Re: [ovs-dev] [PATCH] datapath: Fix sparse warning on BUILD_BUG_ON_INVALID()

Thanks, pushed to the master. On Fri, Dec 20, 2013 at 5:34 PM, Pravin Shelar wrote: > After chatting with Andy offline, it looks good to me. > > Acked-by: Pravin B Shelar > > On Fri, Dec 20, 2013 at 5:19 PM, Pravin Shelar wrote: > > On Fri, Dec 20, 2013 at 4:25 PM, Andy Zhou wrote: > >> Spar

Re: [ovs-dev] [PATCH] datapath: Fix sparse warning on BUILD_BUG_ON_INVALID()

After chatting with Andy offline, it looks good to me. Acked-by: Pravin B Shelar On Fri, Dec 20, 2013 at 5:19 PM, Pravin Shelar wrote: > On Fri, Dec 20, 2013 at 4:25 PM, Andy Zhou wrote: >> Sparse gives the following warnings when compile against Linux kernel >> 3.5: >> >> CHECK /root/projs

Re: [ovs-dev] [PATCH] datapath: Fix sparse warning on BUILD_BUG_ON_INVALID()

On Fri, Dec 20, 2013 at 4:25 PM, Andy Zhou wrote: > Sparse gives the following warnings when compile against Linux kernel > 3.5: > > CHECK /root/projs/ovs/openvswitch/datapath/linux/skbuff-openvswitch.c > include/linux/mm.h:405:9: error: undefined identifier > 'BUILD_BUG_ON_INVALID' > includ

[ovs-dev] [PATCH] datapath: compat: Configure check GRE DEMUX.

RHEL6-openstack kernel has backported gre DEMUX module, Therefore add configure check to detect it. Signed-off-by: Pravin B Shelar Bug #21936 --- acinclude.m4|1 + datapath/linux/compat/gre.c |4 +- datapath/linux/compat/include/net/gre.h | 47 ++

Re: [ovs-dev] [PATCH] connmgr: Log when a packet-in is dropped due to queue overflow.

Looks good the me, On Thu, Dec 5, 2013 at 9:21 AM, Ben Pfaff wrote: > Reported-by: Anton Matsiuk > Signed-off-by: Ben Pfaff > --- > AUTHORS |1 + > ofproto/connmgr.c |9 +++-- > 2 files changed, 8 insertions(+), 2 deletions(-) > > diff --git a/AUTHORS b/AUTHORS > index

[ovs-dev] [PATCH] datapath: Fix sparse warning on BUILD_BUG_ON_INVALID()

Sparse gives the following warnings when compile against Linux kernel 3.5: CHECK /root/projs/ovs/openvswitch/datapath/linux/skbuff-openvswitch.c include/linux/mm.h:405:9: error: undefined identifier 'BUILD_BUG_ON_INVALID' include/linux/mm.h:405:9: error: not a function The same issues may

Re: [ovs-dev] [tcpdump 5/5] [RFC] Add support for capturing all OpenFlow traffic to a .pcap file.

Didn't review closely, I think this feature is very useful in debugging controller - ovs communication, The approach is reasonable and I cannot come up by other method that does not need header synthesis. Maybe others will have better idea. Thanks, Alex Wang, On Fri, Nov 22, 2013 at 1:37 PM,

Re: [ovs-dev] [PATCH] sFlow LAG and Tunnel export

On Fri, Dec 20, 2013 at 03:45:37PM -0800, Neil McKee wrote: > Sorry. Grappling with a new email client. Here it as an attachment, if > that's OK. Let me know if I need to rebase again. Git is happy with this version's formatting. I see a patch failure due to changes in OVS since you made the p

[ovs-dev] [PATCH] ofproto-dpif-monitor: Remove monitor_init().

Commit 881d47a9fa9 (monitor: Replace monitor_seq with periodic wakeup.) removes the global "struct seq" in ofproto-dpif-monitor module. This change makes the monitor_init() no longer needed. This commit removes the monitor_init() from ofproto-dpif-monitor.c. Signed-off-by: Alex Wang --- ofprot

Re: [ovs-dev] [PATCH V2 2/3] bfd: Send FINAL immediately after receiving POLL.

Alex, Thanks, I just pushed this, Jarno On Dec 20, 2013, at 12:47 PM, Alex Wang wrote: > > Thanks Jarno for the review, please see my explanation below, > > > On Fri, Dec 20, 2013 at 11:29 AM, Jarno Rajahalme > wrote: > mport. */ > > -next_wake_time = MIN(bfd_wake_time(mport->bf

Re: [ovs-dev] [tcpdump 4/5] ovs-ofctl: New command "ofp-parse-pcap" to dump OpenFlow from PCAP files.

Patch 2/5 - 4/5 all look good to me, Tested with some sample pcap, works good (including --timestamp), On Fri, Nov 22, 2013 at 1:37 PM, Ben Pfaff wrote: > Based on the number of people who ask about Wireshark support for OpenFlow, > this is likely to be widely useful. > > Signed-off-by: Ben Pf

Re: [ovs-dev] [PATCH v2] util: New wrapper function ovs_vsnprintf().

Sorry, wrong patch. Please ignore. From: Saurabh Shah mailto:ssaur...@nicira.com>> Date: Friday, December 20, 2013 2:58 PM To: "dev@openvswitch.org" mailto:dev@openvswitch.org>> Subject: [ovs-dev] [PATCH v2] util: New wrapper function ovs_vsnprintf(). So that vsnprint

[ovs-dev] [PATCH] util: New wrapper functions ovs_snprintf & ovs_vsnprintf.

From: Saurabh Shah So that snprintf() & vsnprintf() on windows have C99 like semantics. Signed-off-by: Saurabh Shah --- lib/command-line.c |4 ++-- lib/dynamic-string.c |4 ++-- lib/json.c |4 ++-- lib/match.c|2 +- lib/netdev-vport.c |2 +-

[ovs-dev] [PATCH v2] util: New wrapper function ovs_vsnprintf().

So that vsnprintf on windows has C99 like semantics. Signed-off-by: Saurabh Shah --- lib/command-line.c |2 +- lib/dynamic-string.c |4 ++-- lib/util.c | 18 -- lib/util.h |1 + 4 files changed, 20 insertions(+), 5 deletions(-) diff --git a/li

Re: [ovs-dev] [PATCH 2/2] datapath: compat: Add configure check for lockdep_rtnl_is_held()

On Fri, Dec 20, 2013 at 5:30 PM, Pravin B Shelar wrote: > RHEL6-openstack kernel has backported lockdep_rtnl_is_held(). > > Signed-off-by: Pravin B Shelar Acked-by: Jesse Gross ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/li

Re: [ovs-dev] [PATCH 1/2] datapath: compat: Fix skb_has_frag_list definition.

On Fri, Dec 20, 2013 at 5:30 PM, Pravin B Shelar wrote: > RHEL6-openstack kernel has already replaced skb_has_frags > with skb_has_frag_list(). > > Fix compilation error on RHEL6-openstack. > > Signed-off-by: Pravin B Shelar Acked-by: Jesse Gross ___

[ovs-dev] [PATCH 1/2] datapath: compat: Fix skb_has_frag_list definition.

RHEL6-openstack kernel has already replaced skb_has_frags with skb_has_frag_list(). Fix compilation error on RHEL6-openstack. Signed-off-by: Pravin B Shelar --- acinclude.m4 |1 + datapath/linux/compat/include/linux/skbuff.h |2 +- 2 files changed, 2 inse

[ovs-dev] [PATCH 2/2] datapath: compat: Add configure check for lockdep_rtnl_is_held()

RHEL6-openstack kernel has backported lockdep_rtnl_is_held(). Signed-off-by: Pravin B Shelar --- acinclude.m4|1 + datapath/linux/compat/include/linux/rtnetlink.h |2 +- 2 files changed, 2 insertions(+), 1 deletions(-) diff --git a/acinclude.m4 b/acin

Re: [ovs-dev] [PATCH v2] datapath: Check for backported netdev_features_t.

On Tue, Dec 17, 2013 at 10:22 AM, Jesse Gross wrote: > This is apparently used by CentOS 6.5. > > Reported-by: Phil Daws > Reported-by: Madko > Signed-off-by: Jesse Gross > --- > v2: Check for netdev_features_t in netdevice.h instead as this works > on all kernel versions. I applied this p

[ovs-dev] [PATCH] compiler.h: Update documentation

OVS_LOCKS_EXCLUDED doesn't exist. This should be OVS_EXCLUDED. Signed-off-by: Joe Stringer --- lib/compiler.h |6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/compiler.h b/lib/compiler.h index daca32d..e867d72 100644 --- a/lib/compiler.h +++ b/lib/compiler.h @@ -9

Re: [ovs-dev] [PATCH v2] lib/flow: Skip minimask value checks.

On Fri, Dec 20, 2013 at 11:45:55AM -0800, Jarno Rajahalme wrote: > > On Dec 20, 2013, at 8:41 AM, Ben Pfaff wrote: > > > I see that the commit as you pushed it accidentally includes a changelog > > in the commit message. That's not usual practice; careful. > > > > Oops, sorry. Is it still pos

Re: [ovs-dev] [PATCH V2 2/3] bfd: Send FINAL immediately after receiving POLL.

Thanks Jarno for the review, please see my explanation below, On Fri, Dec 20, 2013 at 11:29 AM, Jarno Rajahalme wrote: > mport. */ > > -next_wake_time = MIN(bfd_wake_time(mport->bfd), > cfm_wake_time(mport->cfm)); > > -heap_change(&monitor_heap, heap_max(&monitor_heap), > > +

Re: [ovs-dev] [PATCH] socket-util: Remove unused functions.

compiles and passed make check. Looks good. Acked-by: Andy Zhou On Mon, Dec 9, 2013 at 5:28 PM, Ben Pfaff wrote: > A Windows porter mentioned to me that these functions caused special > trouble in the Windows port. However, they are no longer used, so we > might as well remove them. > > Si

Re: [ovs-dev] [PATCH v2] lib/flow: Skip minimask value checks.

On Dec 20, 2013, at 8:41 AM, Ben Pfaff wrote: > I see that the commit as you pushed it accidentally includes a changelog > in the commit message. That's not usual practice; careful. > Oops, sorry. Is it still possible to remove it? Jarno > On Fri, Dec 20, 2013 at 08:41:02AM -0800, Jarno R

Re: [ovs-dev] [PATCH V2 2/3] bfd: Send FINAL immediately after receiving POLL.

Alex, I’m not familiar with the code, so I’ll ask you to explain some things below, Jarno On Dec 16, 2013, at 9:30 AM, Alex Wang wrote: > Commit 307464a11 (ofproto-dpif-monitor: Use heap to order the mport > wakeup time.) makes bfd only send packet at specified periodic instant. > This fails

Re: [ovs-dev] [PATCH V2 2/3] bfd: Send FINAL immediately after receiving POLL.

I have not reviewed the code, but I tested it once I run into the BFD decay test persistently failing on me. Before, I had to wait about 1.5 secs for the poll sequence to finish at around line 340 in tests/bfd.at, with this patch no such delay is necessary, so it seems to work as advertised.

Re: [ovs-dev] [PATCH v3] linux: Report supported user features to the kernel

On Thu, Dec 19, 2013 at 7:20 AM, Thomas Graf wrote: > Following commit (''netlink: Do not enforce alignment of last Netlink > attribute''), signal the ability to receive unaligned Netlink messages > to the datapath to enable utilization of zerocopy optimizations. > > Opening a datapath is now done

Re: [ovs-dev] [PATCH] ovs-check-dead-ifs: Flush buffer before calling execvp.

On Fri, Dec 20, 2013 at 09:34:36AM -0800, Gurucharan Shetty wrote: > According to Python documentation here for execvp: > http://docs.python.org/2/library/os.html > "The current process is replaced immediately. Open file objects > and descriptors are not flushed, so if there may be data buffered >

[ovs-dev] [PATCH] ovs-check-dead-ifs: Flush buffer before calling execvp.

According to Python documentation here for execvp: http://docs.python.org/2/library/os.html "The current process is replaced immediately. Open file objects and descriptors are not flushed, so if there may be data buffered on these open files, you should flush them using sys.stdout.flush() or os.fsy

Re: [ovs-dev] [PATCH] tests: Make some tests more robust.

On Fri, Dec 20, 2013 at 08:24:31AM -0800, Jarno Rajahalme wrote: > These tests break if OVS internal hash function is changed. Some of > this is due to dependency on the order in which elements are iterated > from hash maps, or the algorithm used is just dependent on the > specific hash values pro

Re: [ovs-dev] [PATCH v2] lib/flow: Skip minimask value checks.

I see that the commit as you pushed it accidentally includes a changelog in the commit message. That's not usual practice; careful. On Fri, Dec 20, 2013 at 08:41:02AM -0800, Jarno Rajahalme wrote: > Pushed, thanks for the review! > > Jarno > > On Dec 19, 2013, at 1:42 PM, Ben Pfaff wrote: >

Re: [ovs-dev] [PATCH] ofp-print: Print durations with at least three decimals.

On Thu, Dec 19, 2013 at 03:53:17PM -0800, Andy Zhou wrote: > On Thu, Dec 19, 2013 at 10:11 AM, Ben Pfaff wrote: > > > Occasionally I run a command like this: > > watch -n.1 ovs-ofctl dump-flows br0 > > to see how flows change over time. Until now, it has been more difficult > > than necessar

Re: [ovs-dev] [PATCH v2] lib/flow: Skip minimask value checks.

Pushed, thanks for the review! Jarno On Dec 19, 2013, at 1:42 PM, Ben Pfaff wrote: > On Thu, Dec 19, 2013 at 01:02:42PM -0800, Jarno Rajahalme wrote: >> We allow zero 'values' in a miniflow for it to have the same map >> as the corresponding minimask. Minimasks themselves never have >> zero

[ovs-dev] [PATCH] tests: Make some tests more robust.

These tests break if OVS internal hash function is changed. Some of this is due to dependency on the order in which elements are iterated from hash maps, or the algorithm used is just dependent on the specific hash values produced for specific inputs (groups). These changes make these test cases

Re: [ovs-dev] [PATCH 7/7] tests/learn.at: Workaround a race

On Fri, Dec 20, 2013 at 07:31:06PM +0900, YAMAMOTO Takashi wrote: > This test seems to assume that the switch completes > processing of the first packet before start processing > the second one. I don't see any code ensuring that. > Workaround the problem by giving 1 second for the upcall. > > Si

Re: [ovs-dev] [PATCH] .gitignore: add /libtool

On Fri, Dec 20, 2013 at 12:46:53PM +0200, Lorand Jakab wrote: > The ./configure script now generates a 'libtool' file in the top-level > directory. Add it to .gitignore > > Signed-off-by: Lorand Jakab Applied, thanks! ___ dev mailing list dev@openvswi

[ovs-dev] [PATCH] .gitignore: add /libtool

The ./configure script now generates a 'libtool' file in the top-level directory. Add it to .gitignore Signed-off-by: Lorand Jakab --- .gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/.gitignore b/.gitignore index d80fc63..d205227 100644 --- a/.gitignore +++ b/.gitignore @@ -40,6

[ovs-dev] [PATCH 3/7] tests/ofproto.at: Avoid stdout/stderr ordering assumptions

Stop assuming the order of outputs from separate streams. (stdout and stderr) Signed-off-by: YAMAMOTO Takashi --- tests/ofproto.at | 13 - 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/tests/ofproto.at b/tests/ofproto.at index c410d29..f6a62cd 100644 --- a/tests/ofpro

[ovs-dev] [PATCH 2/7] tests/ofproto-dpif.at: Portability improvement

The output of "wc -l" have leading spaces on some platforms. (NetBSD, OSX, ...) This fixes a test failure introduced by commit e79a6c83. ("ofproto: Handle flow installation and eviction in upcall.") Signed-off-by: YAMAMOTO Takashi --- tests/ofproto-dpif.at | 8 ++-- 1 file changed, 2 insert

[ovs-dev] [PATCH 7/7] tests/learn.at: Workaround a race

This test seems to assume that the switch completes processing of the first packet before start processing the second one. I don't see any code ensuring that. Workaround the problem by giving 1 second for the upcall. Signed-off-by: YAMAMOTO Takashi --- tests/learn.at | 3 +++ 1 file changed, 3

[ovs-dev] [PATCH 6/7] timeval: Workaround for threaded test failures

BFD tests have the code like the following. # wait for a while to stablize everything. for i in `seq 0 9`; do ovs-appctl time/warp 500; done They no longer work as intended because BFD code is run in a separate monitor thread these days. The loop merely "warp" the time by 5000. The moni

[ovs-dev] [PATCH 5/7] tests/ofproto-dpif.at: Workaround a race

This test seems to assume only the first packets in flows counted as 'miss'. I don't see any code ensuring that. The test would fail if the upcall handler for the flow doesn't run fast enough. Workaround the problem by giving 1 second for the miss upcall. Signed-off-by: YAMAMOTO Takashi --- te

[ovs-dev] [PATCH 4/7] ofproto-dpif-upcall: Reduce log level of "Spent unreasonably long" msg

This message can be caused by a time warp and make tests fail. The message was introduced by commit e79a6c83. ("ofproto: Handle flow installation and eviction in upcall.") Signed-off-by: YAMAMOTO Takashi --- ofproto/ofproto-dpif-upcall.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) di

[ovs-dev] [PATCH 1/7] tests/daemon-py.at: Skip if no python

Signed-off-by: YAMAMOTO Takashi --- tests/daemon-py.at | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/daemon-py.at b/tests/daemon-py.at index 9bc7810..cafa8df 100644 --- a/tests/daemon-py.at +++ b/tests/daemon-py.at @@ -208,6 +208,7 @@ fi AT_CLEANUP AT_SETUP([daemon --detach --moni

Re: [ovs-dev] [PATCH v2] rhel: Fix build failures because of libraries in /usr/lib.

On Fri, Dec 20, 2013 at 12:54 AM, Gurucharan Shetty wrote: >> >> Thanks for looking into this. Does that also work on x86_64? > Yes. The errors were seen in x86_64 machines. > >> The libraries would be installed to lib64 instead, thus not getting >> deleted and thus maybe triggering "unpackaged fi