hw: Mask TCDM write data stability check on reads #125
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Outdated suggestions cannot be applied.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.
pulp-platform/common_cells#200, recently included by a dependency update, may trigger false-positive TCDM stability assertion errors in some cases when using SSRs.
This is because SSR write data is hardwired to the bidirectional data FIFO output. Therefore, on reads, the write data bits may change irrespective of when a read request is pending. This instability is benign, but causes the payload-wide assertion to fail.
This PR, together with pulp-platform/common_cells#219, fixes this issue by masking stability assertion on the write data and mask inside
stream_xbar
andstream_omega_net
and then conditionally asserting write data and mask stability on reads and atomics insnitch_tcdm_interconnect
.A more general discussion is maybe in order on whether SSRs should invest additional logic to instead mask write data on reads, but we should first fix the bug and then ponder design improvements.
TODO:
common_cells
release.common_cells
dependency.