The breakage was introduced by commit: a36de779
("openvswitch: Userspace tunneling.").
Reported-by: Edwin Chiu
Signed-off-by: Saurabh Shah
---
include/windows/netinet/ip.h |1 +
1 file changed, 1 insertion(+)
diff --git a/include/windows/netinet/ip.h b/include/windows/netinet/ip.h
index f3
On 18 November 2014 22:09, Pravin Shelar wrote:
> On Tue, Nov 18, 2014 at 10:54 AM, Joe Stringer
> wrote:
> > When userspace doesn't provide a mask, OVS datapath generates a fully
> > unwildcarded mask for the flow. This is done by taking a copy of the
> > flow key, then iterating across its att
> On Tue, Nov 18, 2014 at 6:54 PM, YAMAMOTO Takashi
> wrote:
>> At least BSD implementation needs this.
>>
>> Signed-off-by: YAMAMOTO Takashi
>> ---
>> lib/ovs-router.c | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/lib/ovs-router.c b/lib/ovs-router.c
>> index 2550288..82a6108 10064
I came around a BSOD that happened when trying to access pidHashLock
from the gOvsSwitchContext, which was NULL. The stop happened in
OvsAcquirePidHashLock function.
To reproduce this BSOD, make sure the extension is enabled and running,
disable it and, after that, execute 'ovs-dpctl.exe show'. Th
> ovs-dpctl crashed due to uninitialized router classifier. To
> fix this issue move ovs router initialization to route table
> module.
does it mean ovs/router/xxx commands will not be available until
vport is created? it seems a little awkward.
YAMAMOTO Takashi
_
On Tue, Nov 18, 2014 at 6:54 PM, YAMAMOTO Takashi
wrote:
> At least BSD implementation needs this.
>
> Signed-off-by: YAMAMOTO Takashi
> ---
> lib/ovs-router.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/lib/ovs-router.c b/lib/ovs-router.c
> index 2550288..82a6108 100644
> --- a/lib
On Tue, Nov 18, 2014 at 10:54 AM, Joe Stringer wrote:
> When userspace doesn't provide a mask, OVS datapath generates a fully
> unwildcarded mask for the flow. This is done by taking a copy of the
> flow key, then iterating across its attributes, setting all values to
> 0xff. This works for most a
> On Wed, Nov 19, 2014 at 11:53:53AM +0900, YAMAMOTO Takashi wrote:
>> hi,
>>
>> > I am unsure of the reason for escaping '-' inside of control
>> > sequences but regardless this patch makes the group documentation
>> > consistent with other portions of the ovs-ofctl(8) in this regard.
>>
>> afai
On Wed, Nov 19, 2014 at 11:53:53AM +0900, YAMAMOTO Takashi wrote:
> hi,
>
> > I am unsure of the reason for escaping '-' inside of control
> > sequences but regardless this patch makes the group documentation
> > consistent with other portions of the ovs-ofctl(8) in this regard.
>
> afaik you sho
This command is useful at least for testing.
Example output:
% ovs-appctl ovs/route/lookup '10.0.0.1'
gateway 172.17.0.254
dev wm0
%
Signed-off-by: YAMAMOTO Takashi
---
lib/ovs-router.c | 27 +++
1 file changed, 27 insertions(+)
diff --git a/lib/ovs-rout
Following the linux version.
Signed-off-by: YAMAMOTO Takashi
---
lib/route-table-bsd.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/lib/route-table-bsd.c b/lib/route-table-bsd.c
index b606059..3608389 100644
--- a/lib/route-table-bsd.c
+++ b/lib/route-table-bsd.c
@@ -30,8 +30,
For userspace tunneling.
Signed-off-by: YAMAMOTO Takashi
---
lib/route-table-bsd.c | 12 +---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/lib/route-table-bsd.c b/lib/route-table-bsd.c
index 3608389..c3119e3 100644
--- a/lib/route-table-bsd.c
+++ b/lib/route-table-bsd.c
At least BSD implementation needs this.
Signed-off-by: YAMAMOTO Takashi
---
lib/ovs-router.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/ovs-router.c b/lib/ovs-router.c
index 2550288..82a6108 100644
--- a/lib/ovs-router.c
+++ b/lib/ovs-router.c
@@ -309,6 +309,7 @@ void
ovs_router_un
hi,
> I am unsure of the reason for escaping '-' inside of control
> sequences but regardless this patch makes the group documentation
> consistent with other portions of the ovs-ofctl(8) in this regard.
afaik you should not escape hyphens.
http://netbsd.gw.com/cgi-bin/man-cgi?mandoc_char++NetBSD
In this patch, we rename the existing OvsFindVportByHvName() to
OvsFindVportByHvNameA() to indicate that the input string is an ASCII
string. We also define a OvsFindVportByHvNameW() that takes as input a
WCHAR string.
This will be used later in HvCreatPort() to check for ports with
duplicate name
In this patch, we update HvCreatePort() to be able to re-add a Hyper-V
port. Specifically, we handle the case where the port had also been
added by OVS userspace, so that when the port was previously deleted
from Hyper-V, we did not deallocate the port.
The key to a vport is its name. We lookup th
A few fixes around locking.
Signed-off-by: Nithin Raju
Acked-by: Ankur Sharma
---
datapath-windows/ovsext/Vport.c | 16 ++--
1 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/datapath-windows/ovsext/Vport.c b/datapath-windows/ovsext/Vport.c
index 2dc5f0a..33aa95e 100
OvsInitVxlanTunnel() need not return a NL_ERROR. In this patch, we
change it to NTSTATUS, and also update the mapping function that maps a
NTSTATUS to NL_ERROR.
Signed-off-by: Nithin Raju
Acked-by: Ankur Sharma
---
datapath-windows/ovsext/Datapath.c |3 ++-
datapath-windows/ovse
In this patch, we add changes to OvsRemoveAndDeleteVport() to allow the
caller to specify if a vport is being deleted because it got deleted on
Hyper-V or if it got deleted from OVS userspace.
The reason we need to make the distinction is to be able to delete the
vport from the relevant hash table
If a Hyper-V port (for which there exists an OVS port), gets deleted and
re-added, we'll call into InitOvsVportCommon() for the port to insert
the port into the 'portIdHashArray' as well as do a few other
initialization in the switch ocntext.
We should not be incrementing 'numHvPorts' at this poin
In this patch, we update the documentation for 'portIdHashArray' to
indicate that a vport would exist in this hash table if and only if it
also exists on the Hyper-V switch.
This functionality to implement this semantic will follow in subsequent
patches.
Signed-off-by: Nithin Raju
---
datapath-
OvsInitVxlanTunnel() need not return a NL_ERROR. In this patch, we
change it to NTSTATUS, and also update the mapping function that maps a
NTSTATUS to NL_ERROR.
Signed-off-by: Nithin Raju
Acked-by: Ankur Sharma
---
datapath-windows/ovsext/Datapath.c |3 ++-
datapath-windows/ovse
In this patch, we update HvCreatePort() to be able to re-add a Hyper-V
port. Specifically, we handle the case where the port had also been
added by OVS userspace, so that when the port was previously deleted
from Hyper-V, we did not deallocate the port.
The key to a vport is its name. We lookup th
In this patch, we add changes to OvsRemoveAndDeleteVport() to allow the
caller to specify if a vport is being deleted because it got deleted on
Hyper-V or if it got deleted from OVS userspace.
The reason we need to make the distinction is to be able to delete the
vport from the relevant hash table
If a Hyper-V port (for which there exists an OVS port), gets deleted and
re-added, we'll call into InitOvsVportCommon() for the port to insert
the port into the 'portIdHashArray' as well as do a few other
initialization in the switch ocntext.
We should not be incrementing 'numHvPorts' at this poin
In this patch, we rename the existing OvsFindVportByHvName() to
OvsFindVportByHvNameA() to indicate that the input string is an ASCII
string. We also define a OvsFindVportByHvNameW() that takes as input a
WCHAR string.
This will be used later in HvCreatPort() to check for ports with
duplicate name
In this patch, we update the documentation for 'portIdHashArray' to
indicate that a vport would exist in this hash table if and only if it
also exists on the Hyper-V switch.
This functionality to implement this semantic will follow in subsequent
patches.
Signed-off-by: Nithin Raju
---
datapath-
A few fixes around locking.
Signed-off-by: Nithin Raju
Acked-by: Ankur Sharma
---
datapath-windows/ovsext/Vport.c | 16 ++--
1 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/datapath-windows/ovsext/Vport.c b/datapath-windows/ovsext/Vport.c
index 2dc5f0a..33aa95e 100
Acked-by: Ankur Sharma
From: dev on behalf of Eitan Eliahu
Sent: Wednesday, November 12, 2014 10:53 AM
To: Nithin Raju
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH 1/7] datapath-windows: fixes in
OvsGetExtInfoIoctl()
Yes, we planned
Acked-by: Ankur Sharma
From: dev on behalf of Nithin Raju
Sent: Wednesday, November 12, 2014 10:56 AM
To: Eitan Eliahu
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH 7/7] datapath-windows: Fixes in HvCreatePort() to
re-add a port
On Nov 12, 2014
Acked-by: Ankur Sharma
From: dev on behalf of Nithin Raju
Sent: Tuesday, November 11, 2014 7:04 PM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH 6/7] datapath-windows: Change return type of
OvsInitVxlanTunnel()
OvsInitVxlanTunnel() need not return
Acked-by: Ankur Sharma
From: dev on behalf of Nithin Raju
Sent: Tuesday, November 11, 2014 7:04 PM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH 5/7] datapath-windows: Changes to
InitOvsVportCommon()
If a Hyper-V port (for which there exists an
Acked-by: Ankur Sharma
From: dev on behalf of Nithin Raju
Sent: Tuesday, November 11, 2014 7:04 PM
To: dev@openvswitch.org
Subject: [ovs-dev] [PATCH 4/7] datapath-windows: Changes to
OvsRemoveAndDeleteVport()
In this patch, we add changes to OvsRem
This is intended as a usable demonstration of how
the NMX selection method extension might may be used.
NMX selection method
Signed-off-by: Simon Horman
---
lib/flow.c | 2 +-
lib/flow.h | 14 ++
ofproto/ofproto-dpif-xlate.c | 32 +
NMX selection method
Signed-off-by: Simon Horman
---
lib/ofp-parse.c | 103 +++
tests/ofproto.at | 4 +-
utilities/ovs-ofctl.8.in | 23 ++-
3 files changed, 127 insertions(+), 3 deletions(-)
diff --git a/lib/ofp-parse.c b/li
Support printing of NMX selection method group experimenter property
NMX selection method
Signed-off-by: Simon Horman
---
lib/flow.c | 47 +++
lib/flow.h | 2 ++
lib/ofp-print.c| 38 ++
tests/ofp
Include NMX selection method experimenter group property in
in group mod request and group desc reply.
NMX selection method
Signed-off-by: Simon Horman
---
lib/nx-match.c| 214 ++
lib/nx-match.h| 2 +
lib/ofp-util.c| 31
This is in preparation for supporting group mod and desc reply
messages with an NMX selection method group experimenter property.
NMX selection method
Signed-off-by: Simon Horman
---
lib/nx-match.c| 52 ---
lib/nx-match.h| 1 +
lib/ofp-util.c| 256 +
NMX selection method
Signed-off-by: Simon Horman
---
ofproto/ofproto.c | 63 +++
1 file changed, 40 insertions(+), 23 deletions(-)
diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c
index 00e668a..b749ca6 100644
--- a/ofproto/ofproto.c
+++ b/of
Only the default existing behaviour is translated.
All other methods are rejected for now.
NMX selection method
Signed-off-by: Simon Horman
---
ofproto/ofproto-dpif-xlate.c | 30 ++
ofproto/ofproto-dpif.c | 6 ++
ofproto/ofproto-dpif.h | 1 +
ofproto
NMX selection method
Signed-off-by: Simon Horman
---
include/openflow/automake.mk | 1 +
include/openflow/netronome-ext.h | 51
2 files changed, 52 insertions(+)
create mode 100644 include/openflow/netronome-ext.h
diff --git a/include/openflow/autom
Add Netronome vendor Id: NMX_VENDOR_ID = 0x1540.
This is based on the Netronome IEEE OUI, 00154D.
And it has been registered with the ONF:
https://rs.opennetworking.org/wiki/display/PUBLIC/ONF+Registry
Signed-off-by: Simon Horman
---
include/openflow/openflow-common.h | 1 +
1 file changed
Hi,
this patch set implements the group select selection method extension that
I circulated some months ago on the d...@openvswtich.org mailing list. For
reference a copy of that proposal (updated for the existence of an Open
Flow 1.5 draft and several errors found during implementation) is at th
Removed the check for previous dump state before calling FreeUserDumpState
as the check is also performed in the latter function.
Signed-off-by: Sorin Vinturis
Acked-by: Nithin Raju
---
datapath-windows/ovsext/Datapath.c | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/
The BSOD is triggered by the incorect extraction of the OVS message
from the received input buffer from userspace, in _FlowNlGetCmdHandler
function.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/issues/54
Acked-by: Nithin Raju
-
This patch series solves a BSOD due to an incorrect handling of the received
OVS message of the _FlowNlGetCmdHandler NL handler. It also removes a duplicate
check of the dump state.
Sorin Vinturis (2):
[PATCH v2 1/2] datapath-windows: BSOD for a transactional NL cmd w/o dump state
[PATCH v2 2/2]
When userspace doesn't provide a mask, OVS datapath generates a fully
unwildcarded mask for the flow. This is done by taking a copy of the
flow key, then iterating across its attributes, setting all values to
0xff. This works for most attributes, as the length of the netlink
attribute typically mat
We actually need to remove the global variable gOvsSwitchContext.
We need to keep a reference to the switch context in POVS_DEVICE_EXTENSION and
vice versa (keep a reference to the control device extension in the switch
context).
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-bo
Hi Nithin,
I would chose option #1, because I would expect some feedback message when
sending any command. For me, an error message is very helpful to get an idea
about what is going on. Otherwise a user could not understand the reason for
not getting the expected output.
Thanks,
Sorin
-O
Hi Nithin,
I have to disagree. Same behavior should take place regardless if the driver is
unloaded or the driver is loaded and the extension is not enabled. On other
platforms there is no two step enabling procedure.
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvsw
On Nov 18, 2014, at 12:15 AM, Sorin Vinturis
wrote:
> I came around a BSOD that happened when trying to access pidHashLock
> from the gOvsSwitchContext, which was NULL. The stop happened in
> OvsAcquirePidHashLock function.
>
> To reproduce this BSOD, make sure the extension is enabled and runn
On Nov 18, 2014, at 6:20 AM, Sorin Vinturis
wrote:
> * that 'ovsMsgReadOp' is needed only in this function to call into
> the
> - * appropraite handler. The handler itself can access the state in
> the
> + * appropriate handler. The handler itself can access the state
> *
> * _FlowNlDumpCmdHandler --
> - *Handler for OVS_FLOW_CMD_GET command.
> + *Handler for OVS_FLOW_CMD_DUMP command.
Thanks for the respin.
If you want to be technically correct, you should say:
"Handler for
hi Sorin,
Pls. see my explanations inline.
>>/* Concurrent netlink operations are not supported. */
>>if (InterlockedCompareExchange((LONG volatile *)&instance->inUse, 1, 0)) {
>>status = STATUS_RESOURCE_IN_USE; @@ -891,7 +895,7 @@
>> ValidateNetlinkCmd(UINT32 devOp,
>>
Removed the check for previous dump state before calling FreeUserDumpState
as the check is also performed in the latter function.
Signed-off-by: Sorin Vinturis
---
datapath-windows/ovsext/Datapath.c | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/datapath-windows/ovsext
The BSOD is triggered by the incorect extraction of the OVS message
from the received input buffer from userspace, in _FlowNlGetCmdHandler
function.
Signed-off-by: Sorin Vinturis
Reported-by: Sorin Vinturis
Reported-at: https://github.com/openvswitch/ovs-issues/issues/54
---
datapath-windows/ov
This patch series solves a BSOD due to an incorrect handling of the received
OVS message of the _FlowNlGetCmdHandler NL handler. It also removes a duplicate
check of the dump state.
Sorin Vinturis (2):
[PATCH 1/2] datapath-windows: BSOD for a transactional NL cmd w/o dump state
[PATCH 2/2] datap
Hi Nithin,
Thanks for your review. I will split this patch into two patch series.
Sorin
-Original Message-
From: Nithin Raju [mailto:nit...@vmware.com]
Sent: Tuesday, 18 November, 2014 08:59
To: Sorin Vinturis
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH] datapath-windows: BSOD
I came around a BSOD that happened when trying to access pidHashLock
from the gOvsSwitchContext, which was NULL. The stop happened in
OvsAcquirePidHashLock function.
To reproduce this BSOD, make sure the extension is enabled and running,
disable it and, after that, execute 'ovs-dpctl.exe show'. Th
59 matches
Mail list logo
