Skip to content

analog M4e-kick (partial): decks for deriving the comparator's clock-edge kick (Part of #10942) - #11092

Merged
gHashTag merged 2 commits into
masterfrom
lane/m4e-kick-10942
Oct 19, 2026
Merged

gHashTag merged 2 commits into
masterfrom
lane/m4e-kick-10942

Conversation

@gHashTag

@gHashTag gHashTag commented Oct 19, 2026 •

Copy link
Copy Markdown
Owner

Updates #10942.

Partial result for the comparator's clock-edge kick (M4e-kick): four ngspice decks under specs/analog/decks/ (sar_kick_qg.cir, sar_kick_edge.cir, sar_kick_aperture.cir, sar_kick_cnode.cir). There is no spec in this PR: the network spec sar_kick_net.t27 follows in the next PR of the lane for #10942. The aperture scan of sar_kick_aperture.cir ran to the end on a remote Docker host (20 of 20 transients) and its table is in a comment on #10942. The decks hold no developer paths: the PDK, the OSDI models and the golden subcircuits come from the environment variables IHP_PDK, IHP_OSDI and SAR_GOLDEN_SUBCKTS.

Checks: decks only, ASCII, no secrets.

🤖 Generated with Claude Code

…ock-edge kick -- gate-charge table, edge waveform, aperture, plate-node capacitance (Part of #10942)

Decks only, no spec yet. sar_kick_qg.cir: Q_G(Vg, Vs, Vd) of the input pair's device, bulk on ground, 50 C, three gate levels (M3c's table is bulk-at-source, 27 C: +38 percent on the edge charge). sar_kick_edge.cir: tail and drain waveforms and the gate charge at the first decision's clock edge, gates held. sar_kick_aperture.cir: how much a step of the input at tau still decides. sar_kick_cnode.cir: the two-port capacitance the idle bottom-plate drivers hang on Bit and Bit_n.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-19 16:14:33 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 38
PRs with All Checks Green 12
READY 2
FAILING 38
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 38 + 0 + 0 = 40, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=57b91bb96334 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…k-edge kick, aperture and gate charge (Closes #10942)

Merge of master plus sealing, gates run on the merged tree with t27c from master: tests clean with vacuous passes 0, seal --verify MATCH, tri types ratchet CLEAN, dupe_scan failure set a subset of master's.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@gHashTag
gHashTag enabled auto-merge (squash) October 19, 2026 16:26
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-10-19 16:26:34 UTC

Summary

Status Count
Total Open PRs 50
PRs with Failing Checks 41
PRs with All Checks Green 9
READY 2
FAILING 41
PENDING 0
NO CHECKS YET 0

These columns do not partition: 2 + 41 + 0 + 0 = 43, and there are 50 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=57b91bb96334 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants