[ovs-dev] [PATCH v2] datapath: Simplify flow mask cache delete.

Currently on mask delete OVS moves last mask to the deleted place to keep compact for per cpu cache. But this generate duplicate entries in mask cache array which results in multiple flow lookups in case we miss flow cache. Following patch simply sets NULL for deleted entry. Signed-off-by: Pravin

Re: [ovs-dev] [PATCH] datapath: Simplify flow mask cache delete.

On Wed, Aug 13, 2014 at 6:55 PM, Andy Zhou wrote: > On Wed, Aug 13, 2014 at 2:43 PM, Pravin B Shelar wrote: >> Currently on mask delete OVS moves last mask to the deleted >> place to keep compact for per cpu cache. But this generate >> duplicate entries in mask cache array which results in >> mul

[ovs-dev] [PATCH v9] Extend OVS IPFIX exporter to export tunnel headers

Extend IPFIX exporter to export tunnel headers when both input and output of the port. Add three other_config options in IPFIX table: enable-input-sampling, enable-output-sampling and enable-tunnel-sampling, to control whether sampling tunnel info, on which direction (input or output). Insert sampl

[ovs-dev] [PATCH 3/4] datapath-windows: add infrastructure for supporting netlink

In this change, we define netlink families and commands supported by the Windows datapath. Only the control family and one command is supported today to get the PID. We also implement the 3 ioctls: read, write and transact. This is bare minimum and can be used to implement the equivalent of a recv

[ovs-dev] [PATCH 4/4] netlink-socket.c: implement get pid support on Windows

To verify if the netlink support in the kernel works, I updated the netlink-socket.c code to get the PID for a given device descriptor. In the existing code, userspace sets the PID, which will not be unique across different processes. So, it is better for the kernel to generate the PID and give it

[ovs-dev] [PATCH 2/4] datapath-windows: Add OvsDatapath.[ch] and OVS_USE_NL_INTERFACE CPP

In this change we create peer to OvsIoctl.[ch] that would interface with userspace using the netlink interface. The new files are called OvsDatpath.[ch]. We are not deleting OvsIoctl.[ch] yet. OvsDatapath.[ch] provides bare minimum functionality to register a pseudo device for communication with us

[ovs-dev] [PATCH 1/4] Add extentions to the standard datapath interface

The datapath interface defined in odp-netlink.h needs some extensions that are platform dependent. Some examples are the name of the communication device on Windows and a set of commands that are specific to Windows. In this change we define a include/odp-netlink-ext.h to in turn include any platf

Re: [ovs-dev] [PATCH] datapath: Simplify flow mask cache delete.

On Wed, Aug 13, 2014 at 2:43 PM, Pravin B Shelar wrote: > Currently on mask delete OVS moves last mask to the deleted > place to keep compact for per cpu cache. But this generate > duplicate entries in mask cache array which results in > multiple flow lookups in case we miss flow cache. > Followin

Re: [ovs-dev] [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers

-Original Message- From: Pravin Shelar [mailto:pshe...@nicira.com] Sent: Thursday, August 14, 2014 8:37 AM To: Wenyu Zhang Cc: Ben Pfaff; Jesse Gross; Romain Lenglet; dev@openvswitch.org Subject: Re: [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers On Wed, Aug 13, 2014 at 5

Re: [ovs-dev] [PATCH V2] bfd: Allow users to set local/remote src/dst MAC address.

Acked-by: Ethan Jackson On Mon, Aug 11, 2014 at 6:34 PM, Alex Wang wrote: > This commit adds options for configuring the MAC addresses > in BFD state machine. Therein, the "bfd_local_src_mac" and > "bfd_local_dst_mac" configure the MAC address of sent BFD > packets. The "bfd_remote_dst_mac" c

[ovs-dev] [DPDK Upcalls v3 1/3] flow: Parse MPLS should return the actual number of labels.

This problem is uncovered by a future patch. Signed-off-by: Ethan Jackson Acked-by: Ben Pfaff --- lib/flow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/flow.c b/lib/flow.c index 2e5ca0a..29b331e 100644 --- a/lib/flow.c +++ b/lib/flow.c @@ -210,7 +210,7 @@ parse_mpls

[ovs-dev] [DPDK Upcalls v3 3/3] dpif-netdev: Streamline miss handling.

This patch avoids the relatively inefficient miss handling processes dictated by the dpif process, by calling into ofproto-dpif directly through a callback. Signed-off-by: Ethan Jackson --- lib/dpif-netdev.c | 293 +++--- lib/dpif-provider.h

[ovs-dev] [DPDK Upcalls v3 2/3] ofproto: Reorganize in preparation for direct dpdk upcalls.

This patch reorganizes ofproto-dpif in preparation for future patches which allow direct upcall processing from dpif-netdev. The main goals are to share as much code as possible between the dpif-linux and dpif-netdev upcall paths. Additionally, to avoid confusing the dpif-netdev fast path, the pa

Re: [ovs-dev] datapath-windows: Renaming files proposal

Hi Samuel, > >OvsNetlink.h -> Netlink.h >OvsPub.h -> Pub.h > >The following will be left intact: >OvsDpInterface.h >Because it is a userspace interaction file and it is a generated file. I think I tried to tell this earlier, so this will be a repeat. If you want to nuke ŒOvs¹ please remove it fro

Re: [ovs-dev] [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers

On Wed, Aug 13, 2014 at 5:25 PM, Wenyu Zhang wrote: >>> Signed-off-by: Wenyu Zhang >>> Acked-by: Romain Lenglet >> >> This patch renumbers OVS_TUNNEL_KEY_ATTR_OAM and >> OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS. Is that really OK? >> > It is not ok. > There are another fixes to kernel part which I was g

Re: [ovs-dev] [PATCH 3/4] ofpbuf: Add ofpbuf_rss()

On Wed, Aug 13, 2014 at 4:42 PM, Daniele Di Proietto wrote: > On 8/13/14, 3:28 PM, "Pravin Shelar" wrote: > >>On Wed, Aug 13, 2014 at 8:49 AM, Daniele Di Proietto >> wrote: >>> >>> >>> On 8/12/14, 5:42 PM, "Pravin Shelar" wrote: >>> On Tue, Jul 22, 2014 at 5:06 PM, Daniele Di Proietto w

Re: [ovs-dev] datapath-windows: Renaming files proposal

On Aug 13, 2014, at 5:32 PM, Samuel Ghinet mailto:sghi...@cloudbasesolutions.com>> wrote: I meant, tell me whether you prefer Eth.h or Ethernet.h I have a slight preference for Ethernet.h, but it's no big deal to me. hi Sam, Since you are making the change, I'll go with your preference. Etherne

Re: [ovs-dev] datapath-windows: Renaming files proposal

I meant, tell me whether you prefer Eth.h or Ethernet.h I have a slight preference for Ethernet.h, but it's no big deal to me. Sam From: Samuel Ghinet Sent: Thursday, August 14, 2014 3:21 AM To: Nithin Raju; Ankur Sharma; Saurabh Shah Cc: dev@openvswitch.org Subjec

Re: [ovs-dev] [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers

>> Signed-off-by: Wenyu Zhang >> Acked-by: Romain Lenglet > > This patch renumbers OVS_TUNNEL_KEY_ATTR_OAM and > OVS_TUNNEL_KEY_ATTR_GENEVE_OPTS. Is that really OK? > It is not ok. There are another fixes to kernel part which I was going to do before pushing this patch to master. Thanks, Prav

Re: [ovs-dev] datapath-windows: Renaming files proposal

On Aug 13, 2014, at 5:21 PM, Samuel Ghinet mailto:sghi...@cloudbasesolutions.com>> wrote: OvsEth.h -> Ethernet.h [TELL ME IF YOU PREFER "Eth.h"] I like Eth.h, just my opinion though. The following will be left intact: OvsDpInterface.h Because it is a userspace interaction file and it is a gen

Re: [ovs-dev] [DPDK Upcalls v2 3/3] dpif-netdev: Streamline miss handling.

> In ofproto-dpif-upcall.c, there are two blank lines above upcall_cb > (horrors!). I'm experimenting with a new form of avant-garde post-post-post-modern dadaist syntax formatting. Only those with the most discerning taste could possibly comprehend its momentous implications. > Did you test wit

Re: [ovs-dev] datapath-windows: Renaming files proposal

Okay, so I'm gonna make a patch which will nuke the Ovs prefix of the following files: OvsActions.c -> Actions.c OvsAtomic.h -> Atomic.h OvsBufferMgmt.c -> BufferMgmt.c OvsBufferMgmt.h -> BufferMgmt.h OvsChecksum.c -> Checksum.c OvsChecksum.h -> Checksum.h OvsDebug.c -> Debug.c OvsDebug.h -> Debu

Re: [ovs-dev] [PATCH 3/4] ofpbuf: Add ofpbuf_rss()

On 8/13/14, 3:28 PM, "Pravin Shelar" wrote: >On Wed, Aug 13, 2014 at 8:49 AM, Daniele Di Proietto > wrote: >> >> >> On 8/12/14, 5:42 PM, "Pravin Shelar" wrote: >> >>>On Tue, Jul 22, 2014 at 5:06 PM, Daniele Di Proietto >>> wrote: ofpbuf_rss() is introduced to get the RSS hash from the NIC.

Re: [ovs-dev] datapath-windows: Renaming files proposal

Sam and Alin are of the opinion that we should nuke 'Ovs' in all files except OvsDpInterface.h. As regards to removing 'Ovs' in OvsPub.h, I said the following to Sam. > However, I would remove "Ovs" from OvsPub.h and OvsNetlink.h. > Would that be ok to you? Totally fine! OvsPub.h is the

Re: [ovs-dev] datapath-windows: Renaming files proposal

I am ok with either approach as long as it is consistent. The current approach was to drop 'Ovs' prefix from OvsPub.h but keep it for OvsDpInterface.h, which is inconsistent and hence I suggested dropping Ovs- from OvsDpInterface.h. If you would like both files to have 'Ovs' prefix, that's fine

Re: [ovs-dev] datapath-windows: Renaming files proposal

Correct, i think userspace interaction headers should have Ovs prefix. Regards, Ankur From: Saurabh Shah Sent: Wednesday, August 13, 2014 4:28 PM To: Ankur Sharma; Samuel Ghinet; Nithin Raju Cc: dev@openvswitch.org Subject: Re: [ovs-dev] datapath-windows: Renaming

Re: [ovs-dev] datapath-windows: Renaming files proposal

If that’s the case, we should also keep the ‘Ovs’ prefix for OvsPub.h. :) Thanks! Saurabh From: Ankur Sharma mailto:ankursha...@vmware.com>> Date: Wednesday, August 13, 2014 at 4:20 PM To: Saurabh Shah mailto:ssaur...@vmware.com>>, Samuel Ghinet mailto:sghi...@cloudbasesolutions.com>>, Nithin R

Re: [ovs-dev] datapath-windows: Renaming files proposal

Hi, The reason we decided to keep Ovs with DpInterface was because this file defines the interface with ovs userspace, i.e it has a different semantics from other header files in ovsext. I would prefer keeping the Ovs prefix, but its fine if you think it should be renamed. Thanks. Regards, An

Re: [ovs-dev] [PATCH 1/2] netdev-dpdk: Move to DPDK 1.7.0

On Tue, Aug 12, 2014 at 9:15 PM, Flavio Leitner wrote: > > I have a couple comments not really related to this > specific patch. See below. > > On Tue, Aug 12, 2014 at 10:43:35AM -0700, Daniele Di Proietto wrote: >> With this commit we move our DPDK support to 1.7.0. >> DPDK binaries (starting wi

Re: [ovs-dev] datapath-windows: Renaming files proposal

+1 for dropping the 'Ovs' prefix for OvsDpInterface. It is odd to have just one file prefixed with Ovs. Thanks! Saurabh From: Samuel Ghinet mailto:sghi...@cloudbasesolutions.com>> Date: Wednesday, August 13, 2014 at 8:29 AM To: Nithin Raju mailto:nit...@vmware.com>> Cc: "dev@openvswitch.org

Re: [ovs-dev] [PATCH 3/4] ofpbuf: Add ofpbuf_rss()

On Wed, Aug 13, 2014 at 8:49 AM, Daniele Di Proietto wrote: > > > On 8/12/14, 5:42 PM, "Pravin Shelar" wrote: > >>On Tue, Jul 22, 2014 at 5:06 PM, Daniele Di Proietto >> wrote: >>> ofpbuf_rss() is introduced to get the RSS hash from the NIC. It works >>>only for >>> packets received by DPDK (othe

Re: [ovs-dev] [PATCHv2 1/2] datapath: Move key_attr_size() to flow_netlink.h.

On Tue, Jul 15, 2014 at 5:15 PM, Joe Stringer wrote: > Signed-off-by: Joe Stringer > --- > v2: First post. > > Side comment: OVS_KEY_ATTR_ETHERTYPE appears twice in this list. I guess > it's not harmful to reserve a bit of extra space, but does anyone know > why it's there twice? > --- > datapat

Re: [ovs-dev] [PATCHv2] dpif: Support flow_get in dpif_operate().

Great, thanks. On Thu, Aug 14, 2014 at 09:43:27AM +1200, Joe Stringer wrote: > Thanks for the review, I plan to push with the following changes (I noticed > that the other dpif-linux operations - PUT,DEL have redundant memsets, so > deleting them too): > > diff --git a/lib/dpif-linux.c b/lib/dpif

Re: [ovs-dev] [PATCHv2] dpif: Support flow_get in dpif_operate().

Thanks for the review, I plan to push with the following changes (I noticed that the other dpif-linux operations - PUT,DEL have redundant memsets, so deleting them too): diff --git a/lib/dpif-linux.c b/lib/dpif-linux.c index e767d9f..c4420f4 100644 --- a/lib/dpif-linux.c +++ b/lib/dpif-linux.c @@

[ovs-dev] [PATCH] datapath: Simplify flow mask cache delete.

Currently on mask delete OVS moves last mask to the deleted place to keep compact for per cpu cache. But this generate duplicate entries in mask cache array which results in multiple flow lookups in case we miss flow cache. Following patch simply sets NULL for deleted entry. Signed-off-by: Pravin

Re: [ovs-dev] [DPDK Upcalls v2 2/3] ofproto: Reorganize in preparation for direct dpdk upcalls.

On Wed, Aug 13, 2014 at 02:35:19PM -0700, Ethan Jackson wrote: > > I'm not sure of the value of the new 'put_actions' member. It only > > gets used in one place. That place can't compose_slow_path() itself? > > It did before this commit. > > So in the final patch, dpif-netdev will use the put_ac

Re: [ovs-dev] [DPDK Upcalls v2 2/3] ofproto: Reorganize in preparation for direct dpdk upcalls.

> In ofproto-dpif-upcall.c, struct upcall seriously needs comments on > the members. Some of them are baffling at first glance (put_actions, > userdata, vsp_adjusted). Done. > I'm not sure of the value of the new 'put_actions' member. It only > gets used in one place. That place can't compose_

Re: [ovs-dev] [PATCH] ovs-numa: Correct printf format specifiers.

Thanks, applied. On Wed, Aug 13, 2014 at 02:20:36PM -0700, Alex Wang wrote: > Thx for the fix, Acked-by: Alex Wang > > > On Wed, Aug 13, 2014 at 2:19 PM, Ben Pfaff wrote: > > > list_size() yields a size_t, not uint64_t, and these types are different on > > 32-bit targets. > > > > Signed-off-b

Re: [ovs-dev] [PATCH] ovs-numa: Correct printf format specifiers.

Thx for the fix, Acked-by: Alex Wang On Wed, Aug 13, 2014 at 2:19 PM, Ben Pfaff wrote: > list_size() yields a size_t, not uint64_t, and these types are different on > 32-bit targets. > > Signed-off-by: Ben Pfaff > --- > lib/ovs-numa.c |4 ++-- > 1 file changed, 2 insertions(+), 2 deletio

[ovs-dev] [PATCH] ovs-numa: Correct printf format specifiers.

list_size() yields a size_t, not uint64_t, and these types are different on 32-bit targets. Signed-off-by: Ben Pfaff --- lib/ovs-numa.c |4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lib/ovs-numa.c b/lib/ovs-numa.c index bcf9b68..44df40f 100644 --- a/lib/ovs-numa.c ++

Re: [ovs-dev] [PATCH] datapath-windows: check source port during tunnel Tx

On Tue, Aug 12, 2014 at 10:59:29PM -0700, Nithin Raju wrote: > In the Windows datapath, Tx tunneling functionality is implemented by checking > if the outport in the action is a tunnel port. If so, the packet is > encapsulated and sent out on the PIF bridge for as second flow lookup. > Basically, w

Re: [ovs-dev] [PATCH] datapath-windows: check source port during tunnel Tx

On Aug 13, 2014, at 1:07 PM, Alin Serdean wrote: > Very sporadic times. > > Will continue my testing and review tomorrow. > > Just wanted to let you know :). Thanks Alin for the review and validation. I also left a VIF <-> VIF VXLAN based ping overnight. Still going strong. -- Nithin

Re: [ovs-dev] [PATCH] datapath-windows: check source port during tunnel Tx

Hi Nithin, At first glance everything seems good. I even tested it out between a HyperV and KVM hypervisor setup. On HyperV: PS C:\package\binaries> .\ovsset.ps1 system@ovs-system: lookups: hit:34 missed:273 lost:0 flows: 0 masks: hit:0 total:0 hit/pkt:0.00 port 1

Re: [ovs-dev] [PATCH v2 4/6] dpif: Meter framework.

On Wed, Aug 13, 2014 at 12:43 PM, Pravin Shelar wrote: > On Wed, Aug 13, 2014 at 9:25 AM, Ben Pfaff wrote: >> On Mon, Aug 11, 2014 at 02:37:45PM -0700, Jarno Rajahalme wrote: >>> Add DPIF-level infrastructure for meters. Allow meter_set to modify >>> the meter configuration (e.g. set the burst s

Re: [ovs-dev] [PATCH] ovs-numa: Add ovs-numa.{c, h} for extracting and storing cpu socket and cpu core info.

On Wed, Aug 13, 2014 at 12:42 PM, Ben Pfaff wrote: > Acked-by: Ben Pfaff > Thx, applied to master! ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

Re: [ovs-dev] [PATCH v2 4/6] dpif: Meter framework.

On Wed, Aug 13, 2014 at 9:25 AM, Ben Pfaff wrote: > On Mon, Aug 11, 2014 at 02:37:45PM -0700, Jarno Rajahalme wrote: >> Add DPIF-level infrastructure for meters. Allow meter_set to modify >> the meter configuration (e.g. set the burst size if unspecified). >> >> Signed-off-by: Jarno Rajahalme >

Re: [ovs-dev] [PATCH] ovs-numa: Add ovs-numa.{c, h} for extracting and storing cpu socket and cpu core info.

On Wed, Aug 13, 2014 at 12:04:19PM -0700, Alex Wang wrote: > Signed-off-by: Alex Wang Acked-by: Ben Pfaff ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

[ovs-dev] [PATCH] ovs-numa: Add ovs-numa.{c, h} for extracting and storing cpu socket and cpu core info.

Signed-off-by: Alex Wang --- Changes since first post: - use readdir() for the buffer safety. - only compile ovs-numa.c in linux. - make ovs_numa_get_n_sockets/cores() return *UNSPEC if there is no sockets or cores extracted. - refine code based on Ben's review. - Address review comments from T

Re: [ovs-dev] [dpdk patch 1/8] ovs-numa: Add ovs-numa.{c, h} for extracting and storing cpu socket and cpu core info.

Thanks Ben, for the review, > > + > > +/* To avoid sparse warning. */ > > +#ifdef __linux__ > ... > > +#endif /* __linux__ */ > > I didn't see the connection to sparse at first. I'm still not sure > there is one. Maybe a better comment: > /* On non-Linux, these functions are defined inline in

Re: [ovs-dev] [PATCH] Prepare ground for extensions to sFlow export

I rebased this with a "git pull --rebase upstream master; git push --force-with-lease", and made a 1-line correction to avoid a compiler warning. "make check" still succeeds. So this is the picture now: https://github.com/sflow/ovs/compare/openvswitch:master...master I'm hoping this can go in.

Re: [ovs-dev] [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers

On Fri, Aug 08, 2014 at 04:02:24AM -0700, Wenyu Zhang wrote: > Extend IPFIX exporter to export tunnel headers when both input and output > of the port. > Add three other_config options in IPFIX table: enable-input-sampling, > enable-output-sampling and enable-tunnel-sampling, to control whether > s

Re: [ovs-dev] [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers

On Wed, Aug 13, 2014 at 10:01 AM, Ben Pfaff wrote: > On Fri, Aug 08, 2014 at 04:02:24AM -0700, Wenyu Zhang wrote: >> Extend IPFIX exporter to export tunnel headers when both input and output >> of the port. >> Add three other_config options in IPFIX table: enable-input-sampling, >> enable-output-s

Re: [ovs-dev] [PATCH v8] Extend OVS IPFIX exporter to export tunnel headers

On Fri, Aug 08, 2014 at 04:02:24AM -0700, Wenyu Zhang wrote: > Extend IPFIX exporter to export tunnel headers when both input and output > of the port. > Add three other_config options in IPFIX table: enable-input-sampling, > enable-output-sampling and enable-tunnel-sampling, to control whether > s

Re: [ovs-dev] [PATCH] ofp-actions: Treat OFPACT_REG_MOVE with reg/metadata src as set action

On Wed, Aug 13, 2014 at 03:34:27PM +0200, Thomas Graf wrote: > This makes the following work: > add-group br0 group_id=1234,type=all, \ > bucket=output:10,move:NXM_NX_REG1[]->NXM_OF_IP_SRC[], \ > bucket=output:11 > add-flow br0 ip actions=load:0x->NXM_N

Re: [ovs-dev] [PATCHv2] dpif: Support flow_get in dpif_operate().

On Wed, Aug 13, 2014 at 12:53:31PM +1200, Joe Stringer wrote: > This cleans up the dpif interface to make it more consistent with the > other dpif operations, and allows flows to be fetched in batches. > > Signed-off-by: Joe Stringer Thanks. I like it. I'd move this memset() in the error case

Re: [ovs-dev] [dpdk patch 1/8] ovs-numa: Add ovs-numa.{c, h} for extracting and storing cpu socket and cpu core info.

On Mon, Aug 11, 2014 at 09:56:36PM -0700, Alex Wang wrote: > Signed-off-by: Alex Wang > + > +/* To avoid sparse warning. */ > +#ifdef __linux__ ... > +#endif /* __linux__ */ I didn't see the connection to sparse at first. I'm still not sure there is one. Maybe a better comment: /* On non-Linux,

Re: [ovs-dev] [PATCH v2 5/6] lib/odp, ofproto xlate: Meter execution support.

On Mon, Aug 11, 2014 at 02:37:46PM -0700, Jarno Rajahalme wrote: > Meter action can drop or modify packets, so the execution framework is > changed to allow for this. Also, a meter action can appear alone > (e.g., to measure traffic that is to be dropped), so the existing drop > implementation is

Re: [ovs-dev] [PATCH v2 4/6] dpif: Meter framework.

On Mon, Aug 11, 2014 at 02:37:45PM -0700, Jarno Rajahalme wrote: > Add DPIF-level infrastructure for meters. Allow meter_set to modify > the meter configuration (e.g. set the burst size if unspecified). > > Signed-off-by: Jarno Rajahalme The dpif_meter_del() convention of using stats->n_bands f

Re: [ovs-dev] [PATCH 2/4] Prepare include headers

On Wed, Aug 13, 2014 at 03:16:07PM +, Alin Serdean wrote: > Preparing the include headers needed to compile dpif-linux.c with MSVC. > > Signed-off-by: Alin Gabriel Serdean Applied, thanks. ___ dev mailing list dev@openvswitch.org http://openvswitch

Re: [ovs-dev] [PATCH] datapath-windows: Create a NBL for each NB when required

Thanks! From: Nithin Raju [nit...@vmware.com] Sent: Wednesday, August 13, 2014 7:04 PM To: Samuel Ghinet Cc: dev@openvswitch.org Subject: Re: [ovs-dev] [PATCH] datapath-windows: Create a NBL for each NB when required Sure. I'll take a look. There may be s

Re: [ovs-dev] [PATCH] datapath-windows: Create a NBL for each NB when required

Sure. I'll take a look. There may be some complications with the existing code if we process an NBL with multiple NBs through tunneling. Maybe your changes fixes some of it. -- Nithin On Aug 13, 2014, at 9:00 AM, Samuel Ghinet wrote: > Hello Nithin, > > I did only a very basic test for thi

Re: [ovs-dev] [PATCH] datapath-windows: Create a NBL for each NB when required

Hello Nithin, I did only a very basic test for this patch. If it's ok with you, could you please take a look on it, as a general overview? I could confirm on the tunneling part in the next 24h. Sam. From: Nithin Raju [nit...@vmware.com] Sent: Wednesday, A

Re: [ovs-dev] [PATCH 1/4] No newline at end of file

On Wed, Aug 13, 2014 at 03:15:21PM +, Alin Serdean wrote: > Adding newline at end of the following files: packet.h, uio.h. > > Signed-off-by: Alin Gabriel Serdean This was obviously correct so I applied it. ___ dev mailing list dev@openvswitch.org

Re: [ovs-dev] [PATCH 1/2] netdev-dpdk: Move to DPDK 1.7.0

On 8/12/14, 5:39 PM, "Pravin Shelar" wrote: >On Tue, Aug 12, 2014 at 10:43 AM, Daniele Di Proietto > wrote: >> With this commit we move our DPDK support to 1.7.0. >> DPDK binaries (starting with dpdk 1.7.0) should be linked with >>--whole-archive >> to include pmd drivers >> >I updated INSTALL.

Re: [ovs-dev] Agenda for IRC neeting for 8/13

Hi Nithin, Thank you for the feedback! We would like to add the following topics: 5. Queue packets inside the kernel and only send information needed to the userspace, instead of sending the whole packet. 6. Spooky hash. 7. Coding styles. 8. Packet buffer management. 9. Fixed sized array 10. Ref

Re: [ovs-dev] [PATCH 3/4] ofpbuf: Add ofpbuf_rss()

On 8/12/14, 5:42 PM, "Pravin Shelar" wrote: >On Tue, Jul 22, 2014 at 5:06 PM, Daniele Di Proietto > wrote: >> ofpbuf_rss() is introduced to get the RSS hash from the NIC. It works >>only for >> packets received by DPDK (otherwise it returns 0, which tells the >>caller to >> compute the hash in

Re: [ovs-dev] [PATCH 2/4] dpif-netdev: Reintroduce ref_cnt for dp_netdev_flow

On 8/12/14, 5:41 PM, "Pravin Shelar" wrote: >On Tue, Jul 22, 2014 at 5:06 PM, Daniele Di Proietto > wrote: >> struct dp_netdev_flow used to have a reference counter. It has been >>replaced by >> RCU. Unfortunately RCU is not enough if we plan to hold a reference to >>the >> dp_netdev_flow for a l

Re: [ovs-dev] [PATCH 2/2] netdev-dpdk: Keep calling rte_eth_tx_burst() until it returns 0

On 8/12/14, 5:40 PM, "Pravin Shelar" wrote: >On Tue, Aug 12, 2014 at 10:43 AM, Daniele Di Proietto > wrote: >> rte_eth_tx_burst() _should_ transmit every packet that it is passed >>unless the >> queue is full. Nontheless some implementation of rte_eth_tx_burst (e.g. >> ixgbe_xmit_pkts_vec()) do

Re: [ovs-dev] datapath-windows: Renaming files proposal

+1 -Mesaj original- De la: dev [mailto:dev-boun...@openvswitch.org] În numele Nithin Raju Trimis: Wednesday, August 13, 2014 6:33 PM Către: Samuel Ghinet Cc: dev@openvswitch.org Subiect: Re: [ovs-dev] datapath-windows: Renaming files proposal On Aug 13, 2014, at 8:29 AM, Samuel Ghinet w

Re: [ovs-dev] Design notes for provisioning Netlink interface from the OVS Windows driver (Switch extension)

Nithin, It might be a bit of misunderstanding on what one or another is trying to say :) I meant concurrency such as: thread1: flow dump start thread1: flow dump next thread2: datapath dump start thread2: datapath dump next thread1: flow dump next This would require, as I see, the buffer to be as

Re: [ovs-dev] datapath-windows: Renaming files proposal

On Aug 13, 2014, at 8:29 AM, Samuel Ghinet wrote: > However, I would remove "Ovs" from OvsPub.h and OvsNetlink.h. > Would that be ok to you? Totally fine! OvsPub.h is the legacy interface and will be nuked as soon as the netlink based interface is ready. OvsNetlink.h will be moved back within

Re: [ovs-dev] datapath-windows: Renaming files proposal

Nithin, Reasonable. I was actually thinking that if I want to remove Ovs-prefixes from all files except this one, some people may not agree with this small inconsistency :) However, I would remove "Ovs" from OvsPub.h and OvsNetlink.h. Would that be ok to you? Sam __

Re: [ovs-dev] [PATCH] datapath-windows: check source port during tunnel Tx

On Aug 13, 2014, at 8:24 AM, Alin Serdean wrote: > I will take a look over the patch after the meeting. Sounds good. thanks, Nithin ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

Re: [ovs-dev] Agenda for IRC neeting for 8/13

Sam/Alin, I saw a bunch of reviews go out this morning. This is great! Ben was mentioning that as much as possible, we should keep review comments on ovs-dev ML. We can certainly discuss stuff about reviews that are better discussed in an interactive medium, but otherwise, I plan to give out com

Re: [ovs-dev] [PATCH] datapath-windows: check source port during tunnel Tx

Hey Nithin, I will take a look over the patch after the meeting. Thanks, Alin. -Mesaj original- De la: dev [mailto:dev-boun...@openvswitch.org] În numele Nithin Raju Trimis: Wednesday, August 13, 2014 8:59 AM Către: dev@openvswitch.org; ssaur...@vmware.com Subiect: [ovs-dev] [PATCH] data

[ovs-dev] [PATCH 4/4] Add files to MSVC automake build

Add dpif-linux.c/h to the lib/automake.mk Signed-off-by: Alin Gabriel Serdean --- lib/automake.mk | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/automake.mk b/lib/automake.mk index e95191d..a561bba 100644 --- a/lib/automake.mk +++ b/lib/automake.mk @@ -324,6 +324,8 @@ endif if WIN32 li

[ovs-dev] [PATCH 2/4] Prepare include headers

Preparing the include headers needed to compile dpif-linux.c with MSVC. Signed-off-by: Alin Gabriel Serdean --- include/windows/automake.mk | 3 +++ include/windows/linux/pkt_sched.h | 26 ++ include/windows/linux/types.h | 0 include/windows/sys/epoll.h

[ovs-dev] [PATCH 3/4] Changes needed to compile dpif-linux.c with MSVC

The following changes are just a start to begin a discussion. Bypass all epoll functionality. IOCP work the other way around than epoll: you get if the operation finished for a HANDLE in contrast of getting the number of file descriptors that are ready for an operation. It also adds a lot of over

[ovs-dev] [PATCH 1/4] No newline at end of file

Adding newline at end of the following files: packet.h, uio.h. Signed-off-by: Alin Gabriel Serdean --- include/windows/netpacket/packet.h | 2 +- include/windows/sys/uio.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/include/windows/netpacket/packet.h b/includ

Re: [ovs-dev] datapath-windows: Renaming files proposal

Sam, OvsDpInterface.h is basically the OVS datapath's interface. It is being generated in a separate directory - sort of public for the datapath - to be sure. I'm inclined to leave the Ovs keyword alone for this 'OVS interface' file. Is that reasonable? thanks, Nithin On Aug 13, 2014, at 8:11

[ovs-dev] [PATCH] Use local time for Inf2Cat

Issue: cannot build properly when local date differs from UTC date. Since Stampinf stamps the driver inf file with the local date set the Inf2Cat to use local date as well. Signed-off-by: Alin Gabriel Serdean --- datapath-windows/Package/package.VcxProj | 20 +++- 1 file changed

[ovs-dev] [PATCH] Add build of ovsext.sln using MSBuild

This commit adds to the automake build system the full build required by the forwarding extension solution. It will help a lot in the future CI to check the full build of the project. Also the documentation was updated. Signed-off-by: Alin Gabriel Serdean --- INSTALL.Windows | 34 +

[ovs-dev] datapath-windows: Renaming files proposal

Hello guys, Just to make sure that everyone agrees to this. I want to add a commit in which to rename all files, by removing the prefix Ovs. E.g. OvsActions.c -> Actions.c Since there is no use in having the prefix "Ovs", and removing it may improve clarity. (there was a discussion on this on th

Re: [ovs-dev] Design notes for provisioning Netlink interface from the OVS Windows driver (Switch extension)

On Aug 13, 2014, at 7:05 AM, Samuel Ghinet wrote: > [QUOTE]The driver does not care about multiple threads concurrently executing > dump as any DeviceIOContol will be self-contained. Also, the user mode > implementation guarantees that there are no simultaneous dumps for a given > socket.[/

Re: [ovs-dev] [PATCH] datapath-windows: Create a NBL for each NB when required

Samuel, Were you able to test cases where the multiple NBL case originated from a VM, and was subject to tunneling? I'll look at your patch. thanks, Nithin On Aug 13, 2014, at 7:25 AM, Samuel Ghinet wrote: > Create a NBL for each NB when required > > ovs/ovs-issues#15 > > All NET_BUFFERs o

[ovs-dev] [PATCH] datapath-windows: Create a NBL for each NB when required

Create a NBL for each NB when required ovs/ovs-issues#15 All NET_BUFFERs of a NET_BUFFER_LIST must go through the pipeline: extract, find flow, execute. Previously, only the first NET_BUFFER of a NET_BUFFER_LIST was going through this pipeline, which was erroneous. OvsPartialCopyToMultipleNBLs i

Re: [ovs-dev] Design notes for provisioning Netlink interface from the OVS Windows driver (Switch extension)

[QUOTE]The driver does not care about multiple threads concurrently executing dump as any DeviceIOContol will be self-contained. Also, the user mode implementation guarantees that there are no simultaneous dumps for a given socket.[/QUOTE] Regarding "DeviceIOContol will be self-contained", I th

Re: [ovs-dev] [PATCH 04/15] datapath-windows: We don't need wrappers for Interlocked ops

Perhaps a neater solution would be to simply have e.g. the OVS_BUFFER_CONTEXT's refCount declared as volatile and ULONG. If the variable is meant to be used (read & written) using atomic operations, we could better declare it volatile. Also, even though sizeof(ULONG) == sizeof(UINT32) and sizeof

[ovs-dev] [PATCH] ofp-actions: Treat OFPACT_REG_MOVE with reg/metadata src as set action

This makes the following work: add-group br0 group_id=1234,type=all, \ bucket=output:10,move:NXM_NX_REG1[]->NXM_OF_IP_SRC[], \ bucket=output:11 add-flow br0 ip actions=load:0x->NXM_NX_REG1[],group:1234 Signed-off-by: Thomas Graf --- lib/ofp-actions

[ovs-dev] Terms and Conditions

I am a former law firm Partner and my legal services business, Contracts 4 Commerce, provides companies with their commercial contracts for a low fixed fee.   I am writing to inform you that we have produced new Terms and Conditions which are for use specifically by businesses in your market s