Hi All
This is regarding the VLAN QinQ (802.1AD ) support in OpenvSwitch .
We are planning to work on VLAN QinQ support in Openvswitch.
Please let us know whether if some people are already working on this feature.
Could anyone please let me know your pointers about this feature.
Thanks,Deba
Fix a regression introduced by commit fce314cd.
("socket-util: Fix definition of LINUX.")
Signed-off-by: YAMAMOTO Takashi
---
lib/socket-util.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/socket-util.c b/lib/socket-util.c
index 5e1be3f..aa0c719 100644
--- a/lib/socket
On Fri, Mar 14, 2014 at 04:12:39PM +0900, Simon Horman wrote:
> Hi Pravin, Hi Jesse, Hi All,
>
> I have updated this patch so that:
>
> * The datapath rejects push MPLS actions in the presence of VLAN tags.
>
> I have done this by white-listing the following:
> - ETH_P_IP (0x0800)
> - ETH_
On Mon, Mar 10, 2014 at 12:32 PM, Pravin Shelar wrote:
> On Thu, Mar 6, 2014 at 1:57 PM, Kyle Mestery
> wrote:
>> On Thu, Feb 27, 2014 at 3:28 PM, Pravin Shelar wrote:
>>>
>>> On Thu, Feb 27, 2014 at 12:41 PM, Kyle Mestery
>>> wrote:
>>> > On Wed, Feb 26, 2014 at 5:07 PM, Pravin Shelar
>>> >
On Fri, Feb 28, 2014 at 1:14 PM, Ben Pfaff wrote:
> diff --git a/datapath/flow.c b/datapath/flow.c
> index c3e3fcb..e9a2a27 100644
> --- a/datapath/flow.c
> +++ b/datapath/flow.c
> @@ -1,5 +1,5 @@
> /*
> - * Copyright (c) 2007-2013 Nicira, Inc.
> + * Copyright (c) 2007-2014 Nicira, Inc.
> *
>
Since daemons can use the cfm status from OVSDB, it is useful
to add a test to guarantee that the update is in time.
Signed-off-by: Alex Wang
---
tests/cfm.at | 38 ++
1 file changed, 38 insertions(+)
diff --git a/tests/cfm.at b/tests/cfm.at
index 7ab4354..
CFM fault variable type has been changed to 'enum cfm_fault_reason' for
long time. However, inside cfm_run(), the old_cfm_fault is still defined
as boolean. This commit fixes the issue.
Found by inspection.
Signed-off-by: Alex Wang
---
lib/cfm.c |4 ++--
1 file changed, 2 insertions(+), 2
Thanks, applied.
On Wed, Mar 19, 2014 at 04:18:21PM -0700, Alex Wang wrote:
> Looks good to me,
>
> Acked-by: Alex Wang
>
>
> On Wed, Mar 19, 2014 at 4:03 PM, Ben Pfaff wrote:
>
> > MSVC 2013 finally implements them, 14 years after they became part of C,
> > so we can use them in generic cod
On Wed, Mar 19, 2014 at 01:35:07PM -0700, Alex Wang wrote:
> With the foundation laid in previous commits, this commit
> removes the 'dispatcher' thread by allowing 'handler'
> threads to read upcalls directly from dpif.
>
> This commit significantly simplifies the flow miss handling
> code and br
This reworks lookup of rules for both table 0 and table action translation.
The result is that Table Mod settings, which can alter the miss-behaviour
of tables, including table 0, on a per-table basis may be honoured.
Previous patches proposed by myself which build on earlier merged patches
by And
On Wed, Mar 19, 2014 at 01:35:06PM -0700, Alex Wang wrote:
> Signed-off-by: Alex Wang
This is a lot of code, but it looks well crafted. I only have a few
comments.
dpif-linux.c
The comment on dpif_linux_refresh_channels() mentions dpif->channels,
which no longer exists.
It looks
Thx Ben,
I'll hold off the push and discuss with Pravin tmr,
On Wed, Mar 19, 2014 at 4:32 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 01:35:04PM -0700, Alex Wang wrote:
> > This commit implements the API functions to allow multiple handler
> > threads read upcall.
> >
> > Also, this commit
On Mar 19, 2014, at 3:32 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 09:16:58AM -0700, Jarno Rajahalme wrote:
>> The flow that created the netdev_flow might have wildcarded TCP flags,
>> or it may not be a TCP flow at all. Fix this by using the freshly
>> extracted flow key to parse TCP flag
On Wed, Mar 19, 2014 at 01:35:05PM -0700, Alex Wang wrote:
> In order to allow handlers directly read upcalls from datapath,
> we need to support per-handler netlink socket for each vport in
> datapath. This commit makes this happen. Also, it is guaranteed
> backward and forward compatibility wit
On Wed, Mar 19, 2014 at 04:33:17PM -0700, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 01:35:05PM -0700, Alex Wang wrote:
> > In order to allow handlers directly read upcalls from datapath,
> > we need to support per-handler netlink socket for each vport in
> > datapath. This commit makes this happe
Pushed, thanks,
Jarno
On Mar 19, 2014, at 3:34 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 09:16:59AM -0700, Jarno Rajahalme wrote:
>> TCP flags are already extracted from the flow, no need to parse them
>> again.
>>
>> Signed-off-by: Jarno Rajahalme
>
> Acked-by: Ben Pfaff
__
On Wed, Mar 19, 2014 at 01:35:04PM -0700, Alex Wang wrote:
> This commit implements the API functions to allow multiple handler
> threads read upcall.
>
> Also, this commit removes the handling priority of DPIF_UC_MISS
> over DPIF_UC_ACTION. So, both misses will be put to the same
> queue. The d
Pushed, thanks,
Jarno
On Mar 19, 2014, at 3:34 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 09:17:00AM -0700, Jarno Rajahalme wrote:
>> This function has no more users, so remove it.
>>
>> Signed-off-by: Jarno Rajahalme
>
> Acked-by: Ben Pfaff
__
Looks good to me,
Acked-by: Alex Wang
On Wed, Mar 19, 2014 at 4:03 PM, Ben Pfaff wrote:
> MSVC 2013 finally implements them, 14 years after they became part of C,
> so we can use them in generic code now.
>
> Signed-off-by: Ben Pfaff
> ---
> CodingStyle |9 -
>
On Wed, Mar 19, 2014 at 4:13 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 04:10:57PM -0700, Justin Pettit wrote:
> > Signed-off-by: Justin Pettit
>
> Acked-by: Ben Pfaff
>
Thanks. Pushed.
--Justin
___
dev mailing list
dev@openvswitch.org
http://o
On Wed, Mar 19, 2014 at 04:10:57PM -0700, Justin Pettit wrote:
> Signed-off-by: Justin Pettit
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Pushed, thanks,
Jarno
On Mar 19, 2014, at 3:30 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 09:16:57AM -0700, Jarno Rajahalme wrote:
>> We used to map ODPP_NONE to port number 0, which is wrong, as
>> ODPP_NONE is a valid value of the flow's in_port.
>>
>> Signed-off-by: Jarno Rajahalme
>
On Wed, Mar 19, 2014 at 01:35:03PM -0700, Alex Wang wrote:
> This commit changes the API in 'dpif-provider.h' to allow multiple
> handler threads call dpif_recv() simultaneously.
>
> Signed-off-by: Alex Wang
Acked-by: Ben Pfaff
___
dev mailing list
de
Signed-off-by: Justin Pettit
---
NEWS |2 +-
debian/changelog |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/NEWS b/NEWS
index 9cff5dd..839b4dd 100644
--- a/NEWS
+++ b/NEWS
@@ -1,4 +1,4 @@
-v2.1.0 - xx xxx
+v2.1.0 - 19 Mar 2014
MSVC 2013 finally implements them, 14 years after they became part of C,
so we can use them in generic code now.
Signed-off-by: Ben Pfaff
---
CodingStyle |9 -
lib/odp-util.c | 11 +--
ofproto/ofproto-dpif-xlate.c | 10 +-
3 files ch
On Tue, Mar 18, 2014 at 11:52 PM, Simon Horman wrote:
> On Tue, Mar 11, 2014 at 04:56:18PM -0700, Andy Zhou wrote:
>> Signed-off-by: Andy Zhou
>>
>> ---
>> v1->v2: No major changes
>
> [snip]
>
>> diff --git a/lib/flow.h b/lib/flow.h
>> index 8165bcf..8e1349d 100644
>> --- a/lib/flow.h
>> +++ b/
On Wed, Mar 19, 2014 at 03:36:12PM -0700, Pravin Shelar wrote:
> On Wed, Mar 19, 2014 at 3:29 PM, Ben Pfaff wrote:
> > On Wed, Mar 19, 2014 at 03:16:37PM -0700, Pravin Shelar wrote:
> >> On Wed, Mar 19, 2014 at 11:08 AM, Ben Pfaff wrote:
> >> > On Tue, Mar 18, 2014 at 01:53:27PM -0700, Pravin wro
On Wed, Mar 19, 2014 at 3:29 PM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 03:16:37PM -0700, Pravin Shelar wrote:
>> On Wed, Mar 19, 2014 at 11:08 AM, Ben Pfaff wrote:
>> > On Tue, Mar 18, 2014 at 01:53:27PM -0700, Pravin wrote:
>> > Polling the atomic variable only every 67,108,864 times throug
On Wed, Mar 19, 2014 at 09:17:00AM -0700, Jarno Rajahalme wrote:
> This function has no more users, so remove it.
>
> Signed-off-by: Jarno Rajahalme
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/d
On Wed, Mar 19, 2014 at 09:16:59AM -0700, Jarno Rajahalme wrote:
> TCP flags are already extracted from the flow, no need to parse them
> again.
>
> Signed-off-by: Jarno Rajahalme
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://ope
On Wed, Mar 19, 2014 at 09:16:58AM -0700, Jarno Rajahalme wrote:
> The flow that created the netdev_flow might have wildcarded TCP flags,
> or it may not be a TCP flow at all. Fix this by using the freshly
> extracted flow key to parse TCP flags.
>
> Signed-off-by: Jarno Rajahalme
Does this nee
On Wed, Mar 19, 2014 at 09:16:57AM -0700, Jarno Rajahalme wrote:
> We used to map ODPP_NONE to port number 0, which is wrong, as
> ODPP_NONE is a valid value of the flow's in_port.
>
> Signed-off-by: Jarno Rajahalme
Acked-by: Ben Pfaff
___
dev mailing
On Wed, Mar 19, 2014 at 03:16:37PM -0700, Pravin Shelar wrote:
> On Wed, Mar 19, 2014 at 11:08 AM, Ben Pfaff wrote:
> > On Tue, Mar 18, 2014 at 01:53:27PM -0700, Pravin wrote:
> > Polling the atomic variable only every 67,108,864 times through the
> > loop seems pretty conservative. (Does that ev
On Wed, Mar 19, 2014 at 03:20:53PM -0700, Pravin Shelar wrote:
> On Wed, Mar 19, 2014 at 11:26 AM, Ben Pfaff wrote:
> > Maybe netdev_rxq_open() should reject calls for which id >=
> > netdev->nr_rx, so that implementations don't have to (I think that all
> > the implementations are technically bro
On Wed, Mar 19, 2014 at 11:13 AM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 01:53:34PM -0700, Pravin wrote:
>> Preparation for multi queue netdev IO.
>>
>> Signed-off-by: Pravin B Shelar
>
> In lib/netdev-provider.h, can you realign the table here?
> *"alloc" "construct"
On Wed, Mar 19, 2014 at 11:26 AM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 01:53:52PM -0700, Pravin wrote:
>> new netdev type like DPDK can support multi-queue IO. Following
>> patch Adds support for same.
>>
>> Signed-off-by: Pravin B Shelar
>
> Needs comment and documentation updates.
>
> In
Thanks, applied to master and branch-2.1 ~
On Wed, Mar 19, 2014 at 3:01 PM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 02:14:43PM -0700, Alex Wang wrote:
> > Commit f23d157c ("ofproto-dpif: Don't poll ports when nothing changes")
> > did not ensure the update of the row of remote maintenance po
On Wed, Mar 19, 2014 at 11:08 AM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 01:53:27PM -0700, Pravin wrote:
>> This patch adds PMD type netdev for netdevice with poll-mode
>> drivers. Since there is no way to get signal on a packet recv
>> from these devices we need to poll them in busy loop. S
On Wed, Mar 19, 2014 at 12:27 PM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 01:54:04PM -0700, Pravin wrote:
>> Following patch adds DPDK netdev-class to userspace datapath. Now
>> OVS can use DPDK port for IO by just configuring DPDK port and then
>> adding dpdk type port to userspace datapath.
>
On Wed, Mar 19, 2014 at 03:01:09PM -0700, Pravin Shelar wrote:
> On Wed, Mar 19, 2014 at 12:32 PM, Ben Pfaff wrote:
> > On Tue, Mar 18, 2014 at 01:54:09PM -0700, Pravin wrote:
> >> DPDK need to set _lcore_id for using multiple core.
> >>
> >> Signed-off-by: Pravin B Shelar
> >
> > In xpthread_set
On Tue, Mar 18, 2014 at 02:14:43PM -0700, Alex Wang wrote:
> Commit f23d157c ("ofproto-dpif: Don't poll ports when nothing changes")
> did not ensure the update of the row of remote maintenance points in ovsdb
> when it changes. This commit makes the update happen by notifying the
> global connect
On Wed, Mar 19, 2014 at 12:32 PM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 01:54:09PM -0700, Pravin wrote:
>> DPDK need to set _lcore_id for using multiple core.
>>
>> Signed-off-by: Pravin B Shelar
>
> In xpthread_setaffinity_np_cpu(), I think that the whole function is
> Linux-specific, so th
skb_zerocopy can copy elements of the frags array between skbs, but it doesn't
orphan them. Also, it doesn't handle errors, so this patch takes care of that
as well.
Signed-off-by: Zoltan Kiss
---
v2: orphan the frags right before touching the frags
diff --git a/include/linux/skbuff.h b/include/
On 19/03/14 20:47, Thomas Graf wrote:
On 03/19/2014 09:38 PM, Zoltan Kiss wrote:
skb_zerocopy can copy elements of the frags array between skbs, but it
doesn't
orphan them. Also, it doesn't handle errors, so this patch takes care
of that
as well.
Signed-off-by: Zoltan Kiss
---
+if (unli
On 03/19/2014 09:38 PM, Zoltan Kiss wrote:
skb_zerocopy can copy elements of the frags array between skbs, but it doesn't
orphan them. Also, it doesn't handle errors, so this patch takes care of that
as well.
Signed-off-by: Zoltan Kiss
---
+ if (unlikely(skb_orphan_frags(to, GFP_ATOMIC
On 19/03/14 20:24, David Miller wrote:
From: Zoltan Kiss
Date: Tue, 18 Mar 2014 21:17:35 +
skb_zerocopy can copy elements of the frags array between skbs, but it doesn't
orphan them. Also, it doesn't handle errors, so this patch takes care of that
as well.
Signed-off-by: Zoltan Kiss
---
skb_zerocopy can copy elements of the frags array between skbs, but it doesn't
orphan them. Also, it doesn't handle errors, so this patch takes care of that
as well.
Signed-off-by: Zoltan Kiss
---
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 03db95a..35c4e85 100644
--- a/inc
Signed-off-by: Alex Wang
---
V4 -> V5:
- rebase.
V3 -> V4:
- add check of handler_id range.
V2 -> V3:
- use OVS_DP_ATTR_USER_FEATURES to inform datapath about the type of
OVS_VPORT_ATTR_UPCALL_PID attribute.
- close epoll fd when destroying all channels.
PATCH -> V2:
- rebase.
major changes
This commit implements the API functions to allow multiple handler
threads read upcall.
Also, this commit removes the handling priority of DPIF_UC_MISS
over DPIF_UC_ACTION. So, both misses will be put to the same
queue. The decision is based on the fact that a lot has changed
since the age when
---
V4 -> V5:
- Rebase to master.
- Minor change based on Thomas Graf's review.
- Add Acked-by: Thomas Graf in PATCH 3/5.
V3 -> V4:
- Implement the recv_set() in dpif-netdev.
- Add check to the range of handler_id in recv() and recv_wait().
- Use reciprocal_div() to reduce the cost of modular ope
In order to allow handlers directly read upcalls from datapath,
we need to support per-handler netlink socket for each vport in
datapath. This commit makes this happen. Also, it is guaranteed
backward and forward compatibility with previous branch.
Signed-off-by: Alex Wang
Acked-by: Thomas Graf
With the foundation laid in previous commits, this commit
removes the 'dispatcher' thread by allowing 'handler'
threads to read upcalls directly from dpif.
This commit significantly simplifies the flow miss handling
code and brings slight improvement to flow setup rate.
Note:
- the flow setup rat
This commit changes the API in 'dpif-provider.h' to allow multiple
handler threads call dpif_recv() simultaneously.
Signed-off-by: Alex Wang
---
V4 -> v5:
- rebase.
V3 -> V4:
- refines the port_get_pid() comments as suggested.
V2 -> V3:
- detach the channel refreshing from recv_set().
Note:
From: Zoltan Kiss
Date: Tue, 18 Mar 2014 21:17:35 +
> skb_zerocopy can copy elements of the frags array between skbs, but it doesn't
> orphan them. Also, it doesn't handle errors, so this patch takes care of that
> as well.
>
> Signed-off-by: Zoltan Kiss
> ---
> net/openvswitch/datapath.c
Hey Thomas,
Thx a lot for the review, again, comments in line,
On Wed, Mar 19, 2014 at 4:37 AM, Thomas Graf wrote:
> On 03/18/2014 01:15 AM, Alex Wang wrote:
>
>> In order to allow handlers directly read upcalls from datapath,
>> we need to support per-handler netlink socket for each vport in
>
On Wed, Mar 19, 2014 at 12:32:22PM -0700, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 01:54:09PM -0700, Pravin wrote:
> > DPDK need to set _lcore_id for using multiple core.
> >
> > Signed-off-by: Pravin B Shelar
>
> In xpthread_setaffinity_np_cpu(), I think that the whole function is
> Linux-spe
On Tue, Mar 18, 2014 at 01:54:09PM -0700, Pravin wrote:
> DPDK need to set _lcore_id for using multiple core.
>
> Signed-off-by: Pravin B Shelar
In xpthread_setaffinity_np_cpu(), I think that the whole function is
Linux-specific, so that the implementation should be enclosed in
#ifdef __linux__.
On Tue, Mar 18, 2014 at 01:54:04PM -0700, Pravin wrote:
> Following patch adds DPDK netdev-class to userspace datapath. Now
> OVS can use DPDK port for IO by just configuring DPDK port and then
> adding dpdk type port to userspace datapath.
>
> Refer to INSTALL.DPDK doc for further info.
>
> This
This function has no more users, so remove it.
Signed-off-by: Jarno Rajahalme
---
v2: rebase
lib/packets.c | 18 +-
lib/packets.h |1 -
2 files changed, 1 insertion(+), 18 deletions(-)
diff --git a/lib/packets.c b/lib/packets.c
index 0c023fc..65ba3f6 100644
--- a/lib/pack
On Mon, Mar 17, 2014 at 3:50 PM, Ben Pfaff wrote:
> On Tue, Mar 11, 2014 at 04:56:17PM -0700, Andy Zhou wrote:
>> Recirculation ID needs to be unique per datapath. Its usage will be
>> tracked by the backer that corresponds to the datapath.
>>
>> In theory, Recirculation ID can be any uint32_t val
On Tue, Mar 18, 2014 at 01:53:52PM -0700, Pravin wrote:
> new netdev type like DPDK can support multi-queue IO. Following
> patch Adds support for same.
>
> Signed-off-by: Pravin B Shelar
Needs comment and documentation updates.
In particular, from reading the code I infer (but I'm not confiden
On Tue, Mar 18, 2014 at 01:53:34PM -0700, Pravin wrote:
> Preparation for multi queue netdev IO.
>
> Signed-off-by: Pravin B Shelar
In lib/netdev-provider.h, can you realign the table here?
*"alloc" "construct" "destruct" "dealloc"
* ---
On Tue, Mar 18, 2014 at 01:53:27PM -0700, Pravin wrote:
> This patch adds PMD type netdev for netdevice with poll-mode
> drivers. Since there is no way to get signal on a packet recv
> from these devices we need to poll them in busy loop. So minimize
> system call overhead this patch uses dpif-th
Thx, applied,
On Wed, Mar 19, 2014 at 10:20 AM, Ben Pfaff wrote:
> On Wed, Mar 19, 2014 at 09:29:43AM -0700, Alex Wang wrote:
> > diff --git a/lib/ovs-rcu.c b/lib/ovs-rcu.c
> > index ac4513b..0f427c8 100644
> > --- a/lib/ovs-rcu.c
> > +++ b/lib/ovs-rcu.c
> > @@ -127,6 +127,8 @@ ovsrcu_quiesce_s
On Wed, Mar 19, 2014 at 09:29:43AM -0700, Alex Wang wrote:
> diff --git a/lib/ovs-rcu.c b/lib/ovs-rcu.c
> index ac4513b..0f427c8 100644
> --- a/lib/ovs-rcu.c
> +++ b/lib/ovs-rcu.c
> @@ -127,6 +127,8 @@ ovsrcu_quiesce_start(void)
> void
> ovsrcu_quiesce(void)
> {
> +ovsrcu_init();
> +
>
On Wed, Mar 19, 2014 at 8:44 AM, Jarno Rajahalme wrote:
>
> On Mar 18, 2014, at 8:19 PM, Pravin Shelar wrote:
>
> On Tue, Mar 18, 2014 at 3:56 PM, Jarno Rajahalme
> wrote:
>
>
> On Mar 18, 2014, at 1:53 PM, Pravin wrote:
>
> DPDK can receive multiple packets but current netdev API does
> not al
*After compiled ovs with clang, I experience the crash:*
[430100.705907] openvswitch: Open vSwitch switching datapath 2.1.90, built
Mar 19 2014 09:22:17
[430101.140122] ovsdb-server[27292]: segfault at 0 ip 0045b3a3 sp
7fffb8a3e100 error 4 in ovsdb-server[40+c7000]
[430101.142000]
The flow that created the netdev_flow might have wildcarded TCP flags,
or it may not be a TCP flow at all. Fix this by using the freshly
extracted flow key to parse TCP flags.
Signed-off-by: Jarno Rajahalme
---
v2: rebase
lib/dpif-netdev.c |7 ---
1 file changed, 4 insertions(+), 3 del
We used to map ODPP_NONE to port number 0, which is wrong, as
ODPP_NONE is a valid value of the flow's in_port.
Signed-off-by: Jarno Rajahalme
---
v2: rebase
lib/flow.c|4 +---
lib/ofp-print.c |2 +-
lib/packets.c | 20
TCP flags are already extracted from the flow, no need to parse them
again.
Signed-off-by: Jarno Rajahalme
---
v2: rebase
lib/dpif-netdev.c |2 +-
lib/dpif.c|2 +-
ofproto/ofproto-dpif-upcall.c |2 +-
ofproto/ofproto-dpif.c|6 ++
4 fil
On Mar 19, 2014, at 8:45 AM, Alex Wang wrote:
> I think we should also remove the 'Note' in the comment.
>
It still seems valid to me:
“Note: While hmap in general is not thread-safe without additional locking,
hmap_is_empty() is.”
Jarno
>
> On Tue, Mar 18, 2014 at 9:00 PM, Ben Pfaff w
On Tue, Mar 11, 2014 at 01:56:02PM -0700, Ben Pfaff wrote:
> v1->v2: Redid all the ovs-atomic changes to fix a bug in the GCC 4.x (x <
> 7) support. This was not a simple fix; all the ovs-atomic patches are
> essentially new. Also, changed the ovs-rcu library to use "consume" rather
> than "acqui
Thanks for the review. Pushed to master,
Jarno
On Mar 18, 2014, at 9:00 PM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 04:42:55PM -0700, Jarno Rajahalme wrote:
>> The fence made classifier_lookup() slower. Access to a size_t 'n' is
>> safe without synchonizing, and if racing with writers ma
I think we should also remove the 'Note' in the comment.
On Tue, Mar 18, 2014 at 9:00 PM, Ben Pfaff wrote:
> On Tue, Mar 18, 2014 at 04:42:55PM -0700, Jarno Rajahalme wrote:
> > The fence made classifier_lookup() slower. Access to a size_t 'n' is
> > safe without synchonizing, and if racing wi
On Mar 18, 2014, at 8:38 PM, Pravin Shelar wrote:
> On Tue, Mar 18, 2014 at 4:04 PM, Jarno Rajahalme
> wrote:
>>
>> On Mar 18, 2014, at 1:53 PM, Pravin wrote:
>>
>>> DPDK netdev need to access ofpbuf while sending buffer. Following
>>> patch changes netdev_send accordingly.
>>
>> Would it
On Mar 18, 2014, at 8:19 PM, Pravin Shelar wrote:
> On Tue, Mar 18, 2014 at 3:56 PM, Jarno Rajahalme
> wrote:
>>
>> On Mar 18, 2014, at 1:53 PM, Pravin wrote:
>>
>>> DPDK can receive multiple packets but current netdev API does
>>> not allow that. Following patch allows dpif-netdev receive
On 03/18/2014 01:15 AM, Alex Wang wrote:
In order to allow handlers directly read upcalls from datapath,
we need to support per-handler netlink socket for each vport in
datapath. This commit makes this happen. Also, it is guaranteed
backward and forward compatibility with previous branch.
Sign
On Tue, Mar 11, 2014 at 04:56:21PM -0700, Andy Zhou wrote:
> Infrastructure to enable megaflow support for bond ports using
> recirculation. This patch adds the following features:
> * Generate RECIRC action when bond can benefit from recirculation.
> * Populate post recirculation rules in table 25
78 matches
Mail list logo
