[cherry-pick 202511] test_qos_sai update for Cisco G200 (#20844)#23970
Open
[cherry-pick 202511] test_qos_sai update for Cisco G200 (#20844)#23970
Conversation
<!-- Please make sure you've read and understood our contributing guidelines; https://github.com/sonic-net/SONiC/blob/gh-pages/CONTRIBUTING.md Please provide following information to help code review process a bit easier: --> <!-- - Please include a summary of the change and which issue is fixed. - Please also include relevant motivation and context. Where should reviewer start? background context? - List any dependencies that are required for this change. --> Summary: Fixes # (issue) Fixed test_qos_sai failures for Cisco G200. 202505 PR: #21162 <!-- - Fill x for your type of change. - e.g. - [x] Bug fix --> - [ ] Bug fix - [ ] Testbed and Framework(new/improvement) - [ ] New Test case - [ ] Skipped for non-supported platforms - [x] Test case improvement - [ ] 202205 - [ ] 202305 - [ ] 202311 - [ ] 202405 - [ ] 202411 - [x] 202505 To fix test_qos_sai failures for Cisco G200. 1, Drop threshold update 1) For egress lossy drop threshold, updated the formula with egress_pool_reserved_memory and profile_reserved_memory, update lossy drop threshold to next mantissa value - 40 buffers. 2) testQosSaiPfcXoffLimit test result flakiness, pkt num to trigger drop is 174379~174380, not 174378, increased lossless_drop_tuning_pkts by 1. 2, testQosSaiPgSharedWatermark 1) Clear PG watermark by debug shell to workaround counterpoll issue. 2) Lossless: Increase upper bound margin to 6, added lower bound margin 3. 3) Lossy: Increase upper bound margin to 14, lower bound margin 6. 3, testQosSaiQWatermarkAllPorts, decreased pkts_num_leak_out from 800 to 200, decrease pkt_count by 8 since lossy drop threshold is not accurate. 4, Other margin adjustments 1) testQosSaiPfcXonLimit, Increased margin to 6. 2) testQosSaiLossyQueue, increase pkt margin to 8. 3) testQosSaiQSharedWatermark lossless, increase pkts_num_margin to 2. 4) testQosSaiQSharedWatermark lossy, increase pkts_num_margin to 9. 5) testQosSaiBufferPoolWatermark lossy, increase extra_cap_margin from 20 to 25. 5, Skipped DWRR cases. G200 SDK implementation changed, it doesn't support port level api set_credit_eir_or_pir, it supports set_credit_pir on each OQ but that is not applicable for WRR test. 6, Masked the failure of ip6tables cmd in ptf, ptf container doesn’t have ip6tables tool installed. The initial test result shows 6 cases passed, 12 failed, 1 error. With this PR, test_qos_sai.py 100% passed. ``` ----------------------------------------- generated xml file: /run_logs/qos/test_qos_sai.xml ------------------------------------------ INFO:root:Can not get Allure report URL. Please check logs ------------------------------------------------------- live log sessionfinish -------------------------------------------------------- 26/09/2025 17:53:15 __init__.pytest_terminal_summary L0067 INFO | Can not get Allure report URL. Please check logs ======================================================= short test summary info ======================================================= PASSED qos/test_qos_sai.py::TestQosSai::testParameter[single_asic] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPfcXoffLimit[single_asic-xoff_1] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPfcXoffLimit[single_asic-xoff_2] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPfcXonLimit[single_asic-xon_1] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPfcXonLimit[single_asic-xon_2] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiBufferPoolWatermark[single_asic-wm_buf_pool_lossless] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiBufferPoolWatermark[single_asic-wm_buf_pool_lossy] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiLossyQueue[single_asic] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiDscpQueueMapping[single_asic] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPgSharedWatermark[single_asic-wm_pg_shared_lossless] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPgSharedWatermark[single_asic-wm_pg_shared_lossy] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiPgHeadroomWatermark[single_asic] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiQSharedWatermark[single_asic-wm_q_shared_lossless] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiQSharedWatermark[single_asic-wm_q_shared_lossy] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiDscpToPgMapping[single_asic] PASSED qos/test_qos_sai.py::TestQosSai::testQosSaiQWatermarkAllPorts[single_asic-wm_q_wm_all_ports] SKIPPED [2] qos/test_qos_sai.py:420: Additional DSCPs are not supported on non-dual ToR ports SKIPPED [4] qos/test_qos_sai.py:482: This test is only for Mellanox. SKIPPED [2] qos/test_qos_sai.py:692: Additional DSCPs are not supported on non-dual ToR ports SKIPPED [4] qos/test_qos_sai.py:775: Unsupported testbed type. SKIPPED [1] qos/test_qos_sai.py:879: Headroom pool size is not enabled on this DUT SKIPPED [2] qos/test_qos_sai.py:979: Shared reservation size test is not supported. SKIPPED [1] qos/test_qos_sai.py:1091: Headroom pool size is not enabled on this DUT SKIPPED [2] qos/test_qos_sai.py:1318: Unsupported testbed type. SKIPPED [2] qos/test_qos_sai.py:1501: Skip this test since separated DSCP_TO_TC_MAP is not applied SKIPPED [1] qos/test_qos_sai.py: Dot1p-queue mapping is only supported on backend. SKIPPED [1] qos/test_qos_sai.py: Dot1p-PG mapping is only supported on backend. SKIPPED [1] qos/test_qos_sai.py: Skip DWRR test on Cisco G200 platform. SKIPPED [3] qos/test_qos_sai.py: Unsupported testbed type. SKIPPED [2] qos/test_qos_sai.py:2047: For DSCP to PG mapping on IPinIP traffic , mellanox device has different behavior to community. For mellanox device, testQosSaiDscpToPgMapping can cover the scenarios / Unsupported testbed type. SKIPPED [2] qos/test_qos_sai.py:2142: Skip this test since separated DSCP_TO_TC_MAP is not applied SKIPPED [1] qos/test_qos_sai.py: Skip DWRR weight change test on Mellanox platform. / Unsupported testbed type. SKIPPED [9] qos/test_qos_sai.py:2502: testQosSaiXonHysteresis case is only supported on cisco-8000 8102/8101/8111. ======================================= 16 passed, 40 skipped, 1 warning in 2612.96s (0:43:32) ======================================== sonic@sonic-ucs-m6-23:/data/tests$ ``` Cisco 8122. <!-- (If it's a new feature, new test case) Did you update documentation/Wiki relevant to your implementation? Link to the wiki page? --> --------- Signed-off-by: Zhixin Zhu <zhixzhu@cisco.com>
Collaborator
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
Contributor
Author
Cherry-pick Conflict Resolution SummaryThere was only 1 conflict in the entire cherry-pick:
|
| Side | Line content |
|---|---|
| 202511 branch (HEAD) | "topo_name not in (constants['QOS_SAI_TOPO'] + ['t2_single_node_max', 't2_single_node_min']) and asic_type not in ['mellanox']" |
| PR #20844 (incoming) | "topo_name not in constants['QOS_SAI_TOPO'] and asic_type not in ['mellanox']" + new line "asic_type in ['cisco-8000'] and platform.startswith('x86_64-8122_')" |
Resolution: Kept the 202511 version of the existing line (with the extra t2_single_node topologies) and added the new Cisco G200 skip condition from PR #20844. Final result:
- "topo_name not in (constants['QOS_SAI_TOPO'] + ['t2_single_node_max', 't2_single_node_min']) and asic_type not in ['mellanox']"
- "asic_type in ['cisco-8000'] and platform.startswith('x86_64-8122_')"The other 5 files (ptfhost_utils.py, qos_param_generator.py, qos_sai_base.py, test_qos_sai.py, sai_qos_tests.py) all merged cleanly with no conflicts.
sonumm
approved these changes
Apr 17, 2026
Contributor
sonumm
left a comment
There was a problem hiding this comment.
The cherry-pick is a faithful port. The single conflict in tests_mark_conditions.yaml was resolved correctly — it preserves the 202511-specific t2_single_node_max/min topology entries while adding the new Cisco G200 DWRR skip. Safe to approve.
vmittal-msft
approved these changes
Apr 17, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description of PR
Summary:
Manual cherry-pick of #20844 to 202511 branch (auto cherry-pick had conflicts).
Fixed test_qos_sai failures for Cisco G200.
Original PR: #20844
202505 PR: #21162
Type of change
Back port request
Approach
What is the motivation for this PR?
Cherry-pick of #20844 to 202511 branch. The auto cherry-pick failed due to a conflict in
tests_mark_conditions.yamlwhere the 202511 branch has additionalt2_single_node_max/mintopology entries not present in the original PR's base.How did you do it?
Manual cherry-pick with conflict resolution. The conflict was in the
testQosSaiDwrrWeightChangeskip conditions — kept both the 202511-specific topology filter update and the new Cisco G200 skip condition from the original PR.How did you verify/test it?
Verified by original PR author on Cisco 8122: test_qos_sai.py 100% passed (16 passed, 40 skipped).
Any platform specific information?
Cisco 8122 (G200).
Supported testbed topology if it's a new test case?
Documentation