Re: [ovs-dev] [PATCH v1] ofp-packet: Silence the 'may be uninitialized' warning.

On Fri, Feb 25, 2022 at 4:43 PM Gregory Rose wrote: > > > > On 2/25/2022 12:02 PM, Mike Pattrick wrote: > > GCC 11.2.1-2.2 emits a false-positive warnings like: > > > > lib/ofp-packet.c: In function 'ofputil_decode_packet_in': > > lib/ofp-packet.c:155:25: warning: 'reason' may be used > >

Re: [ovs-dev] [PATCH v1] ofp-packet: Silence the 'may be uninitialized' warning.

On 2/25/2022 12:02 PM, Mike Pattrick wrote: GCC 11.2.1-2.2 emits a false-positive warnings like: lib/ofp-packet.c: In function 'ofputil_decode_packet_in': lib/ofp-packet.c:155:25: warning: 'reason' may be used uninitialized [-Wmaybe-uninitialized] lib/ofp-packet.c: In function

Re: [ovs-dev] [PATCH ovn v3 02/16] tests: don't bail from OVN_CHECK_PACKETS_CONTAIN prematurily

On Thu, Feb 17, 2022 at 10:18 AM Ihar Hrachyshka wrote: > > When the expected number of packets arrive, OVN_CHECK_PACKETS_CONTAIN > bails out. But it should bail out from waiting only when the expected > packets were observed or when timeout reached, because the whole point > of the function is

Re: [ovs-dev] [PATCH ovn v3 01/16] tests: log more info on OVN_CHECK_PACKETS* failure

On Thu, Feb 17, 2022 at 10:17 AM Ihar Hrachyshka wrote: > > Right now it logs diff between expected and observed only when enough > packets arrived. If less than expected packets arrived, only the > number of packets observed was logged, but not the packets themselves. > This patch makes the

Re: [ovs-dev] [PATCH ovn v2] ovn.at: Fix flaky tests "VLAN transparency, passthru=true, multiple hosts"

On Fri, Feb 4, 2022 at 11:03 AM Xavier Simonart wrote: > > Tests were waiting for ports to be reported up before sending packets. > However, waiting for both ports to be up is not enough to guarantee > that all flows are installed for both ports. > We now wait for flows outputting to patch port

Re: [ovs-dev] [PATCH] ovs-monitor-ipsec: Add force-encapsulation option to force NAT-T

On Thu, 2022-01-20 at 16:33 +0100, Andreas Karis wrote: > Both LibreSwan and OpenSwan allow administrators to unconditionally > force enable NAT-T for ESP. This may help to surmount restrictive > firewalls in scenarios where IP protocol number 50 is blocked, but > where > NAT autodetection fails.

Re: [ovs-dev] [PATCH ovn 2/2] Prepare for post-22.03.0

On 2/24/22 19:00, Numan Siddique wrote: On Thu, Feb 24, 2022 at 5:00 PM Mark Michelson wrote: Signed-off-by: Mark Michelson Acked-by: Numan Siddique Numan I pushed patch 1 to main and branch-22.03 and pushed patch 2 to main. --- NEWS | 3 +++ configure.ac | 2 +-

Re: [ovs-dev] [PATCH ovn] pinctrl: give a 'no such name' response when no DNS record is available

Hi, Can you please fix the below compilation issue and submit another version ? Thanks Numan On Tue, Feb 22, 2022 at 7:39 AM 0-day Robot wrote: > > Bleep bloop. Greetings 294754599--- via dev, I am a robot and I have tried > out your patch. > Thanks for your contribution. > > I encountered

Re: [ovs-dev] ovs/ipsec: StrongSwan report connection update failures to ovs logs

On Mon, Feb 7, 2022 at 2:15 PM Mohammad Heib wrote: > > Currently when the user adds an IPsec tunnel port to the > ovs bridge the ovs-monitor-ipsec script will add this tunnel > IPsec-related configuration to the appropriate file and > submit a request to start the IPsec connection for this port

Re: [ovs-dev] [PATCH ovn v2] Add LTS section to release documentation.

I pushed this to the main branch. On 1/27/22 20:01, Han Zhou wrote: On Thu, Jan 20, 2022 at 1:40 PM Mark Michelson > wrote: > > OVN LTS releases have a lot of ambiguity, so this is intended to codify > LTS support and cadence. > > Signed-off-by: Mark

Re: [ovs-dev] [PATCH ovn v2] northd: Document --dummy-numa option.

On 2/9/22 14:04, Numan Siddique wrote: On Tue, Feb 8, 2022 at 12:30 PM Mark Michelson wrote: Signed-off-by: Mark Michelson LGTM. Acked-by: Numan Siddique Numan I pushed this to the main branch. --- v1 -> v2: * Fixed a mismatched tag in the XML --- northd/ovn-northd.8.xml | 49

Re: [ovs-dev] [PATCH ovn v4] northd: Add feature to log reply and related ACL traffic.

On 2/24/22 21:07, Han Zhou wrote: On Thu, Feb 24, 2022 at 10:10 AM Mark Michelson > wrote: > > It can be desirable for replies to stateful ACLs to be logged. And in > some cases, it can actually be a bit confusing why they aren't logged. > Consider a situation

Re: [ovs-dev] [PATCH v6 ovn 2/2] controller: reconfigure ovs meters for ovn meters updates

On 2/25/22 15:08, Mark Michelson wrote: On 2/25/22 13:44, Han Zhou wrote: On Fri, Feb 25, 2022 at 9:15 AM Lorenzo Bianconi < lorenzo.bianc...@redhat.com> wrote: At the moment ovs meters are reconfigured by ovn just when a meter is allocated or removed while updates for an already allocated

Re: [ovs-dev] [PATCH ovn] Documentation: Add submodule documentation

On 2/25/22 13:22, Mark Michelson wrote: On 2/25/22 01:48, Han Zhou wrote: On Thu, Feb 24, 2022 at 9:31 AM Mark Michelson > wrote:  >  > This provides some missing documentation about the OVS submodule.  >  > It also proposes a currently-unimplemented policy

Re: [ovs-dev] [PATCH v6 ovn 2/2] controller: reconfigure ovs meters for ovn meters updates

On 2/25/22 13:44, Han Zhou wrote: On Fri, Feb 25, 2022 at 9:15 AM Lorenzo Bianconi < lorenzo.bianc...@redhat.com> wrote: At the moment ovs meters are reconfigured by ovn just when a meter is allocated or removed while updates for an already allocated meter are ignored. This issue can be easily

[ovs-dev] [PATCH v1] ofp-packet: Silence the 'may be uninitialized' warning.

GCC 11.2.1-2.2 emits a false-positive warnings like: lib/ofp-packet.c: In function 'ofputil_decode_packet_in': lib/ofp-packet.c:155:25: warning: 'reason' may be used uninitialized [-Wmaybe-uninitialized] lib/ofp-packet.c: In function 'ofputil_decode_packet_in_private':

Re: [ovs-dev] [PATCH v3 08/14] treewide: Avoid offsetting NULL pointers.

On 2/25/22 18:29, Jeffrey Walton wrote: >> @@ -270,7 +277,7 @@ static inline struct ofpbuf *ofpbuf_from_list(const >> struct ovs_list *list) >> static inline bool ofpbuf_equal(const struct ofpbuf *a, const struct ofpbuf >> *b) >> { >> return a->size == b->size && >> -

Re: [ovs-dev] [PATCH v6 ovn 2/2] controller: reconfigure ovs meters for ovn meters updates

On Fri, Feb 25, 2022 at 9:15 AM Lorenzo Bianconi < lorenzo.bianc...@redhat.com> wrote: > > At the moment ovs meters are reconfigured by ovn just when a > meter is allocated or removed while updates for an already > allocated meter are ignored. This issue can be easily verified > with the following

Re: [ovs-dev] [PATCH ovn] ovs: Bump submodule to stable branch-2.17.

On Fri, Feb 25, 2022 at 1:20 PM Mark Michelson wrote: > > On 2/25/22 10:26, Ilya Maximets wrote: > > On 2/25/22 16:19, Numan Siddique wrote: > >> On Fri, Feb 25, 2022 at 2:00 AM Han Zhou wrote: > >>> > >>> On Thu, Feb 24, 2022 at 3:07 PM wrote: > > From: Numan Siddique > >

Re: [ovs-dev] [PATCH v6 ovn 0/2] reconfigure ovs meters for ovn meters updates

Thanks for the work on this Lorenzo. Acked-by: Mark Michelson On 2/25/22 12:15, Lorenzo Bianconi wrote: Changes since v5: - add size parameter to extend_table in order to reduce the size Changes since v4: - add offset parameter to ovn_extend_table_init - rebase on top of ovn master Changes

Re: [ovs-dev] [PATCH ovn] Documentation: Add submodule documentation

On 2/25/22 01:48, Han Zhou wrote: On Thu, Feb 24, 2022 at 9:31 AM Mark Michelson > wrote: > > This provides some missing documentation about the OVS submodule. > > It also proposes a currently-unimplemented policy wherein we attempt > to base OVN releases on a

Re: [ovs-dev] [PATCH ovn] ovs: Bump submodule to stable branch-2.17.

On 2/25/22 10:26, Ilya Maximets wrote: On 2/25/22 16:19, Numan Siddique wrote: On Fri, Feb 25, 2022 at 2:00 AM Han Zhou wrote: On Thu, Feb 24, 2022 at 3:07 PM wrote: From: Numan Siddique The following commits are required for OVN: 498cedc48 reconnect: Fix broken inactivity probe if

[ovs-dev] [PATCH v4 7/7] ci: Add UB Sanitizer.

Note: There still is an UB instance when using SSE4.2 as reported here: https://mail.openvswitch.org/pipermail/ovs-dev/2022-January/390904.html Acked-by: Aaron Conole Signed-off-by: Dumitru Ceara --- v4: - Rebased. v3: - Fix typo in variable name. - Added SSE4.2 UB note to commit log. ---

[ovs-dev] [PATCH v4 6/7] util: Avoid UB when iterating collections.

UB Sanitizer was flagging OBJECT_CONTAINING() as undefined behavior due to the way it's used when iterating through collections, e.g., HMAP_FOR_EACH_INIT(). However, in such cases we don't actually dereference the pointer, we just use its value. Avoid the undefined behavior by casting to 'void

[ovs-dev] [PATCH v6 ovn 2/2] controller: reconfigure ovs meters for ovn meters updates

At the moment ovs meters are reconfigured by ovn just when a meter is allocated or removed while updates for an already allocated meter are ignored. This issue can be easily verified with the following reproducer: $ovn-nbctl meter-add meter0 drop 10 pktps $ovn-nbctl --log --meter=meter0 acl-add

[ovs-dev] [PATCH v6 ovn 1/2] lib: limit table_id space for meters

This is a preliminary patch to move metering management in IP engine. Signed-off-by: Lorenzo Bianconi --- controller/ovn-controller.c | 5 +++-- lib/extend-table.c | 22 +++--- lib/extend-table.h | 4 +++- tests/ovn-controller.at | 6 +++--- tests/ovn.at

[ovs-dev] [PATCH v6 ovn 0/2] reconfigure ovs meters for ovn meters updates

Changes since v5: - add size parameter to extend_table in order to reduce the size Changes since v4: - add offset parameter to ovn_extend_table_init - rebase on top of ovn master Changes since v3: - move full meter management in the IP engine Changes since v2: - move meter ip into lflow ip -

[ovs-dev] [PATCH v4 5/7] ofp-actions: Use aligned structures when decoding ofp actions.

Some openflow actions can be misaligned, e.g., actions whithin OF 1.0 replies to statistics reply messages which have a header of 12 bytes and no additional padding. Also, buggy controllers might incorrectly encode actions. When decoding multiple actions in ofpacts_decode(), make sure that when

[ovs-dev] [PATCH v4 4/7] ofp-errors: Ensure parsed OFPT_ERROR messages are properly aligned.

Trim the ofpbuf to ensure proper alignment. UB Sanitizer report: lib/ofp-print.c:1218:24: runtime error: member access within misaligned address 0x019229d2 for type 'const struct ofp_header', which requires 4 byte alignment 0x019229d2: note: pointer points here 00 00 5a 5a 05 22

[ovs-dev] [PATCH v4 3/7] ofp-actions: Ensure aligned accesses when setting masked fields.

For example is parsing the OVN "eth.dst[40] = 1;" action, which triggered the following warning from UndefinedBehaviorSanitizer: lib/meta-flow.c:3210:9: runtime error: member access within misaligned address 0x00de4e36 for type 'const union mf_value', which requires 8 byte alignment

[ovs-dev] [PATCH v4 2/7] treewide: Avoid offsetting NULL pointers.

This is undefined behavior and was reported by UB Sanitizer: lib/meta-flow.c:3445:16: runtime error: member access within null pointer of type 'struct vl_mf_field' #0 0x6aad0f in mf_get_vl_mff lib/meta-flow.c:3445 #1 0x6d96d7 in mf_from_oxm_header lib/nx-match.c:260 #2

[ovs-dev] [PATCH v4 1/7] treewide: Fix invalid bit shift operations.

UB Sanitizer reports: tests/test-hash.c:59:40: runtime error: shift exponent 64 is too large for 64-bit type 'long unsigned int' #0 0x44c3c9 in get_range128 tests/test-hash.c:59 #1 0x44cb2e in check_hash_bytes128 tests/test-hash.c:178 #2 0x44d14d in test_hash_main

[ovs-dev] [PATCH v4 0/7] Fix UndefinedBehaviorSanitizer reported issues and enable it in CI.

As privately reported by Aaron Conole, and by Jeffrey Walton [0] there's currently a number of undefined behavior instances in the OVS code base. Running the OVS (and OVN) tests with UBSan [1] enabled uncovers these. This series fixes the issues reported by UBSan and, through the last patch,

Re: [ovs-dev] [PATCH v3 08/14] treewide: Avoid offsetting NULL pointers.

On 2/15/22 00:04, Ilya Maximets wrote: > On 1/24/22 15:19, Dumitru Ceara wrote: >> This is undefined behavior and was reported by UB Sanitizer: >> lib/meta-flow.c:3445:16: runtime error: member access within null pointer >> of type 'struct vl_mf_field' >> #0 0x6aad0f in mf_get_vl_mff

Re: [ovs-dev] [PATCH v3 04/14] treewide: Fix invalid bit shift operations.

On 2/14/22 23:04, Ilya Maximets wrote: > On 1/24/22 15:18, Dumitru Ceara wrote: >> UB Sanitizer reports: >> tests/test-hash.c:59:40: runtime error: shift exponent 64 is too large for >> 64-bit type 'long unsigned int' >> #0 0x44c3c9 in get_range128 tests/test-hash.c:59 >> #1

Re: [ovs-dev] [PATCH v3 ovn] northd: introduce exclude-lb-vips-from-garp option for lsp

On Fri, Feb 25, 2022 at 9:25 AM Lorenzo Bianconi wrote: > > The CMS can configure the same lbs on multiple gw routers so the > same VIPs will be adverised by multiple routers through GARPs. > In order to fix the issue, introduce exclude-lb-vips-from-garp > option in the logical_switch_port table

Re: [ovs-dev] [PATCH] tests: Ignore log about failing to set NETLINK_EXT_ACK.

On 25 Feb 2022, at 16:38, Dumitru Ceara wrote: > Since 4a6a4734622e ("netlink-socket: Log extack error messages in > netlink transactions."), tests fail on older systems that don't support > NETLINK_EXT_ACK. It's not really an issue, so we can just ignore the > log. > > CC: Paolo Valerio >

Re: [ovs-dev] [PATCH v2 18/18] python: add unit tests for filtering engine

On 25 Feb 2022, at 16:59, Eelco Chaudron wrote: > On 28 Jan 2022, at 17:04, Adrian Moreno wrote: > >> Add unit test for OFFilter class. >> >> Signed-off-by: Adrian Moreno >> -- > > Changes look good to me. > > Acked-by: Eelco Chaudron Forgot to mention that this concludes my review of this

Re: [ovs-dev] [PATCH v2 18/18] python: add unit tests for filtering engine

On 28 Jan 2022, at 17:04, Adrian Moreno wrote: > Add unit test for OFFilter class. > > Signed-off-by: Adrian Moreno > -- Changes look good to me. Acked-by: Eelco Chaudron ___ dev mailing list d...@openvswitch.org

Re: [ovs-dev] [PATCH v2 17/18] python: add unit tests to datapath parsing

On 28 Jan 2022, at 17:04, Adrian Moreno wrote: > Signed-off-by: Adrian Moreno > --- Changes look good to me. Acked-by: Eelco Chaudron ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

[ovs-dev] [PATCH] tests: Ignore log about failing to set NETLINK_EXT_ACK.

Since 4a6a4734622e ("netlink-socket: Log extack error messages in netlink transactions."), tests fail on older systems that don't support NETLINK_EXT_ACK. It's not really an issue, so we can just ignore the log. CC: Paolo Valerio CC: Eelco Chaudron Signed-off-by: Dumitru Ceara ---

Re: [ovs-dev] [PATCH] ovsdb: raft: Fix inability to join the cluster after interrupted attempt.

On 2/22/22 15:34, Dumitru Ceara wrote: > On 1/28/22 19:51, Ilya Maximets wrote: >> If the joining server re-connects while catching up (e.g. if it crashed >> or connection got closed due to inactivity), the data we sent might be >> lost, so the server will never reply to append request or a

Re: [ovs-dev] [PATCH v2 14/18] python: introduce unit tests

On 28 Jan 2022, at 17:04, Adrian Moreno wrote: > Use pytest to run unit tests as part of the standard testsuite. > > Signed-off-by: Adrian Moreno The changes look good to me. Acked-by: Eelco Chaudron ___ dev mailing list d...@openvswitch.org

Re: [ovs-dev] [PATCH ovn] ovs: Bump submodule to stable branch-2.17.

On 2/25/22 16:19, Numan Siddique wrote: > On Fri, Feb 25, 2022 at 2:00 AM Han Zhou wrote: >> >> On Thu, Feb 24, 2022 at 3:07 PM wrote: >>> >>> From: Numan Siddique >>> >>> The following commits are required for OVN: >>> >>> 498cedc48 reconnect: Fix broken inactivity probe if there is no other

Re: [ovs-dev] [PATCH ovn] ovs: Bump submodule to stable branch-2.17.

On Fri, Feb 25, 2022 at 2:00 AM Han Zhou wrote: > > On Thu, Feb 24, 2022 at 3:07 PM wrote: > > > > From: Numan Siddique > > > > The following commits are required for OVN: > > > > 498cedc48 reconnect: Fix broken inactivity probe if there is no other > reason to wake up. > > 8b5136a89

[ovs-dev] [PATCH v3 ovn] northd: introduce exclude-lb-vips-from-garp option for lsp

The CMS can configure the same lbs on multiple gw routers so the same VIPs will be adverised by multiple routers through GARPs. In order to fix the issue, introduce exclude-lb-vips-from-garp option in the logical_switch_port table in order to advertise in GARPs packets all SNAT/DNAT configured in

Re: [ovs-dev] [RFC] tc: use pedit add actions for ipv4 and ipv6 dec ttl

On 25 Feb 2022, at 13:57, Pieter Jansen van Vuuren wrote: > On 25/02/2022 10:08, Simon Horman wrote: >> Hi Pieter, >> >> nice to hear from you :) > > Hi Simon. Thank you for the insightful feedback. > >> >> On Tue, Feb 22, 2022 at 12:17:40PM +, pieter.jansen-van-vuu...@amd.com >> wrote:

Re: [ovs-dev] [PATCH v1 0/3] Fix offload rule flush race condition

On 2/16/22 16:18, Gaëtan Rivet wrote: > On Wed, Feb 16, 2022, at 14:15, Ilya Maximets wrote: >> On 2/4/22 17:12, Gaetan Rivet wrote: >>> A race condition has been identified during datapath destruction, >>> along with the port offload flushes issued. >>> >>> This series addresses these race

Re: [ovs-dev] [PATCH v5 2/2] ofproto-dpif: avoid unneccesary backer revalidation

On 15 Feb 2022, at 5:34, lic121 wrote: > If lldp didn't change, we are not supposed to trigger backer > revalidation. > Without this patch, bridge_reconfigure() always trigger udpif > revalidator because of lldp. > > This patch also fix lldp memory leak bug: > lldp_create() malloc memory for

Re: [ovs-dev] [RFC] tc: use pedit add actions for ipv4 and ipv6 dec ttl

On 25/02/2022 10:08, Simon Horman wrote: > Hi Pieter, > > nice to hear from you :) Hi Simon. Thank you for the insightful feedback. > > On Tue, Feb 22, 2022 at 12:17:40PM +, pieter.jansen-van-vuu...@amd.com > wrote: >> From: Pieter Jansen van Vuuren >> >> Previously when making use of

Re: [ovs-dev] [PATCH v5 1/2] ofproto-dpif: trigger revalidation when ipfix config set

On 15 Feb 2022, at 5:33, lic121 wrote: > Currently, ipfix conf creation/deletion don't trigger dpif backer > revalidation. This is not expected, as we need the revalidation > to commit ipfix into xlate. So that xlate can generate ipfix > actions. > > This patch covers only new

Re: [ovs-dev] [PATCH v3 8/9] netdev-offload-tc: Check for none offloadable ct_state flag combination

On Thu, Feb 24, 2022 at 03:20:23PM +0100, Eelco Chaudron wrote: > > > On 23 Feb 2022, at 17:23, Marcelo Ricardo Leitner wrote: > > > On Tue, Feb 22, 2022 at 04:26:10PM +0100, Eelco Chaudron wrote: > >> This patch checks for none offloadable ct_state match flag combinations. > >> If they exist

Re: [ovs-dev] [PATCH v2] tc: Keep header rewrite actions order

On 2022-02-25 1:46 PM, Eelco Chaudron wrote: On 20 Feb 2022, at 14:07, Roi Dayan wrote: On 2022-02-20 1:47 PM, Roi Dayan wrote: This is leaving the revalidators in some kind of a loop. Without the patch, I do not see the problem. This is a log snippet:

Re: [ovs-dev] [PATCH v1 2/2] tests/mfex-test: Introduce non datapath testing for avx512.

Bleep bloop. Greetings Kumar Amber, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. build: depbase=`echo tests/test-jsonrpc.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\ gcc -std=gnu99

[ovs-dev] [PATCH v5 5/5] dpif-netdev/mfex: Avoid hashing when opt mfex called.

This patch avoids calculating the software hash of the packet again if the optimized miniflow-extract hit and has already calculated the packet hash. In cases of scalar miniflow extract, the normal hashing calculation is performed. Signed-off-by: Kumar Amber --- lib/dpif-netdev-avx512.c | 6

[ovs-dev] [PATCH v5 4/5] dpif-netdev/mfex: Add ipv6 profile based hashing.

This commit adds IPv6 profile specific hashing which uses fixed offsets into the packet to improve hashing perforamnce. Hash value is autovalidated by MFEX autovalidator. Signed-off-by: Kumar Amber Signed-off-by: Harry van Haaren Co-authored-by: Harry van Haaren --- v4: - Use pre-defined

[ovs-dev] [PATCH v5 3/5] dpif-netdev/mfex: Add ipv4 profile based hashing.

This commit adds IPv4 profile specific hashing which uses fixed offsets into the packet to improve hashing perforamnce. Signed-off-by: Kumar Amber Signed-off-by: Harry van Haaren Co-authored-by: Harry van Haaren --- v4: - Use pre-defined hash length values. v3: - Fix check-patch sign-offs.

[ovs-dev] [PATCH v5 2/5] dpif-netdev/mfex: Add packet hash check to autovalidator.

This patch adds the per profile AVX512 opt hashing to autovalidator for validating the hash values against the scalar hash. Signed-off-by: Kumar Amber --- lib/dpif-netdev-private-extract.c | 12 1 file changed, 12 insertions(+) diff --git a/lib/dpif-netdev-private-extract.c

[ovs-dev] [PATCH v5 1/5] dpif-netdev/mfex: Add AVX512 ipv6 traffic profiles

Add AVX512 Ipv6 optimized profile for vlan/IPv6/UDP and vlan/IPv6/TCP, IPv6/UDP and IPv6/TCP. MFEX autovalidaton test-case already has the IPv6 support for validating against the scalar mfex. Signed-off-by: Kumar Amber Signed-off-by: Harry van Haaren Co-authored-by: Harry van Haaren --- v5:

[ovs-dev] [PATCH v5 0/5] MFEX Optimizations IPv6 + Hashing Optimizations

The patchset introuduces IPv6 optimized MFEX profiles with AVX512 which can deliver upto 20% to 30% gain in performance over the existing scalar data-path. Hashing Optimization are also included which can further improve performance by approximately 10%. --- v5: - Add Ipv6 and TCP packet length

Re: [ovs-dev] [PATCH v3 00/17] Fix undefined behavior in loop macros

On 2/16/22 15:27, Adrian Moreno wrote: > When running builds with UBSan, some undefined behavior was detected in the > iteration of common data data structures in OVS. > Coincidentally, a bug was reported [1] whose root cause was another, this > time undetected, undefined behavior in the

Re: [ovs-dev] [PATCH v3 17/17] sset: add SHORT version of SAFE loop macros

On 2/16/22 15:28, Adrian Moreno wrote: > Add SHORT version of SAFE loop macros and overload the current macro > name to keep backwards compatibility. > > Signed-off-by: Adrian Moreno > --- Acked-by: Dumitru Ceara ___ dev mailing list

Re: [ovs-dev] [PATCH v3 15/17] idlc: support short version of SAFE macros

On 2/16/22 15:27, Adrian Moreno wrote: > In order to be consistent with the rest of the SAFE loop macros, > overload each of the generated *_SAFE macro with a SHORT version that > does not require the user to provide the NEXT variable. > > Signed-off-by: Adrian Moreno > --- Acked-by: Dumitru

Re: [ovs-dev] [PATCH v3 14/17] vtep: use _SAFE iterator if freeing the iterator

On 2/16/22 15:27, Adrian Moreno wrote: > The _SAFE version of the iterator allows the user to free the iterator > structure safely. > > Signed-off-by: Adrian Moreno > --- > vtep/vtep-ctl.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/vtep/vtep-ctl.c

Re: [ovs-dev] [PATCH v3 12/17] hindex: remove the next variable in safe loops

On 2/16/22 15:27, Adrian Moreno wrote: > Using SHORT version of the *_SAFE loops makes the code cleaner and less > error prone. So, use the SHORT version and remove the extra variable > when possible for HINDEX_*_SAFE. > > In order to be able to use both long and short versions without changing >

Re: [ovs-dev] [PATCH v3 13/17] rculist: use multi-variable helpers for loop macros

On 2/16/22 15:27, Adrian Moreno wrote: > Use multi-variable iteration helpers to rewrite rculist loops macros. > > There is an important behavior change compared with the previous > implementation: When the loop ends normally (i.e: not via "break;"), > the object pointer provided by the user is

Re: [ovs-dev] [PATCH v3 11/17] hindex: use multi-variable iterators

On 2/16/22 15:27, Adrian Moreno wrote: > Re-write hindex's loops using multi-variable helpers. > > For safe loops, use the LONG version to maintain backwards > compatibility. > > Signed-off-by: Adrian Moreno > --- Acked-by: Dumitru Ceara ___ dev

Re: [ovs-dev] [PATCH v3 10/17] cmap: use multi-variable iterators

On 2/16/22 15:27, Adrian Moreno wrote: > Re-write cmap's loops using multi-variable helpers. > > For iterators based on cmap_cursor, we just need to make sure the NODE > variable is not used after the loop, so we set it to NULL. > > Signed-off-by: Adrian Moreno > --- Acked-by: Dumitru Ceara

Re: [ovs-dev] [PATCH v3 09/17] hmap: use short version of safe loops if possible

On 2/16/22 15:27, Adrian Moreno wrote: > Using SHORT version of the *_SAFE loops makes the code cleaner and less > error prone. So, use the SHORT version and remove the extra variable > when possible for hmap and all its derived types. > > In order to be able to use both long and short versions

Re: [ovs-dev] [PATCH v3 08/17] hmap: implement UB-safe hmap pop iterator

On 2/16/22 15:27, Adrian Moreno wrote: > HMAP_FOR_EACH_POP iterator has an additional difficulty, which is the > use of two iterator variables of different types. > > In order to re-write this loop in a UB-safe manner, create a iterator > struct to be used as loop variable. > > Signed-off-by:

Re: [ovs-dev] [PATCH v3 06/17] list: use short version of safe loops if possible

On 2/16/22 15:27, Adrian Moreno wrote: > Using the SHORT version of the *_SAFE loops makes the code cleaner > and less error-prone. So, use the SHORT version and remove the extra > variable when possible. > > In order to be able to use both long and short versions without changing > the name of

Re: [ovs-dev] [PATCH v3 07/17] hmap: use multi-variable helpers for hmap loops

On 2/16/22 15:27, Adrian Moreno wrote: > Rewrite hmap's loops using multi-variable helpers. > > For SAFE loops, use the LONG version of the multi-variable macros to > keep backwards compatibility. > > Signed-off-by: Adrian Moreno > --- A couple tiny nits, with those addressed: Acked-by:

Re: [ovs-dev] [PATCH v3 04/17] list: use multi-variable helpers for list loops

On 2/16/22 15:27, Adrian Moreno wrote: > Use multi-variable iteration helpers to rewrite non-safe loops. > > There is an important behavior change compared with the previous > implementation: When the loop ends normally (i.e: not via "break;"), the > object pointer provided by the user is NULL.

Re: [ovs-dev] [PATCH v3 05/17] list: ensure iterator is NULL after pop loop

On 2/16/22 15:27, Adrian Moreno wrote: > After the loop ends, the iterator is not guaranteed to point to a valid > object and should not be used. Make it NULL to avoid undefined behavior. > > Signed-off-by: Adrian Moreno > --- I think this could be squashed together with patch 04/17. For the

Re: [ovs-dev] [PATCH v3 03/17] util: add helpers to overload SAFE macro

On 2/16/22 15:27, Adrian Moreno wrote: > Having both LONG and SHORT versions of the SAFE macros with different > names is not very convenient. Add helpers that facilitate overloading > such macros using a single name. > > In order to work around a known issue in MSVC [1], an indirection layer >

Re: [ovs-dev] [PATCH v3 02/17] util: add safe multi-variable iterators

On 2/16/22 15:27, Adrian Moreno wrote: > Safe version of multi-variable iterator helpers declare an internal > variable to store the next value of the iterator temporarily. > > Two versions of the macro are provided, one that still uses the NEXT > variable for backwards compatibility and a

Re: [ovs-dev] [PATCH v3 01/17] util: add multi-variable loop iterator macros

On 2/16/22 15:27, Adrian Moreno wrote: > Multi-variable loop iterators avoid potential undefined behavior by > using an internal iterator variable to perform the iteration and only > referencing the containing object (via OBJECT_CONTAINING) if the > iterator has been validated via the second

[ovs-dev] [PATCH v1 1/2] tests/mfex: Improve pcap script for complex testing traffic

The mfex pcap generation script is improved for multi-length traffic and also removed hard coded mfex_pcap instead use the script to generate complex traffic for testing. Signed-off-by: Kumar Amber --- tests/automake.mk | 1 - tests/mfex_fuzzy.py | 55

[ovs-dev] [PATCH v1 2/2] tests/mfex-test: Introduce non datapath testing for avx512.

The following patch adds non datapath AVX512 testing for MFEX in c format. Signed-off-by: Kumar Amber --- lib/dpif-netdev-private-extract.c | 4 +- tests/automake.mk | 1 + tests/library.at | 7 tests/test-mfex.c | 70

[ovs-dev] [PATCH v1 0/2] MFEX datapath independent Tests

The Patchset Introduces creating of varied lenght packets for testing MFEX AVX512 implementations as well as Introduces unit-tests which are independet from ovs-datapath. This also allows us to test a lot more scenarios testing only MFEX. Kumar Amber (2): tests/mfex: Improve pcap script for

Re: [ovs-dev] [PATCH v2] tc: Keep header rewrite actions order

On 20 Feb 2022, at 14:07, Roi Dayan wrote: > On 2022-02-20 1:47 PM, Roi Dayan wrote: >>> This is leaving the revalidators in some kind of a loop. Without the patch, >>> I do not see the problem. This is a log snippet: >>> >>> 2022-02-17T15:38:52.990Z|00050|tc|WARN|expected act csum with

Re: [ovs-dev] [PATCH v3] tc: Keep header rewrite actions order

On 20 Feb 2022, at 15:47, Roi Dayan wrote: > From: Chris Mi > > Currently, tc merges all header rewrite actions into one tc pedit > action. So the header rewrite actions order is lost. Save each header > rewrite action into one tc pedit action to keep the order. And only > append one tc csum

Re: [ovs-dev] [PATCH net-next v8] net: openvswitch: IPv6: Add IPv6 extension header support

Hello: This patch was applied to netdev/net-next.git (master) by David S. Miller : On Wed, 23 Feb 2022 16:54:09 -0800 you wrote: > This change adds a new OpenFlow field OFPXMT_OFB_IPV6_EXTHDR and > packets can be filtered using ipv6_ext flag. > > Signed-off-by: Toms Atteka > Acked-by: Pravin B

Re: [ovs-dev] [PATCH ovn 0/2] ci: ovn-kubernetes: Make tests run again.

On 2/24/22 23:51, Numan Siddique wrote: > On Thu, Feb 24, 2022 at 11:06 AM Dumitru Ceara wrote: >> This series addresses two independent issues: >> - ovn-kubernetes needs to be built against the NB/SB schema defined in >> the OVN tree we're running the test for. >> - ovn-kubernetes bumped go

Re: [ovs-dev] [RFC] tc: use pedit add actions for ipv4 and ipv6 dec ttl

Hi Pieter, nice to hear from you :) On Tue, Feb 22, 2022 at 12:17:40PM +, pieter.jansen-van-vuu...@amd.com wrote: > From: Pieter Jansen van Vuuren > > Previously when making use of the tc datapath to achieve decrement ttl, > we would install a filter that matches on the ttl/hoplimit field

Re: [ovs-dev] [PATCH v2 ovn] northd: introduce nat-only-router for nat-addresses option

On Feb 24, Numan Siddique wrote: > On Thu, Feb 24, 2022 at 5:23 PM Lorenzo Bianconi > wrote: > > > > The CMS can configure the same lbs on multiple gw routers so the > > same IPs/mac bindings will be adverised by multiple routers > > through GARPs. > > In order to fix the issue, introduce

[ovs-dev] [PATCH v5 ovn 2/2] controller: reconfigure ovs meters for ovn meters updates

At the moment ovs meters are reconfigured by ovn just when a meter is allocated or removed while updates for an already allocated meter are ignored. This issue can be easily verified with the following reproducer: $ovn-nbctl meter-add meter0 drop 10 pktps $ovn-nbctl --log --meter=meter0 acl-add

[ovs-dev] [PATCH v5 ovn 1/2] lib: limit table_id space for meters

This is a preliminary patch to move metering management in IP engine. Signed-off-by: Lorenzo Bianconi --- controller/ovn-controller.c | 4 ++-- lib/extend-table.c | 6 -- lib/extend-table.h | 3 ++- tests/ovn-controller.at | 6 +++--- tests/ovn.at|

[ovs-dev] [PATCH v5 ovn 0/2] reconfigure ovs meters for ovn meters updates

Changes since v4: - add offset parameter to ovn_extend_table_init - rebase on top of ovn master Changes since v3: - move full meter management in the IP engine Changes since v2: - move meter ip into lflow ip - add new test in ovn-performance.at - manage metering ip full-recompute Changes since