Skip to content

modeld: single jit#37758

Draft
Armandpl wants to merge 66 commits intomasterfrom
modeld-single-jit2
Draft

modeld: single jit#37758
Armandpl wants to merge 66 commits intomasterfrom
modeld-single-jit2

Conversation

@Armandpl
Copy link
Copy Markdown
Contributor

@Armandpl Armandpl commented Apr 2, 2026

  • test vs. compile
  • test vs. onnx (ort)
  • input queues what do: need them to work both for numpy and tg tensors
  • prepare only
    • compile a warp only jit and warm it up?
    • skip frame in some other way?
  • how do we want to handle compile_supercombo in xx? i think validating with compile3 still makes sense
    • refactor later

@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented Apr 2, 2026

Process replay diff report

Replays driving segments through this PR and compares the behavior to master.
Please review any changes carefully to ensure they are expected.

✅ 0 changed, 66 passed, 0 errors

@commaci-public
Copy link
Copy Markdown
Contributor

commaci-public commented Apr 2, 2026

@Armandpl Armandpl force-pushed the modeld-single-jit2 branch from 22fe638 to 98c62ff Compare April 4, 2026 06:22
Armandpl and others added 30 commits April 6, 2026 19:04
This reverts commit 42bd9b6.
This reverts commit ad5422a.
The UNSAFE CI checkout keeps gitignored files (.pkl, .sconsign.dblite),
so stale pkl files from previous commits can persist and be reused
instead of being recompiled. Delete them explicitly before compiling.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Move the stale artifact cleanup from compile_modeld.py into the
SConscript build command. This ensures stale gitignored pkl and chunk
files are deleted even if scons decides to skip the compile step
(due to a stale .sconsign.dblite from UNSAFE CI checkout).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The warp operations must run under IMAGE=0 to avoid QCOM image texture
optimizations that corrupt the output buffer after ~33 frames.
This was accidentally commented out in a855173.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Move PubMaster/SubMaster creation before the model loading step.
During model loading (3.5s+), process_replay may send liveCalibration.
If SubMaster doesn't exist yet, the message is dropped and the warp
transform stays as zeros, producing garbage warped images.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This reverts commit 49e754c.
This reverts commit 8703885.
# Conflicts:
#	selfdrive/modeld/SConscript
#	selfdrive/modeld/modeld.py
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.

3 participants