release candidate 22.07-rc4

DISCUSSION Jiang, YuX 4 participants 6 messages Jul 12, 2022 – Jul 17, 2022 1 message not in the archive

lore

Share & tools

lore

A message that is not in the archive (sent off-list or not yet received): 6125682.jrt8l4j9yI@thomas

Jiang, YuX

author Reviewed-by Jiayu Hu Reviewed-by Chenbo Xia Reviewed-by Maxime Coquelin Tested-by Xingguang He

2022-07-12 09:32 UTC #1 lore raw

Quoted 20 lines

-----Original Message-----
From: Thomas Monjalon <[email protected]>
Sent: Tuesday, July 12, 2022 6:37 AM
To: [email protected]
Subject: release candidate 22.07-rc4
A new DPDK release candidate is ready for testing:
      https://git.dpdk.org/dpdk/tag/?id=v22.07-rc4
There are 22 new patches in this snapshot.
Release notes:
      https://doc.dpdk.org/guides/rel_notes/release_22_07.html
This is the last release candidate.
Only documentation should be updated before the release.
Reviews of deprecation notices are required:
https://patches.dpdk.org/bundle/dmarchand/deprecation_notices/

Quoted 7 lines

You may share some release validation results by replying to this message at [email protected] and by adding tested hardware in the release notes.
Please think about sharing your roadmap now for DPDK 22.11.
Thank you everyone

Update the test status for Intel part. Till now dpdk22.07-rc4 regression is running smooth.
But we find one high issue about "[dpdk-22.07]loopback_virtio_user_server_mode/server_mode_reconnect_with_split_all_path_payload_check: the first packet payload is different from the ohter packets in the same queue", Bad commit id is cd79d1b030bd69df31fd401d1b48c23df0f71d00(which is used to fix the performance drop of split ring)

commit cd79d1b030bd69df31fd401d1b48c23df0f71d00 (HEAD)

Author: Xuan Ding <[email protected]>
Date:   Thu Jul 7 06:55:13 2022 +0000

vhost: fix unnecessary dirty page logging

The dirty page logging is only required in vhost enqueue direction for live migration. This patch removes the unnecessary dirty page logging in vhost dequeue direction. Otherwise, it will result in a performance drop. Some if-else judgements are also optimized to improve performance.

Fixes 6d823bb302c7 ("vhost: prepare sync for descriptor to mbuf refactoring")

Fixes b6eee3e83402 ("vhost: fix sync dequeue offload")

Signed-off-by Xuan Ding <[email protected]>

Reviewed-by Jiayu Hu <[email protected]>

Reviewed-by Chenbo Xia <[email protected]>

Reviewed-by Maxime Coquelin <[email protected]>

Tested-by Xingguang He <[email protected]>

Best regards,
Yu Jiang

Xia, Chenbo

2022-07-12 09:44 UTC #2 lore raw

Quoted 40 lines

-----Original Message-----
From: Jiang, YuX <[email protected]>
Sent: Tuesday, July 12, 2022 5:33 PM
To: Thomas Monjalon <[email protected]>; dev ([email protected]) <[email protected]>
Cc: Devlin, Michelle <[email protected]>; Mcnamara, John <[email protected]>; Ferruh Yigit <[email protected]>; Xia, Chenbo <[email protected]>
Subject: RE: release candidate 22.07-rc4
Quoted 14 lines
-----Original Message-----
From: Thomas Monjalon <[email protected]>
Sent: Tuesday, July 12, 2022 6:37 AM
To: [email protected]
Subject: release candidate 22.07-rc4
A new DPDK release candidate is ready for testing:
	https://git.dpdk.org/dpdk/tag/?id=v22.07-rc4
There are 22 new patches in this snapshot.
Release notes:
	https://doc.dpdk.org/guides/rel_notes/release_22_07.html
Quoted 7 lines
This is the last release candidate.
Only documentation should be updated before the release.
Reviews of deprecation notices are required:
https://patches.dpdk.org/bundle/dmarchand/deprecation_notices/
You may share some release validation results by replying to this
message at
Quoted 6 lines
[email protected] and by adding tested hardware in the release notes.
Please think about sharing your roadmap now for DPDK 22.11.
Thank you everyone
Hi

Quoted 8 lines

Update the test status for Intel part. Till now dpdk22.07-rc4 regression is running smooth.
But we find one high issue about "[dpdk- 22.07]loopback_virtio_user_server_mode/server_mode_reconnect_with_split_al
l_path_payload_check: the first packet payload is different from the ohter packets in the same queue",
Bad commit id is cd79d1b030bd69df31fd401d1b48c23df0f71d00(which is used to fix the performance drop of split ring)

I just found the potential root cause, Maxime helped confirm it and patch will be out soon.

Thanks,
Chenbo

Quoted 25 lines

commit cd79d1b030bd69df31fd401d1b48c23df0f71d00 (HEAD)
Author: Xuan Ding <[email protected]>
Date:   Thu Jul 7 06:55:13 2022 +0000
vhost: fix unnecessary dirty page logging
The dirty page logging is only required in vhost enqueue direction for live migration. This patch removes the unnecessary dirty page logging in vhost dequeue direction. Otherwise, it will result in a performance drop. Some if-else judgements are also optimized to improve
performance.
Fixes 6d823bb302c7 ("vhost: prepare sync for descriptor to mbuf
refactoring")
Fixes b6eee3e83402 ("vhost: fix sync dequeue offload")
Signed-off-by Xuan Ding <[email protected]>
Reviewed-by Jiayu Hu <[email protected]>
Reviewed-by Chenbo Xia <[email protected]>
Reviewed-by Maxime Coquelin <[email protected]>
Tested-by Xingguang He <[email protected]>
Best regards,
Yu Jiang

Jiang, YuX

author

2022-07-14 08:18 UTC #3 lore raw

Quoted 62 lines

-----Original Message-----
From: Xia, Chenbo <[email protected]>
Sent: Tuesday, July 12, 2022 5:45 PM
To: Jiang, YuX <[email protected]>; Thomas Monjalon <[email protected]>; dev ([email protected]) <[email protected]>; Maxime Coquelin <[email protected]>
Cc: Devlin, Michelle <[email protected]>; Mcnamara, John <[email protected]>; Ferruh Yigit <[email protected]>
Subject: RE: release candidate 22.07-rc4
Quoted 36 lines
-----Original Message-----
From: Jiang, YuX <[email protected]>
Sent: Tuesday, July 12, 2022 5:33 PM
To: Thomas Monjalon <[email protected]>; dev ([email protected]) <[email protected]>
Cc: Devlin, Michelle <[email protected]>; Mcnamara, John <[email protected]>; Ferruh Yigit <[email protected]>; Xia, Chenbo <[email protected]>
Subject: RE: release candidate 22.07-rc4
Quoted 14 lines
-----Original Message-----
From: Thomas Monjalon <[email protected]>
Sent: Tuesday, July 12, 2022 6:37 AM
To: [email protected]
Subject: release candidate 22.07-rc4
A new DPDK release candidate is ready for testing:
	https://git.dpdk.org/dpdk/tag/?id=v22.07-rc4
There are 22 new patches in this snapshot.
Release notes:
	https://doc.dpdk.org/guides/rel_notes/release_22_07.html
Quoted 7 lines
This is the last release candidate.
Only documentation should be updated before the release.
Reviews of deprecation notices are required:
https://patches.dpdk.org/bundle/dmarchand/deprecation_notices/
You may share some release validation results by replying to this
message at
Quoted 4 lines
[email protected] and by adding tested hardware in the release notes.
Please think about sharing your roadmap now for DPDK 22.11.
Quoted 8 lines
Quoted 2 lines
Thank you everyone
Hi
Update the test status for Intel part. Till now dpdk22.07-rc4 regression is running smooth.
But we find one high issue about "[dpdk-
22.07]loopback_virtio_user_server_mode/server_mode_reconnect_with_s pli
Quoted 5 lines
t_al
l_path_payload_check: the first packet payload is different from the ohter packets in the same queue", Bad commit id is cd79d1b030bd69df31fd401d1b48c23df0f71d00(which is used to fix the performance drop of split ring)

Quoted 6 lines

I just found the potential root cause, Maxime helped confirm it and patch will be out soon.
Thanks,
Chenbo

All,

Update the test status for Intel part. Till now dpdk22.07-rc4 regression is almost finished, two new bugs are found.
The high issue about "[dpdk-22.07]loopback_virtio_user_server_mode/server_mode_reconnect_with_split_all_path_payload_check: the first packet payload is different from the ohter packets in the same queue".
It has fix link: http://patchwork.dpdk.org/project/dpdk/patch/[email protected]/, Intel validation team has completed validtion test for the patch, no new issue is found.
Other one issue about "vswitch_sample_cbdma:perf_pvp_vhost_async_operation_with_cbdma_2_channels: virtio-user will drop all the packets when sending large 1280 Byte packets".

	It has fix link: https://patches.dpdk.org/project/dpdk/patch/[email protected]/ and tested-by by Intel, but no acked-by yet.

# Basic Intel(R) NIC testing
* Build or compile:
*Build: cover the build test combination with latest GCC/Clang version and the popular OS revision such as Ubuntu20.04, Ubuntu22.04, Fedora36, RHEL8.6 etc.
- All test passed.

* PF/VF(i40e, ixgbe): test scenarios including PF/VF-RTE_FLOW/TSO/Jumboframe/checksum offload/VLAN/VXLAN, etc.
- All test done. No new issue is found

* PF/VF(ice): test scenarios including Switch features/Package Management/Flow Director/Advanced Tx/Advanced RSS/ACL/DCF/Flexible Descriptor, etc.
- All test done. No new issue is found.

* Intel NIC single core/NIC performance: test scenarios including PF/VF single core performance test, RFC2544 Zero packet loss performance test, etc.
- All test done. No new issue is found.

# Basic cryptodev and virtio testing
* Virtio: both function and performance test are covered. Such as PVP/Virtio_loopback/virtio-user loopback/virtio-net VM2VM perf testing/VMAWARE ESXI 7.0u3, etc.
- All test done. Find two new bugs as above.

* Cryptodev:
*Function test: test scenarios including Cryptodev API testing/CompressDev ISA-L/QAT/ZLIB PMD Testing/FIPS, etc.
- All test done. No new new isue is found.
*Performance test: test scenarios including Throughput Performance/Cryptodev Latency, etc.
- All test done. No new new isue is found.

Best regards,
Yu Jiang

Quoted 25 lines

Quoted 25 lines
commit cd79d1b030bd69df31fd401d1b48c23df0f71d00 (HEAD)
Author: Xuan Ding <[email protected]>
Date:   Thu Jul 7 06:55:13 2022 +0000
vhost: fix unnecessary dirty page logging
The dirty page logging is only required in vhost enqueue direction for live migration. This patch removes the unnecessary dirty page logging in vhost dequeue direction. Otherwise, it will result in a performance drop. Some if-else judgements are also optimized to improve
performance.
Fixes 6d823bb302c7 ("vhost: prepare sync for descriptor to mbuf
refactoring")
Fixes b6eee3e83402 ("vhost: fix sync dequeue offload")
Signed-off-by Xuan Ding <[email protected]>
Reviewed-by Jiayu Hu <[email protected]>
Reviewed-by Chenbo Xia <[email protected]>
Reviewed-by Maxime Coquelin <[email protected]>
Tested-by Xingguang He <[email protected]>
Best regards,
Yu Jiang

Jiang, YuX

author

2022-07-15 07:59 UTC #4 lore raw

Quoted 81 lines

-----Original Message-----
From: Jiang, YuX <[email protected]>
Sent: Thursday, July 14, 2022 4:18 PM
To: Xia, Chenbo <[email protected]>; Thomas Monjalon <[email protected]>; dev ([email protected]) <[email protected]>; Maxime Coquelin <[email protected]>
Cc: Devlin, Michelle <[email protected]>; Mcnamara, John <[email protected]>; Ferruh Yigit <[email protected]>
Subject: RE: release candidate 22.07-rc4
Quoted 55 lines
-----Original Message-----
From: Xia, Chenbo <[email protected]>
Sent: Tuesday, July 12, 2022 5:45 PM
To: Jiang, YuX <[email protected]>; Thomas Monjalon <[email protected]>; dev ([email protected]) <[email protected]>; Maxime Coquelin <[email protected]>
Cc: Devlin, Michelle <[email protected]>; Mcnamara, John <[email protected]>; Ferruh Yigit <[email protected]>
Subject: RE: release candidate 22.07-rc4
Quoted 44 lines
-----Original Message-----
From: Jiang, YuX <[email protected]>
Sent: Tuesday, July 12, 2022 5:33 PM
To: Thomas Monjalon <[email protected]>; dev ([email protected]) <[email protected]>
Cc: Devlin, Michelle <[email protected]>; Mcnamara, John <[email protected]>; Ferruh Yigit <[email protected]>; Xia, Chenbo <[email protected]>
Subject: RE: release candidate 22.07-rc4
Quoted 14 lines
-----Original Message-----
From: Thomas Monjalon <[email protected]>
Sent: Tuesday, July 12, 2022 6:37 AM
To: [email protected]
Subject: release candidate 22.07-rc4
A new DPDK release candidate is ready for testing:
	https://git.dpdk.org/dpdk/tag/?id=v22.07-rc4
There are 22 new patches in this snapshot.
Release notes:
	https://doc.dpdk.org/guides/rel_notes/release_22_07.html
Quoted 7 lines
This is the last release candidate.
Only documentation should be updated before the release.
Reviews of deprecation notices are required:
https://patches.dpdk.org/bundle/dmarchand/deprecation_notices/
You may share some release validation results by replying to this
message at
Quoted 6 lines
[email protected] and by adding tested hardware in the release notes.
Please think about sharing your roadmap now for DPDK 22.11.
Thank you everyone
Hi
Update the test status for Intel part. Till now dpdk22.07-rc4 regression is running smooth.
But we find one high issue about "[dpdk-
22.07]loopback_virtio_user_server_mode/server_mode_reconnect_with_s
Quoted 7 lines
pli
Quoted 5 lines
t_al
l_path_payload_check: the first packet payload is different from the ohter packets in the same queue", Bad commit id is cd79d1b030bd69df31fd401d1b48c23df0f71d00(which is used to fix the performance drop of split ring)
Quoted 6 lines
I just found the potential root cause, Maxime helped confirm it and patch will be out soon.
Thanks,
Chenbo
All,

Quoted 10 lines

Update the test status for Intel part. Till now dpdk22.07-rc4 regression is almost finished, two new bugs are found.
The high issue about "[dpdk- 22.07]loopback_virtio_user_server_mode/server_mode_reconnect_with_s plit_all_path_payload_check: the first packet payload is different from the ohter packets in the same queue".
It has fix link:
http://patchwork.dpdk.org/project/dpdk/patch/20220712100447.104162-1-
[email protected]/, Intel validation team has completed validtion test for the patch, no new issue is found.

This patch has been merged into 22.07 main branch, and test passed by Intel Validation team.

Quoted 8 lines

Other one issue about "vswitch_sample_cbdma:perf_pvp_vhost_async_operation_with_cbdma_2 _channels: virtio-user will drop all the packets when sending large 1280 Byte packets".
It has fix link:
https://patches.dpdk.org/project/dpdk/patch/20220714051106.1134222-1-
[email protected]/ and tested-by by Intel, but no acked-by yet.

As discussed by Chenbo and maintainer in yesterday's release meeting, this patch should have minor impact, so the fix could be moved to next release.

BR
Yu Jiang

Quoted 62 lines

# Basic Intel(R) NIC testing
* Build or compile:
*Build: cover the build test combination with latest GCC/Clang version and
the popular OS revision such as Ubuntu20.04, Ubuntu22.04, Fedora36, RHEL8.6 etc.
- All test passed.
* PF/VF(i40e, ixgbe): test scenarios including PF/VF- RTE_FLOW/TSO/Jumboframe/checksum offload/VLAN/VXLAN, etc.
- All test done. No new issue is found
* PF/VF(ice): test scenarios including Switch features/Package Management/Flow Director/Advanced Tx/Advanced RSS/ACL/DCF/Flexible Descriptor, etc.
- All test done. No new issue is found.
* Intel NIC single core/NIC performance: test scenarios including PF/VF single core performance test, RFC2544 Zero packet loss performance test, etc.
- All test done. No new issue is found.
# Basic cryptodev and virtio testing
* Virtio: both function and performance test are covered. Such as PVP/Virtio_loopback/virtio-user loopback/virtio-net VM2VM perf testing/VMAWARE ESXI 7.0u3, etc.
- All test done. Find two new bugs as above.
* Cryptodev:
*Function test: test scenarios including Cryptodev API testing/CompressDev
ISA-L/QAT/ZLIB PMD Testing/FIPS, etc.
- All test done. No new new isue is found.
*Performance test: test scenarios including Throughput
Performance/Cryptodev Latency, etc.
- All test done. No new new isue is found.
Best regards,
Yu Jiang
Quoted 25 lines
Quoted 25 lines
commit cd79d1b030bd69df31fd401d1b48c23df0f71d00 (HEAD)
Author: Xuan Ding <[email protected]>
Date:   Thu Jul 7 06:55:13 2022 +0000
vhost: fix unnecessary dirty page logging
The dirty page logging is only required in vhost enqueue direction for live migration. This patch removes the unnecessary dirty page logging in vhost dequeue direction. Otherwise, it will result in a performance drop. Some if-else judgements are also optimized to improve
performance.
Fixes 6d823bb302c7 ("vhost: prepare sync for descriptor to mbuf
refactoring")
Fixes b6eee3e83402 ("vhost: fix sync dequeue offload")
Signed-off-by Xuan Ding <[email protected]>
Reviewed-by Jiayu Hu <[email protected]>
Reviewed-by Chenbo Xia <[email protected]>
Reviewed-by Maxime Coquelin <[email protected]>
Tested-by Xingguang He <[email protected]>
Best regards,
Yu Jiang

Thinh Tran

2022-07-15 14:18 UTC #5 lore raw

In reply to a message not in the archive

IBM - Power Systems
DPDK v22.07-rc4-1-gb3c781dbde

* Basic PF on Mellanox: No new issues or regressions were seen.
* Performance: not tested.
* OS: RHEL 8.5 kernel: 4.18.0-372.9.1.el8.ppc64le

        with gcc version 8.5.0 20210514 (Red Hat 8.5.0-10)
      RHEL 9.0  kernel: 5.14.0-70.13.1.el9_0.ppc64le
        with gcc version 11.2.1 20220127 (Red Hat 11.2.1-9)

Systems tested:
- IBM Power9 PowerNV 9006-22P
NICs:
- Mellanox Technologies MT28800 Family [ConnectX-5 Ex]
- firmware version: 16.33.1400
- MLNX_OFED_LINUX-5.6-1.0.3.6

- LPARs on IBM Power10 CHRP IBM,9105-22A
NICs:
- Mellanox Technologies MT28800 Family [ConnectX-5 Ex]
- firmware version: 16.33.1400
- MLNX_OFED_LINUX-5.6-1.0.3.6

Regards,
Thinh Tran

Quoted 23 lines · Thomas Monjalon

A new DPDK release candidate is ready for testing:
	https://git.dpdk.org/dpdk/tag/?id=v22.07-rc4
There are 22 new patches in this snapshot.
Release notes:
	https://doc.dpdk.org/guides/rel_notes/release_22_07.html
This is the last release candidate.
Only documentation should be updated before the release.
Reviews of deprecation notices are required:
https://patches.dpdk.org/bundle/dmarchand/deprecation_notices/
You may share some release validation results by replying to this message at [email protected] and by adding tested hardware in the release notes.
Please think about sharing your roadmap now for DPDK 22.11.
Thank you everyone

Wael Abualrub

2022-07-17 14:33 UTC #6 lore raw

In reply to a message not in the archive

Hi,

Quoted 20 lines

-----Original Message-----
From: Thomas Monjalon <[email protected]>
Sent: Tuesday, July 12, 2022 1:37 AM
To: [email protected]
Subject: release candidate 22.07-rc4
A new DPDK release candidate is ready for testing:
	https://git.dpdk.org/dpdk/tag/?id=v22.07-rc4
There are 22 new patches in this snapshot.
Release notes:
	https://doc.dpdk.org/guides/rel_notes/release_22_07.html
This is the last release candidate.
Only documentation should be updated before the release.
Reviews of deprecation notices are required:
https://patches.dpdk.org/bundle/dmarchand/deprecation_notices/

Quoted 7 lines

You may share some release validation results by replying to this message at [email protected] and by adding tested hardware in the release notes.
Please think about sharing your roadmap now for DPDK 22.11.
Thank you everyone

The following is a list of tests that we ran on NVIDIA hardware this release:

- Basic functionality:
Send and receive multiple types of traffic.
- testpmd xstats counter test.
- testpmd timestamp test.
- Changing/checking link status through testpmd.
- RTE flow tests:
Items:
- conntrack
- ecpri
- eth
- flex
- geneve
- geneve_opt
- gre
- gre_key
- gre_option
- gtp
- gtp_psc
- icmp
- icmp6
- integrity
- ipv4
- ipv6
- ipv6_frag_ext
- mark
- meta
- mpls
- nvgre
- phy_port
- port_id
- tag
- tcp
- udp
- vlan
- vxlan
- vxlan_gpe
- esp

Actions:
- age
- conntrack
- count
- dec_tcp_ack
- dec_tcp_seq
- dec_ttl
- drop
- flag
- inc_tcp_ack
- inc_tcp_seq
- jump
- mark
- meter
- modify_field
- nvgre_decap
- nvgre_encap
- of_pop_vlan
- of_push_vlan
- of_set_vlan_pcp
- of_set_vlan_vid
- port_id
- queue
- raw_decap
- raw_encap
- represented_port
- rss
- sample
- set_ipv4_dscp
- set_ipv4_dst
- set_ipv4_src
- set_ipv6_dscp
- set_ipv6_dst
- set_ipv6_src
- set_mac_dst
- set_mac_src
- set_meta
- set_tag
- set_tp_dst
- set_tp_src
- set_ttl
- vxlan_decap
- vxlan_encap
- Some RSS tests.
- VLAN filtering, stripping and insertion tests.
- Checksum and TSO tests.
- ptype tests.
- link_status_interrupt example application tests.
- l3fwd-power example application tests.
- Multi-process example applications tests.
- Hardware LRO tests.
- Regex application tests.
- Buffer Split tests.
- Tx scheduling tests.

PS: We are internally investigating a few test failures, and we'll send fixes if necessary.

Kindest Regards,
Wael Abualrub