cket will enter into another ipf context and causes errors.
>
> This happends when there are multiple CT zones, and frags are
> reassembled in ct(zone=X) might be 'stealed' into the ct(zone=Y).
>
> Signed-off-by: Peng He
> ---
After testing this a bit, it looks goo
On Tue, Dec 7, 2021 at 11:53 AM Flavio Leitner wrote:
>
> Currently 'p' and 'b' and used for packets, so use
> a convention that struct dp_packet is 'p' and
> struct dp_packet_batch is 'b'.
>
> Some comments needed new formatting to not pass the
> 80 column.
>
> Some variables were using 'p' or
On Tue, Dec 7, 2021 at 11:54 AM Flavio Leitner wrote:
>
> The netdev receiving packets is supposed to provide the flags
> indicating if the IP csum was verified and it is OK or BAD,
> otherwise the stack will check when appropriate by software.
>
> If the packet comes with good checksum, then
if the 32bit one has rolled over. This
patch also changes the non-empty check to use bytes, in case the 32bit
packet counter has rolled over for this update.
Fixes: f98e418fbd ("tc: Add tc flower functions")
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=1776816
Signed-off-by: Mik
redhat.com/show_bug.cgi?id=1776816
Signed-off-by: Mike Pattrick
---
lib/tc.c | 16 +++-
1 file changed, 15 insertions(+), 1 deletion(-)
diff --git a/lib/tc.c b/lib/tc.c
index 38a1dfc0e..c710e78e1 100644
--- a/lib/tc.c
+++ b/lib/tc.c
@@ -1702,6 +1702,11 @@ static const struct
e the sporadic failures after 100 iterations
of "conntrack - FTP over IPv6", but I still think this fix makes
sense, and shouldn't break anything.
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
y break the TCP conn_update() in a way that it returns CT_UPDATE_VALID
> without moving to the correct state leading to a false positive.
>
> Signed-off-by: Paolo Valerio
> ---
Looks good to me!
Acked-by: Mike Pattrick
___
dev mailing list
checksum if hwol is disabled or if DPDK
will not handle the IP checksum. This fixes the conntrack - IPv4 FTP
Passive with DNAT" test when run with check-system-tso.
Reported-by: Flavio Leitner
Signed-off-by: Mike Pattrick
---
lib/conntrack.c | 3 ++-
1 file changed, 2 insertions(+), 1 del
On Fri, Apr 8, 2022 at 6:35 AM Eelco Chaudron wrote:
>
>
>
> On 8 Apr 2022, at 11:18, Adrian Moreno wrote:
>
> > On 4/7/22 18:00, Mike Pattrick wrote:
> >> On Thu, Apr 7, 2022 at 7:26 AM Adrian Moreno wrote:
> >>>
> >>> Hi Mike,
> >
On Thu, Apr 7, 2022 at 7:26 AM Adrian Moreno wrote:
>
> Hi Mike,
>
> On 4/6/22 18:48, Mike Pattrick wrote:
> > Verify that a flow isn't adding more than 64kB in actions when an action
> > uses multiple check_pkt_larger statements. Failure to do so can cause
> >
Verify that a flow isn't adding more than 64kB in actions when an action
uses multiple check_pkt_larger statements. Failure to do so can cause
ovs-vswitchd to throw signal 6. Also add a depth check to clone and
check_pkt_larger.
Signed-off-by: Mike Pattrick
---
ofproto/ofproto-dpif-xlate.c | 26
have the side effect of leaving only the dpdk
transmit code under the txq lock.
Signed-off-by: Flavio Leitner
Reviewed-by: David Marchand
Signed-off-by: Mike Pattrick
---
V2:
- mentioned the tx lock change in the commit message.
- fixed packet leak when copy fails.
- moved pkt_cnt = cnt
://bugzilla.redhat.com/show_bug.cgi?id=2213892
Signed-off-by: Mike Pattrick
---
v2: Added additional checks if rule is removed
v3: v2 patch was corrupted in transit
v4: Added check against dpif flow stats
v5: Fixed typos, updated commit message
Changed timestamps to use datapath timestamp more
On Fri, Sep 1, 2023 at 9:14 AM Ilya Maximets wrote:
>
> On 9/1/23 07:06, Mike Pattrick wrote:
> > When the a revalidator thread is updating statistics for an XC_LEARN
> > xcache entry in xlate_push_stats_entry it uses ofproto_flow_mod_learn.
> > The revalidator will upd
://bugzilla.redhat.com/show_bug.cgi?id=2213892
Signed-off-by: Mike Pattrick
---
v2: Added additional checks if rule is removed
v3: v2 patch was corrupted in transit
v4: Added check against dpif flow stats
v5: Fixed typos, updated commit message
Changed timestamps to use datapath timestamp more
%|2%|
netdev | 7.39 Gbps | 4.95 Gbps | 6.24 Gbps |33%| 15%|
The ratios above are the percent reduction in total throughput when
mirroring is used either with or without a filter.
Signed-off-by: Mike Pattrick
---
v3:
- Added more tests
- Refactored mirror wildcard modification
On Fri, Aug 25, 2023 at 10:28 AM Adrian Moreno wrote:
>
>
>
> On 7/18/23 21:38, Mike Pattrick wrote:
> > Currently a bridge mirror will collect all packets and tools like
> > ovs-tcpdump can apply additional filters after they have already been
> > duplicate
On Tue, Sep 5, 2023 at 8:04 AM James Raphael Tiovalen
wrote:
>
> This commit adds a non-exhaustive list of tests for some of the
> functions declared in `lib/byteq`.
>
> These unit tests have been executed via `make check` and they
> successfully passed.
>
> Signed-off-by: James Raphael Tiovalen
ow an error. On the latest GCC version with `-Wc++-compat`,
> it would throw a warning.
>
> Signed-off-by: James Raphael Tiovalen
Did a quick "make check" with ubsan on to test for regressions, looks good!
Acked-by: Mike Pattrick
__
On Sun, Sep 3, 2023 at 7:44 AM James Raphael Tiovalen
wrote:
>
> In commit 40546cd6e51a53048d320fc0ed6a99f1713f5335, the assignment for
> the local `port` variable in the `add_port_to_cache` function was moved
> earlier. This commit reverts the initialization back to after the call
> to
On Mon, Oct 16, 2023 at 4:08 AM Roi Dayan via dev
wrote:
>
>
> On 16/10/2023 11:00, Roi Dayan wrote:
> >
> > On 16/10/2023 10:42, Eelco Chaudron wrote:
> >>
> >>
> >> On 16 Oct 2023, at 9:09, Roi Dayan wrote:
> >>
> >>> On 09/10/2023 15:05, Roi Dayan wrote:
> The cited commit fixed missing
; > Acked-by: Eelco Chaudron
>
> It seems OK as well. But another safer option would have been to move
> the trim_trailing_whitespace = true option in specific sections. Or
> completely remove it since it will be caught by checkpatch.
I think it
On Thu, Oct 26, 2023 at 3:40 AM Dexia Li via dev
wrote:
>
> For userspace datapath, this patch provides vxlan and geneve tunnel tso.
> Only support userspace vxlan or geneve tunnel, meanwhile support
> tunnel outter and inner csum offload. If netdev do not support offload
> features, there is
packets are dropped.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/dp-packet.c| 3 ++
lib/dp-packet.h| 26
lib/netdev-dpdk.c | 12 +++-
lib/netdev-linux.c | 76 --
4 files
Currently when userspace-tso is enabled, netdev-linux interfaces will
indicate support for all offload flags regardless of interface
configuration. This patch checks for which offload features are enabled
during netdev construction.
Signed-off-by: Mike Pattrick
--
v6:
- Removed legacy comment
is sent in order to the netdev.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
v4:
- Various formatting changes
- Fixed memory leak in soft-gso code if packet is flagged
for GSO but incorrectly lacks segment size.
v5:
- Corrected indentation
On Sun, Oct 29, 2023 at 9:59 PM Dexia Li wrote:
>
>
>
> -邮件原件-
> 发件人: Mike Pattrick
> 发送时间: 2023年10月27日 23:04
> 收件人: Dexia Li
> 抄送: ovs-dev@openvswitch.org
> 主题: Re: [ovs-dev] [PATCH v5] userspace: Support vxlan and geneve tso.
>
> On Thu, Oct 26,
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 | 168
lib/dp-packet-gso.h | 23 ++
lib/dp-packet.h | 7
packets are dropped.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/dp-packet.c| 3 ++
lib/dp-packet.h| 26
lib/netdev-dpdk.c | 12 +++-
lib/netdev-linux.c | 76 --
4 files
Currently when userspace-tso is enabled, netdev-linux interfaces will
indicate support for all offload flags regardless of interface
configuration. This patch checks for which offload features are enabled
during netdev construction.
Signed-off-by: Mike Pattrick
---
lib/netdev-linux.c | 156
On Mon, Oct 30, 2023 at 7:43 AM Simon Horman wrote:
>
> On Mon, Oct 30, 2023 at 02:58:36AM -0400, Mike Pattrick wrote:
> > Currently when userspace-tso is enabled, netdev-linux interfaces will
> > indicate support for all offload flags regardless of interface
> > configur
On Wed, Oct 18, 2023 at 5:50 AM Dexia Li wrote:
>
> For userspace datapath, this patch provides vxlan and geneve tunnel tso.
> Only support userspace vxlan or geneve tunnel, meanwhile support
> tunnel outter and inner csum offload. If netdev do not support offload
> features, there is a
On Fri, Oct 13, 2023 at 9:06 AM Ilya Maximets wrote:
>
> On 10/8/23 06:21, Mike Pattrick wrote:
> > Currently a bridge mirror will collect all packets and tools like
> > ovs-tcpdump can apply additional filters after they have already been
> > duplicated by v
On Fri, Oct 13, 2023 at 4:28 AM Eelco Chaudron wrote:
>
>
>
> On 8 Oct 2023, at 7:26, Mike Pattrick wrote:
>
> > Currently a bond will not always revalidate when an active member
> > changes. This can result in counter-intuitive behaviors like the fact
> > that
://bugzilla.redhat.com/show_bug.cgi?id=2214979
Signed-off-by: Mike Pattrick
---
v2: Added a test
v3: Made the test more reliable
v4: Made test much more reliable
---
ofproto/bond.c | 8 +--
tests/system-traffic.at | 50 +
2 files changed, 56 insertions(+), 2
ld you provide the error message with intel 710 ?
It's just the tx_error counter increasing.
-M
>
> Best regards
> Dexia
>
> -邮件原件-
> 发件人: Mike Pattrick
> 发送时间: 2023年9月27日 1:54
> 收件人: Dexia Li
> 抄送: ovs-dev@openvswitch.org; i.maxim...@ovn.org; david.march...@
On Mon, Oct 2, 2023 at 6:07 AM Simon Horman wrote:
>
> On Fri, Sep 08, 2023 at 12:28:24PM -0400, Mike Pattrick wrote:
> > Currently a bridge mirror will collect all packets and tools like
> > ovs-tcpdump can apply additional filters after they have already been
> >
%|2%|
netdev | 7.39 Gbps | 4.95 Gbps | 6.24 Gbps |33%| 15%|
The ratios above are the percent reduction in total throughput when
mirroring is used either with or without a filter.
Signed-off-by: Mike Pattrick
---
v3:
- Added more tests
- Refactored mirror wildcard
Previously the mirror_set() and mirror_get() functions took a large
number of parameters, which was inefficient and difficult to read and
extend. This patch moves most of the parameters into a struct.
Signed-off-by: Mike Pattrick
---
ofproto/ofproto-dpif-mirror.c | 61
%|2%|
netdev | 7.39 Gbps | 4.95 Gbps | 6.24 Gbps |33%| 15%|
The ratios above are the percent reduction in total throughput when
mirroring is used either with or without a filter.
Signed-off-by: Mike Pattrick
---
v3:
- Added more tests
- Refactored mirror wildcard modification
Previously the mirror_set() and mirror_get() functions took a large
number of parameters, which was inefficient and difficult to read and
extend. This patch moves most of the parameters into a struct.
Signed-off-by: Mike Pattrick
---
ofproto/ofproto-dpif-mirror.c | 61
mpatibility for various
> drivers.
> Only setting RTE_MBUF_F_TX_TCP_SEG can support compatibility for various
> drivers tso fuction.
>
>
>
> -邮件原件-
> 发件人: Dexia Li
> 发送时间: 2023年9月18日 17:47
> 收件人: David Marchand
> 抄送: ovs-dev@openvswitch.org; i.maxim...@ov
On Mon, Sep 25, 2023 at 9:19 PM Dexia Li wrote:
>
> For userspace datapath, this patch provides vxlan and geneve tunnel tso.
> Only support userspace vxlan or geneve tunnel, meanwhile support
> tunnel outter and inner csum offload. if netdev do not support offload
> features, there is a software
: James Raphael Tiovalen
Looks good!
Acked-by: Mike Pattrick
> ---
> Revisions:
>
> v1 -> v2: Incorporate Mike's suggestions.
> ---
> tests/automake.mk | 1 +
> tests/library.at | 5 ++
> tests/test-byteq.c | 143 +
&
%|2%|
netdev | 7.39 Gbps | 4.95 Gbps | 6.24 Gbps |33%| 15%|
The ratios above are the percent reduction in total throughput when
mirroring is used either with or without a filter.
Signed-off-by: Mike Pattrick
---
v3:
- Added more tests
- Refactored mirror wildcard
Previously the mirror_set() and mirror_get() functions took a large
number of parameters, which was inefficient and difficult to read and
extend. This patch moves most of the parameters into a struct.
Signed-off-by: Mike Pattrick
Acked-by: Simon Horman
Acked-by: Eelco Chaudron
---
v6
://bugzilla.redhat.com/show_bug.cgi?id=2214979
Signed-off-by: Mike Pattrick
---
v2: Added a test
v3: Made the test more reliable
---
ofproto/bond.c | 8 +--
tests/system-traffic.at | 50 +
2 files changed, 56 insertions(+), 2 deletions(-)
diff --git a/ofproto
On Thu, Oct 19, 2023 at 9:00 AM Eelco Chaudron wrote:
>
>
>
> On 19 Oct 2023, at 4:37, Mike Pattrick wrote:
>
> > Currently a bond will not always revalidate when an active member
> > changes. This can result in counter-intuitive behaviors like the fact
> > that
://bugzilla.redhat.com/show_bug.cgi?id=2214979
Signed-off-by: Mike Pattrick
---
v2: Added a test
v3: Made the test more reliable
v4: Made test much more reliable
v5: Improved test performance
---
ofproto/bond.c | 8 +--
tests/system-traffic.at | 48 +
2 files
nd warning.
> v4 -> v5: Add Eelco's suggestions to rename the test names to not refer
> to platform names.
> ---
Thanks for implementing the requested changes.
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Wed, Aug 16, 2023 at 4:18 PM Ilya Maximets wrote:
>
> On 8/9/23 19:00, Mike Pattrick wrote:
> > Currently a bond will not always revalidate when an active member
> > changes. This can result in counter-intuitive behaviors like the fact
> > that using ovs-appct
://bugzilla.redhat.com/show_bug.cgi?id=2214979
Signed-off-by: Mike Pattrick
---
v2: Added a test
---
ofproto/bond.c | 8 +--
tests/system-traffic.at | 50 +
2 files changed, 56 insertions(+), 2 deletions(-)
diff --git a/ofproto/bond.c b/ofproto/bond.c
index
On Mon, Aug 21, 2023 at 7:15 AM Ilya Maximets wrote:
>
> On 8/18/23 00:46, Mike Pattrick wrote:
> > When the a revalidator thread is updating statistics for an XC_LEARN
> > xcache entry in xlate_push_stats_entry it uses ofproto_flow_mod_learn.
> > The revalidator will upd
. If force is set to false and the rule has been replaced in the
classifier with a more recent rule, then ofproto_flow_mod_learn will
just return.
Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=2213892
Signed-off-by: Mike Pattrick
---
v2: Added additional checks if rule is removed
v3: v2
On Fri, Aug 11, 2023 at 6:56 PM Ilya Maximets wrote:
>
> On 8/4/23 19:58, Mike Pattrick wrote:
> > Diagnosing connectivity issues involving a bond can be complicated by a
> > lack of logging in LACP. It is difficult to determine the health of
> > sending and receving LACP
On Mon, Aug 28, 2023 at 4:19 AM Ilya Maximets wrote:
>
> On 8/25/23 08:12, Mike Pattrick wrote:
> > On Thu, Aug 24, 2023 at 6:01 PM Ilya Maximets wrote:
> >>
> >> On 8/22/23 22:45, Mike Pattrick wrote:
> >>> When the a revalidator thread is updating sta
://bugzilla.redhat.com/show_bug.cgi?id=2213892
Signed-off-by: Mike Pattrick
---
v2: Added additional checks if rule is removed
v3: v2 patch was corrupted in transit
v4: Added check against dpif flow stats
v5: Fixed typos, updated commit message
Changed timestamps to use datapath timestamp more
net/iavf. But to be on the safe
> side, this OVS patch here is a proposal to avoid hitting bugs in some
> other drivers,
>
I didn't test this thoroughly. But the documentation says that these
flags are only required if checksum/TSO is requested.
Acked-by: Mike Pattrick
___
://bugzilla.redhat.com/show_bug.cgi?id=2213892
Signed-off-by: Mike Pattrick
---
v2: Added additional checks if rule is removed
v3: v2 patch was corrupted in transit
v4: Added check against dpif flow stats
v5: Fixed typos, updated commit message
Changed timestamps to use datapath timestamp more
On Thu, Aug 24, 2023 at 6:01 PM Ilya Maximets wrote:
>
> On 8/22/23 22:45, Mike Pattrick wrote:
> > When the a revalidator thread is updating statistics for an XC_LEARN
> > xcache entry in xlate_push_stats_entry it uses ofproto_flow_mod_learn.
> > The revalidator will upd
On Mon, Mar 28, 2022 at 7:12 AM Eelco Chaudron wrote:
>
> This patch will properly initialize offload, as it requires the
> setting to be enabled before starting ovs-vswitchd (or do a
> restart once configured).
>
> Signed-off-by: Eelco Chaudron
> ---
> tests/ofproto-macros.at |6
On Mon, Mar 28, 2022 at 7:11 AM Eelco Chaudron wrote:
>
> This change implements support for the check_pkt_len
> action using the TC police action, which supports MTU
> checking.
>
> Signed-off-by: Eelco Chaudron
Looks good!
Acked
gger the same behavior.
>
> Signed-off-by: Paolo Valerio
This fixes the broken test for me.
Acked-by: Mike Pattrick
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
capitalization and periods
- Summarized info level hash transfers
- Added log if bond members were significantly deviant
Signed-off-by: Mike Pattrick
---
lib/lacp.c | 28 +++---
ofproto/bond.c | 102 +++--
2 files changed, 95 insertions(+), 35
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
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
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
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
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
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
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
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
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
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
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
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
, we now suppress these messages at
the INFO level and display exact byte count at the debug level.
Signed-off-by: Mike Pattrick
--
Since v4:
- Removed prefixes
- Removed unrequired conditional
Since v3:
- Removed log if bond members were significantly deviant
- Normalized log prefixes
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.
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
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
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
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
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
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
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
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
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
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
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
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
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
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
> > > +++
on: Unable to delete Mirror m_br-int
>
> Signed-off-by: Han Ding
I can confirm that this issue does occur, and this patch fixes it.
Acked-by: Mike Pattrick
Small nit but instead of closing and reopening the OVSDB object, maybe
we should just atexit.register the call to ovsdb.close_idl.
Ch
be no functional change with this patch.
Signed-off-by: Flavio Leitner
Co-authored-by: Mike Pattrick
Signed-off-by: Mike Pattrick
---
lib/dp-packet.c| 345 +++
lib/dp-packet.h| 504 ++---
lib/netdev-dummy.c | 8 +-
lib/netdev
This patch refactors several function, variable, and definition names
for consistency and brevity.
cksum is changed to csum, p is the default name for a dp_packet, b is
the default name for a dp_packet_batch, ol instead of hwol, and some
checksum offload related functions are renamed to indicate
101 - 200 of 724 matches
Mail list logo