LLC Config: support concurrent and non-concurrent SPM/FLUSH settings#26
Open
flague wants to merge 1 commit intopulp-platform:masterfrom
Open
LLC Config: support concurrent and non-concurrent SPM/FLUSH settings#26flague wants to merge 1 commit intopulp-platform:masterfrom
flague wants to merge 1 commit intopulp-platform:masterfrom
Conversation
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 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Bug Description
The current version of the LLC config module causes the FSM to remain stuck when writing only the FLUSH register and committing its value (e.g. LLC is currently configured with SPM=0x03, and a flush is sent on all ways).
This is due to an incorrect assignment of the
cfg_o.flushed, in theFsmInitStatestate of the FSM.cfg_o.flushedis updated taking into account only thecfg_i.spmregister value. This causes the flush to be repeated infinite times on some ways.Proposed Solution
The provided solution ensures that in the above described scenario, the
cfg_o.flushedis correctly updated at the beginning of each way-flush.Furthermore, it ensures that when no ways have to be flushed (
to_flush_d =0), thecfg.flushedregister is correctly update with the new SPM configuration.Test Cases
The solution provided has been tested in the following cases:
a. If the ways to be flushed are already in SPM, they won’t be flushed and
cfg.flushedregister will maintain the old SPM configuration.b. If some of the ways to be flushed are not in SPM, they will be flushed, and, at the end of the process, in
FsmEndFlushstate,cfg.flushedwill be restored with the current SPM configuration. This works both for overlapping and non-overlapping SPM and FLUSH ways.A series of corner cases, even though not exhaustive, for non-overlapping, partially overlapping and completely overlapping configs have been tested.