GeezerVison

Walkthrough

Replaces the specialized PrintExpr instruction with an intrinsic-based flow: the compiler emits CallIntrinsic1(IntrinsicFunction1::Print) followed by Pop, the bytecode removes PrintExpr and adds IntrinsicFunction1::Print, and the VM dispatches printing via the intrinsic invoking sys.displayhook.

Changes

Sequence Diagram(s)

sequenceDiagram
  participant Codegen
  participant Bytecode
  participant VM
  participant Sys as sys.displayhook

  Note over Codegen,Bytecode: Compilation
  Codegen->>Bytecode: emit CallIntrinsic1(Print)
  Codegen->>Bytecode: emit Pop

  Note over Bytecode,VM: Execution
  VM->>VM: execute CallIntrinsic1(Print)
  VM->>Sys: invoke displayhook(arg)
  Sys-->>VM: return
  VM->>VM: execute Pop (cleanup)

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

• Attention areas:
  • Both codegen locations emit the exact intrinsic + pop sequence.
  • VM intrinsic dispatcher correctly mirrors previous PrintExpr semantics (argument handling, exceptions).
  • Bytecode enum and stack-effect updates are consistent across consumers (disassembler, tests).

Possibly related PRs

• Remove ImportNameless bytecode #6325 — Similar removal of a dedicated instruction variant and replacement with intrinsic-based handling across codegen/bytecode/frame.
• typing TypeAlias #5945 — Same low-level refactor pattern: drop an Instruction variant and route through CallIntrinsic.
• Instruction::Resume #5944 — Converts specialized instructions into CallIntrinsic1 intrinsics and updates compiler + VM handling.

Suggested reviewers

• youknowone

Poem

🐰 I nibble bytes and hop between lanes,
I swapped the old print for intrinsic chains.
CallIntrinsic hums, then Pop wipes the track,
sys.displayhook answers — no rabbit needs slack! ✨

Pre-merge checks and finishing touches

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

_{Comment @coderabbitai help to get the list of available commands and usage tips.}