On 4 Jul 2018, at 20:06, Tiago Lam wrote:
> A new mutex, 'nonpmd_mp_mutex', has been introduced to serialise
> allocation and free operations by non-pmd threads on a given mempool.
>
> free_dpdk_buf() has been modified to make use of the introduced mutex.
>
> Signed-off-by: Tiago Lam
LGTM,
On 05/07/2018 09:59, Eelco Chaudron wrote:
>
>
> On 4 Jul 2018, at 20:06, Tiago Lam wrote:
>
>> When a dp_packet is from a DPDK source, and it contains multi-segment
>> mbufs, the data_len is not equal to the packet size, pkt_len. Instead,
>> the data_len of each mbuf in the chain should be
On 4 Jul 2018, at 20:06, Tiago Lam wrote:
When a dp_packet is from a DPDK source, and it contains multi-segment
mbufs, the data_len is not equal to the packet size, pkt_len. Instead,
the data_len of each mbuf in the chain should be considered while
distributing the new (provided) size.
To
> Hi,
>
> Here is a joint work from Mellanox and Napatech, to enable the flow hw
> offload with the DPDK generic flow interface (rte_flow).
>
> The basic idea is to associate the flow with a mark id (a unit32_t
> number).
> Later, we then get the flow directly from the mark id, which could
On 5 Jul 2018, at 11:54, Lam, Tiago wrote:
> On 05/07/2018 09:59, Eelco Chaudron wrote:
>>
>>
>> On 4 Jul 2018, at 20:06, Tiago Lam wrote:
>>
>>> When a dp_packet is from a DPDK source, and it contains multi-segment
>>> mbufs, the data_len is not equal to the packet size, pkt_len. Instead,
>>>
On 7/5/2018 12:06 AM, Vishal Deep Ajmera wrote:
OVS reads packets in batches from a given port and packets in the
batch are subjected to potentially 3 levels of lookups to identify
the datapath megaflow entry (or flow) associated with the packet.
Each megaflow entry has a dedicated buffer in
On Wed, Jul 04, 2018 at 07:25:50PM -0700, William Tu wrote:
> On Tue, Jul 3, 2018 at 10:56 AM, Alexei Starovoitov
> wrote:
> > On Thu, Jun 28, 2018 at 07:19:35AM -0700, William Tu wrote:
> >> Hi Alexei,
> >>
> >> Thanks a lot for the feedback!
> >>
> >> On Wed, Jun 27, 2018 at 8:00 PM, Alexei
>
> d71962f ("bpf: allow map helpers access to map values directly") removes
> that limitation from the verifier and should allow you to use map values
> as map keys directly. 4.18-rc1 has it.
>
>> Thanks
>> William
Hi Paul,
Thanks a lot! This is very helpful.
I'm testing it now, works great so
When a dp_packet is from a DPDK source, and it contains multi-segment
mbufs, the data_len is not equal to the packet size, pkt_len. Instead,
the data_len of each mbuf in the chain should be considered while
distributing the new (provided) size.
To account for the above dp_packet_set_size() has
Most helper functions in dp-packet assume that the data held by a
dp_packet is contiguous, and perform operations such as pointer
arithmetic under that assumption. However, with the introduction of
multi-segment mbufs, where data is non-contiguous, such assumptions are
no longer possible. Some
In its current implementation dp_packet_shift() is also unaware of
multi-seg mbufs (that holds data in memory non-contiguously) and assumes
that data exists contiguously in memory, memmove'ing data to perform the
shift.
To add support for multi-seg mbuds a new set of functions was
introduced,
As it stands, OVS_DPDK_START() won't allow other configs to be set
before starting the ovs-vswitchd daemon. This is a problem since some
configs, such as the "dpdk-multi-seg-mbufs=true" for enabling the
multi-segment mbufs, need to be set prior to start OvS.
To support other options,
The following tests are added to the DPDK testsuite to add some
coverage for the multi-segment mbufs:
- Check that multi-segment mbufs are disabled by default;
- Check that providing `other_config:dpdk-multi-seg-mbufs=true` indeed
enables mbufs;
- Using a DPDK port, send a random packet out and
On Wed, Jun 20, 2018 at 10:44:51AM +0300, Ilya Maximets wrote:
> Recent assertion failure fix changed rconn workflow for unreliable
> connections (such as connections from ovs-ofctl) from
>
> |rconn|DBG|br-int<->unix#151: entering ACTIVE
> |rconn|DBG|br-int<->unix#151: connection closed
From: Mark Kavanagh
There are numerous factors that must be considered when calculating
the size of an mbuf:
- the data portion of the mbuf must be sized in accordance With Rx
buffer alignment (typically 1024B). So, for example, in order to
successfully receive and capture a 1500B packet,
From: Mark Kavanagh
dp_packets are created using xmalloc(); in the case of OvS-DPDK, it's
possible the the resultant mbuf portion of the dp_packet contains
random data. For some mbuf fields, specifically those related to
multi-segment mbufs and/or offload features, random values may cause
From: Mark Kavanagh
Currently, jumbo frame support for OvS-DPDK is implemented by
increasing the size of mbufs within a mempool, such that each mbuf
within the pool is large enough to contain an entire jumbo frame of
a user-defined size. Typically, for each user-defined MTU,
'requested_mtu', a
In order to create a minimal environment that allows the tests to get
mbufs from an existing mempool, the following approach is taken:
- EAL is initialised (by using the main dpdk_init()) and a (very) small
mempool is instantiated (mimicking the logic in dpdk_mp_create()).
This mempool
From: Mark Kavanagh
Currently, packets are only copied to a single segment in the function
dpdk_do_tx_copy(). This could be an issue in the case of jumbo frames,
particularly when multi-segment mbufs are involved.
This patch calculates the number of segments needed by a packet and
copies the
Hi Yipeng,
Some further comments below. Mainly to do with readability and understanding of
the changes.
Regards,
Billy.
> -Original Message-
> From: Wang, Yipeng1
> Sent: Friday, June 29, 2018 6:53 PM
> To: d...@openvswitch.org
> Cc: Wang, Yipeng1 ; jan.scheur...@ericsson.com;
>
A new mutex, 'nonpmd_mp_mutex', has been introduced to serialise
allocation and free operations by non-pmd threads on a given mempool.
free_dpdk_buf() has been modified to make use of the introduced mutex.
Signed-off-by: Tiago Lam
Acked-by: Eelco Chaudron
---
lib/netdev-dpdk.c | 30
When enabled with DPDK OvS deals with two types of packets, the ones
coming from the mempool and the ones locally created by OvS - which are
copied to mempool mbufs before output. In the latter, the space is
allocated from the system, while in the former the mbufs are allocated
from a mempool,
On Wed, Jul 04, 2018 at 04:00:26PM +0200, Timothy Redaelli wrote:
> On Tue, 3 Jul 2018 11:32:18 -0700
> Ben Pfaff wrote:
>
> > check_logs ignores some log messages, but it wasn't smart enough to
> > ignore the messages that said that the ignored messages had been
> > rate-limited. This fixes
When enabled with DPDK OvS relies on mbufs allocated by mempools to
receive and output data on DPDK ports. Until now, each OvS dp_packet has
had only one mbuf associated, which is allocated with the maximum
possible size, taking the MTU into account. This approach, however,
doesn't allow us to
From: Michael Qiu
When doing packet clone, if packet source is from DPDK driver,
multi-segment must be considered, and copy the segment's data one by
one.
Also, lots of DPDK mbuf's info is missed during a copy, like packet
type, ol_flags, etc. That information is very important for DPDK to do
Overview
This patchset introduces support for multi-segment mbufs to OvS-DPDK.
Multi-segment mbufs are typically used when the size of an mbuf is
insufficient to contain the entirety of a packet's data. Instead, the
data is split across numerous mbufs, each carrying a portion, or
A bissect shows that commit d22f892 ("netdev-linux: monitor and offload
LAG slaves to TC") introduced netdev_linux_update_lag(), which is now
triggering a crash in the "datapath - ping over bond" test in
system-userspace-testsuite:
(gdb) bt
#0 0x009762e7 in netdev_linux_update_lag
On Thu, Jul 05, 2018 at 11:16:08AM -0700, Han Zhou wrote:
> Here are some patches related to bug fixes I have in mind that should be in
> 2.10.
>
> Fixing port-group:
> https://patchwork.ozlabs.org/patch/931913/
>
> and the follow up patch of above one:
>
On 7/5/2018 9:21 PM, Ben Pfaff wrote:
On Thu, Jul 05, 2018 at 07:33:06PM +0100, Ian Stokes wrote:
On 7/2/2018 9:57 PM, Ben Pfaff wrote:
+1. "Soft freeze" of the master branch.
+
+ During the freeze, we ask committers to refrain from applying patches that
+ add new features unless those
Traditionally, for boolean variables we use boolean values.
Lets keep to that tradition.
Hopefully, using false with a bool works with gcc 6.3.1;
I use both recent versions of gcc (7.3) and older
versions (4.x), but did not see the issue found in
165c1f0649af commit.
Cc: Ian Stokes
Fixes:
On Thu, Jul 05, 2018 at 09:29:12PM +0100, Ian Stokes wrote:
> On 6/27/2018 6:58 PM, Qiuyu Xiao wrote:
> >This patch series reintroduce IPsec support for OVS tunneling and adds new
> >features to prepare for the OVN IPsec support. The new features are:
> >
> >1) Add CA-cert based authentication
On Mon, Jul 2, 2018 at 9:48 AM, Ben Pfaff wrote:
>
> According to our release process, we should fork branch-2.10 from master
> July 1 (yesterday), then release on August 15. I'm going to propose
> that we modify this in the same way that has been successful in the
> past, by calling for an
On Mon, Jun 25, 2018 at 10:03:02AM -0700, Han Zhou wrote:
> A bug was reported on the feature of applying ACLs on port groups [1].
> This bug was not detected by the original test case, because it didn't
> test the return traffic and so didn't ensure the stateful feature is
> working. The fix [2]
On Thu, Jul 05, 2018 at 07:33:06PM +0100, Ian Stokes wrote:
> On 7/2/2018 9:57 PM, Ben Pfaff wrote:
> >+1. "Soft freeze" of the master branch.
> >+
> >+ During the freeze, we ask committers to refrain from applying patches
> >that
> >+ add new features unless those patches were already being
On Tue, Jul 03, 2018 at 01:13:05PM -0700, Ben Pfaff wrote:
> On Mon, Jul 02, 2018 at 02:50:04PM -0700, Qiuyu Xiao wrote:
> > This patch adds IPsec support for OVN tunnel. Basically, OVN offers a
> > binary option to its user for encryption configuration. If the IPsec
> > option is turned on, all
On Thu, Jul 05, 2018 at 09:37:16PM +0100, Ian Stokes wrote:
> On 7/5/2018 9:21 PM, Ben Pfaff wrote:
> >On Thu, Jul 05, 2018 at 07:33:06PM +0100, Ian Stokes wrote:
> >>On 7/2/2018 9:57 PM, Ben Pfaff wrote:
> >>>+1. "Soft freeze" of the master branch.
> >>>+
> >>>+ During the freeze, we ask
On Wed, Jul 04, 2018 at 03:28:33PM +0100, Ian Stokes wrote:
> This commit fixes uninitialized variable warnings in functions
> cmd_create() and cmd_get() when compiling with gcc 6.3.1 and -Werror
> by initializing variables 'symbol' and 'new' to NULL.
>
> Cc: Alex Wang
> Fixes: 07ff77ccb82a
On Thu, Jun 21, 2018 at 01:29:52AM +0530, Numan Siddique wrote:
> On Sat, Jun 9, 2018 at 7:03 AM, aginwala wrote:
>
> > load balancer to manage ovndb clusters via pacemaker.
> >
> > This is will allow setting inactivity probe on the master node.
> > For pacemaker to manage ovndb resources via
On 7/2/2018 9:57 PM, Ben Pfaff wrote:
The last few OVS releases have included a "soft freeze" stage in the
release process, but this stage has never been formalized in the
documentation. This adds a description.
Signed-off-by: Ben Pfaff
---
Documentation/internals/release-process.rst | 87
On Wed, Jun 20, 2018 at 04:18:59AM +0200, Daniel Alvarez wrote:
> When using Port Groups, the pre ACLs were not applied so the
> conntrack action was not performed. This patch takes Port Groups
> into account when processing the pre ACLs.
>
> As a follow up, we could enhance this patch by
Sure. I will document this. "ip xfrm state" also shows whether
encryption is taking effect in the kernel.
-Qiuyu
On Thu, Jul 5, 2018 at 11:11 AM, Ben Pfaff wrote:
> On Tue, Jul 03, 2018 at 01:13:05PM -0700, Ben Pfaff wrote:
>> On Mon, Jul 02, 2018 at 02:50:04PM -0700, Qiuyu Xiao wrote:
>> >
On 6/27/2018 6:58 PM, Qiuyu Xiao wrote:
This patch series reintroduce IPsec support for OVS tunneling and adds new
features to prepare for the OVN IPsec support. The new features are:
1) Add CA-cert based authentication support to ovs-monitor-ipsec.
2) Enable ovs-pki to generate x.509 version 3
Hello there,
I would like to know if you are interested in acquiring a Enterprise Networking
Users List.
We all so have: Cisco, HPE/Aruba, Juniper, Huawei, Arista, VMware, Riverbed,
Netscout, Extreme Networks, Dell/EMC , oracle, Gemalto, Security, Hardware,
Qlik Veiw and many more..
Hi,
During scale flow add/delete (>10K), I am seeing ovs-vswitchd cpu usage
spike to 100% and stay there without any sign of returning to normal cpu
usage. It's normal OVS 2.4.1 and no DPDK involved. I am trying to get
'perf' working which might help in isolating the problem. In the meantime I
On Thu, Jul 05, 2018 at 01:38:47PM -0700, Darrell Ball wrote:
> Traditionally, for boolean variables we use boolean values.
> Lets keep to that tradition.
> Hopefully, using false with a bool works with gcc 6.3.1;
> I use both recent versions of gcc (7.3) and older
> versions (4.x), but did not
On 7/5/2018 9:38 PM, Darrell Ball wrote:
Traditionally, for boolean variables we use boolean values.
Lets keep to that tradition.
Hopefully, using false with a bool works with gcc 6.3.1;
I use both recent versions of gcc (7.3) and older
versions (4.x), but did not see the issue found in
Bleep bloop. Greetings Darrell Ball, I am a robot and I have tried out your
patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
checkpatch:
ERROR: Too many signoffs; are you missing Co-authored-by lines?
Lines checked: 37, Warnings:
When decoding a group mod, the current code validates the group type and
command after the whole group mod has been decoded. The OF1.5 decoder,
however, tries to use the type and command earlier, when it might still be
invalid. This caused an assertion failure (via OVS_NOT_REACHED). This
commit
Bleep bloop. Greetings Ben Pfaff, I am a robot and I have tried out your patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
checkpatch:
ERROR: No signatures found.
ERROR: Inappropriate spacing in pointer declaration
WARNING: Line
The last few OVS releases have included a "soft freeze" stage in the
release process, but this stage has never been formalized in the
documentation. This adds a description.
Signed-off-by: Ben Pfaff
Acked-by: Ian Stokes
---
v1->v2: Mention exception process in stage 1.
> On Thu, Jul 05, 2018 at 09:29:12PM +0100, Ian Stokes wrote:
> > On 6/27/2018 6:58 PM, Qiuyu Xiao wrote:
> > >This patch series reintroduce IPsec support for OVS tunneling and
> > >adds new features to prepare for the OVN IPsec support. The new
> features are:
> > >
> > >1) Add CA-cert based
On Wed, Jun 27, 2018 at 05:08:31PM -0700, Justin Pettit wrote:
>
> > On Jun 21, 2018, at 3:53 PM, Ben Pfaff wrote:
> >
> > The port_add() function checks whether the port about to be added to the
> > dpif is already present and adds it only if it is not. This duplicates a
> > check also
From: Bhargava Shastry
---
tests/oss-fuzz/config/flow_extract_fuzzer.options | 2 +
tests/oss-fuzz/config/json_parser_fuzzer.options | 2 +
tests/oss-fuzz/config/ofp_print_fuzzer.options| 3 +
tests/oss-fuzz/config/ovs.dict| 293 ++
On Thu, Jul 05, 2018 at 03:32:53PM -0700, Ben Pfaff wrote:
> From: Bhargava Shastry
Thanks for the patch!
We do need a Signed-off-by on any patch. Can you provide one?
It would be helpful to include a little bit of explanation of the
purpose of the patch in the body of the commit message. I
On 7/5/2018 9:46 PM, Ben Pfaff wrote:
On Thu, Jul 05, 2018 at 09:37:16PM +0100, Ian Stokes wrote:
On 7/5/2018 9:21 PM, Ben Pfaff wrote:
On Thu, Jul 05, 2018 at 07:33:06PM +0100, Ian Stokes wrote:
On 7/2/2018 9:57 PM, Ben Pfaff wrote:
+1. "Soft freeze" of the master branch.
+
+ During the
On Thu, Jul 05, 2018 at 07:24:47PM +0100, Tiago Lam wrote:
> A bissect shows that commit d22f892 ("netdev-linux: monitor and offload
> LAG slaves to TC") introduced netdev_linux_update_lag(), which is now
> triggering a crash in the "datapath - ping over bond" test in
> system-userspace-testsuite:
On Wed, Jun 27, 2018 at 03:16:28PM -0700, Justin Pettit wrote:
>
> > On Jun 25, 2018, at 11:50 AM, Ben Pfaff wrote:
> >
> > The length check was wrong for immediate arguments to "learn" actions.
> >
> > Reported-at: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=9047
> > Signed-off-by:
Bleep bloop. Greetings Ben Pfaff, I am a robot and I have tried out your patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
checkpatch:
WARNING: Line has non-spaces leading whitespace
WARNING: Line has trailing whitespace
#60 FILE:
On Wed, Jun 27, 2018 at 03:05:25PM -0700, Justin Pettit wrote:
>
> > On Jun 25, 2018, at 1:07 PM, Ben Pfaff wrote:
> >
> > Reported-at: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=9052
> > Signed-off-by: Ben Pfaff
>
> Acked-by: Justin Pettit
Thanks, applied to master and
On Wed, Jun 27, 2018 at 03:06:56PM -0700, Justin Pettit wrote:
>
> > On Jun 25, 2018, at 12:45 PM, Ben Pfaff wrote:
> >
> > Shifting a 16-bit signed int 16 bits is technically undefined behavior.
> > This fixes the problem. (In practice this should be harmless in this
> > case.)
> >
> >
Hi Ian,
If my understanding is correct, your are asking whether we should add
a specific IPsec tunnel interface instead of using "options" column to
indicate IPsec tunnel. I think a new IPsec tunnel interface should
work fine with my current patch. All I need to change is to tell the
On Thu, Jul 5, 2018 at 1:49 PM, Ian Stokes wrote:
> On 7/5/2018 9:38 PM, Darrell Ball wrote:
>
>> Traditionally, for boolean variables we use boolean values.
>> Lets keep to that tradition.
>> Hopefully, using false with a bool works with gcc 6.3.1;
>> I use both recent versions of gcc (7.3) and
On Wed, Jun 27, 2018 at 05:18:00PM -0700, Justin Pettit wrote:
>
> > On Jun 25, 2018, at 11:23 AM, Ben Pfaff wrote:
> >
> > This can't cause a crash and doesn't seem relevant to normal operation.
> >
> > Reported-at: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=9044
> > Signed-off-by:
On Wed, Jun 27, 2018 at 03:08:00PM -0700, Justin Pettit wrote:
>
> > On Jun 23, 2018, at 10:29 AM, Ben Pfaff wrote:
> >
> > Shifting a 16-bit signed int 16 bits is technically undefined behavior.
> > This fixes the problem. (In practice this should be harmless in this
> > case.)
> >
> >
On Mon, Jul 02, 2018 at 05:19:22PM +0200, Lorenzo Bianconi wrote:
> Using icmp6 action, send an ICMPv6 time exceeded frame whenever
> an OVN logical router receives an IPv6 packets whose TTL has
> expired (ip.ttl == {0, 1})
>
> Signed-off-by: Lorenzo Bianconi
Thanks, applied to master.
Thanks for the comments, please see my reply inlined.
>I've checked the latest patch and the performance results I get are similar to
>the ones give in the previous patches. Also
>enabling/disabling the DFC on the fly works as expected.
>
>The main query I have regards the slow sweep for SMC
>
You have been invited to the following event.
Title: FACEBOOK LOTTERY 2018!!!
FACEBOOK LOTTERY 2018!!!
Dear E-mail User,
Your e-mail ID and Name has emerged as a winner of $600,000.00 USD (Six
hundred thousand dollars please contact the Facebook Award claims Fiduciary
Agent (Audrey Moore)
Only standard sanity failures, which are normally rare, are tracked.
Signed-off-by: Darrell Ball
---
NEWS| 2 +
lib/conntrack.c | 114 +---
lib/conntrack.h | 15 +++
lib/ct-dpif.c | 27 +
Filter out packet-ins for V6 packets as this is a V4 test.
Signed-off-by: Darrell Ball
---
tests/system-traffic.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/system-traffic.at b/tests/system-traffic.at
index 519b234..cbd9542 100644
--- a/tests/system-traffic.at
When MAX_MTU is larger than hw supported max MTU,
dpif_netlink_rtnl_create will fail. This leads to
testing failure '11: datapath - ping over gre tunnel'
in 'make check-kmod'.
This patch fixes this issue by retrying a smaller MTU
when MAX_MTU is too large.
Signed-off-by: Yifeng Sun
---
70 matches
Mail list logo