hello,
I made http://www.xrefs.info available to open source community in
the hope of making open source developers more productive.
The site hosts many open source code projects' cross references based
on OpenGrok,
which is a very fast cross reference tool, and easy to use.
openvswitch plays a k
On Fri, Feb 07, 2014 at 02:04:21PM +0900, Simon Horman wrote:
> On Thu, Feb 06, 2014 at 08:50:39PM -0800, Ben Pfaff wrote:
> > Signed-off-by: Ben Pfaff
> > CC: Simon Horman
>
> This looks very nice to me and I believe the only missing piece
> is the kernel datapath support, which I have posted v
> On Feb 6, 2014, at 4:12 PM, Pravin Shelar wrote:
>
>> On Thu, Feb 6, 2014 at 3:13 PM, Jarno Rajahalme
>> wrote:
>>Only the first IP fragment can have a TCP header, check for this.
> Is ip.proto still set to IPPROTO_TCP in this case?
Yes, but I just remembered that the tcp flags are alre
> On Feb 6, 2014, at 6:36 PM, Jesse Gross wrote:
>
>> On Thu, Feb 6, 2014 at 4:09 PM, Pravin Shelar wrote:
>>> On Thu, Feb 6, 2014 at 3:13 PM, Jarno Rajahalme
>>> wrote:
>>>Keep kernel flow stats for each NUMA node rather than each (logical)
>>>CPU. This avoids using the per-CPU allo
On Thu, Feb 06, 2014 at 08:50:39PM -0800, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> CC: Simon Horman
This looks very nice to me and I believe the only missing piece
is the kernel datapath support, which I have posted v2.53 of to the
mailing list for Jesse's review.
Acked-by: Simon Horman
Q: Does Open vSwitch support MPLS?
A: Before version 1.11, Open vSwitch did not support MPLS. That is,
these versions can match on MPLS Ethernet types, but they cannot
match, push, or pop MPLS labels, nor can they look past MPLS labels
into the encapsulated packet.
Open vSwitch versi
Signed-off-by: Ben Pfaff
CC: Simon Horman
---
FAQ | 20
1 file changed, 20 insertions(+)
diff --git a/FAQ b/FAQ
index 75d9e6b..e6e649f 100644
--- a/FAQ
+++ b/FAQ
@@ -,6 +,26 @@ A: Open vSwitch 1.9 and earlier support only OpenFlow
1.0 (plus
When support for a
Hello,
I am sure this is going to be the repeated question. But this time, I tried to
debug what exactly going on with my problem.
I have inserted a flow and it is mentioned below:
ovs-ofctl add-flow s1
in_port=1,ip,actions=push_mpls:0x8847,set_field:100-\>mpls_label,output:2 -O
OpenFlow13
T
On Thu, Feb 6, 2014 at 4:09 PM, Pravin Shelar wrote:
> On Thu, Feb 6, 2014 at 3:13 PM, Jarno Rajahalme wrote:
>> Keep kernel flow stats for each NUMA node rather than each (logical)
>> CPU. This avoids using the per-CPU allocator and removes most of the
>> kernel-side OVS locking ove
On Wed, Feb 05, 2014 at 09:45:37AM +0900, Simon Horman wrote:
> Originally part of "odp: Allow VLAN actions after MPLS actions"
> by Joe Stringer.
>
> Co-authored-by: Joe Stringer
> Signed-off-by: Simon Horman
Applied, thank you both.
___
dev mailing
> On Wed, Feb 05, 2014 at 07:08:11PM +0900, YAMAMOTO Takashi wrote:
>> Signed-off-by: YAMAMOTO Takashi
>
> Would you mind using the ALIGNED_CAST macro? It does the same thing
> but makes it obvious why the cast is present.
i was not aware of the macro. thanks.
i'll take a look later.
YAMAMOTO
On Wed, Feb 05, 2014 at 09:45:03AM +0900, Simon Horman wrote:
> On Tue, Feb 04, 2014 at 12:39:37PM -0800, Ben Pfaff wrote:
> > OpenFlow 1.1 and 1.2 always inserted MPLS labels after VLAN tags.
> > OpenFlow 1.3 and 1.4 insert MPLS labels before VLAN tags.
> > OpenFlow 1.3.4 and 1.5, both in preparat
> On Wed, Feb 05, 2014 at 11:47:17AM +0900, YAMAMOTO Takashi wrote:
>> > On Wed, Jan 15, 2014 at 12:41:20PM +0900, YAMAMOTO Takashi wrote:
>> >> Add a comment about implicit synchronization which
>> >> fat-rwlock seems to rely on.
>> >>
>> >> Signed-off-by: YAMAMOTO Takashi
>> >> ---
>> >> lib/o
On Wed, Feb 05, 2014 at 07:08:11PM +0900, YAMAMOTO Takashi wrote:
> Signed-off-by: YAMAMOTO Takashi
Would you mind using the ALIGNED_CAST macro? It does the same thing
but makes it obvious why the cast is present.
___
dev mailing list
dev@openvswitch.o
On Wed, Feb 05, 2014 at 07:08:12PM +0900, YAMAMOTO Takashi wrote:
> Signed-off-by: YAMAMOTO Takashi
Applied, thanks!
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Feb 05, 2014 at 07:08:10PM +0900, YAMAMOTO Takashi wrote:
> Signed-off-by: YAMAMOTO Takashi
Applied, thanks. (I removed a spurious blank line.)
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Feb 05, 2014 at 11:47:17AM +0900, YAMAMOTO Takashi wrote:
> > On Wed, Jan 15, 2014 at 12:41:20PM +0900, YAMAMOTO Takashi wrote:
> >> Add a comment about implicit synchronization which
> >> fat-rwlock seems to rely on.
> >>
> >> Signed-off-by: YAMAMOTO Takashi
> >> ---
> >> lib/ovs-thread
On Thu, Feb 06, 2014 at 08:12:33AM -0800, Gurucharan Shetty wrote:
> This commit creates events and associates them with
> socket file descriptors to get woken up from poll_block().
> One difference with the Linux implementation is that
> we cannot register for separate network events with poll_blo
On Thu, Feb 06, 2014 at 08:12:31AM -0800, Gurucharan Shetty wrote:
> This is helpful if we want to wait either on 'fd' for POSIX or
> events for Windows.
>
> Signed-off-by: Gurucharan Shetty
Patches 1 and 2 seem OK to me. I wonder whether Linda or anyone else
has comments though.
__
On Thu, Feb 06, 2014 at 08:12:32AM -0800, Gurucharan Shetty wrote:
> For Windows platform, there are different ways to get error numbers
> for differnt situations. For C library functions, errno is set like the way
> they are set for Linux. For Windows API functions, it has to be gotten from
> GetL
Looks good to me~
On Thu, Feb 6, 2014 at 4:16 PM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> ofproto/collectors.c |8 +---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/ofproto/collectors.c b/ofproto/collectors.c
> index 919fa1b..4501d2b 100644
> --- a/of
Before this commit, ovs randomly selects a slave for unassigned
bond entry. If the selected slave is not enabled, the active slave
is chosen instead. In this commit, the slave is selected from the
list of all enabled slaves in a round-robin fashion. This helps
improve the consistency of bond beh
Signed-off-by: Ben Pfaff
---
ofproto/collectors.c |8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/ofproto/collectors.c b/ofproto/collectors.c
index 919fa1b..4501d2b 100644
--- a/ofproto/collectors.c
+++ b/ofproto/collectors.c
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 20
On Tue, Jan 21, 2014 at 11:39:42AM +0530, Arun Sharma wrote:
> Allows active/passive socket operations to use IPv6 network addresses,
> except in-band control which is not supported for IPv6.
>
> tests: Added tests for socket connections over IPv6.
> Also added tests for NetFlow and sFlow using IP
On Thu, Feb 6, 2014 at 3:13 PM, Jarno Rajahalme wrote:
> Only the first IP fragment can have a TCP header, check for this.
>
Is ip.proto still set to IPPROTO_TCP in this case?
> Signed-off-by: Jarno Rajahalme
> ---
> datapath/flow.c |1 +
> 1 file changed, 1 insertion(+)
>
> diff --
On Thu, Feb 6, 2014 at 3:13 PM, Jarno Rajahalme wrote:
> Keep kernel flow stats for each NUMA node rather than each (logical)
> CPU. This avoids using the per-CPU allocator and removes most of the
> kernel-side OVS locking overhead otherwise on the top of perf reports
> and allows
On Thu, Feb 06, 2014 at 12:31:13PM -0800, Pravin wrote:
> From: Pravin Shelar
>
> netdev-linux does not depend on linux kernel version. So
> remove header file include.
>
> Signed-off-by: Pravin B Shelar
Acked-by: Ben Pfaff
___
dev mailing list
dev@
Keep kernel flow stats for each NUMA node rather than each (logical)
CPU. This avoids using the per-CPU allocator and removes most of the
kernel-side OVS locking overhead otherwise on the top of perf reports
and allows OVS to scale better with higher number of threads.
With 9
Only the first IP fragment can have a TCP header, check for this.
Signed-off-by: Jarno Rajahalme
---
datapath/flow.c |1 +
1 file changed, 1 insertion(+)
diff --git a/datapath/flow.c b/datapath/flow.c
index 8be3801..abe6789 100644
--- a/datapath/flow.c
+++ b/datapath/flow.c
@@ -74,6
- Original Message -
From: "Gurucharan Shetty"
To: "Linda Sun"
Cc: "dev"
Sent: Tuesday, February 4, 2014 2:14:13 PM
Subject: Re: [ovs-dev] [PATCH] Add stream-fd implementation for windows. Need
to use send/recv instead of write/read operations. Use the event handle for
polling. check
From: Pravin Shelar
netdev-linux does not depend on linux kernel version. So
remove header file include.
Signed-off-by: Pravin B Shelar
---
lib/netdev-linux.c |1 -
1 file changed, 1 deletion(-)
diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c
index e756d88..9ce1953 100644
--- a/lib/n
This should assist testing of datapath performance, as it allows us to
skip "warming up" the flow limit value.
Signed-off-by: Joe Stringer
---
ofproto/ofproto-dpif-upcall.c | 26 ++
1 file changed, 26 insertions(+)
diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/o
On Tue, Feb 4, 2014 at 11:03 AM, Jarno Rajahalme wrote:
> Keep kernel flow stats for each NUMA node rather than each (logical)
> CPU. This avoids using the per-CPU allocator and removes most of the
> kernel-side OVS locking overhead otherwise on the top of perf report
> and allows OVS to scale be
This should assist testing of datapath performance, as it allows us to
skip "warming up" the flow limit value.
Signed-off-by: Joe Stringer
---
ofproto/ofproto-dpif-upcall.c | 26 ++
1 file changed, 26 insertions(+)
diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/o
This commit creates events and associates them with
socket file descriptors to get woken up from poll_block().
One difference with the Linux implementation is that
we cannot register for separate network events with poll_block().
i.e., we cannot say, we only should be woken up for POLLIN or POLLOUT
For Windows platform, there are different ways to get error numbers
for differnt situations. For C library functions, errno is set like the way
they are set for Linux. For Windows API functions, it has to be gotten from
GetLastError(). For winsock2, errno has to be obtained from WSAGetLastError().
For Windows, WaitForMultipleObjects() does not make use of
file descriptors. So there is no need of passing it to
poll_fd_wait_event(). Since 'fd' is not used, change the
name to poll_wait_event().
Also, WaitForMultipleObjects() cannot take arguments like
POLLIN, POLLOUT etc. So get rid of that to
This is helpful if we want to wait either on 'fd' for POSIX or
events for Windows.
Signed-off-by: Gurucharan Shetty
---
lib/poll-loop.h |2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/poll-loop.h b/lib/poll-loop.h
index cbbaf55..aa8891b 100644
--- a/lib/poll-loop.h
+++ b/lib/poll-lo
38 matches
Mail list logo
