On Sep 24, 2014, at 11:18 PM, Alin Serdean aserd...@cloudbasesolutions.com
wrote:
The following script leverage's the advantages of WMI infrastructure
offered in Hyper-V.
This scripts allows the user to change the
Msvm_EthernetPortAllocationSettingData property of a VM network adapter
This commit updates the DPDK related documentation to reflect
the pmd thread multi-threading work.
Signed-off-by: Alex Wang al...@nicira.com
---
INSTALL.DPDK | 70 ++
vswitchd/vswitch.xml |5 ++--
2 files changed, 39 insertions(+), 36
The message could not be delivered
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
This series modifies the dpif interface for flow commands to use 128-bit unique
identifiers as an alternative to the netlink-formatted flow key, and caches the
mask/actions in the udpif_key. This significantly reduces the cost of
assembling messages between revalidators and the datapath, improving
We previously counted flows that have been installed during the current
dump as duplicates, rather than recognising them as new flows. This
patch separates the counters out for these two cases.
Signed-off-by: Joe Stringer joestrin...@nicira.com
---
v5-v6: No change.
v4: Initial post.
---
Currently, udpif_keys are protected during revalidator_sweep__() as only
one thread accesses the ukey at a time. This is ensured using barriers:
all revalidators will be in the GC phase, so they will only access their
own ukey collection.
A future patch will change the access patterns to allow
Future patches will make use of the 'struct dump_op' in a broader sense,
so this patch renames it to make things a bit clearer.
Signed-off-by: Joe Stringer joestrin...@nicira.com
Acked-by: Ben Pfaff b...@nicira.com
---
v4-v6: No change.
v3: Rebase.
v2: No change.
RFC: First post.
---
Signed-off-by: Joe Stringer joestrin...@nicira.com
Acked-by: Ben Pfaff b...@nicira.com
---
v4-v6: No change.
v3: Rebase.
v2: Call ovsrcu_quiesce() unconditionally.
RFC: Initial Post.
---
ofproto/ofproto-dpif-upcall.c | 61 ++---
1 file changed, 33
Currently, when a revalidator thread first dumps a flow, it creates a
'udpif_key' object and caches a copy of a kernel flow key. This allows
us to perform lookups in the classifier to attribute stats and validate
the correctness of the datapath flow.
This patch sets up this cache from the handler
An upcoming patch will change the access patterns for ukey maps to
increase the number of writers, and shift write-access from revalidator
threads to upcall handler threads. As such, it no longer makes sense to
tie these maps to revalidators in a 1:1 relationship.
This patch separates the ukey
This just makes it a bit easier for someone coming in fresh to do some
searches and figure out what the description means.
Signed-off-by: Joe Stringer joestrin...@nicira.com
---
v6: First post.
---
tests/test-hash.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git
This allows us to ignore most fields of a flow_dump, requiring only the
flow key for looking up the ukey. Fetching flows can also be avoided in
the corner case where a flow is missed from a dump but revalidation is
required.
A future patch will modify the datapath interface to allow datapaths to
Add the 128-bit murmurhash by Austin Appleby, r150 from:
http://code.google.com/p/smhasher/source/browse/trunk/MurmurHash3.cpp
Signed-off-by: Joe Stringer joestrin...@nicira.com
---
The x64 version is approximately 2x faster than x86 in the tests I ran.
v6: Add version that uses 64-bit
This patch shifts the responsibility for determining the hash for a flow
from the revalidation logic down to the dpif layer. This assists in
handling backward-compatibility for revalidation with the upcoming
unique identifier UID patches, and allows ovs-dpctl to automatically
determine flow hashes
If a datapath is created with the flag OVS_DP_F_INDEX_BY_UID, then an
additional table_instance is added to the flow_table, which is indexed
by unique identifiers (UID). Userspace implementations can specify a
UID of up to 128 bits along with a flow operation as shorthand for the
key. This allows
This patch modifies the dpif interface to allow flows to be manipulated
using a 128-bit identifier. This allows revalidator threads to perform
datapath operations faster, as they do not need to serialise the entire
flow key for operations like flow_get and flow_delete. In conjunction
with a future
One of the limiting factors on the number of flows that can be supported
in the datapath is the overhead of assembling flow dump messages in the
datapath. This patch modifies the flow_dump interface to allow
revalidators to skip dumping the key, mask and actions from the
datapath, by making use of
Previously we stored the netlink-formatted version of a flow key and
mask in 'struct udpif_key'. This patch stores the original key,mask
in the 'struct match' format, which reduces the size of a ukey from 1216
bytes to 560. Furthermore, this reduces netlink conversions required and
simplifies mask
This patch enables a user to set importance for a new rule via add-flow
OF1.1+ in the OVS and display the same via dump-flows command OF1.1+ .
The changes are made in accordance with OpenFlow 1.4 specs to implement
Eviction on the basis of importance.
Signed-off-by: Rishi Bamba
Hi Team,
As per the OpenFlow Specification 1.4, section B.14.7, page 203,
Table-desc eviction property ofp_table_mod_prop_eviction to describe the
type of eviction performed by the switch.
As per our findings, there is a function in ofp-print.c,
ofp_print_table_mod, to print table mod
Hello Nithin,
Yes it works if a VM has multiple network adapters. Invocation is different
though. Example for a VM with 2 network adapters:
$vnic = Get-VMNetworkAdapter test_2_1
$vnic[0]
Connect-VMNetworkAdapter -VMNetworkAdapter $vnic[0] -SwitchName external
$vnic[0] |
Hi,
Are you interested in *Storage Virtualization Users* list for your
marketing campaign?
We are a Global Technology Database provider across the globe with Key
decision makers complete contact details like Name, Title, Email, Phone,
Company Details and the Technology insights.
Other
Hi Alin,
Can you please add a short description of how you see this script is used in
the sequence of actions to add or delete a port to the switch?
I assume that the script will set the friendly name of a port even if the port
is not connected (yet), is it correct?
Thank you,
Eitan
Eitan,
I get the following error when I try to apply this patch. Looks like there's a
new line in build-aux/extract-odp-netlink-windows-dp-h.
---
[nithin@pa-dbc1122 openvswitch-review]$ git am
../tmp/ovs-dev-v3-datapath-windows-Event-read-handler.patch
Applying: datapath-windows Event read
On Fri, Sep 26, 2014 at 05:56:25PM +0530, Saloni Jain wrote:
As per the OpenFlow Specification 1.4, section B.14.7, page 203,
Table-desc eviction property ofp_table_mod_prop_eviction to describe the
type of eviction performed by the switch.
As per our findings, there is a function in
Hi Sam,
Thanks a lot for the review.
Please find my replies inline.
Thanks.
Regards,
Ankur
From: Samuel Ghinet sghi...@cloudbasesolutions.com
Sent: Thursday, September 25, 2014 9:14 AM
To: dev@openvswitch.org
Cc: Ankur Sharma
Subject: RE: [ovs-dev]
Hi Sam,
We plan to handle the error cases properly, but not in this series.
Let me know if it sounds fine.
Thanks.
Regards,
Ankur
From: Samuel Ghinet sghi...@cloudbasesolutions.com
Sent: Thursday, September 25, 2014 9:28 AM
To: dev@openvswitch.org
Cc:
On Fri, Sep 26, 2014 at 04:27:06PM +0530, Rishi Bamba wrote:
This patch enables a user to set importance for a new rule via add-flow
OF1.1+ in the OVS and display the same via dump-flows command OF1.1+ .
The changes are made in accordance with OpenFlow 1.4 specs to implement
Eviction on the
On Fri, Sep 26, 2014 at 09:52:14AM -0700, Ben Pfaff wrote:
On Fri, Sep 26, 2014 at 04:27:06PM +0530, Rishi Bamba wrote:
This patch enables a user to set importance for a new rule via add-flow
OF1.1+ in the OVS and display the same via dump-flows command OF1.1+ .
The changes are made in
On Fri, Sep 26, 2014 at 04:27:06PM +0530, Rishi Bamba wrote:
This patch enables a user to set importance for a new rule via add-flow
OF1.1+ in the OVS and display the same via dump-flows command OF1.1+ .
The changes are made in accordance with OpenFlow 1.4 specs to implement
Eviction on the
On Thu, Sep 25, 2014 at 09:20:25PM +, Samuel Ghinet wrote:
The windows ovs kernel uses an OVS_VPORT_TYPE enum that is incompatible with
the userspace counterpart (enum ovs_vport_type from openvswitch.h). We must
use
the same enum type for the netlink communication to work properly.
On Thu, Sep 25, 2014 at 09:21:33PM +, Samuel Ghinet wrote:
Functionality for vport dump.
Later, when we will add more netlink dump commands, some common code will need
to be split to functions.
Notes:
a) the current implementation of vport assumes the datapath feature
multiple upcall
On Thu, Sep 25, 2014 at 09:22:22PM +, Samuel Ghinet wrote:
Contains error codes for netlink transactional errors.
These errors are passed to the error field (INT) of the NL_MSG_ERR struct.
The userspace requires them to be negative values: the nl_msg_nlmsgerr
userspace
function
On Thu, Sep 25, 2014 at 09:48:44AM +0530, Shashwat Srivastava wrote:
Following is our understanding for Delete Flow mechanism:
We have studied the code and have found that OFPFC_DELETE_STRICT specifies
only two fields on which flow is deleted which are OUT_PORT PRIORITY
i.e. flows are
On Thu, Sep 25, 2014 at 01:02:13PM +0900, Simon Horman wrote:
Signed-off-by: Simon Horman simon.hor...@netronome.com
---
I have now had one patch accepted from this address,
I hope there will be more.
I would also like to continue to use ho...@verge.net.au
for work that is not related
Hi Sam,
Please find my replies inline.
Regards,
Ankur
From: Samuel Ghinet sghi...@cloudbasesolutions.com
Sent: Thursday, September 25, 2014 8:30 AM
To: dev@openvswitch.org
Cc: Ankur Sharma
Subject: RE: [ovs-dev] [PATCH v1 05/10] datapath-windows/Netlink:
Thanks for the review Nithin.
new blank line at EOF.
Thanks, will remove.
Also, I am not very convinced that we cannot reuse the READ_IOCTL, but I am
ok with your approach provided the end-to-end (including userspace code) looks
simpler overall. Maybe including the userspace code would make
On Wed, Sep 24, 2014 at 01:43:19PM +0900, Simon Horman wrote:
It is add-groups rather than add-group that takes FILE as an argument.
Signed-off-by: Simon Horman simon.hor...@netronome.com
Applied, thanks!
___
dev mailing list
dev@openvswitch.org
It looks like this wasn't backported, should it be?
On Wed, Sep 24, 2014 at 02:06:46PM -0700, Andy Zhou wrote:
Pushed. Thanks for the fix!
On Tue, Sep 23, 2014 at 9:41 PM, Simon Horman
simon.hor...@netronome.com wrote:
When searching through the valid ids an id should
be used if is not
On Wed, Sep 03, 2014 at 10:28:39AM -0700, Ben Pfaff wrote:
The Open vSwitch make output was still pretty verbose even when
configured with --enable-silent-rules. This cleans it up.
Signed-off-by: Ben Pfaff b...@nicira.com
---
I tested that this yields good output with the following travis
On Wed, Sep 24, 2014 at 11:24:00AM -0700, Jarno Rajahalme wrote:
Aligned 64-bit memory accesses in i586 are atomic. By using an SSE
register we can make such memory accesses in one instruction without
bus-locking. Need to compile with -msse to enable this feature.
Signed-off-by: Jarno
I will back port to branch-2.3. Thanks for the reminder.
On Fri, Sep 26, 2014 at 10:46 AM, Ben Pfaff b...@nicira.com wrote:
It looks like this wasn't backported, should it be?
On Wed, Sep 24, 2014 at 02:06:46PM -0700, Andy Zhou wrote:
Pushed. Thanks for the fix!
On Tue, Sep 23, 2014 at 9:41
On Wed, Sep 24, 2014 at 11:24:01AM -0700, Jarno Rajahalme wrote:
Earlier I have seen the frags checking suspiciously high in perf
reports, but did not understand why. Maybe this explains it:
Previously we checked frags handling before each flow table lookup.
As the nw_frags field is not
On Wed, Sep 24, 2014 at 4:32 PM, Thomas Graf tg...@suug.ch wrote:
On 09/23/14 at 06:32pm, Or Gerlitz wrote:
Indeed.
The idea is to leverage OVS to manage eSwitch (embedded NIC switch) as well
(NOT to offload OVS).
We envision a seamless integration of user environment which is based on OVS
On Wed, Sep 24, 2014 at 11:24:02AM -0700, Jarno Rajahalme wrote:
Before:
$ tests/ovstest test-bitmap benchmark 100
bitmap equal:328 ms
bitmap scan: 8159 ms
After:
$ tests/ovstest test-bitmap benchmark 100
bitmap equal:230 ms
bitmap scan:185 ms
Signed-off-by:
On Wed, Sep 24, 2014 at 11:36:57AM -0700, Jarno Rajahalme wrote:
Correcting myself:
On Sep 24, 2014, at 11:24 AM, Jarno Rajahalme jrajaha...@nicira.com wrote:
-unsigned long *bitmap_allocate1(size_t n_bits);
+/* Initializes bitmap to all-1-bits and returns the bitmap pointer. */
+static
Hi Nithin,
Is there any place where we check nla-nlaLen that is larger NLA_HDRLEN? We
need to add this to the NL attr validation.
I'm not sure what happens if NlAttrGetSize retirns with 0 or -1 :-)
Perhaps we want to add minimal for OVS_VPORT_ATTR_NAME so NlAttrValidate will
catch it.
Thanks,
I wanted to post the latest state of the conntrack nat patches based
on Justin's initial RFC before I travel.
Changes to Justin's patch include the introduction of a new model to
convert nested Netlink attributes to a in-kernel only struct to allow
the zone support patch to keep a reference to
WIP
---
datapath/actions.c | 8
datapath/flow.h | 1 +
datapath/flow_netlink.c | 22 ++
3 files changed, 31 insertions(+)
diff --git a/datapath/actions.c b/datapath/actions.c
index beed5d8..05b465c 100644
--- a/datapath/actions.c
+++
From: Justin Pettit jpet...@nicira.com
An RFC to get back early feedback on exposing Linux's kernel connection
tracker to OVS. The code has a few rough spots that will be addressed
in the next version:
- Need better interface than setting individual flags for the state.
- Need support
WIP
---
datapath/actions.c| 95 +++
datapath/flow.h | 6 +
datapath/flow_netlink.c | 139 ++
datapath/linux/compat/include/linux/openvswitch.h | 54 +
4 files
WIP
---
lib/dpif-netdev.c| 1 +
lib/dpif.c | 1 +
lib/odp-execute.c| 1 +
lib/odp-util.c | 113 +++
lib/ofp-actions.c| 104 +++
lib/ofp-actions.h
The Read event handler is executed when user mode issues a socket
receive on an MC socket associated with the event queue. A new IOCTL
READ command is used to differentiate between transaction based and packet
miss sockets.
An entry for the handler will be added once the Control family (logically
On Sep 26, 2014, at 11:06 PM, Eitan Eliahu elia...@vmware.com
wrote:
The Read event handler is executed when user mode issues a socket
receive on an MC socket associated with the event queue. A new IOCTL
READ command is used to differentiate between transaction based and packet
miss sockets.
Acked-by: Ankur Sharma ankursha...@vmware.com
Regards,
Ankur
From: dev dev-boun...@openvswitch.org on behalf of Eitan Eliahu
elia...@vmware.com
Sent: Friday, September 26, 2014 11:06 PM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v4]
OVSDB has the concept of immutable columns, which are columns whose
values are fixed once a row is inserted. Until now, ovs-vsctl has not
allowed these columns to be modified at all. However, this is a little too
strict, because these columns can be set to any value at the time that the
row is
On Thu, Sep 18, 2014 at 10:55:10AM +0900, Simon Horman wrote:
This patch is a prototype and has several limitations:
* It assumes that no actions follow a select group action
because the resulting packet after a select group action may
differ depending on the bucket used. It may be
On Sep 24, 2014, at 2:43 AM, Sorin Vinturis svintu...@cloudbasesolutions.com
wrote:
Currently, whenever there's a missed packet, the ovs driver allocates
memory and copies the packet even if there's no packet queue setup from
userspace. Then if there's no queue created, the packet is
Moved the structure OVS_MESSAGE to Netlink.h.
This change is done for following reasons.
a. Patch 2 in this series provides a generic API in Netlink.c
for creating netlink message. That API needs OVS_MESSAGE.
Including Datapath.h in Netlink.c/h gives compilation error.
b. OVS_MESSAGE defines
Added NlFillOvsMsg API in Netlink.c This API will be used to populate
netlink message headers.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Acked-by: Nithin Raju nit...@vmware.com
Added an API to retrieve the attribute length.
Added 2 more API for BE16 and BE8 attribute parsing.
Fixed a trailing whitespace issue.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Acked-by: Nithin Raju nit...@vmware.com
Acked-by: Samuel Ghinet sghi...@cloudbasesolutions.com
Tested-by: Ankur Sharma ankursha...@vmware.com
This patch covers basic changes in registering add flow handler.
And declaring FLOW related attribute parsing policies.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Acked-by: Nithin Raju
NlAttrParseNested was using the whole netlink payload for iteration.
This is not correct, as it would lead to exceeding the
nested attribute boundries. Fixed the same in this patch.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Registered FLOW_SET command handler. The same command
handler as FLOW_ADD is good enough to handle FLOW_SET
case as well.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Acked-by: Nithin
This patch covers the changes needed to support FLOW_NEW command.
API _OvsFlowMapNlToFlowPutFlags has a bug, which will be fixed
with the patches for FLOW_DEL.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu
Registered FLOW_DEL command handler. The same command
handler as FLOW_ADD is good enough to handle FLOW_DEL
case as well with minor changes for checking to action
attribute.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by:
Added changes to handle DEL_FLOWS (FLUSH) scenario.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Acked-by: Nithin Raju nit...@vmware.com
Acked-by: Samuel Ghinet
OvsFlowNlNewCmdHandler NEW/DEL/SET/FLUSH commands hence renaming it to
OvsFlowNlCmdHandler.
Signed-off-by: Ankur Sharma ankursha...@vmware.com
Acked-by: Alin Gabriel Serdean aserd...@cloudbasesolutions.com
Acked-by: Eitan Eliahu elia...@vmware.com
Acked-by: Nithin Raju nit...@vmware.com
Acked-by:
The Read event handler is executed when user mode issues a socket
receive on an MC socket associated with the event queue. A new IOCTL
READ command is used to differentiate between transaction based and packet
miss sockets.
An entry for the handler will be added once the Control family (logically
While calculating the hash on a VPORT name, we don't include the NUL character.
We should be doing the same while doing lookup as well.
We set the required minimum length of the name to be 2 so that the string has
at least one valid character.
Signed-off-by: Nithin Raju nit...@vmware.com
Signed-off-by: Nithin Raju nit...@vmware.com
---
datapath-windows/ovsext/Datapath.c |3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/datapath-windows/ovsext/Datapath.c
b/datapath-windows/ovsext/Datapath.c
index b6dc55e..8f8499b 100644
---
Thanks Alin. I tested this script and it works!
It would be great if we can document the usage in INSTALL.Windows - both for
one vNIC case and 2 vNIC case. It does not have to be part of this checkin.
Tested-by: Nithin Raju nit...@vmware.com
Acked-by: Nithin Raju nit...@vmware.com
-- Nithin
Dear user dev@openvswitch.org,
Your e-mail account has been used to send a huge amount of spam messages during
this week.
Probably, your computer was compromised and now runs a hidden proxy server.
We recommend you to follow the instruction in order to keep your computer safe.
Virtually yours,
74 matches
Mail list logo