On Wed, Jan 15, 2014 at 02:10:07PM -0800, Joe Stringer wrote:
> On 15 January 2014 09:48, Ben Pfaff wrote:
> >
> > In the latter case, it's a (probably unlikely) race, and it does not
> > appear that it is too hard to avoid, so maybe we should avoidit.
> >
>
> I agree.
>
> Here's a concept that
Hi Ben,
Is this patch in the right format?
Should I send all patches again?
Thanks!
Alexander Wu
On 2014/1/10 15:04, Alexander Wu wrote:
v5.2:
Merge and solve conflicts while free table_features in
oftable_destroy().
v5.1:
Update mutex for get message, add some comments for get/set
Both mega flow mask's reference counter and per flow table mask list
should only be accessed when holding ovs_mutex() lock. However
this is not true with ovs_flow_table_flush(). The patch fixes this bug.
Reported-by: Joe Stringer
Signed-off-by: Andy Zhou
---
v1->v2:
Atomic ops can prote
Thanks, I will send out V4 based on the comments.
On Wed, Jan 15, 2014 at 2:27 PM, Jesse Gross wrote:
> On Wed, Jan 15, 2014 at 1:06 PM, Andy Zhou wrote:
> > diff --git a/datapath/datapath.c b/datapath/datapath.c
> > index b42fd8b..6ae4ecf 100644
> > --- a/datapath/datapath.c
> > +++ b/datapa
On Wed, Jan 15, 2014 at 03:10:35PM -0800, Jesse Gross wrote:
> On Sat, Jan 11, 2014 at 4:33 PM, Luigi Rizzo wrote:
> > Hi,
> > in porting the ovs kernel code to the FreeBSD kernel,
> > compiling with -Wcast-qual -Werror gives some
> >
> >warning: cast discards qualifiers from pointer target ty
> This annotation would have caught the bug fixed by commit 491a67a0005347130
> (ofproto-dpif-xlate: Avoid recursive acquisition of xlate_rwlock.).
looks good to me.
unless there's a more generic way to detect recursive read-lock.
(i don't know)
YAMAMOTO Takashi
>
> Signed-off-by: Ben Pfaff
>
On Sat, Jan 11, 2014 at 4:33 PM, Luigi Rizzo wrote:
> Hi,
> in porting the ovs kernel code to the FreeBSD kernel,
> compiling with -Wcast-qual -Werror gives some
>
>warning: cast discards qualifiers from pointer target type
>
> due to 'const' being ignored. Another couple of warnings
> are on
On Tue, Jan 14, 2014 at 7:03 PM, David Miller wrote:
> From: Thomas Graf
> Date: Tue, 14 Jan 2014 16:27:49 +
>
>> While the zerocopy method is correctly omitted if user space
>> does not support unaligned Netlink messages. The attribute is
>> still not padded correctly as skb_zerocopy() will
On Wed, Jan 15, 2014 at 1:06 PM, Andy Zhou wrote:
> diff --git a/datapath/datapath.c b/datapath/datapath.c
> index b42fd8b..6ae4ecf 100644
> --- a/datapath/datapath.c
> +++ b/datapath/datapath.c
> @@ -1305,10 +1306,13 @@ static void __dp_destroy(struct datapath *dp)
> list_del_rcu(&dp->lis
On 15 January 2014 09:48, Ben Pfaff wrote:
>
> In the latter case, it's a (probably unlikely) race, and it does not
> appear that it is too hard to avoid, so maybe we should avoidit.
>
I agree.
Here's a concept that you could work from. It reflects what I have in
> mind, and it compiles, but I
Both mega flow mask's reference counter and per flow table mask list
should only be accessed when holding ovs_mutex() lock. However
this is not true with ovs_flow_table_flush(). The patch fixes this bug.
Signed-off-by: Andy Zhou
---
datapath/datapath.c |8 +--
datapath/flow_table.c |
This annotation would have caught the bug fixed by commit 491a67a0005347130
(ofproto-dpif-xlate: Avoid recursive acquisition of xlate_rwlock.).
Signed-off-by: Ben Pfaff
CC: YAMAMOTO Takashi
---
ofproto/ofproto-dpif-xlate.c |1 +
ofproto/ofproto-dpif.h |8 ++--
2 files changed,
On Wed, Jan 15, 2014 at 12:41:24PM +0900, YAMAMOTO Takashi wrote:
> Currently xlate_rwlock is recursively acquired.
> (xlate_send_packet -> ofproto_dpif_execute_actions -> xlate_actions)
> Due to writer-preference in rwlock implementations, this causes
> deadlock if another thread tries to acquire
On Wed, Jan 15, 2014 at 12:41:21PM +0900, YAMAMOTO Takashi wrote:
> Signed-off-by: YAMAMOTO Takashi
Thanks, applied to master and branch-2.1.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Wed, Jan 15, 2014 at 12:41:22PM +0900, YAMAMOTO Takashi wrote:
> No functional changes.
>
> Signed-off-by: YAMAMOTO Takashi
Thanks, applied to master and branch-2.1 (the latter only because it
is a prereq for an actual bugfix in one of your later commits).
On Wed, Jan 15, 2014 at 12:41:23PM +0900, YAMAMOTO Takashi wrote:
> This looks like a leftover of commit 2a73b1d73d.
> ("bridge: Reconfigure in single pass.")
>
> Signed-off-by: YAMAMOTO Takashi
Thanks, applied to master.
___
dev mailing list
dev@openv
On Tue, Jan 14, 2014 at 09:32:02AM -0800, Joe Stringer wrote:
> On 13 January 2014 14:55, Ben Pfaff wrote:
> > I have two concerns about nl_dump_next().
> >
> > The first is that any nonzero status always overwrites the existing
> > status. This means that EOF in one thread overwrites EPROTO (or
On Wed, Jan 15, 2014 at 07:10:04PM +0200, Andrei Andone wrote:
> Is there a routine of announcing whether someone is working on a
> certain feature? I would like to announce you guys when I start if
> nobody else will be working on it by then.
It's best to mention it on ovs-dev. In fact, here's a
On 1/15/14, 8:29 AM, "Junguk Cho" wrote:
>Hi, Can I bind a virtual port to a physical port?
>
>I mean if the packet enters a physical port, it will automatically go to
>a virtual port without flow rule( ex) if the packet arrives at port
>1(physical), it will be sent to port 2(virtial port) ).
No
No. That's what flows are for.
On Jan 15, 2014 8:30 AM, "Junguk Cho" wrote:
> Hi, Can I bind a virtual port to a physical port?
>
> I mean if the packet enters a physical port, it will automatically go to a
> virtual port without flow rule( ex) if the packet arrives at port
> 1(physical), it will
Hi, Can I bind a virtual port to a physical port?
I mean if the packet enters a physical port, it will automatically go to a
virtual port without flow rule( ex) if the packet arrives at port
1(physical), it will be sent to port 2(virtial port) ).
Thanks.
__
Распродажа остатков, хомячков осталось всего 10 штук - Спешите!!!
Хит продаж 2013 года - Говорящий Хомяк-Повторюшка? Лучшая интерактивная игрушка!
Лучший подарок как для детей так и для взрослых!
САЙТ: HOMKA-2014.COM
___
dev mailing list
dev@openvswitch.
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
If VLAN acceleration is used when the kernel receives a packet
then the outer-most VLAN tag will not be present in the packet
when it is received by netdev-linux. Rather, it will be present
in auxdata.
This patch uses recvmsg() instead of recv() to read auxdata for
each packet and if the vlan_tid
If VLAN acceleration is used when the kernel receives a packet
then the outer-most VLAN tag will not be present in the packet
when it is received by netdev-linux. Rather, it will be present
in auxdata.
The purpose of this series is to obtain the vlan_tid and if available the
vlan_tpid from auxdata
This is in preparation for pushing vlan tags
using the TPID provided by the kernel via auxdata.
Signed-off-by: Simon Horman
---
v5
* No change
v4
* First post
---
lib/cfm.c| 2 +-
lib/flow.c | 2 +-
lib/odp-execute.c| 2 +-
lib/packets.c
Update the netdev_class so that struct ofpbuf * is passed to rx_recv()
to provide both the data and size of the data to read a packet into.
On success, update struct ofpbuf size inside netdev_class rx_recv
implementation and return 0. This moves logic from the caller.
On error a positive error cod
27 matches
Mail list logo
