Added an API to retrieve the attribute length.
Added 2 more API for BE16 and BE8 attribute parsing.
Fixed a trailing whitespace issue.
---
datapath-windows/ovsext/Netlink/Netlink.c | 28 +++-
datapath-windows/ovsext/Netlink/Netlink.h | 13 -
datapath-windows/ovs
---
datapath-windows/ovsext/Netlink/Netlink.c | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/datapath-windows/ovsext/Netlink/Netlink.c
b/datapath-windows/ovsext/Netlink/Netlink.c
index 5f07451..5c74ec0 100644
--- a/datapath-windows/ovsext/Netlink/Netlink.c
+++ b/
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.
---
datapath-windows/ovsext/Datapath.c| 4 +++-
datapath-windows/ovsext/Netlink/Netlink.c | 15
This patch covers basic changes in registering add flow handler.
And declaring FLOW related attribute parsing policies.
---
datapath-windows/ovsext/Datapath.c | 18 -
datapath-windows/ovsext/Flow.c | 152 +
datapath-windows/ovsext/Flow.h | 5 ++
3
Added NlFillOvsMsg API in Netlink.c This API will be used to populate
netlink message headers.
---
datapath-windows/ovsext/Netlink/Netlink.c | 38 +++
datapath-windows/ovsext/Netlink/Netlink.h | 14
2 files changed, 52 insertions(+)
diff --git a/datapath-w
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.
---
datapath-windows/include/OvsPub.h | 2 +-
datapath-windows/ovsext/Flow.c| 384 +++---
datapath-w
Registered FLOW_SET command handler. The same command
handler as FLOW_ADD is good enough to handle FLOW_SET
case as well.
---
datapath-windows/ovsext/Datapath.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/datapath-windows/ovsext/Datapath.c
b/datapath-windows/ovsext/D
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 netl
Added changes to handle DEL_FLOWS (FLUSH) scenario.
---
datapath-windows/ovsext/Flow.c | 19 ++-
datapath-windows/ovsext/Flow.h | 2 +-
2 files changed, 11 insertions(+), 10 deletions(-)
diff --git a/datapath-windows/ovsext/Flow.c b/datapath-windows/ovsext/Flow.c
index b95f69b..0
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.
---
datapath-windows/ovsext/Datapath.c | 5 +
datapath-windows/ovsext/Flow.c | 7 +--
2 files changed, 10 insert
This series covers the changes for FLOW_ADD/DEL/SET/FLUSH implementation.
Some fixes were needed in Netlink code as well. Those fixes are
also included in this series.
Ankur Sharma (10):
datapath-windows: move OVS_MESSAGE to Netlink.h
datapath-windows/Netlink: Add NlFillOvsMsg API for creating
Hi Thomas,
While working on 802.1ad feature(provider bridging),will you also incorporate
IEEE 802.1ah (Provider Backbone Bridging)?
Can we Plan to implement IEEE 802.1ah (Provider Backbone Bridging) feature
separtely?
Thanks & Regards,
Hiteshi Madan
-Ben Pfaff wrote: -
To: Hiteshi Mad
On 09/24/14 at 03:00pm, Simon Horman wrote:
> On Fri, Sep 19, 2014 at 03:06:38PM +0100, Thomas Graf wrote:
> > Can we rename & move this to instead?
>
> Sure, how about nla_is_last()?
Sounds great
___
dev mailing list
dev@openvswitch.org
http://openvsw
On 09/24/14 at 03:01pm, Simon Horman wrote:
> > > + /* Only possible type of attributes is OVS_SELECT_GROUP_ATTR_BUCKET */
> > > + for (bucket = nla_data(attr), rem = nla_len(attr); rem > 0;
> > > + bucket = nla_next(bucket, &rem)) {
> > > + uint16_t weight = bucket_weight(bucket);
> >
Hi Thomas,
While working on 802.1ad feature(provider bridging),will you also incorporate
IEEE 802.1ah (Provider Backbone Bridging)?
Can we Plan to implement IEEE 802.1ah (Provider Backbone Bridging) feature
separtely?
Thanks & Regards,
Hiteshi Madan
-Ben Pfaff wrote: -
To: Hiteshi Mad
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 released and
dropped.
The solution was to check for the existence of the user
Send OFPET_BAD_INSTRUCTION/OFPBIC_BAD_TABLE_ID if table is invalid
in goto table instruction.
Signed-off-by: Selvamuthukumar
---
lib/ofp-actions.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/ofp-actions.c b/lib/ofp-actions.c
index 0a5045a..ef6d83f 100644
--- a/lib/o
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
> with SRIOV eSwitch and the grounds for that were very well
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 pids" is not used. A single upcall pid is used now.
c) the vxlan destin
Hi Sam,
Thank you for posting this patch.
Can you please return NT status rather than NDIS status from all NL based
functions?
Thank you,
Eitan
-Original Message-
From: Samuel Ghinet [mailto:sghi...@cloudbasesolutions.com]
Sent: Wednesday, September 24, 2014 7:01 AM
To: dev@openvswitch.o
Contains error codes for netlink transactional errors.
These errors are passed to the "error" field of the NL_MSG_ERR struct.
The userspace requires them to be negative values: the nl_msg_nlmsgerr userspace
function transforms them from negative to positive values.
These error codes correspond to
The transactional get vport command.
This command uses the netlink transactional errors.
Signed-off-by: Samuel Ghinet
---
datapath-windows/ovsext/Datapath.c | 83 ++
1 file changed, 83 insertions(+)
diff --git a/datapath-windows/ovsext/Datapath.c
b/datapath-
The old method of finding vports based on the
ovs port numbers is removed. Now, the lookup of a vport
by ovs port number is done the same way as the lookup by
hyper-v switch port id.
This is done so that the kernel is able to interact with
the userspace correctly when using vport channels.
The pro
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 Vport family implementation
Acked-by: Eitan Eliahu
-Original Message-
From: Samuel Ghinet [mailto:sghi...@cloudbasesolutions.com]
Sent: Wednesday, September 24, 2014 8:41 AM
To: dev@openvswitch.org
Cc: Alin Serdean; Eitan Eliahu; Nithin Raju; Saurabh Shah; Ankur Sharma;
Kaushik Guha
Subject: [PATCH 1/3] datapath-w
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: Samuel Ghinet
Trimis: Wednesday, September 24, 2014 5:01 PM
Către: dev@openvswitch.org
Cc: Alin Serdean; nit...@vmware.com; ssaur...@vmware.com; elia...@vmware.com;
Ankur Sharma; kg...@vmware.com
Subiect: [PATCH v3] datapath-windows
Some minor formatting issues.
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: Samuel Ghinet
Trimis: Wednesday, September 24, 2014 6:41 PM
Către: dev@openvswitch.org
Cc: Alin Serdean; elia...@vmware.com; nit...@vmware.com; ssaur...@vmware.com;
Ankur Sharma; kg...@vmware.com
Subie
Looks good. Please find comments inline.
Thanks,
Eitan
-Original Message-
From: Samuel Ghinet [mailto:sghi...@cloudbasesolutions.com]
Sent: Wednesday, September 24, 2014 8:42 AM
To: dev@openvswitch.org
Cc: Alin Serdean; Eitan Eliahu; Nithin Raju; Ankur Sharma; Saurabh Shah
Subject: [PATCH
On Wed, Sep 24, 2014 at 09:46:48AM +, Selva Muthu Kumar VelmurugaKadavul
wrote:
> Send OFPET_BAD_INSTRUCTION/OFPBIC_BAD_TABLE_ID if table is invalid
> in goto table instruction.
>
> Signed-off-by: Selvamuthukumar
Thanks for the fix. I applied this to master and branch-2.3.
This broke some
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 Rajahalme
---
lib/ovs-atomic-i586.h | 79 ++
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: Kmindg
Signed-off-by: Jarno Rajahalme
---
lib/bitmap.c | 102 ++
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 writeable, it suffices to check them
once, before the first table lookup.
We use the 'counter' as a "lock" providing acquire-release
semantics. Therefore we can use normal, non-atomic access to the
memory accesses between the atomic accesses to 'counter'. The
cmap_node.next needs to be RCU, so that can not be changed.
Also rearrange code to benefit from the fact that
The test cases have been carefully crafted so that we do the same
amount of "overhead" operations in each case. Earlier, with no
mutations, the number of random number generations was different for
hmap and cmap test cases. hmap test was also missing an ignore() call.
Now the numbers look like th
These makes cmap_find 10% faster on GCC 4.7 (-O2 -g).
Signed-off-by: Jarno Rajahalme
---
lib/cmap.c | 44 +++-
1 file changed, 27 insertions(+), 17 deletions(-)
diff --git a/lib/cmap.c b/lib/cmap.c
index ba744cc..fa7c0bb 100644
--- a/lib/cmap.c
+++
Batching the cmap find improves the memory behavior with large cmaps
and can make searches twice as fast:
$ tests/ovstest test-cmap benchmark 200 8 0.1 16
Benchmarking with n=200, 8 threads, 0.10% mutations, batch size 16:
cmap insert:533 ms
cmap iterate:57 ms
batch search: 146 m
This makes the following patch easier and cleans up the code.
Explicit "inline" keywords seem necessary to prevent performance
regression on cmap_find() with GCC 4.7 -O2.
Signed-off-by: Jarno Rajahalme
---
lib/classifier.c |2 +-
lib/cmap.c| 116 +++
Correcting myself:
On Sep 24, 2014, at 11:24 AM, Jarno Rajahalme wrote:
> -unsigned long *bitmap_allocate1(size_t n_bits);
> +/* Initializes bitmap to all-1-bits and returns the bitmap pointer. */
> +static inline void
> +bitmap_init1(unsigned long *bitmap, size_t n_bits)
> +{
> +size_t n_lon
MSVC converts 64 bit read/writes into two instructions (uses 'mov' as
seen through cl //FAs). So there is a possibility that an interrupt can
make a 64 bit read/write non-atomic even when 8 byte aligned. So we cannot use
a simple assignment. Use a full memory barrier function instead.
Reported-by:
Looks good to me:
Acked-by: Jarno Rajahalme
On Sep 24, 2014, at 11:52 AM, Gurucharan Shetty wrote:
> MSVC converts 64 bit read/writes into two instructions (uses 'mov' as
> seen through cl //FAs). So there is a possibility that an interrupt can
> make a 64 bit read/write non-atomic even when 8
Guru, can we differentiate between MSVC 32bit and 64bit builds?
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Gurucharan Shetty
Sent: Wednesday, September 24, 2014 11:53 AM
To: dev@openvswitch.org
Cc: Gurucharan Shetty
Subject: [ovs-dev] [PAT
On Wed, Sep 24, 2014 at 12:54 PM, Eitan Eliahu wrote:
>
> Guru, can we differentiate between MSVC 32bit and 64bit builds?
The ovs-atomic-msvc.h is only included for 32 bit builds (see
ovs-atomic.h). So everything inside it is 32 bit compatible only. If
the suggestion is that we should change the n
I just wanted to be sure that this change does not apply to 64 bit compilation
where a 64 bit aligned move is atomic.
Yes, this is another reason to move to 64 but builds.
Thanks,
Eitan
-Original Message-
From: Gurucharan Shetty [mailto:shet...@nicira.com]
Sent: Wednesday, September 24
On Wed, Sep 24, 2014 at 1:12 PM, Eitan Eliahu wrote:
> I just wanted to be sure that this change does not apply to 64 bit
> compilation where a 64 bit aligned move is atomic.
Okay. That is taken care as of now.
> Yes, this is another reason to move to 64 but builds.
I think we can make an attem
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Nithin Raju
> Sent: Tuesday, September 23, 2014 1:58 PM
> To: Samuel Ghinet
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v2] datapath-windows: fix OVS_VPORT_TYPE
>
> On Sep 23, 2014, at 5:45 AM,
Acked-by: Eitan Eliahu
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v1 01/10] datapath-windows: move OVS_MESSAGE to
Netlink.h
Moved the structur
Hi Ankur, do we really need to align the size of the NL message to 4 bytes? In
this case we don't add any attributes after the OVS header?
Otherwise, looks good.
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, Sep
Pushed. Thanks for the fix!
On Tue, Sep 23, 2014 at 9:41 PM, Simon Horman
wrote:
> When searching through the valid ids an id should
> be used if is not found rather than if it is found.
>
> It appears to me that without this change duplicate recirculation
> ids may used in cases where the last r
NlAttrGetBeXX should be inline. Ankur will make them inline in a successive
patch so we can proceed now with the flow dump.
Acked-by: Eitan Eliahu
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
> -Original Message-
> From: Samuel Ghinet [mailto:sghi...@cloudbasesolutions.com]
> Sent: Wednesday, September 24, 2014 7:01 AM
> To: dev@openvswitch.org
> Cc: Alin Serdean; Nithin Raju; Saurabh Shah; Eitan Eliahu; Ankur Sharma;
> Kaushik Guha
> Subject: [PATCH v3] datapath-windows: Netlin
Hi Eitan,
You are right. We should let the caller deal with alignment.
I'll submit a v2 changing nlMsg->nlmsgLen = NLMSG_ALIGN(NlBufSize(nlBuf)); to
nlMsg->nlmsgLen = (NlBufSize(nlBuf));
Thanks.
Regards,
Ankur
From: Eitan Eliahu
Sent: Wednesday, Septembe
Acked-by: Eitan Eliahu
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v1 04/10] datapath-windows/Netlink: Allow support for
NESTED Attributes in Nl
Acked-by: Eitan Eliahu
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v1 05/10] datapath-windows/Netlink: Fixed
NlAttrParseNested
Nl
Acked-by: Eitan Eliahu
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v1 06/10] datapath-windows/Flow.c : Basic support for
add-flow.
This patch co
Please find comments inline.
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v1 07/10] datapath-windows/Flow.c: FLOW_NEW command
handler
Acked-by: Eitan Eliahu
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH v1 08/10] datapath-windows/Flow.c: FLOW_SET command
handler.
Registered FLOW
Can you please change the name of the handler as it is handled multiple flow
commands?
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ankur Sharma
Sent: Wednesday, September 24, 2014 12:16 AM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH
Hi Nithin,
> Minor: the commit message seems a little inaccurate. One suggestion I have is
> we should probably say something like:
Fixed
> Suggestion: you can annotate the github issue #:
> https://github.com/openvswitch/ovs-issues/issues/2 (If I am not wrong).
I'm not sure how you meant. I real
All NET_BUFFERs of a NET_BUFFER_LIST must go through the pipeline: extract,
find flow, execute. Previously, for NBLs sent from NDIS, we handled only cases
where the NBL had a single NB. Handling NBLs with multiple NBs was not
implemented.
OvsPartialCopyToMultipleNBLs is used to make each NET_BUFFE
Hi Eitan,
1. Git apply whitespaces
$ git apply ovs-dev-datapath-windows-Event-read-handler.patch
ovs-dev-datapath-windows-Event-read-handler.patch:108: trailing whitespace.
* Utility function to fill up information about the state of a port in a reply
ovs-dev-datapath-windows-Event-read-handler.p
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH v1 01/10] datapath-windows: move OVS_MESSAGE to
Netlink.h
Moved the
Thanks Alin for the review. I'll fix and send a new patch.
Eitan
-Original Message-
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, September 24, 2014 4:56 PM
To: Eitan Eliahu; dev@openvswitch.org
Subject: RE: [ovs-dev] [PATCH] datapath-windows Event read hand
Hi Ankur,
As Eitan was mentioning drop:
nlMsg->nlmsgLen = NLMSG_ALIGN(NlBufSize(nlBuf));
I see you declaring the following structure but never used:
/* Structure to hold arguments needed by NlFillOvsMsg */
typedef struct _NL_FILL_ARGS {
UINT16 nlmsgType;
UINT8 genlCmd;
UINT8 vers
Think it would be a good idea to inline the functions.
Also maybe add asserts to nla.
Alin.
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH v1 03/10
Hi Alin,
Yes i'll remove NLMSG_ALIGN.
My bad i missed out removing NL_FILL_ARGS. I'll spin up a v2 with the review
comments.
Thanks.
Regards,
Ankur
From: Alin Serdean
Sent: Wednesday, September 24, 2014 6:14 PM
To: Ankur Sharma; dev@openvswitch.org
Subject: RE:
Hi Alin,
Yes Eitan also pointed it out.
I am panning to send a seperate series which will make a lot of NETLINK APIS
inline.
I did not do it in this patch as i would have complicated it further.
let me know if it sounds fine that i send out a seperate patch handling all the
netlink APIs that ne
Maybe drop ret because it is confusing.
return TRUE;
//goto done;
...
return TRUE;
done:
return FALSE;
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subie
Sounds good to me.
Alin.
-Mesaj original-
De la: Ankur Sharma [mailto:ankursha...@vmware.com]
Trimis: Thursday, September 25, 2014 4:24 AM
Către: Alin Serdean; dev@openvswitch.org
Subiect: RE: [ovs-dev] [PATCH v1 03/10] datapath-windows/Netlink: Added
NlAttrLen API
Hi Alin,
Yes Eitan
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH v1 05/10] datapath-windows/Netlink: Fixed
NlAttrParseNested
NlAttr
$ git apply
ovs-dev-v1-06-10-datapath-windows-Flow.c-Basic-support-for-add-flow..patch
ovs-dev-v1-06-10-datapath-windows-Flow.c-Basic-support-for-add-flow..patch:12:
trailing whitespace.
{ .cmd = OVS_FLOW_CMD_NEW,
warning: 1 line adds whitespace errors.
Acked-by: Alin Gabriel Se
On Wed, Sep 24, 2014 at 02:06:46PM -0700, Andy Zhou wrote:
> Pushed. Thanks for the fix!
Thanks!
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
In OvsFlowNlNewCmdHandler you could drop the following commented lines:
NTSTATUS rc = STATUS_SUCCESS;
POVS_MESSAGE msgIn = (POVS_MESSAGE)usrParamsCtx->inputBuffer;
POVS_MESSAGE msgOut = (POVS_MESSAGE)usrParamsCtx->outputBuffer;
PNL_MSG_HDR nlMsgHdr = &(msgIn->nlMsg);
PGENL_MSG_H
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH v1 08/10] datapath-windows/Flow.c: FLOW_SET command
handler.
Regis
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH v1 09/10] datapath-windows/Flow.c: FLOW_DEL command
handler.
Regis
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: dev [mailto:dev-boun...@openvswitch.org] În numele Ankur Sharma
Trimis: Wednesday, September 24, 2014 10:16 AM
Către: dev@openvswitch.org
Subiect: [ovs-dev] [PATCH v1 10/10] datapath-windows/Flow.c: DEL_FLOWS command
handler.
Add
Just forgot the following:
$ git apply
ovs-dev-v1-10-10-datapath-windows-Flow.c-DEL_FLOWS-command-handler..patch
ovs-dev-v1-10-10-datapath-windows-Flow.c-DEL_FLOWS-command-handler..patch:20:
trailing whitespace.
if ((genlMsgHdr->cmd == OVS_FLOW_CMD_DEL) &&
warning: 1 line adds whitespace err
Tested-by: Alin Gabriel Serdean
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: Sorin Vinturis
Trimis: Wednesday, September 24, 2014 12:43 PM
Către: dev@openvswitch.org
Cc: Alin Serdean; nit...@vmware.com; Sorin Vinturis
Subiect: [PATCH] datapath-windows: Missed packets are no
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 Vport family implementation
Tested-by: Alin Gabriel Serdean
Acked-by: Alin Gabriel Serdean
-Mesaj original-
De la: Samuel Ghinet
Trimis: Thursday, September 25, 2014 2:44 AM
Către: dev@openvswitch.org
Cc: Alin Serdean; nit...@vmware.com; elia...@vmware.com; Ankur Sharma;
ssaur...@vmware.com
Subiect: [PATCH v3] Ha
Hi,
this short series proposes to minor simplifications to ofproto-dpif-rid.
They are not related to each other other than that there is a minor
conflict in applying the second patch if the first one is not present.
Simon Horman (2):
ofproto-dpif-rid: remove struct rid_map
ofproto-dpif-rid: r
struct rid_map only has one member which is a struct hmap.
This allows for a slight simplification of the code by removing
struct rid_map and using a struct hmap directly instead.
Signed-off-by: Simon Horman
---
ofproto/ofproto-dpif-rid.c | 20
1 file changed, 8 insertions(+
The return value of rid_pool_add() is never used
so the code may be slightly simplified by removing it.
Signed-off-by: Simon Horman
---
ofproto/ofproto-dpif-rid.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/ofproto/ofproto-dpif-rid.c b/ofproto/ofproto-dpif-rid.c
inde
Signed-off-by: Simon Horman
---
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 to Netronome.
---
AUTHORS | 1 +
1 file changed, 1 insertion(+)
diff --git a/AUTHORS b/AUTHORS
i
1. ovs-dev-v1-datapath-windows-Event-read-handler.patch:12: new blank line at
EOF.
+
warning: 1 line adds whitespace errors.
2. OvsPostEvent adds elements to the queue can you please find a similar name
for OvsRemoveEventEntry
3. Really don't like that we have to push in:
+#pragma warning( pus
Hi Team,
We are working on the implementation on eviction mechanism on basis of
importance as per openflow specs-1.4. We have added importance parameter
in the flow entry.
Following is our understanding for Delete Flow mechanism:
We have studied the code and have found that OFPFC_DELETE_STRICT
The original motivation for this change was to allow the
helper to be used in files other than actions.c as part
of work on an odp select group action.
It was as pointed out by Thomas Graf that this helper would be best
off living in netlink.h. Furthermore, I think that the generic nature of this
On Wed, Sep 24, 2014 at 09:20:15AM +0100, Thomas Graf wrote:
> On 09/24/14 at 03:00pm, Simon Horman wrote:
> > On Fri, Sep 19, 2014 at 03:06:38PM +0100, Thomas Graf wrote:
> > > Can we rename & move this to instead?
> >
> > Sure, how about nla_is_last()?
>
> Sounds great
Done. I have posted the
On Wed, Sep 24, 2014 at 09:19:42AM +0100, Thomas Graf wrote:
> On 09/24/14 at 03:01pm, Simon Horman wrote:
> > > > + /* Only possible type of attributes is
> > > > OVS_SELECT_GROUP_ATTR_BUCKET */
> > > > + for (bucket = nla_data(attr), rem = nla_len(attr); rem > 0;
> > > > +
Alin, this warning is used to warn about constant in a condition. We shouldn't
have these cases at all. There are other parts in the code which uses the
while(FALSE) condition.
OVS_USE_NL_INTERFACE is used to exclude code which used for the dpif device.
There are many code segments that exclude
-Mesaj original-
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Thursday, September 25, 2014 7:48 AM
Către: Alin Serdean; dev@openvswitch.org
Subiect: RE: [ovs-dev] [PATCH v1] datapath-windows Event read handler
Alin, this warning is used to warn about constant in a condition. W
If we remove OVS_USE_NL_INTERFACE the legacy dpif device should be active
rather the NL device. Basically, we have all code which is used by the legacy
device excluded by using this compilation flag. Are you suggesting that we will
remove all thse code segments?
On the white space I have " git
Why this code you are talking about is different from the code which is already
guarded? Both are for the legacy device so they need to be treated the same?
Did I miss your point?
Thanks,
Eitan
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, September 24, 2014 10:22
It was not guarded until now so was wondering why guard it now :), unless that
you want to use this as a mark for later on.
Alin.
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Thursday, September 25, 2014 8:34 AM
Către: Alin Serdean; dev@openvswitch.org
Subiect: RE: [ovs-dev] [PATCH v1
Yes, I do. Do you see any issue with that?
Thanks,
Eitan
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, September 24, 2014 10:39 PM
To: Eitan Eliahu; dev@openvswitch.org
Subject: RE: [ovs-dev] [PATCH v1] datapath-windows Event read handler
It was not guarded until no
None what so ever. Was just wondering if there was a hidden meaning to it that
I missed :).
Alin.
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Thursday, September 25, 2014 8:41 AM
Către: Alin Serdean; dev@openvswitch.org
Subiect: RE: [ovs-dev] [PATCH v1] datapath-windows Event read ha
We planned to keep the legacy device code in tact until we make progress with
the NL device. We will remove the legacy device soon. Yes, this compilation
variable will help us to find these code segments.
Thank you,
Eitan
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesda
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
connected to a Hyper-V Virtual Switch.
Usage:
import-module .\OVS.psm1
$vnic = Get-VMNetworkA
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 netl
Added NlFillOvsMsg API in Netlink.c This API will be used to populate
netlink message headers.
Signed-off-by: Ankur Sharma
Acked-by: Alin Gabriel Serdean
Acked-by: Eitan Eliahu
---
datapath-windows/ovsext/Netlink/Netlink.c | 38 +++
datapath-windows/ovsext/Netlink/N
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
Acked-by: Alin Gabriel Serdean
Acked-by: Eitan Eliahu
---
datapath-windows/ovsext/Netlink/Netlink.c | 28 ++
1 - 100 of 105 matches
Mail list logo
