[QoS]qos_yaml j2C+ changes for new _vsq thresholds #13069
Merged
Conversation
This file contains 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
|
I can see you request for 202205 branch. How about 202311 branch? |
|
@vmittal-msft , could you please review and comment on above request. |
|
@ansrajpu-git Can you please confirm if this PR has all the latest changes ? |
@vmittal-msft, the latest changes on PR #18239 are tested with ' libsaibcm_7.1.78.4_amd64.deb' for 202205 branch, for which the PR raised #13319. Kindly confirm for master which version should be used. |
|
@ansrajpu-git to update the PR with the new values to match the values in 202205. |
|
@vmittal-msft , updated the qos_params aligning with 202205. Please review. Also closing the PR ##13319 raised separately for 202205 to keep changes at one place. |
8 tasks
kenneth-arista
added a commit
to kenneth-arista/sonic-mgmt
that referenced
this pull request
Jul 15, 2024
Cloned values from sonic-net#13069 for Jericho2.
8 tasks
vmittal-msft
approved these changes
Jul 23, 2024
|
@vmittal-msft , Updated this PR with the latest buffer profile changes #[19653] (sonic-net/sonic-buildimage#19653). |
yxieca
pushed a commit
that referenced
this pull request
Aug 19, 2024
Cloned values from #13069 for Jericho2.
arista-hpandya
pushed a commit
to arista-hpandya/sonic-mgmt
that referenced
this pull request
Oct 2, 2024
What is the motivation for this PR? The new MMU settings to enhance performance for RDMA traffic in production. Hence the qos_params needs to be tweaked according to the set buffer profiles. However, the existing sonic-mgmt LossyQueueTest doesn't fairly verify the buffer threshold for headroom for Lossy traffic. As per the new vsq profile setting the XOFF FADT threshold/PG is way lesser than the Nominal headroom, which limits it to not utilize the headroom buffer completely and send pause frames before reaching the MAX headroom limit. Either the test case needs to be improvised by adding more source ports or a new test case should be added to verify the Lossy queue traffic at PG level
arista-hpandya
pushed a commit
to arista-hpandya/sonic-mgmt
that referenced
this pull request
Oct 2, 2024
Cloned values from sonic-net#13069 for Jericho2.
mssonicbld
pushed a commit
to mssonicbld/sonic-mgmt
that referenced
this pull request
Oct 11, 2024
Cloned values from sonic-net#13069 for Jericho2.
8 tasks
mssonicbld
pushed a commit
that referenced
this pull request
Oct 12, 2024
Cloned values from #13069 for Jericho2.
vikshaw-Nokia
pushed a commit
to vikshaw-Nokia/sonic-mgmt
that referenced
this pull request
Oct 23, 2024
What is the motivation for this PR? The new MMU settings to enhance performance for RDMA traffic in production. Hence the qos_params needs to be tweaked according to the set buffer profiles. However, the existing sonic-mgmt LossyQueueTest doesn't fairly verify the buffer threshold for headroom for Lossy traffic. As per the new vsq profile setting the XOFF FADT threshold/PG is way lesser than the Nominal headroom, which limits it to not utilize the headroom buffer completely and send pause frames before reaching the MAX headroom limit. Either the test case needs to be improvised by adding more source ports or a new test case should be added to verify the Lossy queue traffic at PG level
vikshaw-Nokia
pushed a commit
to vikshaw-Nokia/sonic-mgmt
that referenced
this pull request
Oct 23, 2024
Cloned values from sonic-net#13069 for Jericho2.
ansrajpu-git
added a commit
to ansrajpu-git/sonic-mgmt
that referenced
this pull request
Nov 7, 2024
What is the motivation for this PR? The new MMU settings to enhance performance for RDMA traffic in production. Hence the qos_params needs to be tweaked according to the set buffer profiles. However, the existing sonic-mgmt LossyQueueTest doesn't fairly verify the buffer threshold for headroom for Lossy traffic. As per the new vsq profile setting the XOFF FADT threshold/PG is way lesser than the Nominal headroom, which limits it to not utilize the headroom buffer completely and send pause frames before reaching the MAX headroom limit. Either the test case needs to be improvised by adding more source ports or a new test case should be added to verify the Lossy queue traffic at PG level
8 tasks
bingwang-ms
pushed a commit
that referenced
this pull request
Nov 8, 2024
…hresholds (#15448) * [QoS]qos_yaml j2C+ changes for new _vsq thresholds (#13069) What is the motivation for this PR? The new MMU settings to enhance performance for RDMA traffic in production. Hence the qos_params needs to be tweaked according to the set buffer profiles. However, the existing sonic-mgmt LossyQueueTest doesn't fairly verify the buffer threshold for headroom for Lossy traffic. As per the new vsq profile setting the XOFF FADT threshold/PG is way lesser than the Nominal headroom, which limits it to not utilize the headroom buffer completely and send pause frames before reaching the MAX headroom limit. Either the test case needs to be improvised by adding more source ports or a new test case should be added to verify the Lossy queue traffic at PG level * [Chassis][Voq] Updating J2C+ qos yaml for 400G and 100G profile _egress_pkt count for lossy profile (#14585) Since the dynamic_th-alpha changed from 0 to -4 (400g) & 100g port speed for egress lossy profile. PR #sonic-net/sonic-buildimage#20132 Corresponding changes made in J2C+ qos yaml for t2 -broadcom-dnx * [Qos]qos_yaml updated for 400G
|
@vmittal-msft , @ansrajpu-git , I'm removing the request cherry label as there's separate PR merged to 202405 already. Please review/clarify if any. thanks. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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
Updated qos test params for 400g and 100g port speeds for the new vsq threshold introduced in PR #sonic-net/sonic-buildimage#18239
Summary:
Fixes # (issue)
Type of change
Back port request
Approach
What is the motivation for this PR?
The new MMU settings to enhance performance for RDMA traffic in production.
Hence the qos_params needs to be tweaked according to the set buffer profiles.
However, the existing sonic-mgmt LossyQueueTest doesn't fairly verify the buffer threshold for headroom for Lossy traffic. As per the new vsq profile setting the XOFF FADT threshold/PG is way lesser than the Nominal headroom, which limits it to not utilize the headroom buffer completely and send pause frames before reaching the MAX headroom limit.
Either the test case needs to be improvised by adding more source ports or a new test case should be added to verify the Lossy queue traffic at PG level
How did you do it?
How did you verify/test it?
Executed the qos test cases and verify the results
Any platform specific information?
Supported testbed topology if it's a new test case?
Documentation