Repository navigation
Wait between VIP DMI writes for corrected Capture status - #314
Open
lionelZhaowy wants to merge 4 commits into
Open
lionelZhaowy wants to merge 4 commits into
lionelZhaowy wants to merge 4 commits into
Conversation
AI assistance: OpenAI Codex assisted implementation and validation.
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The VIP previously sent DMI writes without checking their status. Eight idle TCK cycles provided spacing but could still hide a rejected or incomplete operation.
The VIP now checks the write-request scan and a later NOP completion scan. It also checks DTMCS sticky status, because the locked DTM can retain stale DMI scan bits after FAILED.
jtag_writekeeps its existing arguments and stops on non-SUCCESS status.jtag_write_statusreturns status for directed tests. Neither task retries writes or clears sticky errors. Command and SBA polling stay unchanged.make sim-test-jtag-writeruns the directed regression with the locked driver, TAP and CDC. Both the locked and corrected DTM pass all seven profiles: normal latencies 0/4/12, long response delay, request backpressure, FAILED and BUSY. Normal profiles check 32 writes each. Error profiles check sticky rejection, explicit recovery and no duplicate write.The complete SoC/VIP compiles with VCS W-2024.09-SP1 and runs JTAG HelloWorld to a zero return. Local license, clang-format 14, and official
hw-all,sw-all,sim-allchecks also pass. Software uses the CI GCC 12.2.0 release. These results do not claim a full SoC regression.The earlier strict-DTM comparison and fixed inputs are documented in
target/sim/test/JTAG_WRITE_WAIT.md. That comparison still reproduces the original spacing failure. No standalone failure against the current locked DTM is claimed.Prepared with OpenAI Codex assistance. The reported simulations ran with VCS.
The license job now uses Ubuntu 24.04, which supports the action's pinned Python 3.9. All four fork lint jobs pass, including the actual license checker.