Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update Arista7060X6-PE 256x200G mmu configuration. #19570

Merged
merged 3 commits into from
Jul 17, 2024

Conversation

r12f
Copy link
Contributor

@r12f r12f commented Jul 14, 2024

Why I did it

This PR updates the MMU related configurations on Arista 7060X6-PE device with 256x200G breakout.

Work item tracking
  • Microsoft ADO (number only): 28707303

How I did it

This PR updates 3 things:

  1. Updated bcm file for optimal MMU settings.
  2. Updated buffer defaults to accommodate the TH5 architecture with 1 ingress pool + 1 egress pool and updated to optimal value.
  3. Updated PG lookups for buffer setups.

How to verify it

  1. Tested with sonic-mgmt tests with xoff/xon tests with updated QoS parameter: https://github.com/sonic-net/sonic-mgmt/pull/13656/files
  2. Local ixia test is passing in lab.

Both verified using 202311 branch for backporting.

Which release branch to backport (provide reason below if selected)

  • 201811
  • 201911
  • 202006
  • 202012
  • 202106
  • 202111
  • 202205
  • 202211
  • 202305
  • 202311
  • 202405

Tested branch (Please provide the tested image version)

  • 202311

Description for the changelog

Update Arista7060X6-PE 256x200G mmu configuration.

Link to config_db schema for YANG module changes

N/A

A picture of a cute animal (not mandatory but encouraged)

image

@r12f
Copy link
Contributor Author

r12f commented Jul 15, 2024

/azpw ms_conflict

"egress_lossless_profile": {
"pool": "egress_lossless_pool",
"size": "0",
"static_th": "125995684"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why is this larger than the egress_lossless pool size?

Copy link
Contributor Author

@r12f r12f Jul 16, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I checked the configuration and also the update today. This setting is indeed larger than egress pool. I have replied the CSP case to confirm the reason, hopefully will get answers soon.


{%- macro generate_buffer_pool_and_profiles() %}
"BUFFER_POOL": {
"ingress_lossless_pool": {
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

instead of lossless pool, should we say this is ingress_pool?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

currently, the qos tests are using these names for grabbing the configurations, so changing the name will cause them to break.

"mode": "dynamic",
"xoff": "29520896"
},
"egress_lossless_pool": {
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should we just say this is egress_pool?

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i do not know if other repo (swss/sonic-utilitie) make any assumption on the name, we need to check that.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes, things like qos tests are currently assuming the name while grabing the configuration, changing the name will cause these tests to break at this moment...

@Pterosaur
Copy link
Contributor

/azpw ms_conflict

@@ -8,7 +8,7 @@
"size": "165660324",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I believe these values should now be reverted back to reflect the production bcm config, e.g. 121964196 here

Copy link
Contributor Author

@r12f r12f Jul 16, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

not all numbers, these values are the latest recommended ones.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

b62e111#diff-38141bc38cf22f25e51d1afa26023349fcb218e27d0c46e8ada19c6c9081111cL2251-L2252

I'm not sure how 165660324 was derived in the second commit, but I believe this change to the SHARED_LIMIT_CELLS needs to be reflected? (240087 * 254 * 2 = 121964196)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we are aligned from calculation perspective, however if you check the sai api values that shared by broadcom. the value is indeed this....

@r12f
Copy link
Contributor Author

r12f commented Jul 17, 2024

/azpw ms_conflict

@lguohan lguohan merged commit 68136cb into sonic-net:master Jul 17, 2024
22 checks passed
mssonicbld pushed a commit to mssonicbld/sonic-buildimage that referenced this pull request Jul 17, 2024
…et#19570)

Why I did it
This PR updates the MMU related configurations on Arista 7060X6-PE device with 256x200G breakout.

Work item tracking
Microsoft ADO (number only): 28707303
How I did it
This PR updates 3 things:

- Updated bcm file for optimal MMU settings.
- Updated buffer defaults to accommodate the TH5 architecture with 1 ingress pool + 1 egress pool and updated to optimal value.
- Updated PG lookups for buffer setups.

How to verify it

Tested with sonic-mgmt tests with xoff/xon tests with updated QoS parameter: https://github.com/sonic-net/sonic-mgmt/pull/13656/files
Local ixia test is passing in lab.
Both verified using 202311 branch for backporting.
@mssonicbld
Copy link
Collaborator

Cherry-pick PR to 202311: #19604

mssonicbld pushed a commit that referenced this pull request Jul 17, 2024
Why I did it
This PR updates the MMU related configurations on Arista 7060X6-PE device with 256x200G breakout.

Work item tracking
Microsoft ADO (number only): 28707303
How I did it
This PR updates 3 things:

- Updated bcm file for optimal MMU settings.
- Updated buffer defaults to accommodate the TH5 architecture with 1 ingress pool + 1 egress pool and updated to optimal value.
- Updated PG lookups for buffer setups.

How to verify it

Tested with sonic-mgmt tests with xoff/xon tests with updated QoS parameter: https://github.com/sonic-net/sonic-mgmt/pull/13656/files
Local ixia test is passing in lab.
Both verified using 202311 branch for backporting.
arun1355492 pushed a commit to arun1355492/sonic-buildimage that referenced this pull request Jul 26, 2024
…et#19570)

Why I did it
This PR updates the MMU related configurations on Arista 7060X6-PE device with 256x200G breakout.

Work item tracking
Microsoft ADO (number only): 28707303
How I did it
This PR updates 3 things:

- Updated bcm file for optimal MMU settings.
- Updated buffer defaults to accommodate the TH5 architecture with 1 ingress pool + 1 egress pool and updated to optimal value.
- Updated PG lookups for buffer setups.

How to verify it

Tested with sonic-mgmt tests with xoff/xon tests with updated QoS parameter: https://github.com/sonic-net/sonic-mgmt/pull/13656/files
Local ixia test is passing in lab.
Both verified using 202311 branch for backporting.
liushilongbuaa pushed a commit to liushilongbuaa/sonic-buildimage that referenced this pull request Aug 1, 2024
…et#19570)

Why I did it
This PR updates the MMU related configurations on Arista 7060X6-PE device with 256x200G breakout.

Work item tracking
Microsoft ADO (number only): 28707303
How I did it
This PR updates 3 things:

- Updated bcm file for optimal MMU settings.
- Updated buffer defaults to accommodate the TH5 architecture with 1 ingress pool + 1 egress pool and updated to optimal value.
- Updated PG lookups for buffer setups.

How to verify it

Tested with sonic-mgmt tests with xoff/xon tests with updated QoS parameter: https://github.com/sonic-net/sonic-mgmt/pull/13656/files
Local ixia test is passing in lab.
Both verified using 202311 branch for backporting.
@bingwang-ms
Copy link
Contributor

@r12f Please file another PR for 202405 branch to address cherry-pick conflict.

Janetxxx pushed a commit to Janetxxx/sonic-buildimage that referenced this pull request Sep 17, 2024
bingwang-ms pushed a commit that referenced this pull request Sep 18, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

7 participants