File size: 8,859 Bytes
5c93746
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
# Copyright 2024-2025 The Alibaba Wan Team Authors. All rights reserved.
import torch

try:
    import flash_attn_interface
    FLASH_ATTN_3_AVAILABLE = True
except (ImportError, ModuleNotFoundError):
    FLASH_ATTN_3_AVAILABLE = False

try:
    import flash_attn
    FLASH_ATTN_2_AVAILABLE = hasattr(flash_attn, "flash_attn_varlen_func")
except (ImportError, ModuleNotFoundError):
    FLASH_ATTN_2_AVAILABLE = False

import warnings

__all__ = [
    'flash_attention',
    'attention',
]


def _prepare_sdpa_inputs(q, k, v, dtype):
    q = q.transpose(1, 2)
    k = k.transpose(1, 2)
    v = v.transpose(1, 2)

    if q.device.type == 'cpu' and dtype in (torch.float16, torch.bfloat16):
        q = q.float()
        k = k.float()
        v = v.float()
    else:
        q = q.to(dtype)
        k = k.to(dtype)
        v = v.to(dtype)

    return q, k, v


def _build_length_mask(batch_size, q_len, k_len, device, q_lens, k_lens, causal, window_size):
    mask = torch.ones((batch_size, q_len, k_len), dtype=torch.bool, device=device)

    q_idx = torch.arange(q_len, device=device).view(1, q_len, 1)
    k_idx = torch.arange(k_len, device=device).view(1, 1, k_len)

    if q_lens is not None:
        q_lens = q_lens.to(device=device, dtype=torch.long)
        mask = mask & (q_idx < q_lens.view(batch_size, 1, 1))

    if k_lens is not None:
        k_lens = k_lens.to(device=device, dtype=torch.long)
        mask = mask & (k_idx < k_lens.view(batch_size, 1, 1))

    if causal:
        mask = mask & (k_idx <= q_idx)

    if window_size != (-1, -1):
        left, right = window_size
        if left >= 0:
            mask = mask & (k_idx >= q_idx - left)
        if right >= 0:
            mask = mask & (k_idx <= q_idx + right)

    return mask.unsqueeze(1)


def _merge_sdpa_masks(length_mask, attn_mask, dtype):
    if attn_mask is None:
        return length_mask

    if attn_mask.dtype == torch.bool:
        return length_mask & attn_mask

    additive_mask = torch.zeros_like(length_mask, dtype=dtype)
    additive_mask = additive_mask.masked_fill(~length_mask, float('-inf'))
    return additive_mask + attn_mask.to(dtype)


def _sdpa_attention_fallback(
    q,
    k,
    v,
    q_lens=None,
    k_lens=None,
    dropout_p=0.,
    softmax_scale=None,
    q_scale=None,
    causal=False,
    window_size=(-1, -1),
    dtype=torch.bfloat16,
    attn_mask=None,
):
    out_dtype = q.dtype
    batch_size, q_len, k_len = q.size(0), q.size(1), k.size(1)

    q, k, v = _prepare_sdpa_inputs(q, k, v, dtype)

    total_scale = 1.0
    if q_scale is not None:
        total_scale *= q_scale
    if softmax_scale is not None:
        total_scale *= softmax_scale
    if total_scale != 1.0:
        q = q * total_scale

    mask = _build_length_mask(
        batch_size=batch_size,
        q_len=q_len,
        k_len=k_len,
        device=q.device,
        q_lens=q_lens,
        k_lens=k_lens,
        causal=causal,
        window_size=window_size,
    )
    mask = _merge_sdpa_masks(mask, attn_mask, q.dtype)

    out = torch.nn.functional.scaled_dot_product_attention(
        q,
        k,
        v,
        attn_mask=mask,
        is_causal=False,
        dropout_p=dropout_p,
    )

    if q_lens is not None:
        q_valid = (
            torch.arange(q_len, device=out.device).view(1, q_len, 1)
            < q_lens.to(device=out.device, dtype=torch.long).view(batch_size, 1, 1)
        ).unsqueeze(1)
        out = out.masked_fill(~q_valid, 0)

    return out.transpose(1, 2).contiguous().to(out_dtype)


def flash_attention(
    q,
    k,
    v,
    q_lens=None,
    k_lens=None,
    dropout_p=0.,
    softmax_scale=None,
    q_scale=None,
    causal=False,
    window_size=(-1, -1),
    deterministic=False,
    dtype=torch.bfloat16,
    version=None,
):
    """
    q:              [B, Lq, Nq, C1].
    k:              [B, Lk, Nk, C1].
    v:              [B, Lk, Nk, C2]. Nq must be divisible by Nk.
    q_lens:         [B].
    k_lens:         [B].
    dropout_p:      float. Dropout probability.
    softmax_scale:  float. The scaling of QK^T before applying softmax.
    causal:         bool. Whether to apply causal attention mask.
    window_size:    (left right). If not (-1, -1), apply sliding window local attention.
    deterministic:  bool. If True, slightly slower and uses more memory.
    dtype:          torch.dtype. Apply when dtype of q/k/v is not float16/bfloat16.
    """
    half_dtypes = (torch.float16, torch.bfloat16)
    assert dtype in half_dtypes
    if not (FLASH_ATTN_2_AVAILABLE or FLASH_ATTN_3_AVAILABLE):
        warnings.warn(
            'flash_attn is not installed; falling back to scaled_dot_product_attention.',
            stacklevel=2,
        )
        return _sdpa_attention_fallback(
            q=q,
            k=k,
            v=v,
            q_lens=q_lens,
            k_lens=k_lens,
            dropout_p=dropout_p,
            softmax_scale=softmax_scale,
            q_scale=q_scale,
            causal=causal,
            window_size=window_size,
            dtype=dtype,
        )

    assert q.device.type == 'cuda' and q.size(-1) <= 256

    # params
    b, lq, lk, out_dtype = q.size(0), q.size(1), k.size(1), q.dtype

    def half(x):
        return x if x.dtype in half_dtypes else x.to(dtype)

    # preprocess query
    if q_lens is None:
        q = half(q.flatten(0, 1))
        q_lens = torch.tensor(
            [lq] * b, dtype=torch.int32).to(
                device=q.device, non_blocking=True)
    else:
        q = half(torch.cat([u[:v] for u, v in zip(q, q_lens)]))

    # preprocess key, value
    if k_lens is None:
        k = half(k.flatten(0, 1))
        v = half(v.flatten(0, 1))
        k_lens = torch.tensor(
            [lk] * b, dtype=torch.int32).to(
                device=k.device, non_blocking=True)
    else:
        k = half(torch.cat([u[:v] for u, v in zip(k, k_lens)]))
        v = half(torch.cat([u[:v] for u, v in zip(v, k_lens)]))

    q = q.to(v.dtype)
    k = k.to(v.dtype)

    if q_scale is not None:
        q = q * q_scale

    if version is not None and version == 3 and not FLASH_ATTN_3_AVAILABLE:
        warnings.warn(
            'Flash attention 3 is not available, use flash attention 2 instead.'
        )

    # apply attention
    if (version is None or version == 3) and FLASH_ATTN_3_AVAILABLE:
        # Note: dropout_p, window_size are not supported in FA3 now.
        x = flash_attn_interface.flash_attn_varlen_func(
            q=q,
            k=k,
            v=v,
            cu_seqlens_q=torch.cat([q_lens.new_zeros([1]), q_lens]).cumsum(
                0, dtype=torch.int32).to(q.device, non_blocking=True),
            cu_seqlens_k=torch.cat([k_lens.new_zeros([1]), k_lens]).cumsum(
                0, dtype=torch.int32).to(q.device, non_blocking=True),
            max_seqlen_q=lq,
            max_seqlen_k=lk,
            softmax_scale=softmax_scale,
            causal=causal,
            deterministic=deterministic)[0].unflatten(0, (b, lq))
    else:
        assert FLASH_ATTN_2_AVAILABLE
        x = flash_attn.flash_attn_varlen_func(
            q=q,
            k=k,
            v=v,
            cu_seqlens_q=torch.cat([q_lens.new_zeros([1]), q_lens]).cumsum(
                0, dtype=torch.int32).to(q.device, non_blocking=True),
            cu_seqlens_k=torch.cat([k_lens.new_zeros([1]), k_lens]).cumsum(
                0, dtype=torch.int32).to(q.device, non_blocking=True),
            max_seqlen_q=lq,
            max_seqlen_k=lk,
            dropout_p=dropout_p,
            softmax_scale=softmax_scale,
            causal=causal,
            window_size=window_size,
            deterministic=deterministic).unflatten(0, (b, lq))

    # output
    return x.type(out_dtype)


def attention(
    q,
    k,
    v,
    q_lens=None,
    k_lens=None,
    dropout_p=0.,
    softmax_scale=None,
    q_scale=None,
    causal=False,
    window_size=(-1, -1),
    deterministic=False,
    dtype=torch.bfloat16,
    fa_version=None,
    attn_mask=None,
):
    if FLASH_ATTN_2_AVAILABLE or FLASH_ATTN_3_AVAILABLE:
        return flash_attention(
            q=q,
            k=k,
            v=v,
            q_lens=q_lens,
            k_lens=k_lens,
            dropout_p=dropout_p,
            softmax_scale=softmax_scale,
            q_scale=q_scale,
            causal=causal,
            window_size=window_size,
            deterministic=deterministic,
            dtype=dtype,
            version=fa_version,
        )
    else:
        return _sdpa_attention_fallback(
            q=q,
            k=k,
            v=v,
            q_lens=q_lens,
            k_lens=k_lens,
            dropout_p=dropout_p,
            softmax_scale=softmax_scale,
            q_scale=q_scale,
            causal=causal,
            window_size=window_size,
            dtype=dtype,
            attn_mask=attn_mask,
        )