minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[DeepSeek-V4] Add mhc_fused_post_pre kernel (#25976)

Browse Source 
Co-authored-by: Qichao Li <liqichao@baidu.com>
This commit is contained in:
Qichao Li
2026-05-30 02:04:51 -07:00
committed by GitHub
co-authored by Qichao Li
parent 0d9a2a9de3
commit 7c5708cba7
5 changed files with 876 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/environ.py
+1
View File
@@ -657,6 +657,7 @@ class Envs:
SGLANG_OPT_USE_TILELANG_MHC_PRE = EnvBool(True)
SGLANG_OPT_USE_TILELANG_MHC_POST = EnvBool(True)
SGLANG_OPT_USE_TRITON_FUSED_MHC = EnvBool(True)
SGLANG_OPT_FUSE_MHC_POST_PRE = EnvBool(False)
SGLANG_OPT_USE_TILELANG_INDEXER = EnvBool(False)
SGLANG_OPT_USE_AITER_INDEXER = EnvBool(False)
SGLANG_OPT_USE_JIT_INDEXER_METADATA = EnvBool(True)
python/sglang/srt/layers/mhc.py
+497
View File
@@ -896,3 +896,500 @@ def mhc_post(
residual.shape[-1],
)
return out
@tilelang.jit(
pass_configs={
tilelang.PassConfigKey.TL_DISABLE_WARP_SPECIALIZED: True,
tilelang.PassConfigKey.TL_DISABLE_TMA_LOWER: True,
tilelang.PassConfigKey.TL_PTXAS_REGISTER_USAGE_LEVEL: 10,
},
)
def mhc_fused_post_pre_fma_tilelang(
prev_comb_mix,
prev_residual,
prev_post_mix,
hidden_in,
pre_fn,
mixes_partial_out,
sqrsum_partial_out,
cur_residual_out,
hc: int,
hidden_size: int,
num_mix_outputs: int,
n_thr: int = 256,
tile_mix_outputs: int = 1,
split_k: int = 1,
) -> tilelang.JITKernel:
num_tokens = T.dynamic("num_tokens")
split_k = T.dynamic("split_k")
hidden_per_split = (hidden_size + split_k - 1) // split_k
num_mix_output_tiles = (num_mix_outputs + tile_mix_outputs - 1) // tile_mix_outputs
prev_comb_mix: T.Tensor((num_tokens, hc, hc), T.float32)
prev_residual: T.Tensor((num_tokens, hc, hidden_size), T.bfloat16)
prev_post_mix: T.Tensor((num_tokens, hc), T.float32)
hidden_in: T.Tensor((num_tokens, hidden_size), T.bfloat16)
pre_fn: T.Tensor((num_mix_outputs, hc, hidden_size), T.float32)
mixes_partial_out: T.Tensor((split_k, num_tokens, num_mix_outputs), T.float32)
sqrsum_partial_out: T.Tensor((split_k, num_tokens), T.float32)
cur_residual_out: T.Tensor((num_tokens, hc, hidden_size), T.bfloat16)
hidden_iters_per_thread = (hidden_per_split + n_thr - 1) // n_thr
num_warps = n_thr // 32
ENABLE_PDL = is_arch_support_pdl()
# CTA assignment:
# token_idx : this CTA handles one token.
# mix_output_tile_idx : this CTA handles a small tile of mix output columns.
# For HC=4, num_mix_outputs = 24:
# [0:4] -> pre logits
# [4:8] -> post logits
# [8:24] -> comb logits
# hidden_split_idx : this CTA handles one split of the hidden dimension.
#
# Thread assignment inside one CTA:
# Each thread owns several hidden positions in this hidden split:
# hidden_idx = hidden_split_start + hidden_iter * n_thr + thread_idx
#
# For each owned hidden_idx, the thread computes:
# 1. post result: cur_residual[token, :, hidden_idx]
# 2. sqrsum partial for pre RMS
# 3. GEMM partial for several mix output columns
with T.Kernel(
num_tokens,
num_mix_output_tiles,
split_k,
threads=n_thr,
) as (token_idx, mix_output_tile_idx, hidden_split_idx):
thread_idx = T.get_thread_binding()
warp_idx = T.get_warp_idx()
lane_idx = T.get_lane_idx()
warp_partials = T.alloc_shared((num_warps, tile_mix_outputs + 1), T.float32)
post_mix_smem = T.alloc_shared((hc,), T.float32)
comb_mix_smem = T.alloc_shared((hc, hc), T.float32)
post_mix_for_token = T.alloc_local((hc,), T.float32)
comb_mix_for_token = T.alloc_local((hc, hc), T.float32)
mix_acc = T.alloc_local((tile_mix_outputs,), T.float32)
sqrsum_acc = T.alloc_local((1,), T.float32)
cur_residual_values = T.alloc_local((hc,), T.float32)
T.clear(mix_acc)
T.clear(sqrsum_acc)
hidden_split_start = hidden_split_idx * hidden_per_split
if ENABLE_PDL:
T.pdl_sync()
# Load post/comb coefficients for this token.
#
# PyTorch equivalent:
# post = prev_post_mix[token_idx] # [HC]
# comb = prev_comb_mix[token_idx] # [HC, HC]
T.copy(prev_post_mix[token_idx, 0], post_mix_smem)
T.copy(prev_comb_mix[token_idx, 0, 0], comb_mix_smem)
for route_idx in T.unroll(hc):
post_mix_for_token[route_idx] = post_mix_smem[route_idx]
for old_route_idx in T.unroll(hc):
for new_route_idx in T.unroll(hc):
comb_mix_for_token[old_route_idx, new_route_idx] = comb_mix_smem[
old_route_idx, new_route_idx
]
for hidden_iter in T.serial(hidden_iters_per_thread):
hidden_idx = hidden_split_start + hidden_iter * n_thr + thread_idx
if hidden_idx < hidden_size:
# Step A: fused post.
#
# PyTorch equivalent:
# cur_residual =
# post.unsqueeze(-1) * hidden_in.unsqueeze(1)
# + (
# comb.unsqueeze(-1)
# * prev_residual.unsqueeze(2)
# ).sum(dim=1)
#
# Scalar form for this token and this hidden position:
# cur_residual[j, h]
# = post[j] * hidden_in[h]
# + sum_k comb[k, j] * prev_residual[k, h]
for new_route_idx in T.unroll(hc):
cur_residual_values[new_route_idx] = (
post_mix_for_token[new_route_idx]
* hidden_in[token_idx, hidden_idx]
)
for old_route_idx in T.unroll(hc):
cur_residual_values[new_route_idx] += (
comb_mix_for_token[old_route_idx, new_route_idx]
* prev_residual[token_idx, old_route_idx, hidden_idx]
)
# Match the unfused path:
# mhc_post writes bf16 residual,
# then mhc_pre reads bf16 residual.
for route_idx in T.unroll(hc):
cur_residual_values[route_idx] = T.bfloat16(
cur_residual_values[route_idx]
)
# Step B1: pre sqrsum partial.
#
# PyTorch equivalent:
# x_flat = cur_residual.reshape(T, HC * H).float()
# sqrsum = (x_flat * x_flat).sum(dim=-1)
#
# Only mix_output_tile_idx == 0 writes cur_residual and sqrsum,
# otherwise different output-column CTAs would duplicate this work.
if mix_output_tile_idx == 0:
for route_idx in T.unroll(hc):
cur_residual_out[token_idx, route_idx, hidden_idx] = (
cur_residual_values[route_idx]
)
sqrsum_acc[0] += (
cur_residual_values[route_idx]
* cur_residual_values[route_idx]
)
# Step B2: pre GEMM partial.
#
# PyTorch equivalent:
# mixes = F.linear(x_flat, fn)
#
# Scalar form:
# mixes[token, o] +=
# pre_fn[o, route, hidden] * cur_residual[route, hidden]
#
# This CTA computes only tile_mix_outputs columns of mixes.
for tile_col_idx in T.unroll(tile_mix_outputs):
mix_output_idx = (
mix_output_tile_idx * tile_mix_outputs + tile_col_idx
)
if mix_output_idx < num_mix_outputs:
for route_idx in T.unroll(hc):
mix_acc[tile_col_idx] += (
pre_fn[mix_output_idx, route_idx, hidden_idx]
* cur_residual_values[route_idx]
)
# Reduce thread partials inside each warp.
for tile_col_idx in T.unroll(tile_mix_outputs):
mix_acc[tile_col_idx] = T.warp_reduce_sum(mix_acc[tile_col_idx])
if mix_output_tile_idx == 0:
sqrsum_acc[0] = T.warp_reduce_sum(sqrsum_acc[0])
# One lane per warp writes warp-level partials to shared memory.
if lane_idx == 0:
for tile_col_idx in T.unroll(tile_mix_outputs):
warp_partials[warp_idx, tile_col_idx] = mix_acc[tile_col_idx]
if mix_output_tile_idx == 0:
warp_partials[warp_idx, tile_mix_outputs] = sqrsum_acc[0]
T.sync_threads()
# Reduce across warps and write split partials.
#
# The full PyTorch result would be:
# mixes = F.linear(cur_residual.reshape(T, HC * H), fn)
# sqrsum = (cur_residual.float() ** 2).sum(dim=(1, 2))
#
# This kernel is split along hidden, so each CTA writes only:
# mixes_partial_out[hidden_split_idx, token, o]
# sqrsum_partial_out[hidden_split_idx, token]
#
# Later mhc_pre_big_fuse does:
# mixes = mixes_partial_out.sum(dim=0)
# sqrsum = sqrsum_partial_out.sum(dim=0)
# rms = rsqrt(sqrsum / (HC * H) + eps)
# mixes *= rms
# mixes -> pre/post/comb
# layer_input = sum_j pre[j] * cur_residual[j]
if warp_idx == 0:
for tile_col_idx in T.unroll(tile_mix_outputs):
mix_output_idx = mix_output_tile_idx * tile_mix_outputs + tile_col_idx
if mix_output_idx < num_mix_outputs and lane_idx == tile_col_idx:
mix_output_partial = T.alloc_var(T.float32, init=0.0)
for reduce_warp_idx in T.unroll(num_warps):
mix_output_partial += warp_partials[
reduce_warp_idx, tile_col_idx
]
mixes_partial_out[hidden_split_idx, token_idx, mix_output_idx] = (
mix_output_partial
)
if mix_output_tile_idx == 0 and lane_idx == 0:
sqrsum_partial = T.alloc_var(T.float32, init=0.0)
for reduce_warp_idx in T.unroll(num_warps):
sqrsum_partial += warp_partials[reduce_warp_idx, tile_mix_outputs]
sqrsum_partial_out[hidden_split_idx, token_idx] = sqrsum_partial
if ENABLE_PDL:
T.pdl_trigger()
def mhc_fused_post_pre(
x: torch.Tensor,
residual: torch.Tensor,
post_layer_mix: torch.Tensor,
comb_res_mix: torch.Tensor,
fn: torch.Tensor,
hc_scale: torch.Tensor,
hc_base: torch.Tensor,
rms_eps: float,
hc_pre_eps: float,
hc_sinkhorn_eps: float,
hc_post_mult_value: float,
sinkhorn_repeat: int,
n_splits: int = 1,
tile_n: int = 1,
*,
norm_weight: torch.Tensor | None = None,
norm_eps: float | None = None,
) -> tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]:
"""Fuse the boundary between one mHC post step and the next mHC pre step.
The unfused sequence is ``mhc_post -> pre-norm GEMM -> mhc_pre big_fuse``.
This wrapper keeps the numerically sensitive ``mhc_pre_big_fuse`` stage,
including optional RMSNorm, but removes the separate post/pre boundary.
Small token batches use the FMA kernel above to combine ``mhc_post`` and the
pre-norm GEMM in one launch; larger batches keep DeepGEMM for throughput and
only fuse the Python/model-level scheduling boundary.
Returns:
residual_cur: post-mapped residual, shape (..., hc_mult, hidden_size)
post_mix_cur: shape (..., hc_mult, 1)
comb_mix_cur: shape (..., hc_mult, hc_mult)
layer_input_cur: shape (..., hidden_size)
"""
assert residual.dtype == torch.bfloat16
assert x.dtype == torch.bfloat16
assert post_layer_mix.dtype == torch.float32
assert comb_res_mix.dtype == torch.float32
assert fn.dtype == torch.float32
assert hc_scale.dtype == torch.float32
assert hc_base.dtype == torch.float32
hc_mult = residual.shape[-2]
hidden_size = residual.shape[-1]
hc_mult2 = hc_mult * hc_mult
hc_mult3 = hc_mult * 2 + hc_mult2
hc_hidden_size = hc_mult * hidden_size
outer_shape = residual.shape[:-2]
assert x.shape == (*outer_shape, hidden_size)
assert post_layer_mix.shape in (
(*outer_shape, hc_mult, 1),
(*outer_shape, hc_mult),
)
assert comb_res_mix.shape == (*outer_shape, hc_mult, hc_mult)
assert fn.shape == (hc_mult3, hc_hidden_size)
assert hc_scale.shape == (3,)
assert hc_base.shape == (hc_mult3,)
residual_flat = residual.view(-1, hc_mult, hidden_size)
num_tokens = residual_flat.shape[0]
if num_tokens == 0:
# Some DP/EP ranks can receive no tokens; return correctly typed empty
# tensors so later fused layers keep the same contracts as mhc_pre/hc_post.
return (
torch.empty_like(residual),
torch.empty(
(*outer_shape, hc_mult, 1), dtype=torch.float32, device=residual.device
),
torch.empty(
(*outer_shape, hc_mult, hc_mult),
dtype=torch.float32,
device=residual.device,
),
torch.empty(
(*outer_shape, hidden_size),
dtype=torch.bfloat16,
device=residual.device,
),
)
x_flat = x.view(num_tokens, hidden_size)
# The scalar-FMA kernel wins only for small batches where launch
# overhead dominates; beyond the threshold DeepGEMM's tensor-core path wins.
fma_token_threshold = 32
if num_tokens <= fma_token_threshold:
tile_n = 2 if num_tokens < 8 else 3
n_splits = 8 if (num_tokens < 8 and hidden_size <= 4096) else 4
else:
n_splits = _compute_num_split_for_mhc_pre(num_tokens, hc_hidden_size)
gemm_out_mul = torch.empty(
n_splits,
num_tokens,
hc_mult3,
dtype=torch.float32,
device=residual.device,
)
gemm_out_sqrsum = torch.empty(
n_splits,
num_tokens,
dtype=torch.float32,
device=residual.device,
)
residual_cur = torch.empty_like(residual_flat)
if num_tokens <= fma_token_threshold:
# Small-batch path: one TileLang launch computes hc_post, the bf16
# residual write, GEMM partials, and the RMS square-sum partials.
mhc_fused_post_pre_fma_tilelang(
comb_res_mix.view(num_tokens, hc_mult, hc_mult),
residual_flat,
post_layer_mix.view(num_tokens, hc_mult),
x_flat,
fn.view(hc_mult3, hc_mult, hidden_size),
gemm_out_mul,
gemm_out_sqrsum,
residual_cur,
hc_mult,
hidden_size,
hc_mult3,
tile_mix_outputs=tile_n,
split_k=n_splits,
)
else:
# Large-batch path: keep the existing high-throughput TileLang hc_post +
# DeepGEMM pre-norm GEMM decomposition instead of replacing tensor cores.
mhc_post_tilelang(
comb_res_mix.view(num_tokens, hc_mult, hc_mult),
residual_flat,
post_layer_mix.view(num_tokens, hc_mult),
x_flat,
residual_cur,
hc_mult,
hidden_size,
)
if envs.SGLANG_OPT_DEEPGEMM_HC_PRENORM.get():
import deep_gemm
deep_gemm.tf32_hc_prenorm_gemm(
residual_cur.view(num_tokens, hc_hidden_size),
fn,
gemm_out_mul,
gemm_out_sqrsum,
num_splits=n_splits,
)
else:
# Fallback mirrors mhc_pre when DeepGEMM prenorm is disabled.
n_splits = 1
gemm_out_mul_2d = torch.empty(
num_tokens, hc_mult3, dtype=torch.float32, device=residual.device
)
gemm_out_sqrsum_1d = torch.empty(
num_tokens, dtype=torch.float32, device=residual.device
)
mhc_pre_gemm_sqrsum_tilelang(
residual_cur.view(num_tokens, hc_hidden_size),
fn,
gemm_out_mul_2d,
gemm_out_sqrsum_1d,
hc_mult3,
hc_hidden_size,
)
gemm_out_mul = gemm_out_mul_2d.unsqueeze(0)
gemm_out_sqrsum = gemm_out_sqrsum_1d.unsqueeze(0)
post_mix_cur = torch.empty(
num_tokens,
hc_mult,
dtype=torch.float32,
device=residual.device,
)
comb_mix_cur = torch.empty(
num_tokens,
hc_mult2,
dtype=torch.float32,
device=residual.device,
)
layer_input_cur = torch.empty(
num_tokens,
hidden_size,
dtype=torch.bfloat16,
device=residual.device,
)
if norm_weight is not None:
# Final mhc_pre stage: convert GEMM partials into post/comb/layer_input
# and fuse the following RMSNorm when the model passed a norm weight.
assert norm_eps is not None
assert norm_weight.shape == (hidden_size,)
norm_weight_bf = (
norm_weight.bfloat16()
if norm_weight.dtype != torch.bfloat16
else norm_weight
)
if not norm_weight_bf.is_contiguous():
norm_weight_bf = norm_weight_bf.contiguous()
mhc_pre_big_fuse_with_norm_tilelang(
gemm_out_mul,
gemm_out_sqrsum,
hc_scale,
hc_base,
residual_cur,
post_mix_cur,
comb_mix_cur,
layer_input_cur,
norm_weight_bf,
hidden_size,
rms_eps,
hc_pre_eps,
hc_sinkhorn_eps,
hc_post_mult_value,
sinkhorn_repeat,
norm_eps,
n_splits,
hc_mult,
hc_mult3,
)
else:
# Same mhc_pre finalization without the model-layer RMSNorm.
mhc_pre_big_fuse_tilelang(
gemm_out_mul,
gemm_out_sqrsum,
hc_scale,
hc_base,
residual_cur,
post_mix_cur,
comb_mix_cur,
layer_input_cur,
hidden_size,
rms_eps,
hc_pre_eps,
hc_sinkhorn_eps,
hc_post_mult_value,
sinkhorn_repeat,
n_splits,
hc_mult,
hc_mult3,
)
return (
residual_cur.view(*outer_shape, hc_mult, hidden_size),
post_mix_cur.view(*outer_shape, hc_mult, 1),
comb_mix_cur.view(*outer_shape, hc_mult, hc_mult),
layer_input_cur.view(*outer_shape, hidden_size),
)
python/sglang/srt/models/deepseek_v4.py
+262 -48
View File
@@ -2,6 +2,7 @@ from __future__ import annotations
import concurrent.futures
import logging
import time
from contextlib import nullcontext
from typing import (
TYPE_CHECKING,
@@ -61,6 +62,7 @@ from sglang.srt.layers.dp_attention import (
from sglang.srt.layers.layernorm import RMSNorm
from sglang.srt.layers.linear import ColumnParallelLinear, RowParallelLinear
from sglang.srt.layers.logits_processor import LogitsProcessor
from sglang.srt.layers.mhc import mhc_fused_post_pre
from sglang.srt.layers.moe import get_moe_a2a_backend
from sglang.srt.layers.moe.fused_moe_triton import FusedMoE
from sglang.srt.layers.quantization.fp8_kernel import sglang_per_token_group_quant_fp8
@@ -110,6 +112,18 @@ from sglang.srt.utils.hf_transformers_utils import get_rope_config
logger = logging.getLogger(__name__)
_FP8_WO_A_GEMM = envs.SGLANG_OPT_FP8_WO_A_GEMM.get()
_MHC_POST_MULT_VALUE = 2.0
def _is_fused_mhc_post_pre_enabled() -> bool:
# The fused path directly reuses TileLang mhc_post/mhc_pre kernels and their
# tensor layout assumptions, so keep it disabled when either dependency is off.
return (
envs.SGLANG_OPT_FUSE_MHC_POST_PRE.get()
and envs.SGLANG_OPT_USE_TILELANG_MHC_PRE.get()
and envs.SGLANG_OPT_USE_TILELANG_MHC_POST.get()
)
_use_aiter = get_bool_env_var("SGLANG_USE_AITER") and _is_hip
_is_gfx95_supported = is_gfx95_supported()
@@ -976,6 +990,133 @@ class DeepseekV4DecoderLayer(nn.Module):
self.hc_ffn_scale = nn.Parameter(torch.empty(3, dtype=torch.float32))
self.rms_norm_eps = config.rms_norm_eps
self.dsa_enable_prefill_cp = is_dsa_enable_prefill_cp()
self.use_fused_mhc_post_pre = _is_fused_mhc_post_pre_enabled()
self._input_layernorm_weight_bf16 = None
self._post_attention_layernorm_weight_bf16 = None
def refresh_mhc_norm_weight_cache(self):
# Cache bf16 norm weights so the fused path does not allocate/cast per forward.
self._input_layernorm_weight_bf16 = (
self.input_layernorm.weight.data.bfloat16().contiguous()
)
self._post_attention_layernorm_weight_bf16 = (
self.post_attention_layernorm.weight.data.bfloat16().contiguous()
)
def prewarm_mhc_token_counts(
self, token_counts: Tuple[int, ...], device: torch.device
) -> None:
paths = (
(
"attn",
self.hc_attn_fn,
self.hc_attn_scale,
self.hc_attn_base,
self.input_layernorm,
),
(
"ffn",
self.hc_ffn_fn,
self.hc_ffn_scale,
self.hc_ffn_base,
self.post_attention_layernorm,
),
)
with torch.inference_mode():
for num_tokens in token_counts:
for path_name, hc_fn, hc_scale, hc_base, norm in paths:
tic = time.perf_counter()
residual = torch.empty(
(num_tokens, self.hc_mult, self.hidden_size),
dtype=torch.bfloat16,
device=device,
)
y, post, comb, _ = self.hc_pre(
residual,
hc_fn,
hc_scale,
hc_base,
norm=norm,
)
del residual, y, post, comb
torch.cuda.synchronize()
logger.info(
"DeepSeek V4 MHC prewarm path=%s num_tokens=%s completed in %.3fs",
path_name,
num_tokens,
time.perf_counter() - tic,
)
if self.use_fused_mhc_post_pre:
for num_tokens in token_counts:
for path_name, hc_fn, hc_scale, hc_base, norm in paths:
tic = time.perf_counter()
# Dummy inputs matching the fused kernel's expected shapes.
x = torch.empty(
(num_tokens, self.hidden_size),
dtype=torch.bfloat16,
device=device,
)
residual = torch.empty(
(num_tokens, self.hc_mult, self.hidden_size),
dtype=torch.bfloat16,
device=device,
)
post_mix = torch.empty(
(num_tokens, self.hc_mult, 1),
dtype=torch.float32,
device=device,
)
comb_mix = torch.empty(
(num_tokens, self.hc_mult, self.hc_mult),
dtype=torch.float32,
device=device,
)
norm_weight = norm.weight.data.bfloat16().contiguous()
mhc_fused_post_pre(
x,
residual,
post_mix,
comb_mix,
hc_fn,
hc_scale,
hc_base,
self.rms_norm_eps,
self.hc_eps,
self.hc_eps,
_MHC_POST_MULT_VALUE,
self.hc_sinkhorn_iters,
norm_weight=norm_weight,
norm_eps=norm.variance_epsilon,
)
del x, residual, post_mix, comb_mix, norm_weight
torch.cuda.synchronize()
logger.info(
"DeepSeek V4 MHC fused prewarm path=%s num_tokens=%s completed in %.3fs",
path_name,
num_tokens,
time.perf_counter() - tic,
)
def prewarm_mhc_token_count_buckets(
self, max_num_tokens: int, device: torch.device
) -> Tuple[int, ...]:
from sglang.srt.layers.mhc import get_mhc_pre_token_count_representatives
token_counts = get_mhc_pre_token_count_representatives(
max_num_tokens, self.hc_mult * self.hidden_size
)
if not token_counts:
return token_counts
logger.info(
"DeepSeek V4 MHC prewarm max_num_tokens=%s representative token counts: %s",
max_num_tokens,
token_counts,
)
self.prewarm_mhc_token_counts(token_counts, device)
return token_counts
def hc_pre(
self,
@@ -1001,9 +1142,9 @@ class DeepseekV4DecoderLayer(nn.Module):
if x.shape[0] == 0:
y = torch.empty((0, shape[-1]), dtype=dtype, device=x.device)
post = torch.empty((0, self.hc_mult), dtype=dtype, device=x.device)
post = torch.empty((0, self.hc_mult), dtype=torch.float32, device=x.device)
comb = torch.empty(
(0, self.hc_mult, self.hc_mult), dtype=dtype, device=x.device
(0, self.hc_mult, self.hc_mult), dtype=torch.float32, device=x.device
)
return y, post, comb, False
@@ -1023,7 +1164,7 @@ class DeepseekV4DecoderLayer(nn.Module):
rms_eps=self.rms_norm_eps,
hc_pre_eps=self.hc_eps,
hc_sinkhorn_eps=self.hc_eps,
hc_post_mult_value=2.0,
hc_post_mult_value=_MHC_POST_MULT_VALUE,
sinkhorn_repeat=self.hc_sinkhorn_iters,
**norm_kwargs,
)
@@ -1040,7 +1181,7 @@ class DeepseekV4DecoderLayer(nn.Module):
rms_eps=self.rms_norm_eps,
hc_pre_eps=self.hc_eps,
hc_sinkhorn_eps=self.hc_eps,
hc_post_mult_value=2.0,
hc_post_mult_value=_MHC_POST_MULT_VALUE,
sinkhorn_repeat=self.hc_sinkhorn_iters,
)
return y, post.squeeze(-1), comb, False
@@ -1122,27 +1263,60 @@ class DeepseekV4DecoderLayer(nn.Module):
input_ids: torch.Tensor,
forward_batch: ForwardBatch,
input_ids_global: torch.Tensor,
) -> torch.Tensor:
residual = hidden_states
hidden_states, post, comb, norm_fused = self.hc_pre(
hidden_states,
self.hc_attn_fn,
self.hc_attn_scale,
self.hc_attn_base,
norm=self.input_layernorm,
)
if not norm_fused:
if _use_aiter and _is_gfx95_supported:
x_quant, hidden_states = _fused_rmsnorm_fp8_quant(
hidden_states,
self.input_layernorm.weight,
self.rms_norm_eps,
)
else:
hidden_states = self.input_layernorm(hidden_states)
x_quant = None
else:
prev_residual: Optional[torch.Tensor] = None,
prev_post: Optional[torch.Tensor] = None,
prev_comb: Optional[torch.Tensor] = None,
) -> Tuple[
torch.Tensor,
Optional[torch.Tensor],
Optional[torch.Tensor],
Optional[torch.Tensor],
]:
use_fused = self.use_fused_mhc_post_pre
if prev_residual is not None and use_fused:
residual, post, comb, hidden_states = mhc_fused_post_pre(
hidden_states,
prev_residual,
prev_post,
prev_comb,
self.hc_attn_fn,
self.hc_attn_scale,
self.hc_attn_base,
self.rms_norm_eps,
self.hc_eps,
self.hc_eps,
_MHC_POST_MULT_VALUE,
self.hc_sinkhorn_iters,
norm_weight=(
self._input_layernorm_weight_bf16
if self._input_layernorm_weight_bf16 is not None
else self.input_layernorm.weight.data
),
norm_eps=self.input_layernorm.variance_epsilon,
)
x_quant = None
else:
residual = hidden_states
hidden_states, post, comb, norm_fused = self.hc_pre(
hidden_states,
self.hc_attn_fn,
self.hc_attn_scale,
self.hc_attn_base,
norm=self.input_layernorm,
)
if not norm_fused:
if _use_aiter and _is_gfx95_supported:
x_quant, hidden_states = _fused_rmsnorm_fp8_quant(
hidden_states,
self.input_layernorm.weight,
self.rms_norm_eps,
)
else:
hidden_states = self.input_layernorm(hidden_states)
x_quant = None
else:
x_quant = None
hidden_states = self.self_attn(
x=hidden_states,
@@ -1151,35 +1325,58 @@ class DeepseekV4DecoderLayer(nn.Module):
x_quant=x_quant,
)
fused_mhc = try_fused_hc_post_pre(
hidden_states,
residual,
post,
comb,
self.hc_ffn_fn.T,
self.hc_ffn_scale,
self.hc_ffn_base,
self.hc_mult,
self.rms_norm_eps,
self.hc_eps,
2.0,
self.hc_sinkhorn_iters,
_is_gfx95_supported,
)
if fused_mhc is not None:
residual, hidden_states, post, comb, norm_fused = fused_mhc
if use_fused:
fused_mhc = try_fused_hc_post_pre(
hidden_states,
residual,
post,
comb,
self.hc_ffn_fn.T,
self.hc_ffn_scale,
self.hc_ffn_base,
self.hc_mult,
self.rms_norm_eps,
self.hc_eps,
_MHC_POST_MULT_VALUE,
self.hc_sinkhorn_iters,
_is_gfx95_supported,
)
if fused_mhc is not None:
residual, hidden_states, post, comb, norm_fused = fused_mhc
else:
residual, post, comb, hidden_states = mhc_fused_post_pre(
hidden_states,
residual,
post.unsqueeze(-1) if post.ndim == 2 else post,
comb,
self.hc_ffn_fn,
self.hc_ffn_scale,
self.hc_ffn_base,
self.rms_norm_eps,
self.hc_eps,
self.hc_eps,
_MHC_POST_MULT_VALUE,
self.hc_sinkhorn_iters,
norm_weight=(
self._post_attention_layernorm_weight_bf16
if self._post_attention_layernorm_weight_bf16 is not None
else self.post_attention_layernorm.weight.data
),
norm_eps=self.post_attention_layernorm.variance_epsilon,
)
norm_fused = True
else:
hidden_states = self.hc_post(hidden_states, residual, post, comb)
residual = hidden_states # [n, hc, d]
residual = hidden_states
hidden_states, post, comb, norm_fused = self.hc_pre(
hidden_states,
self.hc_ffn_fn,
self.hc_ffn_scale,
self.hc_ffn_base,
norm=self.post_attention_layernorm,
) # -> [n, d]
if not norm_fused:
hidden_states = self.post_attention_layernorm(hidden_states)
)
if not norm_fused:
hidden_states = self.post_attention_layernorm(hidden_states)
_use_cp = self.dsa_enable_prefill_cp and dsa_use_prefill_cp(forward_batch)
_use_tp_moe_gather = (
@@ -1233,9 +1430,13 @@ class DeepseekV4DecoderLayer(nn.Module):
attn_tp_all_gather(gathered, hidden_states.contiguous())
hidden_states = torch.cat(gathered)
hidden_states = self.hc_post(hidden_states, residual, post, comb)
if not use_fused:
hidden_states = self.hc_post(hidden_states, residual, post, comb)
return hidden_states, None, None, None
return hidden_states
# Return the deferred FFN hc_post state; the next layer consumes it with
# cross-layer fusion, and the final layer is completed in DeepseekV4Model.
return hidden_states, residual, post, comb
class DeepseekV4Model(nn.Module):
@@ -1302,6 +1503,7 @@ class DeepseekV4Model(nn.Module):
self.hc_head_scale = nn.Parameter(torch.empty(1, dtype=torch.float32))
self.dsa_enable_prefill_cp = is_dsa_enable_prefill_cp()
self.use_fused_mhc_post_pre = _is_fused_mhc_post_pre_enabled()
if self.dsa_enable_prefill_cp:
self.cp_size = get_attention_cp_size()
@@ -1375,21 +1577,32 @@ class DeepseekV4Model(nn.Module):
# forks alt-streams; later per-layer calls become no-ops.
get_attn_backend()._maybe_upgrade_forward_metadata()
use_fused = self.use_fused_mhc_post_pre
prev_residual, prev_post, prev_comb = None, None, None
last_layer = None
for i in range(self.start_layer, self.end_layer):
layer = self.layers[i]
last_layer = layer
ctx = (
nullcontext()
if not get_global_server_args().disable_piecewise_cuda_graph
else get_global_expert_distribution_recorder().with_current_layer(i)
)
with ctx:
hidden_states = layer(
hidden_states, prev_residual, prev_post, prev_comb = layer(
positions=positions,
hidden_states=hidden_states,
forward_batch=forward_batch,
input_ids=input_ids,
input_ids_global=input_ids_global,
prev_residual=prev_residual,
prev_post=prev_post,
prev_comb=prev_comb,
)
if use_fused and last_layer is not None:
hidden_states = last_layer.hc_post(
hidden_states, prev_residual, prev_post, prev_comb
)
# CP all-gather only on the last PP rank; PP IPC carries CP-split tensors.
if self.pp_group.is_last_rank and dsa_use_prefill_cp(forward_batch):
@@ -1589,6 +1802,7 @@ class DeepseekV4ForCausalLM(nn.Module):
and not self_attn.indexer.compressor.ape_converted
):
self_attn.indexer.compressor.apply_ape_hotfix()
layer.refresh_mhc_norm_weight_cache()
@staticmethod
def remap_weight_name_to_dpsk_hf_format(
python/sglang/srt/models/deepseek_v4_nextn.py
+5 -1
View File
@@ -170,13 +170,17 @@ class DeepseekV4ModelNextN(nn.Module):
hidden_states = cp_split_and_rebuild_data(forward_batch, hidden_states)
positions = cp_split_and_rebuild_position(forward_batch, positions)
hidden_states = self.decoder(
hidden_states, residual, post, comb = self.decoder(
positions=positions,
hidden_states=hidden_states,
forward_batch=forward_batch,
input_ids=input_ids,
input_ids_global=input_ids_global,
)
if residual is not None:
# NextN has a single decoder layer, so no later layer can consume a
# deferred fused hc_post state.
hidden_states = self.decoder.hc_post(hidden_states, residual, post, comb)
if dsa_use_prefill_cp(forward_batch):
hidden_states = cp_all_gather_rerange_output(