Hi,
Thanks Tom for the patch. The patch is clear and concise. We are
looking forward for the testing results of the working patch.
Currently, if a VLAN packet is received by the openvswitch, where the
input switch port is having vlan_mode as:
a. trunk, the packet is forwarded to the respective de
It might be useful to print the port string that was handed to it, since the
other errors in that function do that. Something like:
VLOG_WARN("port %s is a negative value, which is not a valid port number", s);
Assuming that seems reasonable to you and Zoltan thinks it would prevent the
issu
Catched by "learning action - self-modifying flow with hard_timeout"
test case.
The bug introduced by commit b256dc52.
("ofproto-dpif-xlate: Cache xlate_actions() effects.")
Signed-off-by: YAMAMOTO Takashi
---
ofproto/ofproto-dpif-upcall.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
On Apr 17, 2014, at 11:26 AM, Zoltan Kiss wrote:
> On 16/04/14 18:00, Justin Pettit wrote:
>> On April 16, 2014 at 9:00:15 AM, Zoltan Kiss (zoltan.k...@citrix.com) wrote:
>>
>>> My actual problem is that an important rule gets deleted:
>>>
>>> cookie=0x0, duration=1581.083s, table=0, n_packets
Upcoming patches add classifier lookups using miniflows, this is
heavily used for it.
Signed-off-by: Jarno Rajahalme
---
lib/flow.c | 20 ++--
1 file changed, 18 insertions(+), 2 deletions(-)
diff --git a/lib/flow.c b/lib/flow.c
index c6e5e07..03d86ae 100644
--- a/lib/flow.c
+
Add inlined generic accessors for miniflow integer type fields, and a
new miniflow_get_tcp_flags() usinge these. These will be used in a
later patch.
Some definitions also used in lib/packets.h had to be moved there to
resolve circular include dependencies. Similarly, some inline
functions using
Use miniflow as a flow key in the userspace datapath classifier. The
miniflow is expanded for upcalls, but for existing datapath flows, the
key need not be expanded.
Signed-off-by: Jarno Rajahalme
---
lib/dpif-netdev.c | 52 +---
lib/flow.c
Support struct miniflow as a key for datapath flow lookup.
The new classifier interface classifier_lookup_miniflow_first() takes
a miniflow as a key and stops at the first match with no regard to
flow prioritites. This works only if the classifier has no
conflicting rules (as is the case with the
Commit 33ab38d9 (meta-flow: Simplify mf_from_ofp_port_string())
inadvertently removed a check for minus sign at the beginning of a
port number string introduced by commit 05dddba (meta-flow: Don't
allow negative port numbers). This check is still needed, so put it
back, but to ofputil_port_from_st
miniflow_extract() extracts packet headers directly to a miniflow,
which is a compressed form of the struct flow. This does not require
a large struct to be cleared to begin with, and accesses less memory.
These performance benefits should allow this to be used in the DPDK
datapath.
miniflow_extr
Userspace datapath can be made faster by streamlining the flow
extraction code. We already use miniflows internally within the
classifier. This series of patches changes flow extraction to use
miniflow directly, and then also use the extracted miniflow as a
lookup key in the classifier. This mea
This commit reformats the dpif-linux module so that all internal
static functions take 'struct dpif_linux *' as input argument.
This will allow the adding of thread-safety annotations.
Signed-off-by: Alex Wang
---
V9:
- necessary changes for adding thread-safety annotations.
---
lib/dpif-linux.
Signed-off-by: Alex Wang
---
V9:
- add thread-safety annotations.
---
lib/dpif-linux.c | 60 +-
1 file changed, 46 insertions(+), 14 deletions(-)
diff --git a/lib/dpif-linux.c b/lib/dpif-linux.c
index b70ddef..a575b78 100644
--- a/lib/dpif-l
Signed-off-by: Alex Wang
---
V8 -> v9:
- move the fat-lock acquisition inside the if-statement in dpif_linux_run().
- fix an error: dpif_linux_recv_purge() should acquire wrlock.
V7 -> V8:
- rebase.
V6 -> V7:
- rebase.
- set n_handlers to 0 in destroy_all_channels().
- fix the indexing error in
I'd remove the comment about the flow dumper being removed in future.
Doesn't really fit as a code comment.
The comment to upcall_destroy() is rather weird, that function doesn't
know how it's allocated. At any rate, there's only one caller, I'd be
inclined to just get rid of that function and do
All that sounds good to me, go ahead and add my acked by then.
Acked-by: Ethan Jackson
On Thu, Apr 17, 2014 at 4:04 PM, Alex Wang wrote:
> Thx for the review, Ethan,
>
>
> On Thu, Apr 17, 2014 at 3:50 PM, Ethan Jackson wrote:
>>
>> I don't like that dpif_linux_run() has to take the write lock
I don't like that dpif_linux_run() has to take the write lock every
time it's called. Maybe you could make the dpif->refresh_channels
variable atomic, check it, and only take the lock if it's true?
I think a lot of these functions could use annotations saying whether
they require a read/write loc
On Thu, Apr 17, 2014 at 2:48 PM, Jesse Gross wrote:
> On Wed, Apr 16, 2014 at 11:49 PM, Andy Zhou wrote:
>> diff --git a/datapath/actions.c b/datapath/actions.c
>> index 82cfd2d..820075f 100644
>> --- a/datapath/actions.c
>> +++ b/datapath/actions.c
>> +static void execute_hash(struct sk_buff *sk
On Wed, Apr 16, 2014 at 11:49 PM, Andy Zhou wrote:
> diff --git a/datapath/actions.c b/datapath/actions.c
> index 820075f..9307ee1 100644
> --- a/datapath/actions.c
> +++ b/datapath/actions.c
> +static int execute_recirc(struct datapath *dp, struct sk_buff *skb,
> +
On Wed, Apr 16, 2014 at 11:49 PM, Andy Zhou wrote:
> diff --git a/datapath/actions.c b/datapath/actions.c
> index 82cfd2d..820075f 100644
> --- a/datapath/actions.c
> +++ b/datapath/actions.c
> +static void execute_hash(struct sk_buff *skb, const struct nlattr *attr)
> +{
> + struct sw_flow_
On Wed, Apr 16, 2014 at 7:45 PM, Pravin Shelar wrote:
> On Thu, Apr 17, 2014 at 8:12 AM, Jesse Gross wrote:
>> On Wed, Apr 16, 2014 at 7:39 PM, Pravin Shelar wrote:
>>> On Wed, Apr 16, 2014 at 2:09 AM, Andy Zhou wrote:
Thanks for the review. I will send V2.
On Tue, Apr 15, 2014 a
>
> @@ -2481,8 +2436,17 @@ bridge_wait(void)
>>> poll_timer_wait_until(iface_stats_timer);
>>> }
>>>
>>> +/* If the status database transaction is "TXN_INCOMPLETE" in this
>>> run,
>>> + * register a timeout in "STATUS_CHECK_AGAIN_MSEC". Else, wait on
>>> the
>>> + * glo
This commit removes the 'Instant' stats related logic in bridge.c.
Instead, the corresponding status is updated immediately after the
global connectivity sequence number changes.
This change brings the following effects:
1. There is no change to the database update speed, since both the master
This commit adds per 'struct ofproto_dpif' sequence number for
packet-in I/O. Whenever a packet-in is inserted into the 'pins'
queue, the inserting thread will change the sequence number to
wake up the main thread.
Signed-off-by: Alex Wang
---
V4:
- this is a functional change separated from th
This commit adds boolean flag in bfd/cfm module for checking
status change. If there is no status change, the current
update to OVS database will skip the bfd/cfm session.
In the experiment with 5K bfd sessions, when one session is
flapping at rate of every 0.3 second, this patch reduces the
cpu
Someone other than me needs to review this as I co-authored it.
Jarno, do you have time to look at it?
Ethan
On Wed, Apr 16, 2014 at 9:06 PM, Joe Stringer wrote:
> From: Ethan Jackson
>
> Previously, we had a separate flow_dumper thread that fetched flows from
> the datapath to distribute to re
On 16/04/14 18:00, Justin Pettit wrote:
On April 16, 2014 at 9:00:15 AM, Zoltan Kiss (zoltan.k...@citrix.com) wrote:
My actual problem is that an important rule gets deleted:
cookie=0x0, duration=1581.083s, table=0, n_packets=52804,
n_bytes=88968151, idle_age=0, priority=0,in_port=ANY actions=
Thu, Apr 17, 2014 at 05:01:10PM CEST, step...@networkplumber.org wrote:
>On Thu, 17 Apr 2014 14:14:29 +0200
>Jiri Pirko wrote:
>
>> So this can be reused for identification of other "items" as well.
>>
>> Signed-off-by: Jiri Pirko
>> ---
>> drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 2
On Thu, 17 Apr 2014 14:14:29 +0200
Jiri Pirko wrote:
> So this can be reused for identification of other "items" as well.
>
> Signed-off-by: Jiri Pirko
> ---
> drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 2 +-
> drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 2 +-
> drivers/net/et
Le 17/04/2014 14:17, Jiri Pirko a écrit :
Signed-off-by: Jiri Pirko
---
include/linux/if_link.h | 9
ip/Makefile | 3 ++-
ip/iplink_dummyswport.c | 56 +
3 files changed, 67 insertions(+), 1 deletion(-)
create mode 100
On Thu, Apr 17, 2014 at 2:14 PM, Jiri Pirko wrote:
> The the netdevice represents a port in a switch, it will expose
remove one 'the'
> IFLA_PHYS_SWITCH_ID value via rtnl. Two netdevices with a same value
s/a same/the same
> belong to one physical switch.
>
> Signed-off-by: Jiri Pirko
Review
Signed-off-by: Jiri Pirko
---
ip/ipaddress.c | 8
1 file changed, 8 insertions(+)
diff --git a/ip/ipaddress.c b/ip/ipaddress.c
index 43f7296..f794d02 100644
--- a/ip/ipaddress.c
+++ b/ip/ipaddress.c
@@ -470,6 +470,14 @@ int print_linkinfo(const struct sockaddr_nl *who,
f
Signed-off-by: Jiri Pirko
---
include/linux/if_link.h | 9
ip/Makefile | 3 ++-
ip/iplink_dummyswport.c | 56 +
3 files changed, 67 insertions(+), 1 deletion(-)
create mode 100644 ip/iplink_dummyswport.c
diff --git a/include
Signed-off-by: Jiri Pirko
---
lib/utils.c | 46 +-
1 file changed, 13 insertions(+), 33 deletions(-)
diff --git a/lib/utils.c b/lib/utils.c
index 4e9c719..e9e1040 100644
--- a/lib/utils.c
+++ b/lib/utils.c
@@ -743,10 +743,6 @@ char *hexstring_n2a(const
Signed-off-by: Jiri Pirko
---
misc/arpd.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/misc/arpd.c b/misc/arpd.c
index bfe7de9..0839e3f 100644
--- a/misc/arpd.c
+++ b/misc/arpd.c
@@ -36,6 +36,7 @@
#include "libnetlink.h"
#include "utils.h"
+#include "rt_names.h"
Signed-off-by: Jiri Pirko
---
include/linux/if_link.h | 2 ++
ip/ipaddress.c | 8
2 files changed, 10 insertions(+)
diff --git a/include/linux/if_link.h b/include/linux/if_link.h
index f08505c..2932453 100644
--- a/include/linux/if_link.h
+++ b/include/linux/if_link.h
@@ -144,6
Jiri Pirko (5):
iproute2: arpd: use ll_addr_a2n and ll_addr_n2a
iproute2: utils: change hexstring_n2a and hexstring_a2n to do not work
with ":"
iproute2: ipa: show switch id
iproute2: add support for dummyswport
iproute2: ipa: show port id
include/linux/if_link.h | 11 ++
ip
Dummy switch implementation using switchdev interface
Signed-off-by: Jiri Pirko
---
drivers/net/Kconfig | 7 +++
drivers/net/Makefile | 1 +
drivers/net/dummyswitch.c| 126 +++
include/uapi/linux/if_link.h | 9
4 files chang
This will allow to query easily if the vport has netdev. Also it allows
to unexpose netdev_vport_priv and struct netdev_vport.
Signed-off-by: Jiri Pirko
---
net/openvswitch/datapath.c | 2 +-
net/openvswitch/dp_notify.c | 7 ++---
net/openvswitch/vport-internal_dev.c | 53 ++
Signed-off-by: Jiri Pirko
---
include/linux/netdevice.h | 18 +-
net/core/dev.c| 26 ++
2 files changed, 27 insertions(+), 17 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 14bd8b3..bee9673 100644
--- a/in
Benefit from the possibility to work with flows in switch devices and
use the swdev api to offload flow datapath.
Signed-off-by: Jiri Pirko
---
include/linux/sw_flow.h | 16 +++-
net/openvswitch/Makefile | 3 +-
net/openvswitch/datapath.c | 16 +++-
net/openvswitch/hw_offload.c |
This patch adds a couple of ndos which can be used to work with flows.
Note that user can use random port netdevice to access the switch.
Signed-off-by: Jiri Pirko
---
include/linux/netdevice.h | 30
include/linux/switchdev.h | 25
net/core/switchdev
Signed-off-by: Jiri Pirko
---
net/dsa/slave.c | 16
1 file changed, 16 insertions(+)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 02c0e17..22855f3 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -171,6 +171,19 @@ static int dsa_slave_ioctl(struct net_device *dev, s
The the netdevice represents a port in a switch, it will expose
IFLA_PHYS_SWITCH_ID value via rtnl. Two netdevices with a same value
belong to one physical switch.
Signed-off-by: Jiri Pirko
---
include/uapi/linux/if_link.h | 1 +
net/core/rtnetlink.c | 26 +-
2 f
The basic idea is to introduce a generic infractructure to support various
switch chips in kernel.
The current patchset RFC is a heavy rework of the previous ones. The goal is to
be low-intrusive and to use as much as possible from the existing
infrastructure.
Also the idea is to benefit from cur
The goal of this is to provide a possibility to suport various switch
chips. Drivers should implement relevant ndos to do so. So far, only one
ndo for getting physical switch id is in place.
Signed-off-by: Jiri Pirko
---
Documentation/networking/switchdev.txt | 53 +++
After this, flow related structures can be used in other code.
Signed-off-by: Jiri Pirko
---
include/linux/sw_flow.h| 108
net/openvswitch/datapath.c | 45 +++
net/openvswitch/datapath.h | 4 +-
net/openvswitch/flow.c |
Hi Lori, all,
While I've attempted to adapt the layer 3 port patch to GRE tunneling,
and obtained nice results for IP-over-GRE (as I reported last week),
I've had a harder time reusing this framework for -over-GRE, in
particular for MPLS-over-GRE.
Given its name, I maybe shouldn't have expecte
48 matches
Mail list logo
