> Co-Authored-by: Joe Stringer
> Signed-off-by: Simon Horman
>
> ---
>
> Previously part of "Add packet recirculation" (v18)
>
> v2
> * Use OVS_APP_EXIT_AND_WAIT(ovs-ofctl)
> * Add resubmit and goto_table tests
> * Clean up documentation of MPLS unicast and multicast types in packet
> descrip
On Wed, Apr 16, 2014 at 04:32:26PM +0900, YAMAMOTO Takashi wrote:
> > Co-Authored-by: Joe Stringer
> > Signed-off-by: Simon Horman
> >
> > ---
> >
> > Previously part of "Add packet recirculation" (v18)
> >
> > v2
> > * Use OVS_APP_EXIT_AND_WAIT(ovs-ofctl)
> > * Add resubmit and goto_table tes
> @@ -1762,6 +1765,10 @@ compose_output_action__(struct xlate_ctx *ctx,
> ofp_port_t ofp_port,
> } else if (may_receive(peer, ctx)) {
> if (xport_stp_forward_state(peer)) {
> xlate_table_action(ctx, flow->in_port.ofp_port, 0, true,
> true);
> +
On Wed, Apr 16, 2014 at 04:39:24PM +0900, YAMAMOTO Takashi wrote:
> > @@ -1762,6 +1765,10 @@ compose_output_action__(struct xlate_ctx *ctx,
> > ofp_port_t ofp_port,
> > } else if (may_receive(peer, ctx)) {
> > if (xport_stp_forward_state(peer)) {
> > xlate_ta
> Add per-table counters. This resolves some short-comings
> in the data provided in a table stats reply message.
>
> * Lookups and matches are calculated based on table
> accesses rather than datapath flow hits and misses.
>
> * Lookups and matches are credited to the table where they
> occu
> On Wed, Apr 16, 2014 at 04:39:24PM +0900, YAMAMOTO Takashi wrote:
>> > @@ -1762,6 +1765,10 @@ compose_output_action__(struct xlate_ctx *ctx,
>> > ofp_port_t ofp_port,
>> > } else if (may_receive(peer, ctx)) {
>> > if (xport_stp_forward_state(peer)) {
>> > x
2014/04/16 18:54 "YAMAMOTO Takashi" :
>
> > On Wed, Apr 16, 2014 at 04:39:24PM +0900, YAMAMOTO Takashi wrote:
> >> > @@ -1762,6 +1765,10 @@ compose_output_action__(struct xlate_ctx
*ctx, ofp_port_t ofp_port,
> >> > } else if (may_receive(peer, ctx)) {
> >> > if (xport_stp_forw
From: Andy Zhou
When recirculation is used to implement bond, all bond entries are
always populated regardless whether there is traffic going through
them or not. This change cuts down the noise when running
'ovs-appctl bond/show', by skipping '0KB' entries.
Signed-off-by: Andy Zhou
---
ofprot
From: Andy Zhou
Bond recirculation needs to refresh the 'hidden rules' from
time to time. Keep hash entry slave valid to prevent those
hidden rules from being removed.
Signed-off-by: Andy Zhou
---
ofproto/bond.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/ofproto/bond.c b/ofproto/bon
From: Andy Zhou
When recirculation is used to implement bond, the bond entry stats are
collected from the hidden post recirculation rules. This bug causes
double counting of stats to some strenuous bond entries.
Signed-off-by: Andy Zhou
---
ofproto/ofproto-dpif-xlate.c | 2 +-
1 file changed,
From: Andy Zhou
This is a bug causing per hash entry's pr_rule pointer not properly
maintained; they became NULL after each rebalancing. This patch fixes
this bug.
Signed-off-by: Andy Zhou
---
ofproto/bond.c | 38 --
1 file changed, 16 insertions(+), 22 dele
For the series, assuming you fix the Signed-off-by line:
Acked-by: Jarno Rajahalme
On Apr 16, 2014, at 8:13 AM, Andy Zhou wrote:
> From: Andy Zhou
>
> When recirculation is used to implement bond, the bond entry stats are
> collected from the hidden post recirculation rules. This bug causes
Thanks for the quick review. Fixed and pushed.
On Wed, Apr 16, 2014 at 8:47 AM, Jarno Rajahalme wrote:
> For the series, assuming you fix the Signed-off-by line:
>
> Acked-by: Jarno Rajahalme
>
> On Apr 16, 2014, at 8:13 AM, Andy Zhou wrote:
>
>> From: Andy Zhou
>>
>> When recirculation is use
To elaborate a bit more: now I figured out that normally the controller
should take care of removing OpenFlow rules when the port is deleted. I
thought ovs-vswitchd checks the tables for any reference to the deleted
port, but it seems it doesn't.
My actual problem is that an important rule gets
On Tue, Apr 15, 2014 at 4:31 PM, Andy Zhou wrote:
> When sample action returns with an error, the skb has already been
> freed. This patch fix a bug to make sure we don't free it again.
>
> Signed-off-by: Andy Zhou
Acked-by: Jesse Gross
___
dev mailin
On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
> Currently recirculation action can optionally compute hash. This patch
> adds a hash action that is independent of the recirc action, which
> no longer computes hash. For megaflow bond with recirc, the output
> to a bond port action will look li
On April 16, 2014 at 9:00:15 AM, Zoltan Kiss (zoltan.k...@citrix.com) wrote:
> To elaborate a bit more: now I figured out that normally the controller
> should take care of removing OpenFlow rules when the port is deleted. I
> thought ovs-vswitchd checks the tables for any reference to the deleted
On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
> Recirculation implementation for Linux kernel data path.
>
> Signed-off-by: Andy Zhou
> ---
> datapath/actions.c | 49
> -
> datapath/datapath.c | 37 ++--
For the fixed line, the ofctl_monitor.log should have 18 lines.
So, should wait until it reaches 18 lines.
Signed-off-by: Alex Wang
---
tests/ofproto-dpif.at |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ofproto-dpif.at b/tests/ofproto-dpif.at
index b12b4fe..8201d6
Hey Joe,
On Tue, Apr 15, 2014 at 10:31 PM, Alex Wang wrote:
>
>
>
> On Tue, Apr 15, 2014 at 9:19 PM, Joe Stringer wrote:
>
>> (Apologies for the delay on this discussion)
>>
>> On 11 April 2014 16:54, Alex Wang wrote:
>>
>>> On Thu, Apr 10, 2014 at 7:49 PM, Joe Stringer wrote:
>>>
On 11
On Wed, Apr 16, 2014 at 11:03 AM, Pravin Shelar wrote:
> On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
>> Recirculation implementation for Linux kernel data path.
>>
>> Signed-off-by: Andy Zhou
>> ---
>> datapath/actions.c | 49
>> -
>>
On 4/16/14, 2:57 AM, Jesse Gross wrote:
On Fri, Apr 11, 2014 at 4:30 AM, Lori Jakab wrote:
On 4/11/14, 1:47 AM, Jesse Gross wrote:
On Thu, Mar 20, 2014 at 4:37 AM, Lori Jakab wrote:
On 1/6/14, 7:55 PM, Jesse Gross wrote:
On Tue, Dec 24, 2013 at 9:02 AM, Lorand Jakab wrote:
More importantl
Thanks, pushed.
On Wed, Apr 16, 2014 at 9:32 AM, Jesse Gross wrote:
> On Tue, Apr 15, 2014 at 4:31 PM, Andy Zhou wrote:
>> When sample action returns with an error, the skb has already been
>> freed. This patch fix a bug to make sure we don't free it again.
>>
>> Signed-off-by: Andy Zhou
>
> Ac
On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
> diff --git a/datapath/actions.c b/datapath/actions.c
> index 0b66e7c..cb239c8 100644
> --- a/datapath/actions.c
> +++ b/datapath/actions.c
> +static void execute_hash(struct sk_buff *skb, const struct nlattr *attr)
> +{
> + const struct ovs
On Wed, Apr 16, 2014 at 1:34 PM, Jesse Gross wrote:
> On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
>> diff --git a/datapath/actions.c b/datapath/actions.c
>> index 0b66e7c..cb239c8 100644
>> --- a/datapath/actions.c
>> +++ b/datapath/actions.c
>> +static void execute_hash(struct sk_buff *skb
On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
> diff --git a/datapath/flow_netlink.c b/datapath/flow_netlink.c
> index 0a1effa..883d9bf 100644
> --- a/datapath/flow_netlink.c
> +++ b/datapath/flow_netlink.c
> @@ -129,7 +129,8 @@ static bool match_validate(const struct sw_flow_match
> *match,
On Wed, Apr 16, 2014 at 2:14 PM, Jesse Gross wrote:
> On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
>> diff --git a/datapath/flow_netlink.c b/datapath/flow_netlink.c
>> index 0a1effa..883d9bf 100644
>> --- a/datapath/flow_netlink.c
>> +++ b/datapath/flow_netlink.c
>> @@ -129,7 +129,8 @@ stati
Looks good. Thanks!
Acked-by: Andy Zhou
On Tue, Apr 15, 2014 at 1:22 AM, YAMAMOTO Takashi
wrote:
> Signed-off-by: YAMAMOTO Takashi
> ---
> ofproto/ofproto-dpif.c | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c
> i
On Wed, Apr 16, 2014 at 11:13 AM, Andy Zhou wrote:
> On Wed, Apr 16, 2014 at 11:03 AM, Pravin Shelar wrote:
>> On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
>>> diff --git a/datapath/actions.c b/datapath/actions.c
>>> index cb239c8..4182b3d 100644
>>> --- a/datapath/actions.c
>>> +++ b/datap
Signed-off-by: Pritesh Kothari
---
v3: rebase for changes in datapath/vport.c.
v2: Use OVS_GREP_IFELSE instead of upstream kernel versions as Thomas suggested.
---
FAQ | 2 +-
NEWS | 2 +-
acinclude.m4
Acked-by: Ethan Jackson
I suppose my only worry with this is if we should hold off until the
exact match cache is added to the linux kernel datapath as well . . .
Ethan
On Mon, Apr 14, 2014 at 1:14 PM, Jarno Rajahalme wrote:
> This should optimize port masks for megaflows for typical port usag
On Wed, Apr 16, 2014 at 2:25 PM, Jesse Gross wrote:
> On Wed, Apr 16, 2014 at 11:13 AM, Andy Zhou wrote:
>> On Wed, Apr 16, 2014 at 11:03 AM, Pravin Shelar wrote:
>>> On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
diff --git a/datapath/actions.c b/datapath/actions.c
index cb239c8..
On Wed, Apr 16, 2014 at 1:54 PM, Andy Zhou wrote:
> On Wed, Apr 16, 2014 at 1:34 PM, Jesse Gross wrote:
>> On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
>>> diff --git a/datapath/actions.c b/datapath/actions.c
>>> index 0b66e7c..cb239c8 100644
>>> --- a/datapath/actions.c
>>> +++ b/datapath/
On Wed, Apr 16, 2014 at 2:37 PM, Jesse Gross wrote:
> On Wed, Apr 16, 2014 at 1:54 PM, Andy Zhou wrote:
>> On Wed, Apr 16, 2014 at 1:34 PM, Jesse Gross wrote:
>>> On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
diff --git a/datapath/actions.c b/datapath/actions.c
index 0b66e7c..cb23
Thanks for your review! I'll hold on to this until we have figured out when to
enable this feature.
Jarno
> On Apr 16, 2014, at 2:28 PM, Ethan Jackson wrote:
>
> Acked-by: Ethan Jackson
>
> I suppose my only worry with this is if we should hold off until the
> exact match cache is added to
On Wed, Apr 16, 2014 at 2:39 PM, Andy Zhou wrote:
> On Wed, Apr 16, 2014 at 2:37 PM, Jesse Gross wrote:
>> On Wed, Apr 16, 2014 at 1:54 PM, Andy Zhou wrote:
>>> On Wed, Apr 16, 2014 at 1:34 PM, Jesse Gross wrote:
On Tue, Apr 15, 2014 at 6:21 PM, Andy Zhou wrote:
> diff --git a/datapat
On Wed, Apr 16, 2014 at 2:32 PM, Andy Zhou wrote:
> On Wed, Apr 16, 2014 at 2:25 PM, Jesse Gross wrote:
>> On Wed, Apr 16, 2014 at 11:13 AM, Andy Zhou wrote:
>>> On Wed, Apr 16, 2014 at 11:03 AM, Pravin Shelar wrote:
On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
> diff --git a/dat
On Wed, Apr 16, 2014 at 3:07 PM, Jesse Gross wrote:
> On Wed, Apr 16, 2014 at 2:32 PM, Andy Zhou wrote:
>> On Wed, Apr 16, 2014 at 2:25 PM, Jesse Gross wrote:
>>> On Wed, Apr 16, 2014 at 11:13 AM, Andy Zhou wrote:
On Wed, Apr 16, 2014 at 11:03 AM, Pravin Shelar wrote:
> On Wed, Apr 16
Pushed to master. Thanks a lot for Pravin and YAMAMOTO to review this patch.
On Wed, Apr 16, 2014 at 9:53 AM, Pravin Shelar wrote:
> On Wed, Apr 16, 2014 at 6:51 AM, Andy Zhou wrote:
>> Currently recirculation action can optionally compute hash. This patch
>> adds a hash action that is independe
On Mon, Apr 14, 2014 at 11:39 AM, Thomas Graf wrote:
> On 04/12/2014 12:00 AM, Andy Zhou wrote:
>>
>> Implements Linux kernel datapath hash action. Hash action computes
>> hash and stores it into current packet key.
>>
>> Signed-off-by: Andy Zhou
>>
>> hash fix
>
Horse stuff. I will remove it. Th
On Wed, Apr 16, 2014 at 3:22 PM, Andy Zhou wrote:
> On Wed, Apr 16, 2014 at 3:07 PM, Jesse Gross wrote:
>> On Wed, Apr 16, 2014 at 2:32 PM, Andy Zhou wrote:
>>> On Wed, Apr 16, 2014 at 2:25 PM, Jesse Gross wrote:
On Wed, Apr 16, 2014 at 11:13 AM, Andy Zhou wrote:
> On Wed, Apr 16, 201
On Wed, Apr 16, 2014 at 4:15 PM, Jesse Gross wrote:
> On Wed, Apr 16, 2014 at 3:22 PM, Andy Zhou wrote:
>> On Wed, Apr 16, 2014 at 3:07 PM, Jesse Gross wrote:
>>> On Wed, Apr 16, 2014 at 2:32 PM, Andy Zhou wrote:
On Wed, Apr 16, 2014 at 2:25 PM, Jesse Gross wrote:
> On Wed, Apr 16, 20
On Wed, Apr 16, 2014 at 4:46 PM, Andy Zhou wrote:
> On Wed, Apr 16, 2014 at 4:15 PM, Jesse Gross wrote:
>> On Wed, Apr 16, 2014 at 3:22 PM, Andy Zhou wrote:
>>> On Wed, Apr 16, 2014 at 3:07 PM, Jesse Gross wrote:
On Wed, Apr 16, 2014 at 2:32 PM, Andy Zhou wrote:
> On Wed, Apr 16, 2014
Nice results, reducing CPU usage by around 50% in that case :-)
Feedback below, mostly style.
On 12 April 2014 06:20, Alex Wang wrote:
> @@ -1460,15 +1460,17 @@ struct ofproto_class {
> * support CFM, as does a null pointer. */
> int (*set_cfm)(struct ofport *ofport, const struct cfm
On Wed, Apr 16, 2014 at 5:03 PM, Jesse Gross wrote:
> On Wed, Apr 16, 2014 at 4:46 PM, Andy Zhou wrote:
>> On Wed, Apr 16, 2014 at 4:15 PM, Jesse Gross wrote:
>>> On Wed, Apr 16, 2014 at 3:22 PM, Andy Zhou wrote:
On Wed, Apr 16, 2014 at 3:07 PM, Jesse Gross wrote:
> On Wed, Apr 16, 20
As we discussed offline, it would be good to see the refactoring changes
separated from the functional changes. That makes it easier to see what's
changed and refer back in future.
On 12 April 2014 06:20, Alex Wang wrote:
>
> @@ -2450,8 +2373,40 @@ bridge_run(void)
> iface_stats_timer =
Acked-by: Joe Stringer
On 17 April 2014 06:04, Alex Wang wrote:
> For the fixed line, the ofctl_monitor.log should have 18 lines.
> So, should wait until it reaches 18 lines.
>
> Signed-off-by: Alex Wang
> ---
> tests/ofproto-dpif.at |2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
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 at 7:14 AM, Pravin Shelar wrote:
>> On Sat, Apr 12, 2014 at 3:30 AM, Andy Zhou wrote:
>>> Implements Linux kernel datapath hash action. Hash action computes
>>> hash and stores it
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 at 7:14 AM, Pravin Shelar wrote:
>>> On Sat, Apr 12, 2014 at 3:30 AM, Andy Zhou wrote:
Implements Linux kernel data
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 at 7:14 AM, Pravin Shelar wrote:
On Sat, Apr 12, 2014 at
From: Ethan Jackson
Previously, we had a separate flow_dumper thread that fetched flows from
the datapath to distribute to revalidator threads. This patch takes the
logic for dumping and pushes it into the revalidator threads, resulting
in simpler code with similar performance to the current code
What's the advantage of exactly matching dp_hash? May be we could
make it a maskable action.
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,
Thanks for the review, the sense of getting closer is coming to me, ;D
Please see my reply inline,
+/* Checks the status of CFM configured on 'ofport'. Returns 0 if the
>> + * port's CFM status was successfully stored into '*status'. Returns
>> + * negative number if there is no st
On Wed, Apr 16, 2014 at 6:09 PM, Joe Stringer wrote:
> As we discussed offline, it would be good to see the refactoring changes
> separated from the functional changes. That makes it easier to see what's
> changed and refer back in future.
>
Definitely, hope i can master the art of separating p
Signed-off-by: Andy Zhou
V2->V3:
* rename dp_hash to ovs_flow_hash
* Simplify netlink message error checking logic
* other cleanups
---
datapath/actions.c | 19 ++-
datapath/flow.h | 1 +
datapath/flow_netlink.c | 37 +++
Recirculation implementation for Linux kernel data path.
Signed-off-by: Andy Zhou
---
V2 -> v3: * save the input port in OVS_CB
* Allow recirc_id to be masked like any other key attribute.
* DO not force recirc_id to be exact match.
* Needs corresponding user sp
56 matches
Mail list logo