Just trying share an alternative, could we still do the coverage clear
inside 'pmd_thread_main()'. However,
instead of contending for the 'coverage_mutex', use 'ovs_mutex_trylock()'.
Since pmd thread runs in infinite
loop, it can constantly try grabbing the lock and will eventually acquire
it.
So
Hi IIya,
Thx a lot for writing up this fix,
The reason that you did not see the pmd related coverage counter stats
(in "ovs-appctl coverage/show") is that we do not call coverage_clear()
anywhere in pmd_thread_main(). This is in that the coverage_clear()
requires holding the global mutex in the
There is no external users for ofpbuf_use() directly.
Signed-off-by: Andy Zhou
---
lib/ofpbuf.c | 2 +-
lib/ofpbuf.h | 1 -
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/lib/ofpbuf.c b/lib/ofpbuf.c
index 392f843..c190f8b 100644
--- a/lib/ofpbuf.c
+++ b/lib/ofpbuf.c
@@ -47,7 +47,7
Json string are usually created and freed immediately. Thus, there
should not be any downside in creating a larger buffer initially to
avoid the cost of moving strings around in memory due to realloc()
call.
The following script is used as benchmark to measure number of
bytes reallocated:
ovs-vsc
Add the size to its parameter list.
Signed-off-by: Andy Zhou
---
lib/ofpbuf.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/lib/ofpbuf.c b/lib/ofpbuf.c
index 05b513c..9b9d6b2 100644
--- a/lib/ofpbuf.c
+++ b/lib/ofpbuf.c
@@ -33,12 +33,12 @@ ofpbuf_init__(struct
Keep a count of bytes moved due to the realloc() call.
Signed-off-by: Andy Zhou
---
lib/dynamic-string.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/lib/dynamic-string.c b/lib/dynamic-string.c
index a6c8f6c..89ead08 100644
--- a/lib/dynamic-string.c
+++ b/lib/dynamic-string.c
@@
This change reuses the string length that available from 'ds', saving
a strlen() call.
Signed-off-by: Andy Zhou
---
lib/jsonrpc.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/jsonrpc.c b/lib/jsonrpc.c
index ae51b42..282f3bb 100644
--- a/lib/jsonrpc.c
+++ b/lib/js
Remove tabs per coding style
Signed-off-by: Andy Zhou
---
lib/dynamic-string.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/lib/dynamic-string.h b/lib/dynamic-string.h
index 95172d1..f1e0a36 100644
--- a/lib/dynamic-string.h
+++ b/lib/dynamic-string.h
@@ -1,5 +1,5 @@
Make use of ofpbuf_use_ds() to simplify code.
Signed-off-by: Andy Zhou
---
lib/jsonrpc.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/lib/jsonrpc.c b/lib/jsonrpc.c
index 282f3bb..b482777 100644
--- a/lib/jsonrpc.c
+++ b/lib/jsonrpc.c
@@ -240,7 +240,6 @@ jsonrpc_send(st
Add an API to convert a dynamic string object into ofpbuf.
Signed-off-by: Andy Zhou
---
lib/ofpbuf.c | 10 ++
lib/ofpbuf.h | 1 +
2 files changed, 11 insertions(+)
diff --git a/lib/ofpbuf.c b/lib/ofpbuf.c
index 9b9d6b2..392f843 100644
--- a/lib/ofpbuf.c
+++ b/lib/ofpbuf.c
@@ -53,6 +53,
On 10 August 2015 at 11:32, Daniele Di Proietto wrote:
> I'm surprised as well that the tests were passing.
>
> Sorry about this and thanks for finding out
>
> Acked-by: Daniele Di Proietto
Thanks, I applied this patch to master.
I plan to revise the later patches to try to handle support detec
On 11 August 2015 at 11:13, Jarno Rajahalme wrote:
>> -if (!keep) {
>> +if (result == UKEY_DELETE) {
>> delete_op_init(udpif, &ops[n_ops++], ukey);
>> +} else if (result == UKEY_MODIFY) {
>> +ofpbuf_delete(ukey->actions);
>> +
On 10 August 2015 at 18:46, Ethan J. Jackson wrote:
> exit:
> -if (ok) {
> +if (result != UKEY_DELETE) {
> ukey->reval_seq = reval_seq;
> }
> -if (netflow && !ok) {
> +if (netflow && result != UKEY_DELETE) {
> netflow_flow_clear(netflow, &flow);
> }
I
FWIW the "slice" logic in the for loop is meant to ensure that no
revalidator threads concurrently clean the same map.
There's two cases where the main thread might call into this as well
though, one is when reconfiguring threads (which shouldn't conflict,
because it stops all revalidators first),
Thx, applied to master~
On Tue, Aug 11, 2015 at 11:16 AM, Jarno Rajahalme
wrote:
> Acked-by: Jarno Rajahalme
>
> > On Aug 10, 2015, at 7:47 PM, Alex Wang wrote:
> >
> > Signed-off-by: Alex Wang
> > ---
> > ofproto/ofproto-dpif-upcall.c |4
> > 1 file changed, 4 insertions(+)
> >
> > d
Yeah sorry this patch is not great. That second problem is solved in
the follow on. I'd be somewhat inclined to drop this one all
together.
Ethan
On Tue, Aug 11, 2015 at 11:15 AM, Jarno Rajahalme wrote:
>
>> On Aug 11, 2015, at 10:42 AM, Jarno Rajahalme wrote:
>>
>>
>>> On Aug 10, 2015, at 6:
Acked-by: Jarno Rajahalme
> On Aug 10, 2015, at 7:47 PM, Alex Wang wrote:
>
> Signed-off-by: Alex Wang
> ---
> ofproto/ofproto-dpif-upcall.c |4
> 1 file changed, 4 insertions(+)
>
> diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/ofproto-dpif-upcall.c
> index 6385abc..4fed956 10
> On Aug 11, 2015, at 10:42 AM, Jarno Rajahalme wrote:
>
>
>> On Aug 10, 2015, at 6:46 PM, Ethan J. Jackson wrote:
>>
>> From: Ethan Jackson
>>
>> Since revalidator_sweep() doesn't hold the ukey mutex for each full
>> loop iteration, it's theoretically possible that two threads may
>> call
> On Aug 10, 2015, at 6:46 PM, Ethan J. Jackson wrote:
>
> From: Ethan Jackson
>
> There are certain use cases (such as bond rebalancing) where a
> datapath flow's actions may change, while it's wildcard pattern
> remains the same. Before this patch, revalidators would note the
> change, dele
Thx, IIya for the review, please see my reply below,
On Tue, Aug 11, 2015 at 4:41 AM, Ilya Maximets
wrote:
> On 11.08.2015 05:47, Alex Wang wrote:
> > When dpdk configuration changes, all pmd threads are recreated
> > and rx queues of each port are reloaded. After this process,
> > rx queue c
Thx, this is really really useful!!!
Just few minor suggestions below,
On Mon, Aug 10, 2015 at 8:33 PM, Russell Bryant wrote:
> I frequently view the contents of the Logical_Flow table while working
> on OVN. Add a patch that can output the contents of this table in a
> sorted way that makes i
> On Aug 10, 2015, at 6:46 PM, Ethan J. Jackson wrote:
>
> From: Ethan Jackson
>
> Since revalidator_sweep() doesn't hold the ukey mutex for each full
> loop iteration, it's theoretically possible that two threads may
> call ukey_delete() on the same ukey. If this happens, they both will
> at
On Tue, Aug 11, 2015 at 04:24:01PM +, Traynor, Kevin wrote:
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Flavio Leitner
> > Sent: Friday, July 31, 2015 11:30 PM
> > To: dev@openvswitch.org
> > Cc: Flavio Leitner
> > Subject: [ovs-dev] [RFC] d
This looks like a reasonable solution, I'll have a play around with it.
On 11 August 2015 at 05:25, Daniele Di Proietto wrote:
> I'm not an autoconf expert, but having NS_CHECK_EXEC and
> NS_CHECK_EXEC_UNQUOTED for this purpose might be confusing, since
> there already are AT_CHECK and AT_CHECK_U
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Flavio Leitner
> Sent: Friday, July 31, 2015 11:30 PM
> To: dev@openvswitch.org
> Cc: Flavio Leitner
> Subject: [ovs-dev] [RFC] dpdk: support multiple queues in vhost
>
> This RFC is based on the vhost mult
I'm not an autoconf expert, but having NS_CHECK_EXEC and
NS_CHECK_EXEC_UNQUOTED for this purpose might be confusing, since
there already are AT_CHECK and AT_CHECK_UNQUOTED, and the "UNQUOTED"
refers to comparison text.
How about this?
# NS_EXEC([namespace], [command])
#
# Execute 'command' in 'na
On 11.08.2015 05:47, Alex Wang wrote:
> When dpdk configuration changes, all pmd threads are recreated
> and rx queues of each port are reloaded. After this process,
> rx queue could be mapped to a different pmd thread other than
> the one before reconfiguration. However, this is totally
> transp
Hi, I need some help about ovs QOS policing.
I want to know the meaning of the flag VALID_POLICING
In function netdev_linux_set_policing(), whether parameter "kbits_rate"equals
to 0 or not, VALID_POLICING flag will be set.
In my opinion, only when users set ingress_policing_rate to none zero, then
The original message was received at Tue, 11 Aug 2015 14:36:35 +0530 from
[222.51.252.46]
- The following addresses had permanent fatal errors -
dev@openvswitch.org
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listi
When you'll be able to answer her desires?
Good health good times! Improve it swiftly with Cialis Soft. Place an order
in our site!
Make your choice!
A
E
shone and twinkled, and his usually pale face was flushed and
___
de
30 matches
Mail list logo
