Skip to content

[Fwd,Sm100] Extract named barriers #2309

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
drisspg merged 1 commit into from
Mar 9, 2026
+20 −23
Merged

[Fwd,Sm100] Extract named barriers #2309

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 7 additions & 23 deletions flash_attn/cute/flash_fwd_sm100.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,6 @@
# https://github.com/NVIDIA/cutlass/tree/main/examples/77_blackwell_fmha
# https://github.com/NVIDIA/cutlass/blob/main/examples/python/CuTeDSL/blackwell/fmha.py

import enum
import math
from typing import Type, Tuple, Callable, Optional, Literal
from functools import partial
Expand Down Expand Up @@ -50,6 +49,7 @@
from flash_attn.cute.pack_gqa import PackGQA
from flash_attn.cute import mma_sm100_desc as sm100_desc
from flash_attn.cute import blackwell_helpers as sm100_utils
from flash_attn.cute.named_barrier import NamedBarrierFwdSm100
from cutlass.cute import FastDivmodDivisor
from quack.cute_dsl_utils import ParamsBase
from flash_attn.cute.tile_scheduler import (
Expand All @@ -60,22 +60,6 @@
SingleTileVarlenScheduler,
)


class NamedBarrierFwd(enum.IntEnum):
Epilogue = enum.auto() # starts from 1 as barrier 0 is reserved for sync_threads()
TmemPtr = enum.auto()
SoftmaxStatsW0 = enum.auto()
SoftmaxStatsW1 = enum.auto()
SoftmaxStatsW2 = enum.auto()
SoftmaxStatsW3 = enum.auto()
SoftmaxStatsW4 = enum.auto()
SoftmaxStatsW5 = enum.auto()
SoftmaxStatsW6 = enum.auto()
SoftmaxStatsW7 = enum.auto()
# WarpSchedulerWG1 = enum.auto()
# WarpSchedulerWG2 = enum.auto()


class FlashAttentionForwardSm100:

def __init__(
Expand Down Expand Up @@ -814,7 +798,7 @@ def kernel(
storage = smem.allocate(self.shared_storage)

tmem_alloc_barrier = pipeline.NamedBarrier(
barrier_id=int(NamedBarrierFwd.TmemPtr),
barrier_id=int(NamedBarrierFwdSm100.TmemPtr),
num_threads=cute.arch.WARP_SIZE * len(
(self.mma_warp_id,
*self.softmax0_warp_ids,
Expand Down Expand Up @@ -938,7 +922,7 @@ def kernel(
)
# Should put the NamedBarrier inside the pipeline class so we'll just have pipeline_sm_stats
sm_stats_barrier = pipeline_custom.NamedBarrier(
barrier_id=int(NamedBarrierFwd.SoftmaxStatsW0), num_threads=cute.arch.WARP_SIZE * 2
barrier_id=int(NamedBarrierFwdSm100.SoftmaxStatsW0), num_threads=cute.arch.WARP_SIZE * 2
)
pipeline_o_epi = None
if const_expr(not self.use_correction_warps_for_epi):
Expand Down Expand Up @@ -2574,7 +2558,7 @@ def correction_epilogue(
if const_expr(self.use_correction_warps_for_epi):
assert(not self.use_tma_O)
assert(gmem_tiled_copy_O is not None)
cute.arch.barrier(barrier_id=int(NamedBarrierFwd.Epilogue),
cute.arch.barrier(barrier_id=int(NamedBarrierFwdSm100.Epilogue),
number_of_threads=len(self.epilogue_warp_ids) * cute.arch.WARP_SIZE)
mma_tile_coord_v = thr_mma.thr_idx
m_tile_idx = (m_block * self.q_stage + stage) * self.cta_group_size + mma_tile_coord_v
Expand Down Expand Up @@ -2774,20 +2758,20 @@ def offset_kv_smem(self, sX: cute.Tensor, stage: Int32, phase: Int32):
# warp_group_idx = utils.canonical_warp_group_idx(sync=False)
# if warp_group_idx == 0:
# cute.arch.barrier_arrive(
# barrier_id=int(NamedBarrierFwd.WarpSchedulerWG1), number_of_threads=2 * 128,
# barrier_id=int(NamedBarrierFwdSm100.WarpSchedulerWG1), number_of_threads=2 * 128,
# )

# def warp_scheduler_barrier_sync(self):
# cute.arch.barrier(
# barrier_id=int(NamedBarrierFwd.WarpSchedulerWG1) + utils.canonical_warp_group_idx(sync=False),
# barrier_id=int(NamedBarrierFwdSm100.WarpSchedulerWG1) + utils.canonical_warp_group_idx(sync=False),
# number_of_threads=2 * 128
# )

# def warp_scheduler_barrier_arrive(self):
# cur_wg = utils.canonical_warp_group_idx(sync=False)
# next_wg = 1 - cur_wg
# cute.arch.barrier_arrive(
# barrier_id=int(NamedBarrierFwd.WarpSchedulerWG1) + next_wg, number_of_threads=2 * 128,
# barrier_id=int(NamedBarrierFwdSm100.WarpSchedulerWG1) + next_wg, number_of_threads=2 * 128,
# )

@cute.jit
Expand Down
13 changes: 13 additions & 0 deletions flash_attn/cute/named_barrier.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,19 @@ class NamedBarrierFwd(enum.IntEnum):
PEmpty = enum.auto()


class NamedBarrierFwdSm100(enum.IntEnum):
Epilogue = enum.auto() # starts from 1 as barrier 0 is reserved for sync_threads()
TmemPtr = enum.auto()
SoftmaxStatsW0 = enum.auto()
SoftmaxStatsW1 = enum.auto()
SoftmaxStatsW2 = enum.auto()
SoftmaxStatsW3 = enum.auto()
SoftmaxStatsW4 = enum.auto()
SoftmaxStatsW5 = enum.auto()
SoftmaxStatsW6 = enum.auto()
SoftmaxStatsW7 = enum.auto()


class NamedBarrierBwd(enum.IntEnum):
Epilogue = enum.auto()
WarpSchedulerWG1 = enum.auto()
Expand Down