Fix an issue with sysconf returning the wrong last level cache values on Linux running on certain AMD Processors.#109749
Merged
mrsharm merged 2 commits intodotnet:release/9.0-stagingfrom Nov 26, 2024
Conversation
Contributor
|
Tagging subscribers to this area: @dotnet/gc |
Maoni0
approved these changes
Nov 13, 2024
jeffschwMSFT
approved these changes
Nov 13, 2024
Member
jeffschwMSFT
left a comment
There was a problem hiding this comment.
lgtm. we will take for consideration in 9.0.x
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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
.NET 9 port of: #108492 - full details about the problem, solution and how to detect this issue can be found there.
Customer Impact
Certain AMD Processor SKUs suffer from the issue where the output of sysconf, the mechanism to discern the last level cache size, returns the value of the last level cache of the host rather than the VM or container. An example of the processor is: AMD EPYC 7763.
The impact of a larger than expected last level cache size is a larger Gen0 budget and thereby, a larger memory footprint albeit, fewer GCs than if the last level cache size is smaller.
We have provided a new configuration:
DOTNET_GCCacheSizeFromSysConfthat can be set to 1 to revert to the old logic.Regression
The regression in behavior relative to other processors which don't exhibit this behavior.
Testing
Tested with an internal customer.
Risk
High risk as this affects the behavior of how the min and max gen0 budget is calculated for Unix based runtimes that has a significant effect on how the GC behaves.