Re: [ovs-dev] [PATCH v2] Simplify kernel sFlow implementation

On Mon, Aug 15, 2011 at 8:34 PM, pravin shelar wrote: >   This patch makes sampling support more generic/simple by adding >  Datapath SAMPLING action. When sampling is turned on, a SAMPLING >  action is added to all flows for given bridge. >   This patch cleanup sampling upcall by striping it down

Re: [ovs-dev] [prelearning 5/5] ofp-util: Add type-safe functions for serializing actions.

Nifty patch. This has always been a pain point when implementing new actions. Glad to see it all generated now. Ethan On Tue, Aug 16, 2011 at 16:29, Ben Pfaff wrote: > --- >  lib/autopath.c  |    6 +-- >  lib/bundle.c    |    7 +-- >  lib/multipath.c |    6 +-- >  lib/nx-match.c  |   10 +

Re: [ovs-dev] [PATCH 1/1] vswitchd: PVID (Port VLAN ID) tagging feature

On Wed, Aug 17, 2011 at 6:33 AM, Philippe Jung wrote: > Le 03/08/2011 18:30, Ben Pfaff a écrit : >> On Sat, Jul 30, 2011 at 01:41:04AM +0200, Philippe Jung wrote: >>> This patchs adds the PVID (Port VLAN ID) tagging feature to the >>> switch ports: >>> - When an untagged packet is received by a sw

Re: [ovs-dev] [prelearning 4/5] ofp-parse: Refactor action parsing to improve compiler warnings.

Looks good. Ethan On Tue, Aug 16, 2011 at 16:29, Ben Pfaff wrote: > When a new action is added, compiler warnings show most of the places that > need new code to handle that action.  The action parsing code in > ofp-parse.c was the one remaining missing case.  This commit fixes that. > --- >  li

Re: [ovs-dev] [prelearning 3/5] ofp-util: Further abstract definitions of action properties.

Looks good, I would be inclined to s/false/0 and s/true/1 in ofp-util.def. That would allow us to align it against NXAST_RESUBMIT_TABLE without going over 79 characters. Doesn't really matter though. Ethan On Tue, Aug 16, 2011 at 16:29, Ben Pfaff wrote: > This commit primarily moves the OFPAT

Re: [ovs-dev] [prelearning 2/5] ofp-util: New functions for parsing and formatting OpenFlow port numbers.

Looks good, Ethan On Tue, Aug 16, 2011 at 16:29, Ben Pfaff wrote: > These functions were previously used only in ofp-parse.c and ofp-print.c, > but they are more generally useful and future commits will add more users. > --- >  lib/ofp-parse.c |   39 ++--- >  lib/ofp-

Re: [ovs-dev] [ovs-discuss] QoS in OVS compared to tc

> As a follow-up question to this, does OVS use tc at the backend to configure > queues? I ask this because lets say I want to configure queues using tc and > then somehow have them reflected in the OVS configuration database, so that > later I can map flows to these queues (e.g. using Openflow). I

Re: [ovs-dev] [prelearning 1/5] packets: Add more utility functions for IPv4 and IPv6 addresses.

I would be inclined to implement a "clz" function similar to what log_2_floor() does in util. We could use it both for log_2_floor() and ip_count_cidr_bits(). That way we would only have to deal with the ugly GNUC ifdefing in one place. Seems fine otherwise. Ethan On Tue, Aug 16, 2011 at 16:29

Re: [ovs-dev] [prelearning 0/5] cleanup prior to adding learning action

Thanks for sending this out. I'll get started reviewing it. Ethan On Tue, Aug 16, 2011 at 16:29, Ben Pfaff wrote: > I'm working on a new action that will allow a controller to do > more advanced learning than is currently possible with OFPP_NORMAL. > This series is a set of cleanups that I made

[ovs-dev] Prices further reduced - all optin lists

For more info and samples please reply to barbara.vods...@gmx.com $199 for all lists below ALL USA data, 100% optin, Excel format (separate file for each state), Unlimited use rights 1) Physicians (34 specialties): 788k records, 17k emails, 200k fax numbers 2) Chiropractors:108,421 to

[ovs-dev] [prelearning 4/5] ofp-parse: Refactor action parsing to improve compiler warnings.

When a new action is added, compiler warnings show most of the places that need new code to handle that action. The action parsing code in ofp-parse.c was the one remaining missing case. This commit fixes that. --- lib/ofp-parse.c | 351 +-- 1

[ovs-dev] [prelearning 5/5] ofp-util: Add type-safe functions for serializing actions.

--- lib/autopath.c |6 +-- lib/bundle.c|7 +-- lib/multipath.c |6 +-- lib/nx-match.c | 10 + lib/ofp-parse.c | 124 --- lib/ofp-util.c | 59 -- lib/ofp-util.h | 24 +++ 7 files chang

[ovs-dev] [prelearning 3/5] ofp-util: Further abstract definitions of action properties.

This commit primarily moves the OFPAT_ACTION and NXAST_ACTION invocations into a new file ofp-util.def. This allows multiple places in the source to use them. This commit also adds a new function ofputil_action_code_from_name(). The following commit will add the first user. --- lib/automake.mk

[ovs-dev] [prelearning 2/5] ofp-util: New functions for parsing and formatting OpenFlow port numbers.

These functions were previously used only in ofp-parse.c and ofp-print.c, but they are more generally useful and future commits will add more users. --- lib/ofp-parse.c | 39 ++--- lib/ofp-print.c | 57 - lib/ofp-util.

[ovs-dev] [prelearning 1/5] packets: Add more utility functions for IPv4 and IPv6 addresses.

We had these functions scattered around the source tree anyway. packets.h is a good place to centralize them. I do plan to introduce some additional callers. --- lib/classifier.c | 22 ++--- lib/ofp-util.c | 16 ++-- lib/packets.c| 65 +

[ovs-dev] [prelearning 0/5] cleanup prior to adding learning action

I'm working on a new action that will allow a controller to do more advanced learning than is currently possible with OFPP_NORMAL. This series is a set of cleanups that I made along the way. Ben Pfaff (5): packets: Add more utility functions for IPv4 and IPv6 addresses. ofp-util: New functions

Re: [ovs-dev] [PATCH 1/1] vswitchd: PVID (Port VLAN ID) tagging feature

Le 03/08/2011 18:30, Ben Pfaff a écrit : > On Sat, Jul 30, 2011 at 01:41:04AM +0200, Philippe Jung wrote: >> This patchs adds the PVID (Port VLAN ID) tagging feature to the >> switch ports: >> - When an untagged packet is received by a switch port from an >> external device, the packet is tagged

[ovs-dev] [PATCH 3/3] ofproto: Update 'struct facet''s comments.

Some of the comments in the definition of 'struct facet' had become out of date. --- ofproto/ofproto-dpif.c | 11 ++- 1 files changed, 2 insertions(+), 9 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index a181122..3fd95ea 100644 --- a/ofproto/ofproto-dpif.c

[ovs-dev] [PATCH 1/3] ofproto: Fix over accounting of byte counters.

The update_stats() function in ofproto was attributing more bytes to facets than they had actually accrued. This could potentially throw off bond load balancing. Found by inspection. --- ofproto/ofproto-dpif.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/ofproto/ofpr

[ovs-dev] [PATCH 2/3] ofproto: Remove extra_bytes parameter of facet_account().

It no longer has interesting users. --- ofproto/ofproto-dpif.c | 19 --- 1 files changed, 8 insertions(+), 11 deletions(-) diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofproto-dpif.c index 3aa0e78..a181122 100644 --- a/ofproto/ofproto-dpif.c +++ b/ofproto/ofproto-dpif.c @@ -272

[ovs-dev] [PATCH 0/3] Facet accounting and comment updates.

The first patch of this series fixes a bug I noticed involving facet_account(). The second patch in the series would have fixed it as well. I wrote them as two patches in case we want to backport it to 1.2. I don't have a strong opinion either way on that matter. Ethan Jackson (3): ofproto: F

Re: [ovs-dev] [PATCH] ofproto: Clear packet and byte counters on flow additions.

> I noticed that the comment on the 'accounted_bytes' member still > refers to an "account_cb" callback that doesn't exist anymore.  I > guess it should talk about facet_account() now. I'll merge this one and send out a separate patch. Ethan ___ dev mai

Re: [ovs-dev] [PATCH] ofproto: Clear packet and byte counters on flow additions.

On Tue, Aug 16, 2011 at 01:57:18PM -0700, Ethan Jackson wrote: > When a flow is added to the flow table, its packet and byte > counters should be reset. This patch efficiently approximates this > behavior. It also does some minor code refactoring. > > Bug #3183. Seems reasonable. Thanks. I no

[ovs-dev] [PATCH] ofproto: Clear packet and byte counters on flow additions.

When a flow is added to the flow table, its packet and byte counters should be reset. This patch efficiently approximates this behavior. It also does some minor code refactoring. Bug #3183. --- ofproto/ofproto-dpif.c | 24 +++- 1 files changed, 19 insertions(+), 5 deletion

[ovs-dev] (no subject)

Hello, Do You Need A Loan?If Yes Apply For Further Procedure. Thanks___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

Re: [ovs-dev] Error create module with kernel 3.0.x

On Tue, Aug 16, 2011 at 5:10 PM, Anthony Demard wrote: > Package: openvswitch-datapath-source > > Version: 1.1.1-1 > > Install OK --> aptitude install openvswitch-datapath-source > > But after install i launch the commad --> module-assistant auto-install > openvswitch-datapath > And error... >    

[ovs-dev] Error create module with kernel 3.0.x

Package: openvswitch-datapath-source Version: 1.1.1-1 Install OK --> aptitude install openvswitch-datapath-source But after install i launch the commad --> module-assistant auto-install openvswitch-datapath And error...     │ checking for Linux 2.6 build directory...