[proposal] support llc/mba isolation for AMD arch

[zwzhang0107]

What is your proposal:
support llc/mba isolation for AMD

Why is this needed:
The memory bandwidth schema format of resctrl on AMD is different with Intel.
L3 isolation formats are same.
Intel use percentage for MB configuration, however AMD use absolutely value for MB configuration.

Intel Format:
L3:0=00f;1=00f
MB:0=100;1=100

AMD Format
L3:0=ffff;1=ffff;2=ffff;3=ffff;4=ffff;5=ffff;6=ffff;7=ffff;8=ffff;9=ffff;10=ffff;11=ffff;12=ffff;13=ffff;14=ffff;15=ffff
MB:0=2048;1=2048;2=2048;3=2048;4=2048;5=2048;6=2048;7=2048;8=2048;9=2048;10=2048;11=2048;12=2048;13=2048;14=2048;15=2048

For AMD format, 8 means the MB limit of 1GB/s (e.g. 48.means 6GB/s), and 2048 means unlimited MB.

Is there a suggested solution, if so, please add it:
Measure the MB extreme limit offline, and convert to absolute value for AMD.

The theoretical value of MB on AMD EPYC(TM) MILAN is 25GB/s for each CCD.
There will be a discount for actual value due to the real environment limit.

[jasonliu747]

/area koordlet

[hormes]

/help

[koordinator-bot]

@hormes:
This request has been marked as needing help from a contributor.

Please ensure the request meets the requirements listed here.

If this request no longer meets these requirements, the label can be removed
by commenting with the /remove-help command.

In response to this:

/help

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[jasonliu747]

/assign