Skip to content

Feat/multi merkle tree gpu resident - #951

Draft
ColoCarletti wants to merge 22 commits into
mainfrom
feat/multi-merkle-tree-gpu-resident
Draft

Feat/multi merkle tree gpu resident#951
ColoCarletti wants to merge 22 commits into
mainfrom
feat/multi-merkle-tree-gpu-resident

Conversation

@ColoCarletti

Copy link
Copy Markdown
Collaborator

No description provided.

Per table the batched prover still ran a redundant HOST LDE FFT (main+aux)
to feed prep trees / retained_main / later phases, leaving the GPU idle.
This completes the device port:
- aux/main: resident device expand first; host-expand only for the
  small-table fallback / Retain. A per-table commit-stream sync closes an
  async race the old host FFT delay had hidden.
- OOD (phase 4): device-recompute device-only + read the trace OOD via the
  GPU barycentric fast path (no host coset eval, no D2H download).

Byte-identical proof (roots unchanged); all batched tests pass. e22 mainnet
prove 284s -> 155s (-45%), GPU util 17% -> 44%. Budget test updated: phase 4
is now a full device expansion.
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