On Mon, Apr 13, 2015 at 11:15 AM, Pravin Shelar wrote:
> On Thu, Apr 9, 2015 at 11:15 AM, Jesse Gross wrote:
>> diff --git a/datapath/linux/compat/gso.h b/datapath/linux/compat/gso.h
>> index 337d13a..dce4c67 100644
>> --- a/datapath/linux/compat/gso.h
>> +++ b/
On Mon, Apr 13, 2015 at 12:40 PM, Jesse Gross wrote:
> On Mon, Apr 13, 2015 at 11:15 AM, Pravin Shelar wrote:
>> On Thu, Apr 9, 2015 at 11:15 AM, Jesse Gross wrote:
>>> diff --git a/datapath/linux/compat/gso.h b/datapath/linux/compat/gso.h
>>> index 337d13a..dce4c6
On Tue, Apr 14, 2015 at 9:37 AM, Rogers, Gerald wrote:
> There are some deployments that are using or require older OS
> distributions using qemu versions prior to version 2.1, thus would require
> the support for vHost-cuse. Most general reason is qualification cycles
> for new OS versions. The
On Thu, Apr 16, 2015 at 11:33 AM, Alex Wang wrote:
> With the latest change of separating vports into their own modules,
> we also need to explicitly export rpl_skb_gso_segment to avoid linker
> error.
>
> VMware-BZ: #1432578
> VMware-BZ: #1308175
>
> Signed-off-by: Ale
On Tue, Apr 14, 2015 at 8:46 PM, Pravin B Shelar wrote:
> diff --git a/datapath/linux/compat/stt.c b/datapath/linux/compat/stt.c
> new file mode 100644
> index 000..f294ecd
> --- /dev/null
> +++ b/datapath/linux/compat/stt.c
> +static struct sk_buff *handle_offloads(struct sk_buff *skb, int mi
On Thu, Apr 16, 2015 at 2:10 AM, Gray, Mark D wrote:
>
>
>> -Original Message-
>> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Tahhan,
>> Maryam
>>
>> > On Tue, Apr 14, 2015 at 9:37 AM, Rogers, Gerald
>> >
>> > wrote:
>> > > There are some deployments that are using or requ
On Thu, Apr 16, 2015 at 3:35 PM, Pravin Shelar wrote:
> On Thu, Apr 16, 2015 at 3:15 PM, Jesse Gross wrote:
>> On Tue, Apr 14, 2015 at 8:46 PM, Pravin B Shelar wrote:
>>> diff --git a/datapath/linux/compat/stt.c b/datapath/linux/compat/stt.c
>>> new file mode 100644
On Thu, Apr 16, 2015 at 7:17 PM, Pravin B Shelar wrote:
> +static struct sk_buff *push_stt_header(struct sk_buff *head, __be64 tun_id,
> + __be16 s_port, __be16 d_port,
> + __be32 saddr, __be32 dst,
> +
On Fri, Apr 17, 2015 at 1:05 PM, Pravin Shelar wrote:
> On Fri, Apr 17, 2015 at 12:34 PM, Jesse Gross wrote:
>> On Thu, Apr 16, 2015 at 7:17 PM, Pravin B Shelar wrote:
>>> +static struct sk_buff *push_stt_header(struct sk_buff *h
On Fri, Apr 17, 2015 at 1:26 PM, Pravin Shelar wrote:
> On Fri, Apr 17, 2015 at 1:11 PM, Jesse Gross wrote:
>> On Fri, Apr 17, 2015 at 1:05 PM, Pravin Shelar wrote:
>>> On Fri, Apr 17, 2015 at 12:34 PM, Jesse Gross wrote:
>>>> On Thu, Apr 16, 2015 at 7:17 PM
On Mon, Apr 20, 2015 at 10:47 AM, Ben Pfaff wrote:
> On Mon, Apr 20, 2015 at 10:07:54AM -0700, Alex Wang wrote:
>> On Mon, Apr 20, 2015 at 1:17 AM, Thomas Graf
>> wrote:
>>
>> > On 04/19/15 at 10:15pm, Alex Wang wrote:
>> > > With the latest change of separating vports into their own modules,
>>
On Mon, Apr 20, 2015 at 1:29 PM, Pravin B Shelar wrote:
> diff --git a/datapath/linux/compat/stt.c b/datapath/linux/compat/stt.c
> new file mode 100644
> index 000..209bf1a
> --- /dev/null
> +++ b/datapath/linux/compat/stt.c
> +static void update_headers(struct sk_buff *skb, bool head,
> +
On Mon, Apr 20, 2015 at 3:54 PM, Pravin Shelar wrote:
> On Mon, Apr 20, 2015 at 3:36 PM, Jesse Gross wrote:
>> On Mon, Apr 20, 2015 at 1:29 PM, Pravin B Shelar wrote:
>>> diff --git a/datapath/linux/compat/stt.c b/datapath/linux/compat/stt.c
>>> new file mode 100644
On Mon, Apr 20, 2015 at 7:33 PM, Pravin Shelar wrote:
> On Mon, Apr 20, 2015 at 5:56 PM, Jesse Gross wrote:
>> On Mon, Apr 20, 2015 at 3:54 PM, Pravin Shelar wrote:
>>> On Mon, Apr 20, 2015 at 3:36 PM, Jesse Gross wrote:
>>>> On Mon, Apr 20, 2015 at 1:29 PM
On Tue, Apr 21, 2015 at 3:20 PM, Alex Wang wrote:
> No longer need this compat file, we can use the upstream version
> of the function.
>
> Signed-off-by: Alex Wang
There's also a check in acinclude.m4 to define HAVE_LOG_LOG2, can you
remove that as well?
On Tue, Apr 21, 2015 at 7:23 PM, Pravin Shelar wrote:
> On Tue, Apr 21, 2015 at 6:33 PM, Ethan Jackson wrote:
>> I really don't think this should be configurable, I can't imagine a
>> situation in which a user would have enough knowledge to tweak the
>> knob correctly. I think the only reason we
On Wed, Apr 22, 2015 at 10:03 AM, Pravin Shelar wrote:
> On Mon, Apr 20, 2015 at 9:54 PM, Jesse Gross wrote:
>> On Mon, Apr 20, 2015 at 7:33 PM, Pravin Shelar wrote:
>>> On Mon, Apr 20, 2015 at 5:56 PM, Jesse Gross wrote:
>>>> On Mon, Apr 20, 2015 at 3:54 PM, Prav
On Wed, Apr 22, 2015 at 11:09 AM, Pravin Shelar wrote:
> On Wed, Apr 22, 2015 at 10:52 AM, Jesse Gross wrote:
>> On Wed, Apr 22, 2015 at 10:03 AM, Pravin Shelar wrote:
>>> On Mon, Apr 20, 2015 at 9:54 PM, Jesse Gross wrote:
>>>> On Mon, Apr 20, 2015 at 7:33 PM,
On Wed, Apr 22, 2015 at 1:15 AM, mark.d.gray wrote:
> Pravin Shelar wrote:
>>
>> On Tue, Apr 21, 2015 at 7:30 PM, Jesse Gross wrote:
>>
>>> On Tue, Apr 21, 2015 at 7:23 PM, Pravin Shelar
>>> wrote:
>>>
>>>> On Tue, Apr 21, 2015 at 6:
On Wed, Apr 22, 2015 at 3:19 PM, Pravin Shelar wrote:
> On Wed, Apr 22, 2015 at 11:31 AM, Jesse Gross wrote:
>> On Wed, Apr 22, 2015 at 11:09 AM, Pravin Shelar wrote:
>>> On Wed, Apr 22, 2015 at 10:52 AM, Jesse Gross wrote:
>>>> On Wed, Apr 22, 2015 at 10:03 AM,
On Wed, Apr 22, 2015 at 3:38 PM, Pravin Shelar wrote:
> On Wed, Apr 22, 2015 at 3:24 PM, Jesse Gross wrote:
>> On Wed, Apr 22, 2015 at 3:19 PM, Pravin Shelar wrote:
>>> On Wed, Apr 22, 2015 at 11:31 AM, Jesse Gross wrote:
>>>> On Wed, Apr 22, 2015 at 11:09 AM,
On Wed, Apr 22, 2015 at 8:22 PM, Pravin Shelar wrote:
> On Wed, Apr 22, 2015 at 7:48 PM, Jesse Gross wrote:
>> On Wed, Apr 22, 2015 at 3:38 PM, Pravin Shelar wrote:
>>> On Wed, Apr 22, 2015 at 3:24 PM, Jesse Gross wrote:
>>>> On Wed, Apr 22, 2015 at 3:19 PM, Prav
On Wed, Apr 22, 2015 at 8:37 PM, Pravin Shelar wrote:
> On Wed, Apr 22, 2015 at 8:29 PM, Jesse Gross wrote:
>> On Wed, Apr 22, 2015 at 8:22 PM, Pravin Shelar wrote:
>>> On Wed, Apr 22, 2015 at 7:48 PM, Jesse Gross wrote:
>>>> On Wed, Apr 22, 2015 at 3:38 PM, Prav
On Wed, Apr 22, 2015 at 8:55 PM, Pravin Shelar wrote:
> On Wed, Apr 22, 2015 at 8:50 PM, Jesse Gross wrote:
>> On Wed, Apr 22, 2015 at 8:37 PM, Pravin Shelar wrote:
>>> On Wed, Apr 22, 2015 at 8:29 PM, Jesse Gross wrote:
>>>> On Wed, Apr 22, 2015 at 8:22 PM, Prav
On Tue, Apr 21, 2015 at 3:20 PM, Alex Wang wrote:
> diff --git a/datapath/linux/compat/flow_dissector.c
> b/datapath/linux/compat/flow_dissector.c
> index a68f84f..c6644a5 100644
> --- a/datapath/linux/compat/flow_dissector.c
> +++ b/datapath/linux/compat/flow_dissector.c
> @@ -232,4 +232,5 @@ u3
; rpl_ or ovs_ prefixed.
>
> How about it~?
>
> Thanks,
> Alex Wang,
>
> On Fri, Apr 24, 2015 at 2:20 PM, Jesse Gross wrote:
>>
>> On Tue, Apr 21, 2015 at 3:20 PM, Alex Wang wrote:
>> > diff --git a/datapath/linux/compat/flow_dissector.c
>> > b/d
On Tue, Apr 21, 2015 at 3:20 PM, Alex Wang wrote:
> This commit adds check in datapath/Makefile to make sure that all public
> functions in linux/compat/ are all rpl_ prefixed and exported, except
> those in compat/build-aux/export-check-whitelist.
>
> Signed-off-by: Alex Wang
This looks good to
On Thu, Apr 23, 2015 at 9:06 PM, Neil McKee wrote:
> I looked into what it would take to have the kernel pass up the actions when
> it samples a packet. The changes seemed straightforward (see attached
> diff) and worked well. I prefer this approach because:
>
> (1) These really are the action
On Fri, Apr 24, 2015 at 11:25 AM, Pravin Shelar wrote:
> On Thu, Apr 23, 2015 at 12:11 PM, Pravin Shelar wrote:
>> On Thu, Apr 23, 2015 at 11:55 AM, Jesse Gross wrote:
>>
>>> I'm a little nervous about orphaning the packets in this case since it
>>> can po
On Thu, Apr 23, 2015 at 1:19 AM, Mark D. Gray wrote:
> On 04/22/15 22:18, Jesse Gross wrote:
>>
>> On Wed, Apr 22, 2015 at 1:15 AM, mark.d.gray
>> wrote:
>>>
>>> Pravin Shelar wrote:
>>>>
>>>>
>>>> On Tue, Apr 21, 2015
On Sat, Apr 25, 2015 at 1:08 AM, Alex Wang wrote:
> No longer need this compat file, we can use the upstream version
> of the function.
>
> Signed-off-by: Alex Wang
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://ope
t_init_net(struct net *net, struct rpl_pernet_operations *pnet);
> +#define compat_exit_net rpl_compat_exit_net
> +void rpl_compat_exit_net(struct net *net, struct rpl_pernet_operations
> *pnet);
These compat functions also aren't upstream, so maybe the same comment
about making th
; Signed-off-by: Alex Wang
> ---
> PATCH->V2:
> - Add check for EXPORT_SYMBOL_GPL(). The exported function must have
> rpl_ or ovs_ as prefix.
I would probably expand the export check to also look for
EXPORT_SYMBOL(), in case those sneak in as well.
Otherwise:
When the UDP tunnel compat code was written, it backported some
functions that were slated to be in the next kernel release, then
called 3.20. However, this was ultimately released as 4.0 instead.
Signed-off-by: Jesse Gross
---
datapath/linux/compat/include/net/udp_tunnel.h | 4 ++--
datapath
it as
if it were VXLAN, which was fixed in the 4.0 release. Since our
UDP tunnel compat layer already takes care of this, we continue
using the OVS Geneve implementation until 4.0.
Reported-by: Alex Wang
Signed-off-by: Jesse Gross
---
datapath/linux/compat/geneve.c | 5 +
dat
On Sun, Apr 26, 2015 at 6:28 PM, Simon Horman
wrote:
> On Sat, Apr 25, 2015 at 05:40:29PM -0700, Jesse Gross wrote:
>> When the UDP tunnel compat code was written, it backported some
>> functions that were slated to be in the next kernel release, then
>> called 3.20. However
version of the module. It retains the
version string, however, which should provide the same information
in most cases.
Signed-off-by: Jesse Gross
---
datapath/datapath.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/datapath/datapath.c b/datapath/datapath.c
index c0af9ad
> Signed-off-by: Pravin B Shelar
> Signed-off-by: Jesse Gross
I haven't looked through this yet but this currently doesn't compile
after Alex's latest patches:
Should prefix stt_sock_add with rpl_ or ovs_.
Makefile:791: recipe for target 'check-export-symbol' fa
On Mon, Apr 27, 2015 at 2:00 PM, Ben Pfaff wrote:
> On Mon, Apr 27, 2015 at 01:29:52PM -0700, Jesse Gross wrote:
>> An increasing number of distributions ship with GCC 4.9 (including
>> Fedora and Ubuntu) that has -Werror=date-time. This causes kernel
>> compilation to fail b
bool ecn_ce;
> +};
I just noticed that the ecn_ce field appears to be write only. We
should set this on the outer IP header of the reassembled packet if
any of the individual segments have it set.
These are pretty minor changes though and I am otherwise happy with
this. Thanks
> port 0: ovs-system (internal)
> port 1: br0 (internal)
> port 4: gre_sys (gre)
>
> Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Fri, May 1, 2015 at 12:40 PM, Pravin B Shelar wrote:
> For kernel version 3.12 to 3.18, GRE uses compat code to
> transmit packets which used fix_segment to segment packets.
> but ovs_gso_cb->fix_segment is not initialized for GRE tunnels.
> Following patches fixes it by resetting fix_segment.
ing fix_segment.
>
> Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Fri, May 1, 2015 at 5:44 PM, Pravin B Shelar wrote:
> OVS correctly define skb_gso_segment() to handle MPLS and VLAN
> segmentation correctly. But OVS also uses __skb_gso_segment() in
> some cases. Following patch defines compat __skb_gso_segment()
> to handle all segmentation cases.
>
> Signed
On Sun, May 3, 2015 at 11:56 AM, Pravin B Shelar wrote:
> RHEL7 backported nf_hookfn from newer kernel. Handle compatibility
> by checking nf_hookfn declaration.
>
> Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
___
dev mail
On Fri, May 1, 2015 at 7:00 PM, Pravin B Shelar wrote:
> diff --git a/datapath/linux/compat/gso.c b/datapath/linux/compat/gso.c
> index 552e748..f8a70d0 100644
> --- a/datapath/linux/compat/gso.c
> +++ b/datapath/linux/compat/gso.c
> @@ -219,6 +219,9 @@ static struct sk_buff *tnl_skb_gso_segment(s
On Mon, May 4, 2015 at 12:48 PM, Pravin Shelar wrote:
> On Mon, May 4, 2015 at 11:26 AM, Jesse Gross wrote:
>> What was the change that bumps up the backport requirement from 3.16 to 3.19?
>
> MPLS offload is supported on 3.19 or newer.
OK, I remember - you made some critical b
way prior extending its life-span
> * beyond v3.16.
The above should be v3.19, not v3.16.
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Tue, May 5, 2015 at 2:01 PM, Joe Stringer wrote:
> Signed-off-by: Joe Stringer
> ---
> acinclude.m4 | 18 +-
> 1 file changed, 9 insertions(+), 9 deletions(-)
Can you also update NEWS and the table in the FAQ?
> diff --git a/acinclude.m4 b/acinclude.m4
> index e9d0ed9..cd2
On Fri, May 8, 2015 at 9:48 AM, Joe Stringer wrote:
> Signed-off-by: Joe Stringer
> ---
> v2: Address Jesse's feedback.
> Added travis build for 4.0.
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://openvs
On Wed, May 13, 2015 at 7:00 AM, Beurer, Maria (Coriant - DE/Munich)
wrote:
> Hello,
> We use openVSwitch in our commercial product and therefore need to know the
> ECCN for it.
> The ECCN is an international Export control and compliance number that is
> necessary for EVERY Export. Most of the
On Wed, May 13, 2015 at 6:05 AM, Guillermo Ontañón
wrote:
> On Wed, May 13, 2015 at 1:00 AM, Ben Pfaff wrote:
>> On Tue, May 12, 2015 at 10:21:19AM +0200, Duarte Nunes wrote:
>> Second, is it possible to use a better data structure? Perhaps one
>> could, for example, use a mutex per hash chain,
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> This patchset is against the current net-next. I'm intentionally not sending
> it to netdev yet, I'd like to gather feedback on design decisions first,
> especially on the ovs part in the second patchset.
>
> This patchset adds support for tunne
d-off-by: Pravin B Shelar
Acked-by: Jesse Gross
I noticed two others though (in addition to the old SCTP ones):
CHECK /home/jesse/openvswitch/datapath/linux/datapath.c
/home/jesse/openvswitch/datapath/linux/datapath.c:1523:28: warning:
symbol 'i' shadows an earlier one
/home/jesse/openvsw
On Fri, May 15, 2015 at 3:04 PM, Neil McKee wrote:
> I understand that the (optional) kernel datapath changes will need to
> be submitted as a kernel
> patch on netdev. I can try that, but I'm guessing I won't get much
> attention there without
> input from someone they have heard of. How does
On Thu, May 21, 2015 at 10:37 AM, Jiri Benc wrote:
> Hi,
>
> currently, ovs switches LRO off in user space:
>
> commit 24b019f808211a95078efd916064af0975ca5733
> Author: Justin Pettit
> Date: Fri Aug 26 23:34:40 2011 -0700
>
> datapath: Disable LRO from userspace instead of the kernel.
>
>
tream patch and I
nothing that should require special backporting comes to mind, so
looks good to me.
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Geneve options are variable length and up to 124 bytes long, which means
that they can't be easily manipulated by the integer string functions
like we do for other fields. This adds a few helper routines to make
these operations easier.
Signed-off-by: Jesse Gross
---
lib/dynamic-string.c
we have sets of
tables for the type of each attribute linked together by pointers.
This allows the mask generation function to automatically traverse
the nested attributes and always get the right types.
Signed-off-by: Jesse Gross
---
lib/odp-util.c | 151
Even though userspace does not yet support Geneve options,
the kernel does and there is some basic support for decoding
those attributes. This adds the ability to print Geneve
attributes that might potentially come from the kernel.
Signed-off-by: Jesse Gross
---
lib/odp-util.c | 206
re not present.
Signed-off-by: Jesse Gross
---
lib/odp-util.c | 455 +---
tests/odp.at| 18 +--
tests/tunnel.at | 82 +-
3 files changed, 420 insertions(+), 135 deletions(-)
diff --git a/lib/odp-util.c b/lib/odp-util.c
index 09a1ff2
On Tue, May 26, 2015 at 4:59 PM, Ben Pfaff wrote:
> diff --git a/datapath/linux/compat/include/linux/if_vlan.h
> b/datapath/linux/compat/include/linux/if_vlan.h
> index 05c2876..0750bbe 100644
> --- a/datapath/linux/compat/include/linux/if_vlan.h
> +++ b/datapath/linux/compat/include/linux/if_vla
On Wed, May 27, 2015 at 2:07 PM, Hauke Mehrtens wrote:
> On 05/27/2015 10:14 PM, Jesse Gross wrote:
>> On Tue, May 26, 2015 at 4:59 PM, Ben Pfaff wrote:
>>> diff --git a/datapath/linux/compat/include/linux/if_vlan.h
>>> b/datapath/linux/compat/include/linux/if_vlan.h
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> When the packet ovs gets from a NIC driver is CHECKSUM_PARTIAL and it goes
> through decapsulation, csum_start may point before the skb->data after the
> decapsulation. This triggers a BUG_ON in skb_checksum_help.
>
> As we don't need to calcula
On Wed, May 27, 2015 at 3:03 PM, Andy Zhou wrote:
> On Wed, May 27, 2015 at 10:48 AM, Jesse Gross wrote:
>> When formatting netlink attributes if no mask is present a wildcarded
>> attribute is synthesized for the purposes of later processing. In
>> the case of nested at
On Wed, May 27, 2015 at 3:15 PM, Andy Zhou wrote:
> On Wed, May 27, 2015 at 10:48 AM, Jesse Gross wrote:
>> diff --git a/lib/util.c b/lib/util.c
>> index bcf7700..3dc06d0 100644
>> --- a/lib/util.c
>> +++ b/lib/util.c
>> +int
>> +parse_int_string(const ch
On Wed, May 27, 2015 at 3:18 PM, Andy Zhou wrote:
> On Wed, May 27, 2015 at 10:48 AM, Jesse Gross wrote:
>> Even though userspace does not yet support Geneve options,
>> the kernel does and there is some basic support for decoding
>> those attributes. This adds the ab
On Wed, May 27, 2015 at 3:25 PM, Ben Pfaff wrote:
> On Wed, May 27, 2015 at 10:48:40AM -0700, Jesse Gross wrote:
>> Geneve options are variable length and up to 124 bytes long, which means
>> that they can't be easily manipulated by the integer string functions
>> l
On Wed, May 27, 2015 at 3:18 PM, Andy Zhou wrote:
> On Wed, May 27, 2015 at 10:48 AM, Jesse Gross wrote:
>> Even though userspace does not yet support Geneve options,
>> the kernel does and there is some basic support for decoding
>> those attributes. This adds the ab
need for
casts.
Reported-by: Joe Stringer
Signed-off-by: Jesse Gross
---
lib/odp-util.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/lib/odp-util.c b/lib/odp-util.c
index c3497ea..3204d16 100644
--- a/lib/odp-util.c
+++ b/lib/odp-util.c
@@ -2925,7 +2925,7 @@ scan_vxlan
On Fri, May 29, 2015 at 10:20 AM, Joe Stringer wrote:
> On 28 May 2015 at 18:39, Jesse Gross wrote:
>> On Wed, May 27, 2015 at 3:18 PM, Andy Zhou wrote:
>>> On Wed, May 27, 2015 at 10:48 AM, Jesse Gross wrote:
>>>> Even though userspace does not yet support Geneve
On Fri, May 29, 2015 at 10:43 AM, Ben Pfaff wrote:
> I'm seeing the following new warnings from the Geneve patch checked in
> recently:
>
> ../lib/odp-util.c:2936:30: error: cast from 'uint8_t *' (aka 'unsigned
> char *')
> to 'struct geneve_opt *' increases required alignment from
On Fri, May 29, 2015 at 10:50 AM, Joe Stringer wrote:
> On 29 May 2015 at 10:46, Jesse Gross wrote:
>> Clang complains about the fact that we use a byte array to scan
>> Geneve attributes into since there are different alignment requirements:
>>
>> lib/odp-util.
On Fri, May 29, 2015 at 10:40 AM, Ben Pfaff wrote:
> On Wed, Feb 25, 2015 at 09:12:38AM -0800, Ben Pfaff wrote:
>> Don't panic, this is not a remote buffer overflow, only a bug in a program
>> parsing its input from the command line or a file.
>>
>> Found by inspection.
>>
>> Signed-off-by: Ben Pf
On Fri, May 29, 2015 at 3:15 PM, Ben Pfaff wrote:
> parse_int_string() returns an error if the field overflows, not if there
> is a bad hex digit.
>
> Signed-off-by: Ben Pfaff
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswi
allowed by NXM/OXM. This is useful for future large fields,
such as Geneve options.
Signed-off-by: Jesse Gross
---
build-aux/extract-ofp-fields | 20 +++
lib/meta-flow.c | 59 +---
2 files changed, 32 insertions(+), 47 deletions(-)
diff
ture means we only have to do this work in one place.
Signed-off-by: Jesse Gross
---
lib/flow.c| 48 ++--
lib/flow.h| 17 +--
lib/learning-switch.c | 8 ++--
lib/ofp-print.c | 39 +---
lib/
ture means we only have to do this work in one place.
Signed-off-by: Jesse Gross
---
v2: Add missing forward declaration for struct match in flow.h.
---
lib/flow.c| 48 ++--
lib/flow.h| 18 +---
lib/learning-switch.c | 8 ++-
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
> index 2af6ffbf2f2e..78e96a120120 100644
> --- a/net/openvswitch/flow.h
> +++ b/net/openvswitch/flow.h
> -struct ovs_key_ipv4_tunnel {
> +struct ovs_key_ip_tunnel {
> __be64 t
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> diff --git a/include/uapi/linux/openvswitch.h
> b/include/uapi/linux/openvswitch.h
> index 4d26da40b01f..ba7ae3b05308 100644
> --- a/include/uapi/linux/openvswitch.h
> +++ b/include/uapi/linux/openvswitch.h
> @@ -267,6 +267,7 @@ enum {
>
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> diff --git a/net/openvswitch/vport.h b/net/openvswitch/vport.h
> index b22d45775eb5..ca9ebb9eb0f7 100644
> --- a/net/openvswitch/vport.h
> +++ b/net/openvswitch/vport.h
> +static inline struct dst_entry *ovs_tunnel6_route_lookup(struct sock *sk,
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> Other users of IPv6 tunneling (e.g. openvswitch) may not have a net_device
> that represents the tunnel.
>
> Signed-off-by: Jiri Benc
I think this will probably not make people happy upstream, so we'll
need to wait for the netdevice tunnel pat
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> diff --git a/include/net/vxlan.h b/include/net/vxlan.h
> index 0082b5d33d7d..a4ed7241c56a 100644
> --- a/include/net/vxlan.h
> +++ b/include/net/vxlan.h
> +int vxlan6_xmit_skb(struct dst_entry *dst, struct sock *sk,
> + struct
On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
> Signed-off-by: Jiri Benc
> ---
> net/openvswitch/vport-vxlan.c | 133
> +++---
> 1 file changed, 99 insertions(+), 34 deletions(-)
No particular objection to this although it's obviously going to need
to l
On Mon, Jun 1, 2015 at 2:55 PM, Ben Pfaff wrote:
> On Mon, Jun 01, 2015 at 01:49:30PM -0700, Jesse Gross wrote:
>> We have a special flow_metadata structure to represent the parts
>> of a packet that aren't carried in the payload itself. This is
>> used in the case
On Mon, Jun 1, 2015 at 7:08 PM, Andy Zhou wrote:
> Covert tabs into spaces. Found by inspection.
>
> Signed-off-by: Andy Zhou
Acked-by: Jesse Gross
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Mon, Jun 1, 2015 at 6:54 PM, Andy Zhou wrote:
[...]
>> -return xasprintf("%s: bad syntax for %s", s, mf->name);
>> +if (err == ERANGE) {
>> +return xasprintf("%s: value too large for %u-byte field %s",
>> + s, mf->n_bytes, mf->name);
> This error message
On Tue, Jun 2, 2015 at 8:03 AM, Ben Pfaff wrote:
> On Tue, Jun 02, 2015 at 07:58:48AM -0700, Ben Pfaff wrote:
>> On Mon, Jun 01, 2015 at 03:35:23PM -0700, Jesse Gross wrote:
>> > On Mon, Jun 1, 2015 at 2:55 PM, Ben Pfaff wrote:
>> > > On Mon, Jun 01, 2015 at 01:49
On Tue, Jun 2, 2015 at 8:58 AM, Jiri Benc wrote:
> On Mon, 1 Jun 2015 14:40:41 -0700, Jesse Gross wrote:
>> On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
>> > diff --git a/include/uapi/linux/openvswitch.h
>> > b/include/uapi/linux/openvswitch.h
>> > ind
On Tue, Jun 2, 2015 at 8:37 AM, Jiri Benc wrote:
> On Mon, 1 Jun 2015 14:22:48 -0700, Jesse Gross wrote:
>> On Thu, May 14, 2015 at 11:10 AM, Jiri Benc wrote:
>> > diff --git a/net/openvswitch/flow.h b/net/openvswitch/flow.h
>> > index 2af6ffbf2f2e..78e96a120120 100644
On Sun, Jun 7, 2015 at 10:36 AM, Ben Pfaff wrote:
> On Mon, Jun 01, 2015 at 01:49:30PM -0700, Jesse Gross wrote:
>> We have a special flow_metadata structure to represent the parts
>> of a packet that aren't carried in the payload itself. This is
>> used in the case
this no longer makes sense given the way that we format
the flow itself. In fact, the code is not actually invoked any
more, so we can simply remove it.
This retains the special case for tunnels as a safety measure but it
should not matter in practice.
Signed-off-by: Jesse Gross
---
lib/odp-util.c
On Mon, Jun 8, 2015 at 3:50 PM, Pravin B Shelar wrote:
> Some of code is backported from following commit.
>
> commit 13dd4a9738e99684a56b10ce2f1a5ee2d2ec2f9f
> Author: Joe Stringer
> Date: Tue Mar 24 16:16:18 2015 -0700
>
> compat: Fix RHEL7 build.
>
> Tested against 3.10.0-229.el7.x86
On Thu, Jun 11, 2015 at 10:57 AM, Ben Pfaff wrote:
> On Wed, Jun 10, 2015 at 08:29:30PM -0700, Jesse Gross wrote:
>> We formerly converted tunnel attributes to their flow representation
>> before formatting but now perform all operations directly on the
>> netlink attribute
continue to use the length for non-variable fields to ensure that
we don't accept something that can never match.
Signed-off-by: Jesse Gross
---
lib/nx-match.c | 17 +
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/lib/nx-match.c b/lib/nx-match.c
index 21
annotated with not only their size
but whether the field can be smaller than that. The following
patches will change OpenFlow parsing based on that.
Signed-off-by: Jesse Gross
---
build-aux/extract-ofp-fields | 23 +++
lib/meta-flow.h | 1 +
2 files changed, 16
that are generally
sensible without specific knowledge of what is being transmitted.
(Of course, other implementations, such as controllers, may know
exactly the expected length of the field and are free to encode
it that way even if it has leading zeros.)
Signed-off-by: Jesse Gross
---
lib/meta
When reading in hex strings that form NXM fields, we don't need to
enforce size constraints if the fields are variable length.
Instead, we can set the header size based on the string length.
Signed-off-by: Jesse Gross
---
lib/nx-match.c | 20 ++--
1 file changed, 18 inser
implementation
already required callers to pass in the exact (fixed) size of
the field or it would not work properly, so there is no danger
that this will change the behavior for non-variable length
fields.
Signed-off-by: Jesse Gross
---
lib/nx-match.c | 20 ++--
1 file changed, 18
and
the userspace implementation of Geneve does not currently support
options anyways.
Signed-off-by: Jesse Gross
---
lib/flow.c| 7 ++-
lib/packets.h | 2 +-
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/lib/flow.c b/lib/flow.c
index 9f3c7dc..dad94c3 100644
--- a
801 - 900 of 3414 matches
Mail list logo
