I am seeing a segfault with ovs-vswitchd when a VM attached to a OVS
bridge is shutdown.
I am running openvswitch-2.0.0 on linux kernel:3.10.30
Is this a known issue with any fix in the git repository?
Core was generated by `ovs-vswitchd --pidfile --detach'.
Program terminated with signal 11, S
In dpif_netdev_flow_del() and dp_netdev_port_input(), the
referenced 'netdev_flow' is not un-referenced. This causes
the leak of the struct's memory.
This commit fixes the above issue by calling dp_netdev_flow_unref()
after using the reference.
Signed-off-by: Alex Wang
---
lib/dpif-netdev.c |
This commit makes dp_netdev_flow_unref() and dp_netdev_actions_unref()
invoke the ovs_refcount_destroy() before freeing the corresponding
pointer.
Signed-off-by: Alex Wang
---
lib/dpif-netdev.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 7
On Fri, Feb 21, 2014 at 5:19 AM, Ben Pfaff wrote:
> On Thu, Feb 20, 2014 at 12:45:49PM +0800, Kmindg G wrote:
>> On Thu, Feb 20, 2014 at 3:20 AM, Ben Pfaff wrote:
>> > Until now, when it processes OFPT_PORT_MOD message, Open vSwitch has waited
>> > for ports to notify it that their status has cha
On Wed, Feb 19, 2014 at 03:36:25PM -0800, Gurucharan Shetty wrote:
> There is no direct mapping for the ioctl function in
> Windows. As of now, af_inet_ioctl() is being used for Linux
> and BSD. So, don't try to compile it for Windows.
>
> Signed-off-by: Gurucharan Shetty
Acked-by: Ben Pfaff
On Wed, Feb 19, 2014 at 03:36:24PM -0800, Gurucharan Shetty wrote:
> In windows there is no clear way to distinguish between a
> socket fd and a file fd.
>
> We use the function, describe_fd() mostly for debugging.
> For now, return a generic statement.
>
> Co-authored-by: Linda Sun
> Signed-off
On Wed, Feb 19, 2014 at 03:36:23PM -0800, Gurucharan Shetty wrote:
> Windows defines the 'optval' argument as char * instead of void *.
>
> Signed-off-by: Gurucharan Shetty
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://openvswitc
On Wed, Feb 19, 2014 at 03:36:22PM -0800, Gurucharan Shetty wrote:
> There is no corresponding function for Windows.
> open() does not work on directories.
> There is a function _commit(fd), but that is only meant
> for files.
>
> Signed-off-by: Gurucharan Shetty
Acked-by: Ben Pfaff
___
On Wed, Feb 19, 2014 at 03:36:21PM -0800, Gurucharan Shetty wrote:
> Signed-off-by: Gurucharan Shetty
This is only used in daemon.c, which isn't compiled on Windows. Can we
just move it into there (as a static function)?
Alternatively, I'm pretty sure that Windows does have /dev/null under
anot
On Wed, Feb 19, 2014 at 03:36:20PM -0800, Gurucharan Shetty wrote:
> For Windows sockets, one has to call closesocket() to
> close the sockets.
>
> Signed-off-by: Gurucharan Shetty
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://op
On Wed, Feb 19, 2014 at 03:36:19PM -0800, Gurucharan Shetty wrote:
> Don't try to compile Unix socket related functions for Windows.
>
> Signed-off-by: Gurucharan Shetty
Can we check for AF_UNIX instead of for _WIN32?
___
dev mailing list
dev@openvswit
On Wed, Feb 19, 2014 at 03:36:18PM -0800, Gurucharan Shetty wrote:
> drain_rcvbuf() is currenlty called from netlink-socket.c and
> netdev-linux.c. As of now, I don't see it being used for Windows.
>
> Signed-off-by: Gurucharan Shetty
Is this because of the MSG_DONTWAIT here too? If so, then I
On Wed, Feb 19, 2014 at 03:36:17PM -0800, Gurucharan Shetty wrote:
> Windows send() does not have a MSG_DONTWAIT.
> So, use the get_socket_error() function.
>
> Co-authored-by: Linda Sun
> Signed-off-by: Linda Sun
> Signed-off-by: Gurucharan Shetty
The use of MSG_DONTWAIT here is just caution,
> On Jan 21, 2014, at 9:34 AM, Ben Pfaff wrote:
>
> Signed-off-by: Ben Pfaff
> ---
> FAQ | 17 +
> 1 file changed, 17 insertions(+)
>
> diff --git a/FAQ b/FAQ
> index 75d9e6b..dac8fe1 100644
> --- a/FAQ
> +++ b/FAQ
> @@ -1454,6 +1454,23 @@ A: These flows drop the ARP packets t
Signed-off-by: Gurucharan Shetty
---
lib/signals.c | 69 -
lib/signals.h |7 --
2 files changed, 76 deletions(-)
diff --git a/lib/signals.c b/lib/signals.c
index 27da5d6..85e5c79 100644
--- a/lib/signals.c
+++ b/lib/signals.c
@@ -
On Tue, Feb 18, 2014 at 10:52 AM, Alex Wang wrote:
> This commit removes the 'dispatcher' thread by allowing 'handler'
> threads to read upcalls directly from dpif. vport in dpif will
> open netlink sockets for each handler and will use the 5-tuple
> hash from the missed packet to choose which so
We are registering an interest in SIGHUP to reopen
log files. But there is an 'ovs-appctl vlog/reopen'
command that does the same and is used in the
logrotate config for the distributions.
So remove the redundant functionality.
Signed-off-by: Gurucharan Shetty
---
vswitchd/ovs-vswitchd.8.in |
On Wed, Feb 19, 2014 at 03:36:16PM -0800, Gurucharan Shetty wrote:
> Couple of return values need changes.
> * EAI_NODATA is the same as EAI_NONAME. So we prevent duplicate cases.
> * Windows does not have a EAI_SYSTEM.
>
> Signed-off-by: Gurucharan Shetty
To make this a little less Win32 specif
On Wed, Feb 19, 2014 at 03:36:15PM -0800, Gurucharan Shetty wrote:
> Windows does not have a inet_aton(). But does have a
> inet_pton().
>
> Signed-off-by: Gurucharan Shetty
inet_aton() isn't in POSIX, but inet_pton() is. I think it's better if
we just change every use of inet_aton() to use ine
On Wed, Feb 19, 2014 at 03:36:14PM -0800, Gurucharan Shetty wrote:
> Windows does not have a getrlimit() function. As such,
> there is no limit to number of file descriptors that
> can be opened. So, set an aritificial limit of 1024.
> This may be increased in the future if the limit is too
> low.
On Wed, Feb 19, 2014 at 03:36:13PM -0800, Gurucharan Shetty wrote:
> Co-authored-by: Linda Sun
> Signed-off-by: Linda Sun
> Signed-off-by: Gurucharan Shetty
Acked-by: Ben Pfaff
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/l
On Wed, Feb 19, 2014 at 03:36:12PM -0800, Gurucharan Shetty wrote:
> And add more users.
>
> Signed-off-by: Gurucharan Shetty
Bitwise operators don't make sense to me here:
> +if (error == EINPROGRESS
> +#ifdef _WIN32
> +| WSAEALREADY | WSAEWOULDBLOCK
> +#endif
Otherwise,
Acked-by:
On Wed, Feb 12, 2014 at 01:50:47PM -0800, Linda Sun wrote:
> Use send/recv for socket stream instead of read/write.
> Use event handle for polling on socket stream.
> Check windows specific return code.
>
> Signed-off-by: Linda Sun
Guru, will you review this (and apply it if you are happy)?
On Thu, Feb 20, 2014 at 12:45:49PM +0800, Kmindg G wrote:
> On Thu, Feb 20, 2014 at 3:20 AM, Ben Pfaff wrote:
> > Until now, when it processes OFPT_PORT_MOD message, Open vSwitch has waited
> > for ports to notify it that their status has changed before it sends a
> > port status update to control
On Fri, Feb 21, 2014 at 02:12:28AM +0530, Nipun Gupta wrote:
> Also, we'd like to know about Bundles extensions. It seemed pretty
> interesting too.
That one is a very big project.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/l
Thanks for the suggestion.
Also, we'd like to know about Bundles extensions. It seemed pretty
interesting too.
Nipun Gupta
Computer Science and Engineering
IIT Delhi
India
On Fri, Feb 21, 2014 at 1:29 AM, Ben Pfaff wrote:
> On Thu, Feb 20, 2014 at 11:44:00PM +0530, Nipun Gupta wrote:
> > I am
On Thu, Feb 20, 2014 at 11:44:00PM +0530, Nipun Gupta wrote:
> I am CS Dual Degree (B.Tech+M.Tech) student in IIT Delhi, currently in my
> 3rd year. I am looking forward to work on a 6-7 week project, along with a
> batchmate of mine in a team of two, to extend OVS with OF1.3+ specs.
> So we looked
Hi everyone,
I am CS Dual Degree (B.Tech+M.Tech) student in IIT Delhi, currently in my
3rd year. I am looking forward to work on a 6-7 week project, along with a
batchmate of mine in a team of two, to extend OVS with OF1.3+ specs.
So we looked at some extensions required for OF1.4 and got interest
On Thu, Feb 20, 2014 at 11:11:00AM +0900, YAMAMOTO Takashi wrote:
> Avoid the following warning caused by the combination
> of NetBSD's htons implementation and gcc bug.
> Now --enable-Werror build succeeds on NetBSD-6.
>
> lib/ofp-util.c: In function 'ofputil_match_from_ofp10_match':
> li
On Thu, Feb 20, 2014 at 01:48:10PM +0900, Simon Horman wrote:
> $logfile is already prefixed by "$sandbox/" and suffixed by ".log"
> so do not duplicate this prefix and suffix combination when appending
> $logfile to $logs.
>
> Cc: YAMAMOTO Takashi
> Signed-off-by: Simon Horman
Applied, thanks!
Thanks! I didn't know about the existence of megaflows feature.
I'll search more on that topic.
If you have some hints, they will be welcome.
best regards
Rafael Duarte Vencioneck
2014-02-19 17:48 GMT-03:00 Ben Pfaff :
> On Wed, Feb 19, 2014 at 05:12:22PM -0300, Rafael Duarte Vencioneck wrote
31 matches
Mail list logo
