch have similar issues which my initial patch ha
Hi Ben,
All your code review comments have been taken care for MPLS. I did
retest the code after incorporating code review comments. I also
modified the code for QinQ (common comments which apply to QinQ code
as well), added additional automation tests in odp.at for QinQ. In
addition, patch to sup
On Tue, May 29, 2012 at 11:14 AM, Ben Pfaff wrote:
> Thanks so much for applying my comments. Will you please post the
> newest version of the MPLS patch? At this time it would be good to
> get your
> Signed-off-by: Ravi Kerur
> in the commit log, since it is getting cl
May 29, 2012 at 12:12 PM, ravi kerur wrote:
>> Additional changes are needed here esp. in NIC driver(looked into
>> e1000e driver code from Intel) as it requires handling of new
>> eth_types. For these changes I probably need to work with you and
>> Jesse on how best we
Jesse, Does this make sense?
Thanks,
Ravi
On Tue, May 29, 2012 at 1:32 PM, ravi kerur wrote:
> Thanks Jesse. What I have tested is 2 Ubuntu 11.04 and 12.04 physical
> systems connected via a ethernet switch. I have tested scp downloads,
> iperf + netperf. Flows are setup such that MP
Thanks Jesse, I will check the code in vport-netdev.c file and get
back with the diffs. I did modify the code to get earlier version
working. Anyways will get back to you.
On Tue, May 29, 2012 at 3:32 PM, Jesse Gross wrote:
> On Tue, May 29, 2012 at 1:32 PM, ravi kerur wrote:
>> Tha
ght of keeping you guys in loop, please let me know if its the
right approach?
Thanks,
Ravi
On Thu, May 31, 2012 at 10:16 AM, ravi kerur wrote:
> Hi Jesse,
>
> Attached latest code diffs. It includes changes to handle offload as
> suggested + incorporates Ben's code review c
is, I don't think will be handled in ovs offload code.
Anyways, I will wait until you return and we can work on this.
Thanks,
Ravi
On Thu, May 31, 2012 at 5:03 PM, Jesse Gross wrote:
> On Thu, May 31, 2012 at 3:44 PM, ravi kerur wrote:
>> Hi Jesse, Ben,
>>
>> I
On Thu, May 31, 2012 at 6:54 PM, Jesse Gross wrote:
> On Thu, May 31, 2012 at 6:04 PM, ravi kerur wrote:
>> Thanks Jesse, I don't think I was clear. For mpls and qinq, offloading
>> goes through ovs emulation code in vport-netdev.c. In addiition to
>> TSO, there is chec
PM, Ben Pfaff wrote:
> On Thu, May 31, 2012 at 10:16:51AM -0700, ravi kerur wrote:
>> I have generated single patch for OF 1.1 and 1.3 changes assuming it
>> would be easier for code-review but as I suggested earlier, I will
>> probably have 2 separate commits. Please let me k
Hi, I see some discussion on ovs + gre but not sure what the solution
is. I am facing some what similar issue w.r.t configuration and flow
matching.
On host-1, gre0 interface is configured via
ovs-vsctl add-br br0
ovs-vsctl add-port br0 eth0
ovs-vsctl add-port br0 gre0 -- set interface gre0 type=
_type(0x0800),ipv4(src=10.0.0.3,dst=10.0.0.7,proto=1,tos=0,ttl=64,frag=no),icmp(type=8,code=0),
packets:5, bytes:490, used:0.728s, actions:set(tun_id(0x6f)),2
On Mon, Jun 4, 2012 at 8:07 PM, Jesse Gross wrote:
> On Tue, Jun 5, 2012 at 5:26 AM, ravi kerur wrote:
>> Hi, I see some dis
)
2. match IP/GRE/IP_src/IP_dst and actions=1(eth0 port number)
On receive side:
1. match IP/GRE/IP_src/IP_dst and actions=??? to strip/decap gre header
2. match IP/ICMP/IP_src/IP_dst and actions=NORMAL
Thanks,
Ravi
On Tue, Jun 5, 2012 at 10:54 AM, Justin Pettit wrote:
> On Jun 5, 2012
d every line
>> though.)
>>
>> I did notice that compose_dec_mpls_ttl() decrements "ttl" twice (it
>> has --ttl in two places).
>>
>> I believe that Pravin is going to review the kernel portions of this
>> code now. Once he's happy, I'll su
On Tue, Jun 12, 2012 at 12:13 PM, Ben Pfaff wrote:
> On Tue, Jun 12, 2012 at 11:40:22AM -0700, ravi kerur wrote:
>> On Tue, Jun 12, 2012 at 11:11 AM, Ben Pfaff wrote:
>> > Ravi, I'm more or less happy with the userspace code here. I mostly
>> > have style kind
I am facing an issue in flow matching for incoming vlan packets in ovs
userspace. In lib/flow.c and flow_extract function, eth->eth_type(line
no 363) is eth_type_ip. From logs, I see
Jun 12 14:43:43 rkerur-thinkpad-t500 ovs-vswitchd:
in_port(1),eth(src=00:27:13:67:b9:9b,dst=00:24:7e:68:ae:46),eth_
Thanks Pravin for the review comments. Inline
On Tue, Jun 12, 2012 at 5:31 PM, Pravin Shelar wrote:
> Hi Ravi,
>
> I see following issues with datapath mpls support:
>
> Actions:
> I do not see skb->csum updated when mpls action is performed on packet.
csum calculation is usually done in the k
Inline
On Wed, Jun 13, 2012 at 5:55 AM, Jesse Gross wrote:
> On Jun 13, 2012, at 12:12 PM, ravi kerur wrote:
>
>> Thanks Pravin for the review comments. Inline
>>
>> On Tue, Jun 12, 2012 at 5:31 PM, Pravin Shelar wrote:
>>> Hi Ravi,
>>>
>>>
Yes kernel flow_extract finds the tag when kernel module is loaded and
works fine. Its only with userspace + vlan.
On Wed, Jun 13, 2012 at 8:22 AM, Jesse Gross wrote:
> On Wed, Jun 13, 2012 at 6:49 AM, ravi kerur wrote:
>> I am facing an issue in flow matching for incoming vlan packe
On Wed, Jun 13, 2012 at 8:49 AM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 12:04 AM, ravi kerur wrote:
>> On Wed, Jun 13, 2012 at 5:55 AM, Jesse Gross wrote:
>>> On Jun 13, 2012, at 12:12 PM, ravi kerur wrote:
>>>> On Tue, Jun 12, 2012 at 5:31 PM, Pravin Sh
ecause its not yet clear to me why performance would be bad.
Finally, are performance numbers for OVS published or documented?
On Wed, Jun 13, 2012 at 10:19 AM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 1:32 AM, ravi kerur wrote:
>> On Wed, Jun 13, 2012 at 8:49 AM, Jesse Gross wrote:
On Wed, Jun 13, 2012 at 7:58 PM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 4:51 AM, ravi kerur wrote:
>> Ok got it. At least I am sensing that OVS will/should be optimized for
>> both core and edge cases. I have taken care of comments from Pravin
>> and I think we are wa
On Wed, Jun 13, 2012 at 8:21 PM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 9:25 AM, Pravin Shelar wrote:
>> On Wed, Jun 13, 2012 at 5:55 AM, Jesse Gross wrote:
>>> I spent a fair amount of time thinking about this today and concluded that
>>> neither option is very attractive as is.
>>>
>>>
On Wed, Jun 13, 2012 at 9:24 PM, ravi kerur wrote:
> On Wed, Jun 13, 2012 at 7:58 PM, Jesse Gross wrote:
>> On Thu, Jun 14, 2012 at 4:51 AM, ravi kerur wrote:
>>> Ok got it. At least I am sensing that OVS will/should be optimized for
>>> both core and edge cases. I
On Wed, Jun 13, 2012 at 8:38 PM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 9:37 AM, Ben Pfaff wrote:
>> On Wed, Jun 13, 2012 at 09:55:36PM +0900, Jesse Gross wrote:
>>> If we go with what you have now, I'm fairly confident that we will
>>> regret it in the future. The kernel code used to mor
On Thu, Jun 14, 2012 at 3:58 AM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 1:24 PM, ravi kerur wrote:
>> On Wed, Jun 13, 2012 at 7:58 PM, Jesse Gross wrote:
>>> On Thu, Jun 14, 2012 at 4:51 AM, ravi kerur wrote:
>>>> Ok got it. At least I am sensing that OV
On Thu, Jun 14, 2012 at 4:12 AM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 2:42 PM, ravi kerur wrote:
>> On Wed, Jun 13, 2012 at 8:38 PM, Jesse Gross wrote:
>>> On Thu, Jun 14, 2012 at 9:37 AM, Ben Pfaff wrote:
>>>> On Wed, Jun 13, 2012 at 09:55:36PM +0900, Jes
On Thu, Jun 14, 2012 at 8:22 AM, Jesse Gross wrote:
> On Jun 14, 2012, at 10:13 PM, ravi kerur wrote:
>
>> On Thu, Jun 14, 2012 at 3:58 AM, Jesse Gross wrote:
>>> On Thu, Jun 14, 2012 at 1:24 PM, ravi kerur wrote:
>>>> On Wed, Jun 13, 2012 at 7:58 PM, Jesse G
I am getting following errors when installing ovs on kernel 3.3 and
Ubuntu 12.04, it works fine on kernel 3.0. I checked out latest git
code and while installation get following errors. Installation on
kernel 3.3 had worked perfectly before. Does ovs-vsctl need more
parameters now?
root@:~/tmp_ovs
n/openvswitch/db.sock:
receive error: Connection reset by peer
2012-06-19T21:52:05Z|00013|reconnect|WARN|unix:/usr/local/var/run/openvswitch/db.sock:
connection dropped (Connection reset by peer)
^C2012-06-19T21:52:10Z|00014|fatal_signal|WARN|terminating with signal
2 (Interrupt)
On Tue, Jun 19, 2012 at
How odd.
>
> What's in the ovsdb-server log?
>
> On Tue, Jun 19, 2012 at 02:55:44PM -0700, ravi kerur wrote:
>> root@:~/tmp_ovs/openvswitch# ovs-vsctl -vjsonrpc --no-wait init
>> 2012-06-19T21:51:58Z|2|jsonrpc|DBG|unix:/usr/local/var/run/openvswitch/db.sock:
>>
-06-19T22:39:32Z|3|memory|INFO|
On Tue, Jun 19, 2012 at 3:17 PM, Ben Pfaff wrote:
> If you add --log-file to the ovsdb-server command line then you'll get
> a log file in the log directory.
>
> On Tue, Jun 19, 2012 at 03:14:49PM -0700, ravi kerur wrote:
>> Odd about err
Tue, Jun 19, 2012 at 03:45:32PM -0700, ravi kerur wrote:
>> this is all it has...
>>
>> root@:~/tmp_ovs/openvswitch# cat
>> /usr/local/var/log/openvswitch/ovsdb-server.log
>> 2012-06-19T22:39:22Z|1|vlog|INFO|opened log file
>> /usr/local/var/log/openvswitch/o
On Tue, Jun 19, 2012 at 11:55 AM, Jesse Gross wrote:
> On Thu, Jun 14, 2012 at 5:29 PM, ravi kerur wrote:
>> On Thu, Jun 14, 2012 at 8:22 AM, Jesse Gross wrote:
>>> On Jun 14, 2012, at 10:13 PM, ravi kerur wrote:
>>>
>>>> On Thu, Jun 14, 2012 at 3:58 AM
its/s on a 1G nic. If I understand correctly, part of the issue
is that resubmit action is in userspace, but the bigger question is
why first config on the Rx side doesn't work?
Inputs appreciated.
On Sat, Jun 9, 2012 at 10:03 PM, Jesse Gross wrote:
> On Jun 10, 2012, at 1:14 PM, ravi
On Mon, Jun 25, 2012 at 3:33 PM, Jesse Gross wrote:
> On Thu, Jun 21, 2012 at 2:54 PM, ravi kerur wrote:
>> I would like to restart this discussion as I am having trouble with
>> configuration + performance. I would like send just tcp traffic via
>> gre tunnel and would li
starting new thread for this...
The problem is that ovs-vswitchd crashes when eth_type = 0x8847/0x8848
and happens only in the automation i.e "make check", or "make
dist-check", since user-space code has been successfully tested
manually and I couldn't find a way to run gdb during automation not
s
Thanks Ben, I don't remember seeing core file under
tests/testsuite.dir/0449 directory. I will double-check, if its
available it will be definitely helpful.
On Tue, Jun 26, 2012 at 5:10 PM, Ben Pfaff wrote:
> On Tue, Jun 26, 2012 at 05:08:43PM -0700, ravi kerur wrote:
>> The probl
fproto-dpif.at:363: ovs_wait_cond () { ovs-appctl -t ovs-ofctl exit
}
On Tue, Jun 26, 2012 at 6:08 PM, Ethan Jackson wrote:
> Make sure you've enabled core files
>
> ulimit -a
> ulimit -c unlimited
>
> Ethan
>
> On Tue, Jun 26, 2012 at 5:49 PM, ravi kerur wrote:
>>
Never mind, I figured it out. Thanks for the inputs.
On Wed, Jun 27, 2012 at 11:11 AM, ravi kerur wrote:
> Thanks Ethan. Core file generation is enabled. However, I do not see
> any core files under failed test directory i.e
> tests/testsuite.dir/0449. Is the generated core file ava
erspace just installs a flow to send packets up.
>
> This would mean that the kernel only needs at a minimum push and pop
> actions. As userspace would always know the contents of the packet it
> can pass it down MPLS in their entirety with the kernel having to
> compute anything.
>
Packet recycling for
ttl operations looks like an overkill to me as you might be aware that
packet recycling has performance impact.
On Mon, Jul 2, 2012 at 7:15 PM, Jesse Gross wrote:
> On Sat, Jun 30, 2012 at 8:56 AM, ravi kerur wrote:
>> Can you take a look at latest diffs I sent out on
On Thu, Jul 5, 2012 at 7:37 PM, Jesse Gross wrote:
> On Thu, Jul 5, 2012 at 7:48 AM, ravi kerur wrote:
> > I understand code review is time consuming and please note it's a bit
> time
> > consuming to me as well as I have to retest/rebase everytime and I am not
> &g
On Fri, Jul 6, 2012 at 3:31 PM, Jesse Gross wrote:
> On Thu, Jul 5, 2012 at 9:45 PM, ravi kerur wrote:
> > As mentioned earlier, in the best interest of the time(both mine and
> yours)
> > I will have to re-evaluate whether to continue working on both mpls and
> qinq
&g
46 matches
Mail list logo
