> diff --git a/tests/atlocal.in b/tests/atlocal.in
> index 825d289..bada7e2 100644
> --- a/tests/atlocal.in
> +++ b/tests/atlocal.in
> @@ -93,13 +93,19 @@ FreeBSD|NetBSD)
> ;;
> esac
>
> -# Check for MINGW platform.
> +# Check for platform.
> case `uname` in
> MINGW*)
> IS_WIN32="yes
Hey Flavio,
I'm thinking of adding a kernel version dependency to the rpm spec file...
Don't know if it is possible to specify something similar to this ?
Requires: kernel >= 3.10.0-123.8.1.el7.x86_64
Since that is the version of kernel on which ovs kmod can run
successfully...
Thanks,
Alex Wan
On Thu, Nov 13, 2014 at 11:56:14AM -0800, Jarno Rajahalme wrote:
> Previously, accurate iteration required writers to be excluded during
> iteration. This patch adds an rculist to struct cls_subtable, and a
> corresponding list node to struct cls_rule, which makes iteration more
> straightforward,
Acked-by: Jarno Rajahalme -
Thanks!
Jarno
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> With this patch setters invoke procedures only if values have changed.
> Also rstp_set_bridge_address__() keeps the existing priority in the
> bridge_identifier.
>
> Signed-off-by
Acked-by: Jarno Rajahalme
Thanks!
Jarno
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> Other ports should stop forwarding and learning when a port receives a
> superior BPDU carrying a Proposal flag.
> Without this patch this does not happen and other ports keep execu
Daniele,
See my comments below,
Thanks!
Jarno
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> When the condition associated with a global transition is met, it
> supersedes all other exit conditions including UCT.
It would be nice if you’d explain in a sentence or two what
On Thu, Nov 13, 2014 at 08:48:10AM -0800, Ben Pfaff wrote:
> On Thu, Nov 13, 2014 at 11:53:27AM +0900, Simon Horman wrote:
> > Hi,
> >
> > this series is intended to complete a full implementation of
> > EXT-350, draft OpenFlow 1.5 groups including. It provides the missing
> > components of that i
Daniele,
See comments inline.
Thanks!
Jarno
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> All MAC addresses previously learned on a Root Port can be moved to an
> Alternate Port that becomes the new Root Port; i.e., Dynamic Filtering
> Entries for those addresses may
Acked-by: Jarno Rajahalme
Thanks!
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> A Config BPDU always conveys a Designated Port Role.
>
> Signed-off-by: Daniele Venturino
> ---
> lib/rstp-state-machines.c | 13 -
> 1 file changed, 8 insertions(+), 5 deletions(-)
>
>
On Thu, Nov 13, 2014 at 11:56:13AM -0800, Jarno Rajahalme wrote:
> There is no point in adding duplicate information into prefix tries.
>
> Also, since the lower-priority duplicate rules are not visible to
> lookups, they do not need to be in staged lookup indices directly
> either (the head rule
Assuming you fix the commit message:
Acked-by: Jarno Rajahalme
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> Invoke move_rstp__() in rstp_port_set_administrative_bridge_port__() if
> port is not initializing.
>
Explain in the commit message why this is needed. And lose the
Small nits below, otherwise:
Acked-by: Jarno Rajahalme
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> With this commit, RSTP is able to flush from the MAC learning table
> entries pertaining to a single port. Before this commit the whole table
> was flushed every time a
This patch also changes STP, so the subject line should start with “stp, rstp:”
and the commit message should not have indentation. Otherwise:
Acked-by: Jarno Rajahalme
Thanks!
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> There is a difference between a port with STP/RSTP prot
On Wed, Nov 5, 2014 at 2:23 PM, Michio Honda wrote:
> Parsing packets against all the flow-key fields and lookup
> is expensive.
> This patch enables to implement least-possible flow
> key extraction and/or packet lookup for a specific pattern
> of flows installed in the datapath (e.g., all the fl
This patch needs a proper commit message, e.g., why the change is needed?
Thanks,
Jarno
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> Signed-off-by: Daniele Venturino
> ---
> lib/rstp.c | 14 ++
> 1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/lib/rs
Apart from the indentation in commit message:
Acked-by: Jarno Rajahalme
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
>Set rcdv_tcn and return OTHER_INFO when a
>TOPOLOGY_CHANGE_NOTIFICATION_BPDU is received.
>
> Signed-off-by: Daniele Venturino
> ---
> lib/rstp-state-
“rename … in …” -> “rename … as …”. Also, there is extra indentation in the
commit message. Otherwise:
Acked-by: Jarno Rajahalme
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
>All other similar port_role_transition_sm states are named like this.
>
> Signed-off-by: Daniele Vent
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
>If the DesignatedBridgeID Bridge Address component is equal to that
>component of the Bridge's own bridge priority vector skip to the next
>port.
>
git adds indentation so the email format should not have any.
Otherw
This patch needs a proper commit message. Explain what was wrong, and why the
fix is correct.
Regards,
Jarno
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
> Signed-off-by: Daniele Venturino
> ---
> lib/rstp-state-machines.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
>
On Nov 6, 2014, at 7:31 AM, Daniele Venturino wrote:
>The bridge_port_id is never conveyed in Configuration Messages, but is
>used as a tie-breaker within a Bridge. This patch extends the
>comparison to this fifth field in a rstp_priority_vector.
The commit message in th
Assuming this is a good commit message:
“If the Unknown value of the Port Role parameter is received, the
state machines should treat the RST BPDU as if it were a
Configuration BPDU.”
Also, I think this is more readable with this incremental:
diff --git a/lib/rstp-state-machines.c b/lib/rstp-sta
odp_execute_actions() already handles hash execution part.
Signed-off-by: Pravin B Shelar
---
lib/dpif-netdev.c | 29 +
1 files changed, 1 insertions(+), 28 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 65df19b..2afe71d 100644
--- a/lib/dpif
This patch needs a proper commit message. The user visible changes are in the
corresponding configuration settings. Maybe briefly introduce them in the
commit message?
More comments inline below.
I have a working copy with the white-space changes, so if you provide the
commit message and the a
Also updates ovs-router README documentation.
Signed-off-by: Pravin B Shelar
---
README-native-tunneling.md |2 ++
tests/atlocal.in |8 +++-
tests/tunnel-push-pop.at |3 +++
3 files changed, 12 insertions(+), 1 deletions(-)
diff --git a/README-native-tunneling.md b/R
Acked-by: Jarno Rajahalme
From CodingStyle:
"Use the int_t and uint_t types from for exact-width
integer types. Use the PRId, PRIu, and PRIx macros from
for formatting them with printf() and related functions."
I’ll merge this to master with these portability/style changes:
diff --git a/v
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 modifies the dpif to allow revalidators to skip
dumping the key, mask and actions from the datapath, by making use of
the unique flow
On Nov 6, 2014, at 7:30 AM, Daniele Venturino wrote:
> Hi.
> We tested the RSTP implementation against the IXIA IxANVL validation software.
> This allowed us to fix some small bugs, and the implementation now passes
> such tests with success!
> Please find attached the patch series.
>
>
> It a
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 allow datapaths to
s
Простая ХИТРОСТЬ для безупречной чистоты... Подробности здесь
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
This patch adds a new functions classifier_defer() and
classifier_publish(), which control when the classifier modifications
are made available to lookups. By default, all modifications are made
available to lookups immediately. Modifications made after a
classifier_defer() call MAY be 'deferred'
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
Signed-off-by: Joe Stringer
---
v10: First post.
---
lib/dpctl.c| 41 +
lib/odp-util.c | 35 +++
lib/odp-util.h |2 ++
3 files changed, 74 insertions(+), 4 deletions(-)
diff --git a/lib/dpctl.c b/lib/dpctl.c
inde
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
Previously, flows were manipulated by userspace specifying a full,
unmasked flow key. This adds significant burden onto flow
serialization/deserialization, particularly when dumping flows.
This patch adds an alternative way to refer to flows using a
variable-length "unique flow identifier" (UFID).
Almost all classifier users already exclude concurrent modifications,
or are single-threaded, hence the classifier internal mutex can be
removed. Due to this change, ovs-router.c and tnl-ports.c need new
mutexes, which are added.
Suggested-by: Ben Pfaff
Signed-off-by: Jarno Rajahalme
---
lib/c
This patch shifts the responsibility for determining the hash for a flow
from the revalidation logic down to the dpif layer. This assists in
handling backward-compatibility for revalidation with the upcoming
unique identifier "UFID" patches.
A 128-bit UFID was selected to minimize the likelihood o
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
---
v8-v10: No change.
v7: Rebase.
v4-v6: No change.
v3: Rebase.
v2: No change.
RFC: First post.
---
ofproto/ofproto-dp
Previously, accurate iteration required writers to be excluded during
iteration. This patch adds an rculist to struct cls_subtable, and a
corresponding list node to struct cls_rule, which makes iteration more
straightforward, and allows the iterators to remain ignorant of the
internals of the cls_
From user mode an OVS_IOCTL_TRANSACT NL packet with no previous
dump state is sent to the OVS datapath. The packet is handled by
the dispatch routine and the OvsFlowNlGetCmdHandler() NL command
handler is invoked. The OVS datapath crashes when trying to get
the attributes from the NL header of the
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
There is no point in adding duplicate information into prefix tries.
Also, since the lower-priority duplicate rules are not visible to
lookups, they do not need to be in staged lookup indices directly
either (the head rule is).
Finally, now that cmap operations return the number of elements in th
classifier_remove() was recently changed to take a const struct
cls_rule *. Make the corresponding change to classifier_replace() and
classifier_insert(). This simplifies existing calling sites in
ofproto.
Signed-off-by: Jarno Rajahalme
---
lib/classifier.c | 14 --
lib/classifi
This function previously hid the 'is_mask' parameter from the callers,
which actually have better knowledge about whether it is serializing a
mask or not. Expose this parameter to the callers. This allows the same
function to be called to serialize masked keys as well as masked keys.
To serialize
This patch modifies the dpif interface to allow flows to be manipulated
using a 128-bit identifier. This allows revalidator threads to perform
datapath operations faster, as they do not need to serialise the entire
flow key for operations like flow_get and flow_delete. In conjunction
with a future
Add the 128-bit murmurhash by Austin Appleby, r150 from:
http://code.google.com/p/smhasher/source/browse/trunk/MurmurHash3.cpp
Signed-off-by: Joe Stringer
Acked-by: Ben Pfaff
---
v9-v10: No change.
v8: Build fix for 32-bit (non-static function).
v7: Replace getblock*() with get_unaligned_u*().
Signed-off-by: Joe Stringer
Acked-by: Ben Pfaff
---
v4-v10: No change.
v3: Rebase.
v2: Call ovsrcu_quiesce() unconditionally.
RFC: Initial Post.
---
ofproto/ofproto-dpif-upcall.c | 61 ++---
1 file changed, 33 insertions(+), 28 deletions(-)
diff --git a/ofp
This patch adds a new 'ovs-appctl revalidator/purge' command which
flushes all flows from all datapaths, and updates the revalidator
udpif_key cache at the same time.
Update the ofproto-dpif fragment tests which may fail when ukeys are
created from handler threads.
Signed-off-by: Joe Stringer
CC
The standard mode for printing flows doesn't always provide the full
range of infarmation that is available, particularly with the upcoming
UFID changes. Turn on more detail.
Suggested-by: Justin Pettit
Signed-off-by: Joe Stringer
---
v10: No change.
v9: First post.
---
utilities/bugtool/ovs-bu
This series modifies the dpif interface for flow commands to use 128-bit unique
identifiers as an alternative to the netlink-formatted flow key, and caches the
mask/actions in the udpif_key. This significantly reduces the cost of
assembling messages between revalidators and the datapath, improving
On Thu, Nov 13, 2014 at 10:32:43AM -0800, Pravin Shelar wrote:
> On Wed, Nov 12, 2014 at 7:13 PM, YAMAMOTO Takashi
> wrote:
> >> +AT_CHECK([ovs-appctl ovs/route/add 1.1.2.92/24 br0], [0], [OK
> >> +])
> >
> > this command is available only for linux, isn't it?
>
> Yes, currently it is since other
how are you doing today
i am Susan a single lady i am really looking to meet a honest caring and
loving man who will love me and make me happy please if you are the one
just email me at
susanscott555...@yahoo.com i will send u my pics and tell u more about
myself
yours Susan
On Wed, Nov 12, 2014 at 7:13 PM, YAMAMOTO Takashi
wrote:
>> +AT_CHECK([ovs-appctl ovs/route/add 1.1.2.92/24 br0], [0], [OK
>> +])
>
> this command is available only for linux, isn't it?
>
Yes, currently it is since other platform do not implement route-table
module. I will Document it.
> YAMAMO
Use wire_reason() to translate between different OF versions.
OF1.3 and earlier will still use OFPR_ACTION instead of
OFPR_GROUP.
Async config is initiated and checked againt OF version to
make sure OFPR_GROUP is only reported for OF1.4+. Controllers
running OF1.3+ cannot override the mask and ena
On Thu, Nov 13, 2014 at 11:53:27AM +0900, Simon Horman wrote:
> Hi,
>
> this series is intended to complete a full implementation of
> EXT-350, draft OpenFlow 1.5 groups including. It provides the missing
> components of that implementation by adding support for bucket actions.
> Other portions of
OK, I'll wait until v2 to review patch 2, then. Thanks for letting us
know.
On Thu, Nov 13, 2014 at 08:47:58AM +, Shu Shen wrote:
> It appears the patch is failing a testcase 'ofproto - event filtering
> (OpenFlow 1.3)'.
> I have a fix running through Travis now and will submit a V2 of the p
On Wed, Nov 12, 2014 at 04:29:59PM -0800, Shu Shen wrote:
> Signed-off-by: Shu Shen
Applied, thanks!
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Thu, Nov 13, 2014 at 10:59:41AM -0200, Flavio Leitner wrote:
> On Thu, Nov 13, 2014 at 12:28:39PM +0100, Thomas Graf wrote:
> > v2 with the already merged patches omitted and comments addressed:
> > - Dropped convertion to automake subdir and use of prefix
> >header variable as per Ben's su
On Thu, Nov 13, 2014 at 12:28:39PM +0100, Thomas Graf wrote:
> v2 with the already merged patches omitted and comments addressed:
> - Dropped convertion to automake subdir and use of prefix
>header variable as per Ben's suggestion.
> - Fixes trailing whitespace as spotted by Flavio.
> - Remo
On Thu, Nov 13, 2014 at 12:28:39PM +0100, Thomas Graf wrote:
> v2 with the already merged patches omitted and comments addressed:
> - Dropped convertion to automake subdir and use of prefix
>header variable as per Ben's suggestion.
> - Fixes trailing whitespace as spotted by Flavio.
> - Remo
On 11/13/14 at 10:27am, Flavio Leitner wrote:
> On Thu, Nov 13, 2014 at 12:10:17PM +0100, Thomas Graf wrote:
> > Providing config.h seems pointless too. Whatever was available on the
> > build machine can't be taken for granted on another build machine.
>
> But it tells what features/flags were us
On Thu, Nov 13, 2014 at 12:10:17PM +0100, Thomas Graf wrote:
> On 11/12/14 at 04:06pm, Flavio Leitner wrote:
> > On Wed, Nov 12, 2014 at 03:06:09PM +0100, Thomas Graf wrote:
> > > Ignore both pkgconfig and headers in RHEL package for now as there
> > > is no -devel package yet.
> >
> > Now we have
As headers are installed by 'make install' now, the explicit install
instructions can be be removed from the spec file.
The spec file has been including plus all headers in lib/ so
far which is dangerous:
* config.h is a description of the capabilities of the build machine
where the package
Signed-off-by: Thomas Graf
---
configure.ac | 1 +
lib/.gitignore | 1 +
lib/automake.mk| 3 ++-
lib/libsflow.pc.in | 11 +++
4 files changed, 15 insertions(+), 1 deletion(-)
create mode 100644 lib/libsflow.pc.in
diff --git a/configure.ac b/configure.ac
index ee12216..8
From: Rob Adams
Add rule to generate pkgconfig .pc file from configure.
Install pkg-config file to $(libdir)/pkgconfig
Signed-off-by: Rob Adams
---
AUTHORS | 1 +
Makefile.am | 2 ++
configure.ac | 1 +
lib/.gitignore | 1 +
lib/automake.
Signed-off-by: Thomas Graf
---
configure.ac | 1 +
ovsdb/.gitignore | 1 +
ovsdb/automake.mk| 3 +++
ovsdb/libovsdb.pc.in | 11 +++
4 files changed, 16 insertions(+)
create mode 100644 ovsdb/libovsdb.pc.in
diff --git a/configure.ac b/configure.ac
index d3b3d59..ee1221
Signed-off-by: Thomas Graf
---
configure.ac | 1 +
ofproto/.gitignore | 1 +
ofproto/automake.mk | 2 ++
ofproto/libofproto.pc.in | 11 +++
4 files changed, 15 insertions(+)
create mode 100644 ofproto/libofproto.pc.in
diff --git a/configure.ac b/configure.ac
in
Suggested-by: Ben Pfaff
Signed-off-by: Thomas Graf
---
include/openflow/automake.mk| 3 ++-
include/openvswitch/automake.mk | 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/openflow/automake.mk b/include/openflow/automake.mk
index 4194996..512991e 100644
--- a/
v2 with the already merged patches omitted and comments addressed:
- Dropped convertion to automake subdir and use of prefix
header variable as per Ben's suggestion.
- Fixes trailing whitespace as spotted by Flavio.
- Removed redundant header install instructions in Fedora spec as
suggeste
On 11/12/14 at 04:06pm, Flavio Leitner wrote:
> On Wed, Nov 12, 2014 at 03:06:09PM +0100, Thomas Graf wrote:
> > Ignore both pkgconfig and headers in RHEL package for now as there
> > is no -devel package yet.
>
> Now we have the openvswitch-fedora.spec installing the headers
> during ``make insta
It appears the patch is failing a testcase 'ofproto - event filtering (OpenFlow
1.3)'.
I have a fix running through Travis now and will submit a V2 of the patch
tomorrow morning.
Sorry for letting the failed test case slip through.
Thanks,
Shu
-Original Message-
From: Shu Shen
Sent:
Hi Eitan,
I think this issue should be handled both ways, in the extension as well as
control device interface. The extension should not crash in any case, even if
the user level tools are not correctly used.
Thanks,
Sorin
From: EitanEliahu [mailto:notificati...@github.com]
Sent: Wednesday, 12
71 matches
Mail list logo
