[SOLVED] AMD/Zen - Fixed performance counters garbage after S3

[cyring]

Discussed in https://github.com/cyring/CoreFreq/discussions/298

^{Originally posted by **cyring** November 11, 2021} # AMD/Matisse This is an unsolved issue about counters _corruption_ after resuming processor from S3 (Suspend To Ram). Looking into the UCC column, delta value of `18446744073709551615`, the positive value of `-1`, let me think that the counter is stuck to reset. UCC is bound to the x86 register `MSR_IA32_APERF` https://github.com/cyring/CoreFreq/blob/07069d5c80acb48f55f1b944b1424f6109175f6e/corefreqk.c#L10548 **The same test on a 5300U does not produce such issue!** Searching among the AMD specification updates has not revealed an errata. ## Before S3   ## After S3   ## Unexpected workaround Strangely, the effect tends to disappear over several periods and the `MSR_IA32_APERF` per CPU starts to count again, with occasional corruptions.

[cyring]

• The issue is also happening to Lucienne/5300U

  
  rdmsr -aX 0x000000e7
  82635D58
  FFFFFFFE6F53B476
  FFFFFFFE67BB95B8
  FFFFFFFE63EBCB4E
  FFFFFFFE71D93866
  FFFFFFFE69EEF454
  FFFFFFFE61B03508
  FFFFFFFE63ACEE04

rdmsr -aX 0x000000e8 58DA1ED6 FFFFFFFD651E4C86 2CBB9199C 2CA06BAEE 2D692AD05 2D104EB8F 2CC5DE25B 2CF718AE4

* Resetting counters is the workaround which will be applied to all Zen & EPYC families