From: gyang
With megaflow support, there is API to convert mask to nlattr key based
format. This change introduces API to do the reverse conversion. We
leverage the existing odp_flow_key_to_flow() API to resue the code.
Signed-off-by: Guolin Yang
---
lib/odp-util.c | 295 +
fix a bug introduced by commit e1ec7dd4608876283038c417065c4f9978255fa3.
(ofproto-dpif: Implement multi-threaded miss handling.)
Signed-off-by: YAMAMOTO Takashi
---
ofproto/ofproto-dpif-upcall.c | 41 ++---
1 file changed, 22 insertions(+), 19 deletions(-)
di
Reported-by: Ben Pfaff
Signed-off-by: Ethan Jackson
---
ofproto/ofproto-dpif-xlate.c |5 +
1 file changed, 5 insertions(+)
diff --git a/ofproto/ofproto-dpif-xlate.c b/ofproto/ofproto-dpif-xlate.c
index 8be8088..a7f051d 100644
--- a/ofproto/ofproto-dpif-xlate.c
+++ b/ofproto/ofproto-dpif
I'll have a look thanks.
Ethan
On Fri, Aug 16, 2013 at 9:23 AM, Ben Pfaff wrote:
> valgrind reports a couple of memory leaks for current master:
>
> ==25376== 84 bytes in 1 blocks are definitely lost in loss record 163 of
> 175
> ==25376==at 0x4D7B308: malloc (vg_replace_malloc.c:26
valgrind reports a couple of memory leaks for current master:
==25376== 84 bytes in 1 blocks are definitely lost in loss record 163 of 175
==25376==at 0x4D7B308: malloc (vg_replace_malloc.c:263)
==25376==by 0x80E2364: xmalloc (util.c:103)
==25376==by 0x8096244: mac_lear
Thanks, applied.
On Fri, Aug 16, 2013 at 08:51:18AM +0800, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
> Thanks a lot.
>
> On Fri, Aug 16, 2013 at 8:39 AM, Ben Pfaff wrote:
> > The OpenFlow OFPFF_CHECK_OVERLAP flag requires us to check whether the flow
> > being inserted overlaps with any
Thanks, applied.
On Fri, Aug 16, 2013 at 08:58:40AM +0800, Ethan Jackson wrote:
> Acked-by: Ethan Jackson
>
> On Fri, Aug 16, 2013 at 8:55 AM, Ben Pfaff wrote:
> > The 'pins' list contains packet-ins, not flow-mods.
> >
> > Introduced by commit ada3a58d1f8 (ofproto-dpif: Make packet_ins thread
Acked-by: Ethan Jackson
On Fri, Aug 16, 2013 at 8:55 AM, Ben Pfaff wrote:
> The 'pins' list contains packet-ins, not flow-mods.
>
> Introduced by commit ada3a58d1f8 (ofproto-dpif: Make packet_ins thread
> safe.).
>
> Signed-off-by: Ben Pfaff
> ---
> ofproto/ofproto-dpif.c |4 ++--
> 1 file
The 'pins' list contains packet-ins, not flow-mods.
Introduced by commit ada3a58d1f8 (ofproto-dpif: Make packet_ins thread
safe.).
Signed-off-by: Ben Pfaff
---
ofproto/ofproto-dpif.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/ofproto/ofproto-dpif.c b/ofproto/ofpr
Acked-by: Ethan Jackson
Thanks a lot.
On Fri, Aug 16, 2013 at 8:39 AM, Ben Pfaff wrote:
> The OpenFlow OFPFF_CHECK_OVERLAP flag requires us to check whether the flow
> being inserted overlaps with any existing flows. That isn't efficiently
> implemented and typically requires us to compare the
Thanks for rebasing.
On Wed, Aug 14, 2013 at 5:04 PM, Alex Wang wrote:
> This commit changes the "ovs-appclt coverage/show" command to show the
> per-second, per-minute and per-hour rates of function invocation. More
> importantly, this makes using coverage counter an easy way to monitor
> the
The OpenFlow OFPFF_CHECK_OVERLAP flag requires us to check whether the flow
being inserted overlaps with any existing flows. That isn't efficiently
implemented and typically requires us to compare the new flow against most
or all of the existing flows. We don't have to do that work if
OFPFF_CHECK
On Fri, Aug 16, 2013 at 08:09:02AM +0900, YAMAMOTO Takashi wrote:
> > On Fri, Aug 16, 2013 at 07:40:20AM +0900, YAMAMOTO Takashi wrote:
> >> is there any reason why ovs-appctl bridge/dump-flows reports
> >> "priority" twice for each flows?
> >
> > There are two possibilities. I have not looked to
Applied, thanks.
On Thu, Aug 15, 2013 at 04:05:15PM -0700, Alex Wang wrote:
> This looks good, thanks Ben,
>
>
> On Thu, Aug 15, 2013 at 3:49 PM, Ben Pfaff wrote:
>
> > I think that the documentation can be improved. What do you think of
> > this incremental?
> >
> > diff --git a/vswitchd/vsw
> On Fri, Aug 16, 2013 at 07:40:20AM +0900, YAMAMOTO Takashi wrote:
>> is there any reason why ovs-appctl bridge/dump-flows reports
>> "priority" twice for each flows?
>
> There are two possibilities. I have not looked to see which is the
> case.
>
> One possibility is that this is a bug. If so
This looks good, thanks Ben,
On Thu, Aug 15, 2013 at 3:49 PM, Ben Pfaff wrote:
> I think that the documentation can be improved. What do you think of
> this incremental?
>
> diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
> index 8c7d082..5bbe943 100644
> --- a/vswitchd/vswitch.xml
>
I think that the documentation can be improved. What do you think of
this incremental?
diff --git a/vswitchd/vswitch.xml b/vswitchd/vswitch.xml
index 8c7d082..5bbe943 100644
--- a/vswitchd/vswitch.xml
+++ b/vswitchd/vswitch.xml
@@ -160,16 +160,18 @@
+ type='{"type":
On Fri, Aug 16, 2013 at 07:40:20AM +0900, YAMAMOTO Takashi wrote:
> is there any reason why ovs-appctl bridge/dump-flows reports
> "priority" twice for each flows?
There are two possibilities. I have not looked to see which is the
case.
One possibility is that this is a bug. If so, then we shou
On Fri, Aug 16, 2013 at 07:25:18AM +0900, YAMAMOTO Takashi wrote:
> > On Thu, Aug 15, 2013 at 08:25:44PM +0900, YAMAMOTO Takashi wrote:
> >> move per-backer wait calls (dpif_wait/udpif_wait) from ofproto_wait
> >> to ofproto_type_wait.
> >>
> >> this fixes O(N^2) behaviour (eg. poll slot consumpti
is there any reason why ovs-appctl bridge/dump-flows reports
"priority" twice for each flows?
YAMAMOTO Takashi
duration=45786s, priority=32768, n_packets=0, n_bytes=0,
dl_dst=30:31:3a:38:30:3a,dl_type=0x88cc,actions=CONTROLLER:51
table_id=254, duration=45787s, priority=0, n_packets=0, n_bytes=0,
> On Thu, Aug 15, 2013 at 08:25:44PM +0900, YAMAMOTO Takashi wrote:
>> move per-backer wait calls (dpif_wait/udpif_wait) from ofproto_wait
>> to ofproto_type_wait.
>>
>> this fixes O(N^2) behaviour (eg. poll slot consumption) for
>> dpif-netdev where N is the number of bridges.
>>
>> Signed-off-b
> There are two solutions I can think of:
> 1. use "(2000 / bfd->cfg_min_rx + 10)" to compensate for the jitter.
Sure I think this is fine as it gives us a little bit of wiggle room.
Seems a bit cleaner than option (2).
Ethan
>
>
> 2. if I can share the "push_timer" in ofproto-dpif.c with bfd.c
On Thu, Aug 15, 2013 at 08:25:44PM +0900, YAMAMOTO Takashi wrote:
> move per-backer wait calls (dpif_wait/udpif_wait) from ofproto_wait
> to ofproto_type_wait.
>
> this fixes O(N^2) behaviour (eg. poll slot consumption) for
> dpif-netdev where N is the number of bridges.
>
> Signed-off-by: YAMAMO
Only parse the encap key field if eth_type is 802.1Q and
VLAN_TAG_PRESENT bit is set. Add a few more eror checks and logs.
Signed-off-by: Andy Zhou
---
datapath/flow.c | 66 ---
1 file changed, 44 insertions(+), 22 deletions(-)
diff --git a/
When there is no incoming data traffic at the interface for a period,
BFD decay allows the bfd session to increase the min_rx. This is
helpful in that some interfaces may be idle for long a time. And cpu
consumption can be reduced by processing fewer bfd control packets.
Signed-off-by: Alex Wang
On Wed, Aug 14, 2013 at 7:21 PM, Ethan Jackson wrote:
> > - do not allow bfd to call bfd_may_decay when decay_min_rx
> > is less than rmt_min_tx.
>
> Why? I don't think we should need this. If the rmt_min_tx is greater
> than our min_rx, according to the bfd protocol we'll simply use the
> rm
Thanks guys. I pushed patch to master
On Thu, Aug 15, 2013 at 12:49 PM, Thomas Graf wrote:
> On Thu, Aug 15, 2013 at 12:33:06PM -0700, Chris Wright wrote:
> > Andrei Andone reported an oops on CentOS 6.4 when adding a device to an
> > ovs instance. The problem is easy to reproduce and generate
On Wed, Aug 14, 2013 at 11:24:38AM -0700, Ben Pfaff wrote:
> A little work with gdb shows that 501 failed due to misaligned data
> access in a "packet out" OpenFlow command in packet_set_ipv4(). It
> looks like handle_packet_out() in ofproto.c needs to properly align
> the L3 header. It seems lik
Memory access tends to be faster when data is properly aligned.
Signed-off-by: Ben Pfaff
---
lib/packets.c | 16 +++-
ofproto/ofproto-provider.h |1 +
ofproto/ofproto.c |4 ++--
ofproto/pktbuf.c |9 -
4 files changed, 18 insertion
This fixes the same problem for IPv6 headers treated for other headers in
the previous commit.
Signed-off-by: Ben Pfaff
---
lib/csum.c|7 ---
lib/csum.h|2 +-
lib/flow.c| 10 +-
lib/packets.c | 32 ++--
lib/packets.h | 32
This series fixes bus errors on RISC machines due to misaligned
data, e.g. see this sparc build:
https://buildd.debian.org/status/fetch.php?pkg=openvswitch&arch=sparc&ver=1.9.2%2Bgit20130813-1&stamp=1376471499
See also http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=719736
I intend to b
At one time, the ofproto-dpif implementation of the 'rule_execute' member
function required, indirectly, at least struct(ofp10_packet_in) bytes of
headroom in the packet passed into it. (This allowed constructing an
OFPT_PACKET_IN without allocating and copying a new ofpbuf.) This
restriction has
Ethernet headers are 14 bytes long, so when the beginning of such a header
is 32-bit aligned, the following data is misaligned. The usual trick to
fix that is to start the Ethernet header on an odd-numbered 16-bit
boundary. That trick works OK for Open vSwitch, but there are two
problems:
- O
On Thu, Aug 15, 2013 at 12:33:06PM -0700, Chris Wright wrote:
> Andrei Andone reported an oops on CentOS 6.4 when adding a device to an
> ovs instance. The problem is easy to reproduce and generates the
> following stack trace:
>
> kernel tried to execute NX-protected page - exploit attempt? (uid
On Aug 15, 2013, at 2:33 PM, Chris Wright wrote:
> Andrei Andone reported an oops on CentOS 6.4 when adding a device to an
> ovs instance. The problem is easy to reproduce and generates the
> following stack trace:
>
> kernel tried to execute NX-protected page - exploit attempt? (uid: 0)
> BUG:
Andrei Andone reported an oops on CentOS 6.4 when adding a device to an
ovs instance. The problem is easy to reproduce and generates the
following stack trace:
kernel tried to execute NX-protected page - exploit attempt? (uid: 0)
BUG: unable to handle kernel paging request at 88033afa49c0
IP:
On Wed, Aug 14, 2013 at 10:56 AM, Jesse Gross wrote:
> On Mon, Jul 29, 2013 at 3:48 PM, Pravin B Shelar
> wrote:
> > diff --git a/datapath/vport-lisp.c b/datapath/vport-lisp.c
> > index 847cb39..9ffa74f 100644
> > --- a/datapath/vport-lisp.c
> > +++ b/datapath/vport-lisp.c
> > +static void handl
move per-backer wait calls (dpif_wait/udpif_wait) from ofproto_wait
to ofproto_type_wait.
this fixes O(N^2) behaviour (eg. poll slot consumption) for
dpif-netdev where N is the number of bridges.
Signed-off-by: YAMAMOTO Takashi
---
ofproto/ofproto-dpif.c | 4 ++--
1 file changed, 2 insertions(+
From: Jesse Gross
Date: Wed, 14 Aug 2013 16:19:04 -0700
> Three bug fixes that are fairly small either way but resolve obviously
> incorrect code. For net/3.11.
Pulled, thanks Jesse.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailm
This patch is the rework of :
http://openvswitch.org/pipermail/dev/2013-August/030705.html
It includes the configuration of ofproto-dpif-upcall module.
On Thu, Aug 15, 2013 at 12:23 AM, Alex Wang wrote:
> This commit adds a new column "n-handler-threads" to ovsdb Open_vSwitch
> TABLE. This is
This commit adds a new column "n-handler-threads" to ovsdb Open_vSwitch
TABLE. This is used to set the number of upcall handler threads created
by the ofproto-dpif-upcall module.
Signed-off-by: Alex Wang
---
ofproto/ofproto-dpif.c | 20
ofproto/ofproto-provider.h |
41 matches
Mail list logo
