- g/x/w may each be bf16 or pre-quantized fp8 matching fmt; a pre-quantized operand skips its quantize kernel
- snapshot sx/sw/sg before the ring finalize overwrites the aliased scale slot so the gemm dequantizes with the quantize scale
- forward carries its scale to backward so gradients reuse the forward's scale
- grad_input/grad_weight forced bf16; a pre-quantized g dequantizes before the bias-sum
- regression test: two delayed steps with a changing amax must not leak the scale ratio
- Finalize scale rings inside the quantize kernels: a last-block epilogue (threadfence + counter elect) folds amax into hist, reduces the window and publishes the next scale on device, zero extra launches; _ScaleRing packs [hist | scale | counter] into one CUDA buffer.
- Split FP8QuantizeParams out of FP8Params so each operator owns its fields; linear_forward/backward_fp8 take optional ring arguments.
- Drop the inference weight-quantization cache; the optimizer bumps the weight version every step, so a cache would miss anyway.
- Zero amax scratch via empty + cudaMemsetAsync instead of torch::zeros, cutting a ~50us fill_ dispatch per quantize.
- Stage crosswise-B operands K-major with cp.async (contract >= 8192) and PRMT-transpose per k_seg region in smem, interleaved with the MMAs; the sync LDG + byte-scatter path it replaces was long-scoreboard bound (ncu 4.6 vs 0.4 stalls/issue).
- Load crosswise-A direct with an in-register PRMT transpose; its operands are typically L2-resident and the staging round trip measured as a net loss.
- Enable grouped rasterization for the congruous NT forward (shared B stripe keeps the weight operand hot in L2) and make the smem budget layout-aware (Fp8GemmSmem) while holding two CTAs per SM.
- Annotate ops/fp8.py return types; drop weight-cache and decorator tests, hoist their imports to module level.
e2e 12L/dim1024/B4xT512 fused AdamW: fp8 137.8ms/step vs bf16 210.3ms, 1.53x. Kernel vs cuBLASLt _scaled_mm: fwd 1.03-1.09x, dX 1.33-1.47x, dW 1.30-1.39x (from 1.10/1.42-1.49/1.52-1.56x), before the pre-transposed copies cuBLASLt needs for dX/dW. fp8 train step vs bf16: 1.34x at 2048 tokens (was 1.25x), 1.08x at 512.
- last-block epilogue (threadfence + counter elect) folds amax into hist[idx], reduces the window and publishes the next scale on device — zero extra launches per linear layer
- _ScaleRing packs [hist | scale | counter] into one CUDA buffer; the eager hist-write / max / scale-copy chain and update() are gone
- split FP8QuantizeParams out of FP8Params so each operator owns its fields; linear_forward/backward_fp8 take optional ring arguments
- e2e 12L/dim1024/B4xT512 (fused AdamW): fp8 137.8ms/step vs bf16 210.3ms, 1.53x; fwd 1.82x, bwd 1.50x
- collapse FP8TensorMeta's 12 slots + 6 copy-paste methods into three _ScaleRing objects (hist/idx/scale/initialized + update/seed)
- skip meta allocation entirely on the DynamicScaling path (zero rings, scales measured inline)
- drop write-only FP8State._last_device and unused E4M3_MAX alias
- linear_forward_fp8 accepts pre-quantized w8 (matching fmt) and skips the weight quantize; amax_w returns 0 on that path since no bf16 values are seen
- bias is now fused into the GEMM epilogue for both dtypes, replacing the separate torch-level add (one elementwise kernel per linear removed)
- FP8Params.bias becomes void* with a new bias_scale slot: null scale = raw bf16 bias, non-null = fp8 storage dequantized in the epilogue after the operand scaling and before any output quantization
- ops/fp8.py relaxes the w dtype check to bf16-or-fp8 and passes bias_scale through
- regression test covers w8/b8, w8/bf16-bias and the amax_w = 0 contract vs an explicit quantization reference
- backward used to read the global fp8 flag at loss.backward() time, so calling it outside fp8_autocast silently fell back to bf16 mm (953 ms cublas per step, 49.9% of the model step)
- _LinearFp8(torch.autograd.Function) now owns the fwd/bwd pair: forward captures fmt/recipe/meta on ctx inside the autocast region, backward reads only ctx (scales from the meta rings, masks from ctx.needs_input_grad), so backward is fp8 wherever it runs
- register the aten::linear impl on AutogradCUDA (replaces torch's generated linear formula that calls aten::linear_backward into the bf16 fallback) and keep the CUDA key for inference_mode
- drop the aten::linear_backward override and fp8_linear_backward (dead paths)
- regression test asserts the fp8 backward fires outside the autocast region and grads match the bf16 reference by direction/norm (E5M2 noise)
- model step (0.67B, CE loss, batch 4x1024): backward GEMMs 953 -> 618 ms (1.54x), full step ~1.2x
- add skip_no_fp8 (CUDA + fp8_mm kernel + cc 8.9+) to tests/conftest.py
- use skip_no_cuda / skip_no_kernel / skip_no_fp8 directly in test modules
- drop _GPU alias and tests.extension.conftest re-exports
- remove unused imports (Union in hf_adapter, make_grpo_config in data conftest)
- route dX/dW through the fused 128x64 fast kernel via contiguous transposes
- drop the legacy 64x64 kernel, cutting dX 1.55->0.38 ms and dW 1.28->0.26 ms
- sync all threads after cp.async.wait_group to fix sporadic NaN in large GEMMs
- add fp8_mm_prequant_fp8 custom op for FP8-in/FP8-out GEMM
- keep training attention on dense 4d tensors
- use packed 3d tensors with KV cache for inference
- extend CUDA rotary embedding to packed 3d inputs
- adapt torch, CUDA and FlashAttention backend dispatch
- store page-table, request-row, and cache-location indices as int32
- preserve CUDA graph replay with bit-exact logits and KV cache coverage
- improve B=1 decode latency by 1-6% across 1K-32K contexts on L20
- Eliminate core/ directory into cache/, runtime/, network/ subpackages plus flat modules
- Split cache.py (647 lines) into cache/{buffer,strategy,pool}.py by layer
- Add explicit ContiguousStrategy, make AllocationStrategy a real ABC
- Move TaskCacheState to cache/strategy.py, drop string forward references
- Rename api/ to network/, server.py to app.py
- Move sample.py into runtime/ alongside executor and graph
- Simplify TaskCacheManager.__init__ to single pool param
- Expose pool.strategy and pool.req_pool as public properties
- Fix KVCache import in attention_backend.py (TYPE_CHECKING guard)
- Fix steady-state decode reading uninitialized position_ids on first step
- TaskCacheRegistry -> TaskCacheManager (independent, held by scheduler)
- TaskCacheState co-locates 5 parallel dicts into one dataclass
- AllocationStrategy base class + PagedStrategy subclass (page_size is a parameter)
- _rollback() helper for unified cleanup (no duplicate free paths)
- Task._kv_len + prefill_done property (explicit, no output_tokens proxy)
- Steady-state detection single-sourced in TaskCacheManager.bind()
- PagePool is now pure physical layer (no task knowledge)
- Removed dead _page_to_hash dict in RadixCache
- merge _generate_streaming/_generate_non_streaming into single _generate() with stream flag
- delete dead GenerationRequest class and generate_with_request method
- inline _next_token helper into generate_async
- replace flash-attn double-checked locking with functools.lru_cache
- extract _write_and_gather_kv helper shared by TorchNative/FlashAttn backends
- inline _kv_cache_is_contiguous into its sole call site in FlashAttnBackend
- change default backend priority from flash>cuda>torch to cuda>flash>torch
- add ASTR_BACKEND env var to override default backend at resolve time
- add supports_graph() static method to AttentionBackend ABC, override in CudaBackend
- replace isinstance(get_backend(), CudaBackend) with get_backend().supports_graph() in executor
- add torch.cuda.is_available() guard to CudaBackend.supports()
- PagedKV::host_kv_len now returns max_context_len instead of max_seq_len
- Eliminates grid-z instability for CUDA graph capture/replay
- Restore skip_no_kernel re-export accidentally removed by ruff --fix
- Remove unused local variable b in attention_backend.py
- Remove unused variable rank0_sd in test_broadcast_state_dict.py
- Remove unused imports across test files
- Replace per-.cu-file static cached tensors with workspace-managed pre-allocated buffers
- InferenceWorkspace now owns decode_o_part / decode_ml_part (mirrors FlashInfer's workspace pattern)
- KVCache carries the buffers through the backend -> C++ kernel chain
- C++ kernels accept optional pre-allocated buffers; fallback to alloc_split_partials for backward compat
- Pre-allocates once at Executor init, zero allocation in the decode hot loop
- Prerequisite for CUDA-graph capture (all kernel addresses are stable)
- add InferenceWorkspace with fixed-shape per-step buffers (input_ids, decode mask, KV bind metadata) for CUDA-graph capture
- bind_tasks derives seq_lens from the pool's own _task_len tracking, dropping the seq_lens parameter
- update decode metadata in-place (position_ids, seq_lens, kv_indptr) instead of re-allocating per step
- task_extend advances _task_len in contiguous mode so the pool tracks current length
- skip log_softmax when logprobs are not requested
- Add broadcast_state_dict to sync state_dict from rank-0 to all ranks
- Fix create_ref_model returning None on non-rank-0 under FSDP
- Fix sync_old_model only updating old_model on rank-0 under FSDP
- Split skip_no_cuda/skip_no_kernel markers and hoist to top-level conftest
- Add distributed tests for broadcast_state_dict and create_ref_model
ViperEkura