Skip to content

Release Linx ISA v0.57 integration - #66

Merged
zhoubot merged 1 commit into
mainfrom
codex/release-0.57
Jul 17, 2026
Merged

Release Linx ISA v0.57 integration#66
zhoubot merged 1 commit into
mainfrom
codex/release-0.57

Conversation

@zhoubot

@zhoubot zhoubot commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator

Integrates this repository with the standalone Linx ISA v0.57 release built from the v0.56.5 baseline. Removes active legacy v0.56 naming and aligns implementation, tests, and documentation with the v0.57 scalar/PTO contract, including TPREFETCH adjacency to TLOAD/TSTORE where applicable.\n\nValidation was run from the LinxISA superproject release branch; repository-specific build and regression evidence is recorded in the commits and superproject release notes.

Update the outer-cube superscalar design references so the active documentation points only at the released ISA profile.

Constraint: Active architecture documentation must not route implementers to v0.56.\nConfidence: high\nScope-risk: narrow\nDirective: Keep design references synchronized with the superproject release profile.\nTested: documentation diff validation.\nNot-tested: pyCircuit build because no executable source changed.

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request updates the Davinci superscalar design documentation to refine instruction mappings and descriptions. Specifically, it clarifies that VFMOD and VLRELU are Davinci-local operations without canonical LinxISA v0.57 origins, updates the VPRELU description, and renames TTRANS to TTRANSPOSE. The review feedback identifies a typo and redundancy in the updated VPRELU description, where 'TEPL' should be 'TPRELU' and the version reference should be corrected to v0.56.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

| VSUBC | TSUBC | Td, Ts1, Ts2, Ts3 | Td[i,j] = Ts1[i,j] − Ts2[i,j] + Ts3[i,j] | same | 16 |
| VFMA | — | Td, Ts1, Ts2, Ts3 | Td[i,j] = Ts1[i,j] × Ts2[i,j] + Ts3[i,j] | FP64/32/16/BF16 | 16 |
| VPRELU | TPRELU | Td, Ts1, Ts2 | Td[i,j] = Ts1[i,j]≥0 ? Ts1[i,j] : Ts1[i,j]×Ts2[i,j] | FP32/16/BF16 | 16 |
| VPRELU | TPRELU | Td, Ts1, Ts2 | Inherited canonical v0.57 TEPL operation retained unchanged in v0.57 | FP32/16/BF16 | 16 |

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

There is a typo in the description for VPRELU. TEPL should be TPRELU. Additionally, the phrase 'Inherited canonical v0.57 ... retained unchanged in v0.57' is redundant and likely meant to refer to inheriting the operation from the v0.56 baseline.

Suggested change
| VPRELU | TPRELU | Td, Ts1, Ts2 | Inherited canonical v0.57 TEPL operation retained unchanged in v0.57 | FP32/16/BF16 | 16 |
| VPRELU | TPRELU | Td, Ts1, Ts2 | Inherited canonical v0.56 TPRELU operation retained unchanged in v0.57 | FP32/16/BF16 | 16 |

@zhoubot
zhoubot merged commit 92783ed into main Jul 17, 2026
7 checks passed
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.

1 participant