jgrusewski 4842a04011 refactor: eliminate ALL bf16 from entire workspace — pure f32/TF32 pipeline ...

Complete bf16 elimination across all crates (ml, ml-core, ml-dqn, ml-ppo,
ml-supervised). Zero half::bf16, __nv_bfloat16, or CudaSlice<half::bf16>
references remain (verified by grep).

CUDA: All 60+ .cu kernels and .cuh headers converted to native float.
  - Half-precision intrinsics (__hmul, __hadd, __hdiv) → float operators
  - atomicAddBF16 → native atomicAdd(float)
  - bf16 wrapper functions → f32 identity passthroughs

Rust: All CudaSlice<half::bf16> → CudaSlice<f32> across 90+ files.
  - htod_f32_to_bf16/dtoh_bf16_to_f32 → htod_f32/dtoh_f32 (direct, no conversion)
  - Deleted bf16 mirror infrastructure (DuelingWeightSetBf16, alloc_bf16_mirror, etc.)
  - Renamed params_bf16→params_flat, d_value_logits_bf16→d_value_logits, etc.
  - Fixed .to_f32() sed damage on Decimal::to_f32() and rng.f32()

FxCache: Single f32 disk format (was bf16/f64 dual-version).
  - Deleted --bf16 CLI flag from precompute_features
  - PVC cache files need regeneration via precompute_features

TF32 tensor cores activated via cublasLtMatmul CUBLAS_COMPUTE_32F — no
explicit TF32 types needed. Storage is pure f32 everywhere.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-04-10 18:52:21 +02:00

15 KiB

Raw Blame History

View Raw