On Fri, Dec 2, 2022 at 11:59 AM Ilya Maximets wrote:
>
> On 12/2/22 11:36, Maxime Coquelin wrote:
> >
> >
> > On 12/2/22 11:09, David Marchand wrote:
> >> On Wed, Nov 30, 2022 at 9:30 PM Ilya Maximets wrote:
> >>> Shouldn't this be 0x7f instead?
> >>> 0x3f doesn't enable bit #6, which is
On Fri, Dec 2, 2022 at 10:20 AM Maxime Coquelin
wrote:
>
>
>
> On 12/2/22 15:59, Mike Pattrick wrote:
> > On Fri, Dec 2, 2022 at 5:37 AM Maxime Coquelin
> > wrote:
> >>
> >>
> >>
> >> On 12/2/22 11:09, David Marchand wrote
On Fri, Dec 2, 2022 at 5:37 AM Maxime Coquelin
wrote:
>
>
>
> On 12/2/22 11:09, David Marchand wrote:
> > On Wed, Nov 30, 2022 at 9:30 PM Ilya Maximets wrote:
> >> Shouldn't this be 0x7f instead?
> >> 0x3f doesn't enable bit #6, which is responsible for dumping
> >> shared huge
On Wed, Nov 30, 2022 at 7:27 AM Ilya Maximets wrote:
>
> On 11/25/22 18:19, Adrian Moreno wrote:
> > Hi Mike,
> >
> > Sorry it took that long to review this patch.
> >
> > On 3/25/22 23:17, Mike Pattrick wrote:
> >> Add new option --dump-hugep
ink this is a reasonable change, 100MB has mostly been relegated
to embedded devices for a long time.
Acked-by: Mike Pattrick
Slightly related, I noticed that we're missing support for 20 and
56Gbps, we should probably add those as well.
-M
> ---
> NEWS | 4
&
offload, so this
patch enables the feature. However, Linux socket interface
remains disabled because the API doesn't allow enabling
those two features without enabling TSO too.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 15
offload alone, so the
support is not enabled.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 17 +++---
lib/dp-packet.c | 15 ++
lib/dp-packet.h | 60
From: Flavio Leitner
This patch introduces a command to display the current checksum offload
status by port, allowing the user to gain insight into where checksum
offloading is active.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Reviewed-by: David
This is a subset of the larger TSO patchset with various checksum
improvements. This set includes additional documentation, new appctl
command "dpif-netdev/offload-show" to display interface offload
support, and improvements to tracking when an updated checksum is
required.
In a simple iperf test
From: Flavio Leitner
Document the implementation of netdev hardware offloading
in userspace datapath.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
Documentation/automake.mk| 1 +
Documentation
On Wed, Nov 23, 2022 at 5:03 AM Adrian Moreno wrote:
>
I believe the commit message was cut out.
> Signed-off-by: Adrian Moreno
> ---
> python/ovs/flow/kv.py| 17 ++---
> python/ovs/flow/ofp.py | 7 ---
> python/ovs/tests/test_ofp.py | 15 +++
> 3
aliases
> so it's easy to map OXM/NXM names to their fields and decoding
> information.
>
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
> ---
> build-aux/gen_ofp_field_decoders | 15 +++
> 1 file changed, 15 insertions(+)
>
> diff --git a/bu
On Wed, Nov 9, 2022 at 3:32 PM David Marchand wrote:
>
> The count of received multicast packets has been computed internally,
> but not exposed to ovsdb. Fix this.
>
> Signed-off-by: David Marchand
Acked-by: Mike Pattrick
> ---
> Strictly speaking, nothing was broken s
On Fri, Nov 4, 2022 at 2:46 AM Peng He wrote:
>
> The tail of the struct dp_netdev_flow contains a whole netdev_flow_key
> struct.
>
> We need to first minus the size of netdev_flow_key then add back
> the real size of this netdev_flow_key.
>
> Signed-off-by: Peng He
> ---
> lib/dpif-netdev.c |
should be done by the caller anyway.
>
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
that already takes that into account from ofp_act.py to
> kv.py and use it for datapath action "clone".
>
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Mon, Oct 17, 2022 at 9:19 AM Adrian Moreno wrote:
>
> Create a small helper script and check that flows tested in odp.at are
> parseable.
>
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvs
On Mon, Oct 17, 2022 at 9:19 AM Adrian Moreno wrote:
>
> Create a small helper script and check that flows used in ofp-actions.at
> are parseable.
>
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
___
dev mailing list
d...@
ed. Also, if a KVDecoder does
> need this default behavior, it can be explicitly configured specifying
> it's default decoder.
>
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
>
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
e_output,
> "drop": decode_flag,
> "controller": decode_controller,
> +"CONTROLLER": decode_controller,
This is beyond the scope of this patch, but I was surprised to see
that format_CONTROLLER in ofp-actions.c
On Mon, Oct 17, 2022 at 9:18 AM Adrian Moreno wrote:
>
> Fix the following erros in odp decoding:
> - Missing push_mpls action
> - Typos in collector_set_id, tp_src/tp_dst and csum
> - Missing two fields in vxlan match
>
> Signed-off-by: Adrian Moreno
Ac
ec.
>
> Fixes: be15ec48d766 ("lib: Use a more accurate value for CPU count
> (sched_getaffinity).")
> Cc: david.march...@redhat.com
> Signed-off-by: Adrian Moreno
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Fri, Nov 4, 2022 at 11:45 AM Adrian Moreno wrote:
>
> Currently, things like the number of handler and revalidator threads are
> calculated based on the number of available CPUs. However, this number
> is considered static and only calculated once, hence ignoring events
> such as cpus being
% | 7.39 Gbps |
Signed-off-by: Mike Pattrick
---
Documentation/ref/ovs-tcpdump.8.rst | 4 +++
NEWS| 2 ++
lib/flow.h | 11 ++
ofproto/ofproto-dpif-mirror.c | 53 +++--
ofproto/ofproto-dpif-mirror.h
On Thu, Oct 27, 2022 at 6:00 AM Adrian Moreno wrote:
>
>
>
> On 10/26/22 18:41, Adrian Moreno wrote:
> >
> >
> > On 10/21/22 18:57, Mike Pattrick wrote:
> >> On Fri, Sep 30, 2022 at 11:31 AM Adrian Moreno wrote:
> >>>
> >>> Curre
te_in_count:600 duration:6s
> bands:
>
> Fix sed matches to correctly handle that scenario.
> Repeating the [0-9\.] twice because it is hard to write a shorter
> portable version with sed.
>
> Signed-off-by: Ilya Maximets
Acked-by: Mike Pattrick
__
On Mon, Oct 17, 2022 at 9:18 AM Adrian Moreno wrote:
>
> We currently auto-generate a dictionary of field names and decoders.
> However, sometimes fields can be specified by their cannonical NXM or
> OXM names.
>
> Modify gen_ofp_field_decoders to also generate a dictionary of aliases
> so it's
On Fri, Sep 30, 2022 at 11:31 AM Adrian Moreno wrote:
>
> Currently, things like the number of handler and revalidator threads are
> calculated based on the number of available CPUs. However, this number
> is considered static and only calculated once, hence ignoring events
> such as cpus being
the
support is not enabled.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 17 ---
lib/dp-packet.c | 15 ++
lib/dp-packet.h | 60
From: Flavio Leitner
Document the implementation of netdev hardware offloading
in userspace datapath.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
Documentation/automake.mk| 1 +
Documentation
this
patch enables the feature. However, Linux socket interface
remains disabled because the API doesn't allow enabling
those two features without enabling TSO too.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 15
From: Flavio Leitner
This patch introduces a command to display the current checksum offload
status by port, allowing the user to gain insight into where checksum
offloading is active.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Reviewed-by: David
This is a subset of the larger TSO patchset with various checksumming
improvements. This set includes additional documentation, new appctl
command "dpif-netdev/offload-show" to display interface offload
support, and improvements to tracking when an updated checksum is
required.
Flavio Leitner
it currently uses 720 bytes, this patch
reduces that to 288. In my testing of a resubmit loop, this patch
increases the number of recursions possible before a segfault by over
20%.
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=2104779
Signed-off-by: Mike Pattrick
---
include/openvswitch
the
support is not enabled.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 17 ---
lib/dp-packet.c | 15 ++
lib/dp-packet.h | 60
this
patch enables the feature. However, Linux socket interface
remains disabled because the API doesn't allow enabling
those two features without enabling TSO too.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 15
This is a subset of the larger TSO patchset with various checksumming
improvements. This set includes additional documentation, new appctl
command "dpif-netdev/offload-show" to display interface offload
support, and improvements to tracking when an updated checksum is
required.
Flavio Leitner
From: Flavio Leitner
This patch introduces a command to display the current checksum offload
status by port, allowing the user to gain insight into where checksum
offloading is active.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Reviewed
From: Flavio Leitner
Document the implementation of netdev hardware offloading
in userspace datapath.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
Documentation/automake.mk| 1 +
Documentation
On Wed, Oct 19, 2022 at 12:27 PM Ilya Maximets wrote:
>
> On 10/19/22 15:48, Mike Pattrick wrote:
> > On Wed, Oct 19, 2022 at 9:30 AM Flavio Leitner wrote:
> >>
> >>
> >> Hi Mike,
> >>
> >> Thanks for the patch.
> >>
> >
n stack usage.
>
> BTW, I think we use Reported-at: tag instead of Bugzilla:.
You're right! I don't know where I came up with that tag.
-M
>
>
> fbl
>
> On Wed, Oct 19, 2022 at 09:01:46AM -0400, Mike Pattrick wrote:
> > Previously the minimum thread stack size was
here is to set a minimum of 4MB on systems with
more than 4GB of total ram.
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=2104779
Signed-off-by: Mike Pattrick
---
lib/ovs-thread.c | 46 --
lib/ovs-thread.h | 1 +
2 files changed, 45 insertions
application won't notice it. This patch expose the error to the OVSDB
> record, and reverts the dpif port changes upon failure.
>
> Signed-off-by: Han Zhou
This seems reasonable to me.
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitc
the DPDK transmit path.")
> Signed-off-by: Robin Jarry
Looks good to me.
Acked-by: Mike Pattrick
> ---
>
> Notes:
> v2 -> v3: fixed double count of tx_failure_drops
>
> lib/netdev-dpdk.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
>
om sys/types.h to netinet/in.h since
> struct ip6_addr is defined there. Fixed one instance where
> this is not respected at the moment.
>
Looks reasonable to me.
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
errorp ? " (" : "", errorp ? errorp : "",
> + errorp ? ")" : "");
Wouldn't it just be easier to wrap the whole VLOG_INFO_RL() in an if
statement than have this piecewise string format construction?
I guess th
expected 0
530. vconn.at:21: 530. tcp vconn - refuse connection (vconn.at:21):
FAILED (vconn.at:21)
This was observed from a CI system, and isn't a common case.
Signed-off-by: Mike Pattrick
---
tests/test-vconn.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tests/test-vconn.c b/tests/te
test is added to catch the double lock case.
>
> Reported-at: https://github.com/openvswitch/ovs-issues/issues/259
> Reported-by: Daniel Ding
> Fixes: adcf00ba35a0 ("ofproto/bond: Implement bond megaflow using
> recirculation")
> Signed-off-by: Ilya Maximets
> ---
Ne
a Maximets
> ---
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Tue, Sep 13, 2022 at 5:21 AM David Marchand
wrote:
>
> At some point in OVS history, some virtio features were announced as
> supported (ECN and UFO virtio features).
>
> The userspace TSO code, which has been added later, does not support
> those features and tries to disable them.
>
> This
reated is True:
> -_del_taps[sys.platform](mirror_interface)
> except Exception:
> print("Unable to tear down the ports and mirrors.")
> print("Please use ovs-vsctl to remove the ports and mirrors
> created.")
Nit: This
On Wed, Aug 3, 2022 at 4:58 AM Robin Jarry wrote:
>
> Packets that could not be transmitted because the TXQ are full should be
> taken into account in the global ovs_tx_failure_drops as it was the case
> before commit 29b94e12d57d ("netdev-dpdk: Refactor the DPDK transmit
> path.").
>
> Also, the
, we now suppress these messages at
the INFO level and display exact byte count at the debug level.
Signed-off-by: Mike Pattrick
--
Since v5:
- Implemented reverse xmas tree
- Adjusted spacing and wrapping
- Corrected info log units
Since v4:
- Removed prefixes
- Removed unrequired
On Tue, Aug 30, 2022 at 8:52 AM Ilya Maximets wrote:
>
> On 6/22/22 16:28, Mike Pattrick wrote:
> > Add additional logging for debug and info level with a focus on code
> > related to bond members coming up, go down, and changing.
> >
> > Several existing log messag
but it might be possible to trigger the self-lockup
> issue on the main thread. I'll try to work on the test while
> waiting for review.
>
Hello Ilya,
I've tested this and confirmed that it solves the problem for me.
Acked-by: Mike Pattrick
I created a reproducer that doesn't require
only because specific state machine we are using there.
>
> transition_ukey should be as simple as possible, not related to any specific
> state machine used.
>
>
> Mike Pattrick 于2022年8月16日 周二22:05写道:
>>
>> On Tue, Aug 16, 2022 at 4:04 AM Peng He wrote:
>>
instead and allow caller's to use kfree_skb.
Signed-off-by: Mike Pattrick
Link: https://bugzilla.redhat.com/show_bug.cgi?id=2109957
---
Changes in v2:
- Corrected bugzilla link
---
net/openvswitch/datapath.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/openvswitch
Frames sent to userspace can be reported as dropped in
ovs_dp_process_packet, however, if they are dropped in the netlink code
then netlink_attachskb will report the same frame as dropped.
This patch checks for error codes which indicate that the frame has
already been freed.
Signed-off-by: Mike
n;
+}
VLOG_ABORT("Invalid ukey transition %d->%d (last transitioned from "
"thread %u at %s)", ukey->state, dst, ukey->state_thread,
ukey->state_where);
>
> Mike Pattrick 于2022年8月15日周一 20:16写道:
>>
&
On Sat, Jun 4, 2022 at 11:19 AM Peng He wrote:
>
> There is a race condition between the revalidator threads and
> the handler/pmd threads.
>
> revalidator PMD threads
> push_dp_ops deletes a key and tries
> to del the dp magaflow.
>
we're installing extra dependencies
> that backtrack flake8 down to 4.1 or even 3.9.
>
> Signed-off-by: Ilya Maximets
Looks good to me!
Acked-by: Mike Pattrick
> ---
> python/setup.py | 4 ++--
> tests/test-ovsdb.py | 2 +-
> utilities/bugtool/ovs
On Fri, Jul 1, 2022 at 6:53 AM Han Ding wrote:
>
>
> When the bridge protocols is not included Openflow10, printing an error
> message "version negotiation failed" when doing "Restoring saved flows".
>
> Signed-off-by: Han Ding
Looks good to me!
Acked-by:
86_64-linux-gnu/libc.so.6+0x29e3f)
> #14 0x562594eed024 in _start (vswitchd/ovs-vswitchd+0x787024)
>
> Fixes: 526df7d8543f ("tunnel: Provide framework for tunnel extensions for
> VXLAN-GBP and others")
> Signed-off-by: Ilya Maximets
Acked-by: Mike Pattrick
> ---
>
On Wed, Aug 3, 2022 at 11:12 AM Ilya Maximets wrote:
>
> On 8/3/22 16:58, Mike Pattrick wrote:
> > On Thu, Jul 28, 2022 at 2:27 PM Ilya Maximets wrote:
> >>
> >> $NC_EOF_OPT should be used to avoid some netcat implementations
> >> to wait indefinitely.
meter to tc police action")
> Signed-off-by: Ilya Maximets
Should we also add these flags to the other system tests?
Acked-by: Mike Pattrick
> ---
> tests/system-offloads-traffic.at | 8
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/tests/syst
Signed-off-by: Mike Pattrick
---
net/openvswitch/datapath.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 029f9c3e1c28..3eee4b0a2005 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
://bugzilla.redhat.com/show_bug.cgi?id=2109946
Signed-off-by: Mike Pattrick
---
net/openvswitch/datapath.c | 16 +---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 7e8a39a35627..029f9c3e1c28 100644
--- a/net
e may
> fail, most of the examples in python documentation include
> handling of a NULL case.
>
> Signed-off-by: Ilya Maximets
Looks good to me!
Acked-by: Mike Pattrick
> ---
> python/ovs/_json.c | 10 +-
> 1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --
On Thu, Jun 23, 2022 at 6:32 AM yangchang wrote:
>
> change 'negotations' to 'negotiations'
>
> Signed-off-by: yangchang
Good catch!
Acked-by: Mike Pattrick
> ---
> lib/lacp.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/lib/lacp.h b
ding a new flag to recalculate the checksum only for last frag.
Should be "only for the first frag."
Acked-by: Mike Pattrick .
>
> Fixes: bc7a5acdff08 ("datapath: add ipv6 'set' action")
> Signed-off-by: Salem Sol
> ---
> lib/packets.c
this
patch enables the feature. However, Linux socket interface
remains disabled because the API doesn't allow enabling
those two features without enabling TSO too.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
v5:
- Added David's proposed
the
support is not enabled.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
v5:
- Refactor was mostly removed, except for valid->good
- Reset unsupported offload flags in send_prepare
- Moved send_prepare from process_upcall to netdev_upcall
From: Flavio Leitner
Add a new command to show the offloading features of
each data path port.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Reviewed-by: David Marchand
---
lib/dpif-netdev-unixctl.man | 6
lib/dpif-netdev.c | 58
From: Flavio Leitner
Document the implementation of netdev hardware offloading
in userspace datapath.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
Documentation/automake.mk| 1 +
Documentation/topics/index.rst | 1
This is a subset of the larger TSO patchset with various checksumming
improvements. This set includes additional documentation, new appctl
command "dpif-netdev/offload-show" to display interface offload
support, and improvements to tracking when an updated checksum is
required.
Flavio Leitner
It looks like one of the test files fell out of this patch, I'll
re-submit with that corrected.
-M
On Wed, Jul 13, 2022 at 2:14 PM Mike Pattrick wrote:
>
> From: Flavio Leitner
>
> The netdev receiving packets is supposed to provide the flags
> indicating if the IP cs
this
patch enables the feature. However, Linux socket interface
remains disabled because the API doesn't allow enabling
those two features without enabling TSO too.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
v5:
- Added David's proposed
From: Flavio Leitner
Add a new command to show the offloading features of
each data path port.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Reviewed-by: David Marchand
---
lib/dpif-netdev-unixctl.man | 6
lib/dpif-netdev.c | 58
the
support is not enabled.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
v5:
- Refactor was mostly removed, except for valid->good
- Reset unsupported offload flags in send_prepare
- Moved send_prepare from process_upcall to netdev_upc
From: Flavio Leitner
Document the implementation of netdev hardware offloading
in userspace datapath.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
Documentation/automake.mk| 1 +
Documentation/topics/index.rst | 1
This is a subset of the larger TSO patchset with various checksumming
improvements. This set includes additional documentation, new appctl
command "dpif-netdev/offload-show" to display interface offload
support, and improvements to tracking when an updated checksum is
required.
Flavio Leitner
On Thu, Jul 7, 2022 at 5:53 AM Amber, Kumar wrote:
>
> Hi Flavio, Mike,
>
>
>
> I did have a quick look over the patch-set
> https://patchwork.ozlabs.org/project/openvswitch/list/?series=307485
>
>
>
> I have some comments over the patches mentioned below:
>
> 1. Can a Cover letter
ad.
>
> setuptools < 59.0 doesn't have setuptools.errors and so, in this case,
> distutils.errors is still used.
>
> Signed-off-by: Timothy Redaelli
LGTM!
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.
On Thu, Jul 7, 2022 at 7:03 AM David Marchand wrote:
>
> Hello Mike,
>
>
> I did not review carefully yet.
> Just two quick comments.
>
>
> On Fri, Jul 1, 2022 at 5:58 AM Mike Pattrick wrote:
> >
> > From: Flavio Leitner
> >
> > This provides a
On Wed, Jul 6, 2022 at 9:02 AM David Marchand wrote:
>
> On Mon, Jul 4, 2022 at 10:24 PM David Marchand
> wrote:
> > > diff --git a/ofproto/ofproto-dpif-upcall.c b/ofproto/ofproto-dpif-upcall.c
> > > index 57f94df54..164738503 100644
> > > --- a/ofproto/ofproto-dpif-upcall.c
> > > +++
is not desired for some deployment.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
Documentation/topics/userspace-tso.rst | 21 ++--
NEWS | 4 +
lib/netdev-linux.c | 133 -
lib
From: Flavio Leitner
It seems csum is more common and shorter.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
Reviewed-by: David Marchand
--
v2: changed title per suggestion
---
lib/dp-packet.h | 72
this
patch enables the feature. However, Linux socket interface
remains disabled because the API doesn't allow enabling
those those features without enabling TSO too.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 16
enough memory and finally do the work.
Finally each batch is sent in order to the netdev.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/automake.mk | 2 +
lib/dp-packet-gso.c | 172
lib/dp
From: Flavio Leitner
Rename to better represent their flags.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
Reviewed-by: David Marchand
---
lib/conntrack.c| 4 ++--
lib/dp-packet.h| 14 +++---
lib/ipf.c
packets are dropped.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/dp-packet.c| 1 +
lib/dp-packet.h| 26 +++
lib/netdev-dpdk.c | 13 ++--
lib/netdev-linux.c | 83 ++
4 files
the
support is not enabled.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 12 ++---
lib/dp-packet.c | 12 +
lib/dp-packet.h | 70 +
lib/dpif.h
be no functional change with this patch.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
--
v2: Corrected missing conversion in packet.c
---
lib/dp-packet.c| 345 +++
lib/dp-packet.h| 504
From: Flavio Leitner
Add a new command to show the offloading features of
each data path port.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/dpif-netdev-unixctl.man | 5
lib/dpif-netdev.c | 58
From: Flavio Leitner
Document the implementation of netdev hardware offloading
in userspace datapath.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
Documentation/automake.mk | 1 +
Documentation/topics/index.rst| 1
From: Flavio Leitner
Rename to dp_packet_ol_tcp_seg, because that is less
redundant and allows other protocols.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
---
lib/dp-packet.h| 2 +-
lib/netdev-linux.c | 2 +-
lib
From: Flavio Leitner
This helps to identify when it is about the flags or
the packet itself.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
Reviewed-by: David Marchand
---
lib/conntrack.c | 8
lib/dp
From: Flavio Leitner
The name correlates better with the flag names.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
Reviewed-by: David Marchand
--
v2: changed title per suggestion
---
lib/conntrack.c| 8
lib
From: Flavio Leitner
Use the 'NETDEV_OFFLOAD_' prefix in the flags to indicate
we are talking about hardware offloading capabilities.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Reviewed-by: David Marchand
Acked-by: Maxime Coquelin
---
lib
From: Flavio Leitner
It seems csum is more common and shorter.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
Acked-by: Maxime Coquelin
Reviewed-by: David Marchand
--
v2: changed title per suggestion
---
lib/dp-packet.h | 72
401 - 500 of 664 matches
Mail list logo