liangsu9988 commited on
Commit
b6972b2
·
verified ·
1 Parent(s): 88b8ef2

Add torch211-cxx11-cu130-aarch64-linux aarch64 artifact

Browse files
build/torch211-cxx11-cu130-aarch64-linux/__init__.py CHANGED
@@ -58,6 +58,16 @@ def _legacy_linear_fake(
58
  return None
59
 
60
 
 
 
 
 
 
 
 
 
 
 
61
  @torch.library.register_fake(add_op_namespace_prefix("quantize_fp4_sfa_fp16"))
62
  def _quant_fake(x: torch.Tensor, packed: torch.Tensor, sfa: torch.Tensor, is_sfb: bool = False) -> None:
63
  return None
@@ -191,6 +201,45 @@ def fp4_w4a16_linear_bf16(
191
  )
192
 
193
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
  def nvfp4_gemm_residual_bf16(
195
  a_packed: torch.Tensor,
196
  b_packed: torch.Tensor,
@@ -296,8 +345,10 @@ __all__ = [
296
  "fp4_w4a16_linear_bf16",
297
  "fp4_w4a4_gemv_warpsplit_bf16",
298
  "nvfp4_gemm_bf16",
 
299
  "nvfp4_gemm_bias_gelu_bf16",
300
  "nvfp4_gemm_bias_gelu_nvfp4",
 
301
  "nvfp4_gemm_residual_bf16",
302
  "nvfp4_gemm_streamk_bf16",
303
  "nvfp4_gemm_streamk_bias_bf16",
 
58
  return None
59
 
60
 
61
+ @torch.library.register_fake(add_op_namespace_prefix("nvfp4_gemm_bias_bf16"))
62
+ def _bias_fake(a, b, sfa, sfb, bias, out) -> None:
63
+ return None
64
+
65
+
66
+ @torch.library.register_fake(add_op_namespace_prefix("nvfp4_gemm_bias_residual_bf16"))
67
+ def _bias_residual_fake(a, b, sfa, sfb, bias, residual, out) -> None:
68
+ return None
69
+
70
+
71
  @torch.library.register_fake(add_op_namespace_prefix("quantize_fp4_sfa_fp16"))
72
  def _quant_fake(x: torch.Tensor, packed: torch.Tensor, sfa: torch.Tensor, is_sfb: bool = False) -> None:
73
  return None
 
201
  )
202
 
203
 
204
+ def nvfp4_gemm_bias_bf16(
205
+ a_packed: torch.Tensor,
206
+ b_packed: torch.Tensor,
207
+ sfa: torch.Tensor,
208
+ sfb: torch.Tensor,
209
+ bias: torch.Tensor,
210
+ *,
211
+ out: torch.Tensor | None = None,
212
+ ) -> torch.Tensor:
213
+ """SM110 NVFP4 GEMM with a fused per-column BF16 bias."""
214
+ if out is None:
215
+ out = torch.empty(
216
+ (a_packed.shape[0], b_packed.shape[0]),
217
+ device=a_packed.device,
218
+ dtype=torch.bfloat16,
219
+ )
220
+ ops.nvfp4_gemm_bias_bf16(a_packed, b_packed, sfa, sfb, bias, out)
221
+ return out
222
+
223
+
224
+ def nvfp4_gemm_bias_residual_bf16(
225
+ a_packed: torch.Tensor,
226
+ b_packed: torch.Tensor,
227
+ sfa: torch.Tensor,
228
+ sfb: torch.Tensor,
229
+ bias: torch.Tensor,
230
+ residual: torch.Tensor,
231
+ *,
232
+ out: torch.Tensor | None = None,
233
+ ) -> torch.Tensor:
234
+ """SM110 NVFP4 GEMM with fused BF16 bias and residual add."""
235
+ if out is None:
236
+ out = torch.empty_like(residual)
237
+ ops.nvfp4_gemm_bias_residual_bf16(
238
+ a_packed, b_packed, sfa, sfb, bias, residual, out
239
+ )
240
+ return out
241
+
242
+
243
  def nvfp4_gemm_residual_bf16(
244
  a_packed: torch.Tensor,
245
  b_packed: torch.Tensor,
 
345
  "fp4_w4a16_linear_bf16",
346
  "fp4_w4a4_gemv_warpsplit_bf16",
347
  "nvfp4_gemm_bf16",
348
+ "nvfp4_gemm_bias_bf16",
349
  "nvfp4_gemm_bias_gelu_bf16",
350
  "nvfp4_gemm_bias_gelu_nvfp4",
351
+ "nvfp4_gemm_bias_residual_bf16",
352
  "nvfp4_gemm_residual_bf16",
353
  "nvfp4_gemm_streamk_bf16",
354
  "nvfp4_gemm_streamk_bias_bf16",
build/torch211-cxx11-cu130-aarch64-linux/_fp4_gemm_cuda_8a66d8b.abi3.so ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35bd2043a1a1880f362954fe72a22651017d30a585d38ed5d2ffe86de286b45a
3
+ size 1118808
build/torch211-cxx11-cu130-aarch64-linux/_ops.py CHANGED
@@ -1,6 +1,6 @@
1
  import torch
2
- from . import _fp4_gemm_cuda_b46a817
3
- ops = torch.ops._fp4_gemm_cuda_b46a817
4
 
5
  def add_op_namespace_prefix(op_name: str):
6
- return f"_fp4_gemm_cuda_b46a817::{op_name}"
 
1
  import torch
2
+ from . import _fp4_gemm_cuda_8a66d8b
3
+ ops = torch.ops._fp4_gemm_cuda_8a66d8b
4
 
5
  def add_op_namespace_prefix(op_name: str):
6
+ return f"_fp4_gemm_cuda_8a66d8b::{op_name}"
build/torch211-cxx11-cu130-aarch64-linux/metadata.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "name": "fp4-gemm",
3
- "id": "_fp4_gemm_cuda_b46a817",
4
  "version": 1,
5
  "license": "Apache-2.0",
6
  "python-depends": [],
@@ -13,20 +13,16 @@
13
  "digest": {
14
  "algorithm": "sha256",
15
  "files": {
16
- "__init__.py": "pJlGLDlOKnju+m59WyZJjpynMTvkPQTF6w+z4YfCwSY=",
17
- "_fp4_gemm_cuda_b46a817.abi3.so": "ScrXtLeiNtFCOMmagyvcEGxehJTIzongAMO8dbRcwGE=",
18
- "_ops.py": "PqcxyYzP4Co0y9NUZPOomwASQWuPPNn5QjtKrfTiZ2k=",
19
  "fp4_gemm/__init__.py": "v6p5XMfQzddhi1fLSAw4HX9CyS0rQsidvu9VsT01xi4="
20
  }
21
  },
22
  "provenance": {
23
  "kernel": {
24
- "sha": "b46a81771fdabe26f2c2494dc2ecf8492c88c45a",
25
  "dirty": false
26
- },
27
- "validation": {
28
- "torch": "2.11.0+cu130",
29
- "cuda": "13.0"
30
  }
31
  }
32
  }
 
1
  {
2
  "name": "fp4-gemm",
3
+ "id": "_fp4_gemm_cuda_8a66d8b",
4
  "version": 1,
5
  "license": "Apache-2.0",
6
  "python-depends": [],
 
13
  "digest": {
14
  "algorithm": "sha256",
15
  "files": {
16
+ "__init__.py": "+Kk/VNnIWwe9nszQYvL2qtU3DpKxbqHVfklg5WtVuB4=",
17
+ "_fp4_gemm_cuda_8a66d8b.abi3.so": "Nb0gQ6GhiA82KVT+cqImUQF9MKWF047V0v/obeKGtFo=",
18
+ "_ops.py": "/hna1MwmSjR4pewWf12X5smsDRcaiHuoR2Gri/PrBFc=",
19
  "fp4_gemm/__init__.py": "v6p5XMfQzddhi1fLSAw4HX9CyS0rQsidvu9VsT01xi4="
20
  }
21
  },
22
  "provenance": {
23
  "kernel": {
24
+ "sha": "8a66d8b",
25
  "dirty": false
 
 
 
 
26
  }
27
  }
28
  }