Partial
JIT frontend: riscv64 (RV64I/M/A)
Priority 1 - CPU Architectures & Execution Engines / 1.1 Execution engines
$0 / $10,5000% funded
$0 / $10,500 - 0%
As a firmware developer, I want vemu to finish its JIT frontend: riscv64 (RV64I/M/A) model, so that I can depend on speed for RISC-V for production firmware, not just the common path.
Why it matters
speed for RISC-V
Summary
RV64 - > IR
Scope of work
Partially modeled. Complete the missing mechanics of JIT frontend: riscv64 (RV64I/M/A). What it is: RV64 - > IR. Why it matters: speed for RISC-V.
Current state
Status: Partially implemented. Notes / evidence: vemu-jit-frontend-riscv64: RV64I/M + A (LR/SC/AMO) + compressed + control flow lowered to IR; CSR/privileged/FP/vector deferred to interpreter; block chaining + jit_parity/differential tests.
Blocked by 2
Fund these first - this work can't be completed until they ship.
- Done
- Done
Activity log
Feature defined and added to the roadmap
Funding goal set to $10,500
Implementation status: Partial