Hi
I found that OVS DPDK fails to build with DPDK 16.07 after commit
01961bb. I am not sure if this is because of some expectations about
DPDK configuration or problems in the code. I could fix this fairly
easily with 4 additions to lib/dpdk.c (patch below), but am unsure
about whether this is
Ciara Loftus writes:
> The rte_pdump header file was not included in the file that requires it.
> Fix this.
>
> Fixes: 01961bbdd34a ("dpdk: New module with some code from netdev-dpdk.")
> Signed-off-by: Ciara Loftus
> ---
I'm not sure why I
The rte_pdump header file was not included in the file that requires it.
Fix this.
Fixes: 01961bbdd34a ("dpdk: New module with some code from netdev-dpdk.")
Signed-off-by: Ciara Loftus
---
lib/dpdk.c| 4
lib/netdev-dpdk.c | 3 ---
2 files changed, 4
>
> Hi
> I found that OVS DPDK fails to build with DPDK 16.07 after commit
> 01961bb. I am not sure if this is because of some expectations about
> DPDK configuration or problems in the code. I could fix this fairly
> easily with 4 additions to lib/dpdk.c (patch below), but am unsure
> about
On 2016-10-13 13:10, Russell Bryant wrote:
On Thu, Oct 13, 2016 at 6:00 AM, Stephen Finucane
wrote:
On 2016-10-12 20:20, Russell Bryant wrote:
On Sat, Oct 8, 2016 at 12:30 PM, Stephen Finucane
wrote:
I'm going to dripfeed the conversion patches to avoid
2016-10-13 10:27 GMT-07:00 Ciara Loftus :
> The rte_pdump header file was not included in the file that requires it.
> Fix this.
>
> Fixes: 01961bbdd34a ("dpdk: New module with some code from netdev-dpdk.")
> Signed-off-by: Ciara Loftus
>
Oops,
> From: "Andy Zhou"
> To: "Ben Pfaff"
> Cc: "ovs dev" , "Numan Siddique" ,
> "Babu Shanmugam" ,
> "Lance Richardson" , "Justin Pettit" ,
> "Russell Bryant"
This change set adds a new optional "access control" specifier to
remote connection descriptors used by ovsdb.
Examples:
--remote=ptcp:ro:0:192.168.0.10
--remote=punix:ro:asocket.sock
--remote=pssl:ro:0:192.168.0.10
--remote=tcp:ro:192.168.0.99:
This change set adds a new optional "access control" specifier to
remote connection descriptors used by ovsdb.
Examples:
--remote=ptcp:ro:0:192.168.0.10
--remote=punix:ro:asocket.sock
--remote=pssl:ro:0:192.168.0.10
--remote=tcp:ro:192.168.0.99:
The commit message says:
openvswitch: Allow external IPsec tunnel management.
OVS GRE IPsec tunnel support has multiple issues, Therefore
it was deprecated in OVS 2.6.
Following patch removes support for GRE IPsec and allows external
IPsec tunnel management for any
> From: "Ben Pfaff"
> To: "Andy Zhou"
> Cc: "ovs dev" , "Numan Siddique" ,
> "Babu Shanmugam" ,
> "Lance Richardson" , "Justin Pettit" ,
> "Russell Bryant"
On Wed, Oct 12, 2016 at 01:51:39PM -0400, Russell Bryant wrote:
> 1) Add support to ovsdb-server for read-only remotes. The port reachable
> by ovn-controller would only accept read-only connections.
Andy, is this something that you can put on your to-do list? I guess
that it is not a huge
On Thu, Oct 13, 2016 at 11:26 AM, Ben Pfaff wrote:
> On Wed, Oct 12, 2016 at 01:51:39PM -0400, Russell Bryant wrote:
> > 1) Add support to ovsdb-server for read-only remotes. The port reachable
> > by ovn-controller would only accept read-only connections.
>
> Andy, is this
On Thu, Oct 13, 2016 at 03:08:46PM -0400, Lance Richardson wrote:
> Doing 'make check TESTSUITEFLAGS="1876-1881"' executes these test
> cases:
> 1876: ovsdb-server/read-only ptcp connection ok
> 1877: ovsdb-server/read-only punix connection ok
> 1878:
> From: "Ben Pfaff"
> To: "Lance Richardson"
> Cc: dev@openvswitch.org, jpe...@ovn.org
> Sent: Thursday, October 13, 2016 5:08:42 PM
> Subject: Re: [ovs-dev] [RFC] ovsdb: implement read-only remote connection type
>
> On Thu, Oct 13, 2016 at 03:08:46PM -0400,
On Wed, Oct 12, 2016 at 10:57 PM, Babu Shanmugam
wrote:
>
>
> On Thursday 13 October 2016 07:26 AM, Andy Zhou wrote:
>
>
>
> On Sun, Oct 9, 2016 at 12:02 AM, Babu Shanmugam
> wrote:
>
>>
>>
>> On Friday 07 October 2016 05:33 AM, Andy Zhou wrote:
>>
>>>
This patch adds action 'nd_ra' specific inner actions 'put_nd_ra',
'put_nd_opt_prefix', 'put_nd_opt_mtu', 'put_nd_opt_sll'.
Action put_nd_ra will set following fields in Router Advertisement (RA) packet:
- cur_hop_limit(8-bit unsigned integer),
- "Managed address configuration" and "Other
This patch add support for building Router Solicitation responder flows, which
will be used to reply Router Advertisement packet from ovn-controller for
VM/VIF port.
It adds table Logical_Router_RA_Prefix_Config to define a set of ND Prefix
informations configurations(such as valid_lifetime,
This patch add lflow support for Router Advertisement packet forward.
Forward RA packets to router port in lswitch l2_lkup table.
Signed-off-by: Zongkai LI
---
ovn/northd/ovn-northd.c | 26 ++
1 file changed, 26 insertions(+)
diff --git
Since ovs_nd_mtu_opt and ovs_nd_prefix_opt is introducted, rename
ovs_nd_opt to ovs_nd_lla_opt to specify it's Source/Target Link-layer
Address Option.
Signed-off-by: Zongkai LI
---
lib/flow.c| 30 +++
lib/odp-execute.c | 22 -
This serial patches try to support Router Solicitation (RS) message responder,
which will reply Router Advertisement (RA) message, for OVN.
There will be logical flows in table ls_in_rs_rsp to work as RS responder, with:
- match: ip6.dst == ff02::2 && nd_rs
(nd_rs: icmp6.type == 133 &&
This patch introduces methods to compose a Router Advertisement (RA) packet,
introduces flags for RA. RA packet composed structures against specification
in RFC4861.
Caller can use compse_nd_ra_with_sll_mtu_opts to compose a RA packet with
Source Link-layer Address Option and MTU Option.
Caller
On Thu, Oct 13, 2016 at 9:32 AM, Ben Pfaff wrote:
> On Thu, Oct 13, 2016 at 09:10:58AM -0400, Russell Bryant wrote:
> > On Thu, Oct 13, 2016 at 6:00 AM, Stephen Finucane
> wrote:
> >
> > > On 2016-10-12 20:20, Russell Bryant wrote:
> > >
> > >> On Sat, Oct 8,
I should have also noted who was already looking at each of these items ...
On Thu, Oct 13, 2016 at 10:02 AM, Numan Siddique
wrote:
> We may have to add one more item in the task breakdown list. Please see
> below
>
>
> On Wed, Oct 12, 2016 at 11:21 PM, Russell Bryant
On Thu, Oct 13, 2016 at 1:46 PM, Stephen Finucane wrote:
>
> ..and sorry for not setting the scene: I forget how active this mailing
> list is :)
>
> Let me know if you've any questions.
>
No problem. I think it's more my fault. I've been out a lot lately so
there's a bunch
I think this looks good, thanks.
I have a few comments inline to simplify the code a little bit more.
2016-10-13 2:58 GMT-07:00 Bhanuprakash Bodireddy <
bhanuprakash.bodire...@intel.com>:
> This patch increases the number of packets processed in a batch during a
> lookup from 16 to 32.
Acked-by: Daniele Di Proietto
2016-10-13 2:58 GMT-07:00 Bhanuprakash Bodireddy <
bhanuprakash.bodire...@intel.com>:
> Aligning the 'keys' array seems to have positive performance impact.
>
> Signed-off-by: Bhanuprakash Bodireddy
>
2016-10-06 3:25 GMT-07:00 Mark Kavanagh :
> The DPDK pdump sample app was renamed from 'dpdk_pdump' to
> 'dpdk-pdump'. Update references to same within
> INSTALL.DPDK-ADVANCED.md.
>
> Add an additional sample command line that shows how to capture all
> traffic
Acked-by: Daniele Di Proietto
2016-10-13 2:58 GMT-07:00 Bhanuprakash Bodireddy <
bhanuprakash.bodire...@intel.com>:
> Add comments in dp_netdev_input__() to explain the reason behind
> clearing the flow batches before packet_batch_execute().
>
> Signed-off-by:
Thanks for v2.
I had a few extra comments on patch one
Sorry for nitpicking on this and sorry I wasn't clear enough on the signoff
requirements.
From CONTRIBUTING.md:
Co-authored-by: Author Name
Git can only record a single person as the author of a
Hi, Mark
I have reviewed your RFC patch, I found one place could have issue, see
below:
+if (v > ETHER_MAX_LEN) {
+b->mbuf.data_len =
+(uint16_t)(b->mbuf.buf_len - b->mbuf.data_off);
+} else {
+b->mbuf.data_len = (uint16_t)v;
+}
+
Here, what will
On Wed, Oct 12, 2016 at 01:36:44PM -0700, Pravin Shelar wrote:
> Sorry for jumping in a bit late. I have couple of high level comments below.
>
> On Thu, Oct 6, 2016 at 10:10 AM, Rony Efraim wrote:
> > From: Joe Stringer [mailto:j...@ovn.org] Sent: Thursday, October 06, 2016
This patch series is aimed at improving the performance of OVS-DPDK
dpcls.
With few thousand flows installed, the EMC becomes inefficient due
to thrashing and the bottleneck moves to the dpcls. In EMC disabled
case, through VTune we found that significant performance degradation
is due to LLC
This patch adds comments to mf_get_next_in_map() to make it more
comprehensible.
Signed-off-by: Jarno Rajahalme
Acked-by: Bhanuprakash Bodireddy
Acked-by: Antonio Fischetti
---
lib/flow.h | 32
This patch checks if trash is non-zero and only then resets the flowmap
bit and increment the pointer by set bits as found in trash.
Signed-off-by: Bhanuprakash Bodireddy
Co-authored-by: Antonio Fischetti
Acked-by: Jarno Rajahalme
On 2016-10-12 20:20, Russell Bryant wrote:
On Sat, Oct 8, 2016 at 12:30 PM, Stephen Finucane
wrote:
I'm going to dripfeed the conversion patches to avoid overloading
peoples (there are ~25 of them so far). This is the first batch.
Stephen Finucane (9):
dist-docs: Add
prefetching the data in to the caches isn't improving the performance in
cmap_find_batch(). Moreover its found that there is slight improvement
in performance with out prefetching.
This patch removes prefetching from cmap_find_batch().
Signed-off-by: Bhanuprakash Bodireddy
Aligning the 'keys' array seems to have positive performance impact.
Signed-off-by: Bhanuprakash Bodireddy
Co-authored-by: Antonio Fischetti
---
lib/dpif-netdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
By reordering the data elements in dp_netdev_port structure, pad bytes
can be reduced and there by saving a cache line.
Before: structure size:136, holes:3, sum padbytes:15, cachelines:3
After: structure size:128, holes:2, sum padbytes:7, cachelines:2
Signed-off-by: Bhanuprakash Bodireddy
This patch increases the number of packets processed in a batch during a
lookup from 16 to 32. Processing batches of 32 packets improves
performance and also one of the internal loops can be avoided here.
Signed-off-by: Antonio Fischetti
Co-authored-by: Bhanuprakash
mhash_add__() is expensive and should be only called with valid input.
zero-valued 'data' will not affect the 'hash' value and expensive hash
computation can be skipped when input is zero.
This patch will validate the input in mhash_add__ to save some cpu
cycles.
Signed-off-by: Bhanuprakash
Add comments in dp_netdev_input__() to explain the reason behind
clearing the flow batches before packet_batch_execute().
Signed-off-by: Bhanuprakash Bodireddy
Co-authored-by: Antonio Fischetti
---
lib/dpif-netdev.c | 9 +
By reordering the elements in ovsdb_table_schema structure, pad bytes
can be reduced and also a cache line is saved.
Before: structure size:72, holes:2, sum padbytes:10, cachelines:2
After: structure size:64, holes:1, sum padbytes:2, cachelines:1
Signed-off-by: Bhanuprakash Bodireddy
By reordering the elements in nl_dump structure, pad bytes can be
reduced there by saving a cache line.
Before: structure size:72, holes:1, sum padbytes:4, cachelines:2
After: structure size:64, holes:0, sum padbytes:0, cachelines:1
Signed-off-by: Bhanuprakash Bodireddy
By reordering the data elements in dpif_upcall structure, pad bytes can
be reduced and also a cache line. Also dp_packet should be the first
member of the structure because rte_mbuf, the first member of dp_packet
should be aligned atleast on a 64-byte boundary.
Before: structure size:768,
By reordering the elements in clock structure, pad bytes
can be reduced and also a cache line is saved.
Before: structure size:136, holes:3, sum padbytes:18, cachelines:3
After: structure size:120, holes:1, sum padbytes:2, cachelines:2
Signed-off-by: Bhanuprakash Bodireddy
On Wed, Oct 12, 2016 at 11:56:51AM -0400, Russell Bryant wrote:
> On Fri, Oct 7, 2016 at 1:36 PM, Aaron Conole wrote:
>
> > During testing it was found that systemd would consider the openvswitch
> > service as a part of networking component, but the dependent services of
> >
xu.binb...@zte.com.cn writes:
> Hi Aaron,
>
> Thanks for your suggestion, I'll resubmit a patch for DPDK, and take a look
> at
> lib/netdev-bsd.c to confirm that the additional patch is needed or not.
>
> Thanks,
>
> Aaron Conole 写于 2016/10/12 21:26:14:
>
>> 发件人: Aaron
On Thu, Oct 13, 2016 at 09:10:58AM -0400, Russell Bryant wrote:
> On Thu, Oct 13, 2016 at 6:00 AM, Stephen Finucane wrote:
>
> > On 2016-10-12 20:20, Russell Bryant wrote:
> >
> >> On Sat, Oct 8, 2016 at 12:30 PM, Stephen Finucane
> >> wrote:
> >>
> >> I'm
Below is the proposal to add native DNS support in OVN for internal DNS
resolution. This will be useful If a VM sends a DNS lookup request for
another VM belonging to the same virtual network.
- hostname of the logical ports will be stored in the north db lsp row
- ovn-northd will add the
On Thu, Oct 13, 2016 at 6:00 AM, Stephen Finucane wrote:
> On 2016-10-12 20:20, Russell Bryant wrote:
>
>> On Sat, Oct 8, 2016 at 12:30 PM, Stephen Finucane
>> wrote:
>>
>> I'm going to dripfeed the conversion patches to avoid overloading
>>> peoples (there
We may have to add one more item in the task breakdown list. Please see
below
On Wed, Oct 12, 2016 at 11:21 PM, Russell Bryant wrote:
> Hello, I'm back to looking at southbound database security concerns in
> OVN. A previous thread discussing approaches was here:
>
>
Your message was undeliverable due to the following reason(s):
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
Dear Ben ,
Thank you for your response.
Could you please throw some light on ,any specific reason why OVS stopped
support for ovs-monitor-ipsec and decided to remove support for ipsec
altogether in ovs.
Regards
Muthukrishnan
9952012433
From: Ben Pfaff
54 matches
Mail list logo