Dear *Sir/Madam*
The *Asian Economic and Social Society and Pak Publishing Group* are
organizing a 2nd International Conference on Economics, Finance and
Management Outlooks
20-21 December, 2014,* Pearl International Hotel, Kuala Lumpur, Malaysia.*
Website: http://www.pakrdw.com/?ic=details&id
On Wed, Aug 27, 2014 at 03:03:53PM -0500, Jesse Gross wrote:
> On Wed, Aug 27, 2014 at 11:51 AM, Ben Pfaff wrote:
> > On Wed, Aug 27, 2014 at 10:26:14AM +0900, Simon Horman wrote:
> >> On Fri, Aug 22, 2014 at 08:30:08AM -0700, Ben Pfaff wrote:
> >> > On Fri, Aug 22, 2014 at 09:19:41PM +0900, Simon
On Wed, Aug 27, 2014 at 09:51:59AM -0700, Ben Pfaff wrote:
> On Wed, Aug 27, 2014 at 10:26:14AM +0900, Simon Horman wrote:
> > On Fri, Aug 22, 2014 at 08:30:08AM -0700, Ben Pfaff wrote:
> > > On Fri, Aug 22, 2014 at 09:19:41PM +0900, Simon Horman wrote:
> > > > I have been working with Netronome on
On 28 August 2014 10:34, Pravin Shelar wrote:
>
>
> This is weird. I could not reproduce this on newer kernel. But I found
> related bug. Can you try attached patch? It may be related.
> I will submit formal patch later.
>
> --Pravin.
>
This patch doesn't fix the issue.
Another slight variatio
On Tue, Aug 26, 2014 at 11:43 PM, Joe Stringer wrote:
> I've been having a bit of trouble with ovs after this patch was applied,
> ovs-vswitchd locks up on Debian 7.6 (VBox VM). Kernel is 3.2.0-4-amd64 #1
> SMP Debian 3.2.60-1+deb7u3 x86_64 GNU/Linux.
>
> I have a fairly typical phy bridge + int b
On Thu, Aug 21, 2014 at 9:18 AM, Jiri Pirko wrote:
> diff --git a/include/linux/switchdev.h b/include/linux/switchdev.h
> new file mode 100644
> index 000..ba77a68
> --- /dev/null
> +++ b/include/linux/switchdev.h
It should be in include/net/ instead, since it never
goes out of networking.
__
Acked-by: Eitan Eliahu
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Eitan Eliahu
Sent: Wednesday, August 27, 2014 1:47 PM
To: Nithin Raju
Cc: dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH] netlink-socket.c: use read/write ioctl instead
of ReadFile/Wr
On 08/27/14 at 07:48am, Ben Pfaff wrote:
> Are you sure? I think that the only real bug here is that if the
> timeout is negative *and* the "until" string is invalid, then there's a
> memory leak of the error message for the negative timeout. We should
> still fix that, and the same code change d
Yes, I'll open a gerrit issue for that.
Thnks,
Eitan
From: Nithin Raju
Sent: Wednesday, August 27, 2014 1:16 PM
To: Eitan Eliahu
Cc: dev@openvswitch.org; aserd...@cloudbasesolutions.com
Subject: Re: [ovs-dev] [PATCH] netlink-socket.c: use read/write ioctl i
On August 26, 2014 at 8:57:41 AM, Justin Pettit (jpet...@nicira.com) wrote:
>
> > On Aug 26, 2014, at 8:48 AM, Ben Pfaff wrote:
> >
> >> On Mon, Aug 25, 2014 at 06:13:14PM +0200, Thomas Graf wrote:
> >> Can you setup a mailing list for the reports on openvswitch.org
> >> and set an environment va
On Aug 27, 2014, at 11:19 AM, Eitan Eliahu wrote:
>
> +if (!DeviceIoControl(sock->handle, OVS_IOCTL_READ,
> + NULL, 0, tail, sizeof tail, &bytes, NULL))
> ...
> +memcpy(ofpbuf_data(buf), tail, retval);
> +ofpbuf_set_size(buf, r
On Wed, Aug 27, 2014 at 11:51 AM, Ben Pfaff wrote:
> On Wed, Aug 27, 2014 at 10:26:14AM +0900, Simon Horman wrote:
>> On Fri, Aug 22, 2014 at 08:30:08AM -0700, Ben Pfaff wrote:
>> > On Fri, Aug 22, 2014 at 09:19:41PM +0900, Simon Horman wrote:
>> What we would like to do is to provide something ge
hi,
I'm sending out the meeting minutes for future reference.
1. Netlink implementation:
--
Status wise, discussed the following:
a) Nithin created issues on github/ovs-issues to distribute the work across
multiple developers.
b) Ankur committed the patches for the netli
+if (!DeviceIoControl(sock->handle, OVS_IOCTL_READ,
+ NULL, 0, tail, sizeof tail, &bytes, NULL))
...
+memcpy(ofpbuf_data(buf), tail, retval);
+ofpbuf_set_size(buf, retval);
We probably want to send down the "struct iovec iov" (ra
On Wed, Aug 27, 2014 at 10:26:14AM +0900, Simon Horman wrote:
> On Fri, Aug 22, 2014 at 08:30:08AM -0700, Ben Pfaff wrote:
> > On Fri, Aug 22, 2014 at 09:19:41PM +0900, Simon Horman wrote:
> > > I have been working with Netronome on examining the possibilities of
> > > providing (richer) load balan
On Wed, Aug 27, 2014 at 04:01:17PM +, Samuel Ghinet wrote:
> For future reviews, could you please reduce or mark somehow the patch, so it
> would be easier for me to find your comments?
> Something like,
>
> >All NET_BUFFERs of a NET_BUFFER_LIST must go through
> >the pipeline: extract, find
C99 only requires compilers to support four types for bit-fields: signed
int, unsigned int, int, and _Bool. "int" should not be used because it
is implementation-defined whether it is signed. In practice, we have found
that compilers (in particular, GCC, Clang, and MSVC 2013) support any
integer
On Wed, Aug 27, 2014 at 08:54:47AM -0700, Gurucharan Shetty wrote:
> On Tue, Aug 26, 2014 at 4:08 PM, Ben Pfaff wrote:
> > On Tue, Aug 26, 2014 at 04:04:09PM -0700, Jarno Rajahalme wrote:
> >>
> >> On Aug 26, 2014, at 3:53 PM, Ben Pfaff wrote:
> >>
> >> > On Tue, Aug 26, 2014 at 03:42:33PM -0700,
You need to commit the patch also. It needs to do the diff between git ls-files
and the actual files.
-Mesaj original-
De la: Saurabh Shah [mailto:ssaur...@vmware.com]
Trimis: Wednesday, August 27, 2014 6:59 PM
Către: Alin Serdean; Samuel Ghinet; dev@openvswitch.org
Cc: Nithin Raju
Subie
Hello Saurabh,
Thanks for the review!
For future reviews, could you please reduce or mark somehow the patch, so it
would be easier for me to find your comments?
Something like,
>All NET_BUFFERs of a NET_BUFFER_LIST must go through
>the pipeline: extract, find flow, execute. Previously,
>only th
Hi Alin,
I was getting an error after reconf which I now realize was due to missing
files in the distribution.
make[3]: Entering directory `/home/Administrator/work/temp/ovs/datapath'
make[3]: Leaving directory `/home/Administrator/work/temp/ovs/datapath'
The distribution is missing the followin
On Tue, Aug 26, 2014 at 4:08 PM, Ben Pfaff wrote:
> On Tue, Aug 26, 2014 at 04:04:09PM -0700, Jarno Rajahalme wrote:
>>
>> On Aug 26, 2014, at 3:53 PM, Ben Pfaff wrote:
>>
>> > On Tue, Aug 26, 2014 at 03:42:33PM -0700, Jarno Rajahalme wrote:
>> >> Finally, use change the storage type of 'values_i
The Windows datapath supports a READ/WRITE ioctl instead of ReadFile/WriteFile.
In this change, we update the following:
- WriteFile() in nl_sock_send__() to use DeviceIoControl(OVS_IOCTL_WRITE)
- ReadFile() in nl_sock_recv__() to use DeviceIoControl(OVS_IOCTL_READ)
The WriteFile() call in nl_sock
On Tue, Aug 26, 2014 at 08:37:18PM -0700, Nithin Raju wrote:
> I didn't realize earlier while defining OvsDpInterfaceExt.h that
> there are special values defined in netlink-protocol.h for nlmsg_type.
> For Eg. NLMSG_ERROR is defined to be 2. In this patch, we update the
> values of the family IDs
On Tue, Aug 26, 2014 at 08:03:47PM +, Samuel Ghinet wrote:
>
> This patch includes the file renaming and accommodations needed for the file
> renaming to build the forwarding extension for Hyper-V.
>
> This patch is also a follow-up for the thread:
> http://openvswitch.org/pipermail/dev/2014-
On Aug 26, 2014, at 5:13 PM, Ben Pfaff wrote:
> On Wed, Aug 27, 2014 at 12:11:34AM +, Saurabh Shah wrote:
>>
>> I reverted uint8_t -> uint64_t and was able to move past compile flow.h.
>
> OK, so much for that idea. Jarno, let's stick with uint64_t then?
Reverted back to uint64_t on mast
On Tue, Aug 26, 2014 at 08:17:03PM -0700, Nithin Raju wrote:
> While we work out whether nl_sock_join_mcgroup() will be the mechanism
> to support VPORT events, it is easy to add support for
> nl_lookup_genl_mcgroup() and make progress on the other commands.
>
> In this patch, we implement support
On Tue, Aug 26, 2014 at 08:20:51PM -0700, Nithin Raju wrote:
> A typo crept in while respinning get_sock_pid_from_kernel() in the previous
> patch. Fixing it now. Also, get_sock_pid_from_kernel() doesn't need an OUT
> argument. Fixing that too.
>
> Signed-off-by: Nithin Raju
Applied, thanks!
___
Hi Eithan,
Will add them anyway for the moment.
Alin.
-Mesaj original-
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Wednesday, August 27, 2014 6:08 PM
Către: Alin Serdean; Saurabh Shah; dev@openvswitch.org
Subiect: RE: [ovs-dev] [PATCH] Add build of ovsext.sln using MSBuild
Thanks Alin. We probably don't need Win8 but definitely Release and Debug
builds of Win8.1.
Eitan
-Original Message-
From: Alin Serdean [mailto:aserd...@cloudbasesolutions.com]
Sent: Wednesday, August 27, 2014 7:46 AM
To: Eitan Eliahu; Saurabh Shah; dev@openvswitch.org
Subject: RE: [ovs-
On Tue, Aug 26, 2014 at 06:16:12PM -0700, Madhu Challa wrote:
> error string should be freed in all cases.
>
> Signed-off-by: Madhu Challa
This patch was wordwrapped so I applied it by hand.
The "if" isn't necessary so I removed it.
I adjusted the commit message to primarily describe the chang
On Tue, Aug 26, 2014 at 06:00:42PM -0700, Madhu Challa wrote:
> call odp_execute_actions__ only if there are any subactions to prevent null
> pointer dereference in nl_attr_get().
>
> Signed-off-by: Madhu Challa
Should not be necessary because we don't generally deal with untrusted
ODP actions i
On Wed, Aug 27, 2014 at 12:48:04AM +0200, Thomas Graf wrote:
> An error returned from ovsdb_row_from_json() currently results in
> leaking of row.
>
> Signed-off-by: Thomas Graf
I applied a patch from yinpeijun a few minutes ago that fixed the same
problem.
__
On Wed, Aug 27, 2014 at 12:41:50AM +0200, Thomas Graf wrote:
> Although the check is present already, a missing !error branch
> in the next condition prevents the error from being reported back.
>
> Signed-off-by: Thomas Graf
Are you sure? I think that the only real bug here is that if the
time
Hi Eithan,
Yes. Also Win8 or Win8.1
I still to format it accordingly and will send out the patch.
Alin.
-Mesaj original-
De la: Eitan Eliahu [mailto:elia...@vmware.com]
Trimis: Wednesday, August 27, 2014 5:40 PM
Către: Alin Serdean; Saurabh Shah; dev@openvswitch.org
Subiect: RE: [ovs-
On Wed, Aug 27, 2014 at 09:52:54AM +0800, \y00206114 wrote:
> From: yinpeijun
>
> Signed-off-by: yinpeijun
Thanks, I applied this to master.
I changed the commit message to primarily describe the problem fixed and
only secondarily the tool:
From: yinpeijun
Date: Wed, 27 Aug 2014 09:5
Hi Alin,
Should we able to specify Debug or Release build when building with MSBuild.exe?
Thanks,
Eitan
-Original Message-
From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean
Sent: Wednesday, August 27, 2014 7:33 AM
To: Saurabh Shah; dev@openvswitch.org
Subject: Re: [o
Hi Saurabh,
Could you please try to do the autoreconf again.
./boot.sh
./configure
Thanks.
-Mesaj original-
De la: Saurabh Shah [mailto:ssaur...@vmware.com]
Trimis: Wednesday, August 27, 2014 3:01 AM
Către: Samuel Ghinet; dev@openvswitch.org
Cc: Alin Serdean; Nithin Raju
Subiect: RE: [
Hi Saurabh,
At the end of the day today. Was on short PTO last week.
Alin.
-Mesaj original-
De la: Saurabh Shah [mailto:ssaur...@vmware.com]
Trimis: Wednesday, August 27, 2014 2:46 AM
Către: Alin Serdean; dev@openvswitch.org
Subiect: Re: [ovs-dev] [PATCH] Add build of ovsext.sln using M
For the out of tree OVS module, the network stack recursion limit are
some times lower than the default value enforced by dev.c.
example that the default
This patch implements a lower limit, than the limit enforced by dev,c,
to accommodate some OVS use cases that involves higher stack usage.
For e
Since kernel stack is limited in size, it is not wise to using
recursive function with large stack frames.
This patch provides an alternative implementation of recirc action
without using recursion.
A per CPU fixed sized, 'deferred action stack' is used to store both
recirc and sample actions whe
Future pathces will make use of those functions.
Signed-off-by: Andy Zhou
---
datapath/Modules.mk | 1 +
datapath/actions.c | 57 +
datapath/actions.h | 54 ++
datapath/datapath.c | 4
da
Split ovs_dp_packet_flow_lookup() into its own API. In preparation for
the next patch.
Signed-off-by: Andy Zhou
---
datapath/datapath.c | 31 ++-
datapath/datapath.h | 1 +
2 files changed, 23 insertions(+), 9 deletions(-)
diff --git a/datapath/datapath.c b/datapath
Hello,
I'm updating information on Open vSwitch, so I believe you may help me.
I'd like to know, what Neutron releases Open vSwitch is in trunk.
Thank you for your help.
Best regards,
Irina.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.o
On 08/26/14 at 03:23pm, Junguk Cho wrote:
> What I thought is to generally use metadata as matching rule in datapath.
> It is not just between tables.
You may be looking for the 'mark' field of the flow. It is known in
the datapath and used across different Linux kernel subsystems.
___
On 08/26/14 at 06:00pm, Madhu Challa wrote:
> call odp_execute_actions__ only if there are any subactions to prevent null
> pointer dereference in nl_attr_get().
>
> Signed-off-by: Madhu Challa
> ---
> lib/odp-execute.c |3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/lib/odp-exec
Signed-off-by: Joe Stringer
---
lib/dpif-linux.c | 91 +-
1 file changed, 42 insertions(+), 49 deletions(-)
diff --git a/lib/dpif-linux.c b/lib/dpif-linux.c
index f4e78a1..a5f4182 100644
--- a/lib/dpif-linux.c
+++ b/lib/dpif-linux.c
@@ -110,6
When handler threads are installing ukeys, they hold the ukey mutex for
the duration of flow setup. If a revalidator thread dumps this flow
while the handler holds the lock, it will attempt to lock it using
ovs_mutex_trylock(), then if it cannot grab the lock, skip the flow.
Attempting to lock on
This patch modifies the way that ukey lookup is done, to use a 128-bit
hash of the flow key as Netlink attributes. This hash is generated at
flow installation time, and passed down to the dpif so that datapaths
which support UID can index based on this ID rather than the flow.
For backward compati
Signed-off-by: Joe Stringer
---
lib/dpif-netdev.c | 210 ---
lib/flow.h |6 ++
tests/dpif-netdev.at|3 +
tests/ofproto-dpif.at | 20 +++--
tests/ofproto-macros.at |1 +
5 files changed, 165 insertions(+), 75 delet
Currently, udpif_keys are protected during revalidator_sweep__() as only
one thread accesses the ukey at a time. This is ensured using barriers:
all revalidators will be in the GC phase, so they will only access their
own ukey collection.
A future patch will change the access patterns to allow the
Future patches will make use of the 'struct dump_op' in a broader sense,
so this patch renames it to make things a bit clearer.
Signed-off-by: Joe Stringer
Acked-by: Ben Pfaff
---
v2: Added ack.
RFC: First post.
---
ofproto/ofproto-dpif-upcall.c | 95 +
Add the 128-bit murmurhash by Austin Appleby, for 32-bit systems from:
http://code.google.com/p/smhasher/source/browse/trunk/MurmurHash3.cpp
Signed-off-by: Joe Stringer
---
There is also a version for 64-bit systems which I haven't tried yet,
mostly because this version provides the same code for
One of the limiting factors on the number of flows that can be supported
in the datapath is the overhead of assembling flow dump messages in the
datapath. This patch adds a new alternative to dumping the key, mask and
actions from the datapath, which is a 128-bit unique identifier for the
flow. For
This allows us to ignore most fields of a flow_dump, requiring only the
flow key for looking up the ukey. Fetching flows can also be avoided in
the corner case where a flow is missed from a dump but revalidation is
required.
A future patch will modify the datapath interface to make these cached
fi
Signed-off-by: Joe Stringer
Acked-by: Ben Pfaff
---
v2: Call ovsrcu_quiesce() unconditionally.
Added ack.
RFC: Initial Post.
---
ofproto/ofproto-dpif-upcall.c | 61 ++---
1 file changed, 33 insertions(+), 28 deletions(-)
diff --git a/ofproto/ofproto-dpi
An upcoming patch will change the access patterns for ukey maps to
increase the number of writers, and shift write-access from revalidator
threads to upcall handler threads. As such, it no longer makes sense to
tie these maps to revalidators in a 1:1 relationship.
This patch separates the ukey map
Currently, when a revalidator thread first dumps a flow, it creates a
'udpif_key' object and caches a copy of a kernel flow key. This allows
us to perform lookups in the classifier to attribute stats and validate
the correctness of the datapath flow.
This patch sets up this cache from the handler
RFCv2:
- Revised early patches from v1 feedback
- Rebased
RFCv1:
http://openvswitch.org/pipermail/dev/2014-August/044383.html
Joe Stringer (12):
revalidator: Use 'cmap' for storing ukeys.
revalidator: Protect ukeys with a mutex.
udpif: Separate udpif_key maps from revalidators.
upcall: Re
59 matches
Mail list logo
