From ec19eae876f64c900f8340ac12388af21765fe88 Mon Sep 17 00:00:00 2001 From: lcy-seso Date: Mon, 19 Oct 2026 14:43:56 +0000 Subject: [PATCH] [Docs] Document six newly implemented ops on the API pages TileOPs main now exports 205 ops; six of them had no API page naming them, so check_api_pages.py printed them as undocumented. Each goes on its family page: FusedAddRMSNormPerTokenQuantFwdOp and FusedQKNormRoPEFwdOp on Normalization, GemmINT8W8A8FwdOp on GEMM, MLAPagedFwdOp and MLAKVCacheWriteFwdOp on Attention, and SelectiveScanFwdOp on Mamba. The API index states the new coverage. --- docs/api/attention.md | 12 ++++++++++++ docs/api/index.md | 8 ++++---- docs/api/linear-algebra.md | 8 ++++++++ docs/api/mamba.md | 8 ++++++++ docs/api/normalization.md | 12 ++++++++++++ 5 files changed, 44 insertions(+), 4 deletions(-) diff --git a/docs/api/attention.md b/docs/api/attention.md index cf1658fa..8630c563 100644 --- a/docs/api/attention.md +++ b/docs/api/attention.md @@ -53,6 +53,12 @@ what runs when you call `op(...)`. heading_level: 3 members: ["__init__", "forward"] +::: tileops.attention.MLAPagedFwdOp + options: + show_root_heading: true + heading_level: 3 + members: ["__init__", "forward"] + ::: tileops.attention.MLAVarlenFwdOp options: show_root_heading: true @@ -122,3 +128,9 @@ what runs when you call `op(...)`. show_root_heading: true heading_level: 3 members: ["__init__", "forward"] + +::: tileops.attention.MLAKVCacheWriteFwdOp + options: + show_root_heading: true + heading_level: 3 + members: ["__init__", "forward"] diff --git a/docs/api/index.md b/docs/api/index.md index 5cf3f254..677b8b7e 100644 --- a/docs/api/index.md +++ b/docs/api/index.md @@ -25,16 +25,16 @@ a tool rather than an op, comes last. The Benchmarks pages use the same order. | [Elementwise](elementwise.md) | unary and binary maps, activations, dropout, and the in-place forms | | [RoPE](rope.md) | rotary position embedding — NeoX and interleaved layouts, Llama 3.1, YaRN, LongRoPE | | [Reduction](reduction.md) | sums, extrema, arg-reductions, cumulative scans, softmax | -| [Normalization](normalization.md) | RMSNorm, LayerNorm, GroupNorm, BatchNorm and the fused variants | +| [Normalization](normalization.md) | RMSNorm, LayerNorm, GroupNorm, BatchNorm and the fused variants, including RMSNorm with per-token quantization and the q/k RMSNorm with RoPE | | [Pooling](pool.md) | average, max and adaptive pooling, with and without indices, plus the chunked sequence mean | | [Convolution](convolution.md) | forward convolution over 1D, 2D and 3D inputs | -| [GEMM](linear-algebra.md) | dense matmul — plain, batched and grouped, with the fp8, W4A16 and W4A8 variants | +| [GEMM](linear-algebra.md) | dense matmul — plain, batched and grouped, with the fp8, W4A16, W4A8 and INT8 W8A8 variants | | [Quantization & Dequantization](quantization.md) | INT8, FP8 and INT4 quantization, and INT8 dequantization | -| [Attention](attention.md) | forward and backward attention, including the paged and decode kernels, merging attention states, and writing and gathering the paged KV cache | +| [Attention](attention.md) | forward and backward attention, including the paged and decode kernels, merging attention states, writing and gathering the paged KV cache, and writing the MLA latent cache | | [MoE](moe.md) | the routed mixture-of-experts FFN and its separately callable stages | | [Top-k & Sampling](sampling.md) | top-k selection, logits masks (top-k, top-p, min-p) and sampling, including chain speculative sampling | | [Linear Attention](linear-attention.md) | DeltaNet, Gated DeltaNet, Kimi Delta Attention and Gated Linear Attention | -| [Mamba](mamba.md) | the causal convolution and the SSD scan, their decode steps, and the chunked forms | +| [Mamba](mamba.md) | the Mamba-1 selective scan, the causal convolution and the SSD scan, their decode steps, and the chunked forms | | [FFT](fft.md) | the discrete transform | | [Transform](transform.md) | the fast Walsh-Hadamard transform along the last axis | | [mHC](mhc.md) | Manifold-Constrained Hyper-Connections — the pre/post pair around a layer | diff --git a/docs/api/linear-algebra.md b/docs/api/linear-algebra.md index f2e6d9d9..d84e4674 100644 --- a/docs/api/linear-algebra.md +++ b/docs/api/linear-algebra.md @@ -39,6 +39,14 @@ what runs when you call `op(...)`. merge_init_into_class: false show_signature_annotations: false +::: tileops.gemm.GemmINT8W8A8FwdOp + options: + show_root_heading: true + heading_level: 3 + members: ["__init__", "forward"] + merge_init_into_class: false + show_signature_annotations: false + ## Batched GEMM ::: tileops.gemm.BmmFwdOp diff --git a/docs/api/mamba.md b/docs/api/mamba.md index 3bb566f1..fb82e523 100644 --- a/docs/api/mamba.md +++ b/docs/api/mamba.md @@ -13,6 +13,14 @@ what runs when you call `op(...)`. heading_level: 3 members: ["__init__", "forward"] +## Selective scan + +::: tileops.mamba.SelectiveScanFwdOp + options: + show_root_heading: true + heading_level: 3 + members: ["__init__", "forward"] + ## Causal convolution ::: tileops.mamba.CausalConv1dFwdOp diff --git a/docs/api/normalization.md b/docs/api/normalization.md index e8bbbcbb..fc446dae 100644 --- a/docs/api/normalization.md +++ b/docs/api/normalization.md @@ -33,6 +33,18 @@ what runs when you call `op(...)`. heading_level: 3 members: ["__init__", "forward"] +::: tileops.norm.FusedAddRMSNormPerTokenQuantFwdOp + options: + show_root_heading: true + heading_level: 3 + members: ["__init__", "forward"] + +::: tileops.norm.FusedQKNormRoPEFwdOp + options: + show_root_heading: true + heading_level: 3 + members: ["__init__", "forward"] + ## Adaptive LayerNorm ::: tileops.norm.AdaLayerNormFwdOp