[FEAT] [ROCm] Add AITER int8 scaled gemm kernel (#15433)

Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
This commit is contained in:
TJian 2025-03-29 18:33:56 +08:00 committed by GitHub
parent 73aa7041bf
commit 4965ec42d2
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 202 additions and 5 deletions

View File

@ -20,6 +20,23 @@ from vllm.model_executor.layers.quantization.utils.w8a8_utils import (
sparse_cutlass_supported) sparse_cutlass_supported)
from vllm.platforms import current_platform from vllm.platforms import current_platform
# AITER only supports per-channel-per-channel INT8 gemm
# and per-tensor-per-tensor INT8 GEMM.
# It does not support mix precision MM and mix quantization scheme.
ROCM_AITER_SUPPORTED_INT8_MODEL = [
"neuralmagic/Llama-3.2-1B-quantized.w8a8",
"nm-testing/tinyllama-oneshot-w8a8-channel-dynamic-token-v2"
]
# TritonScaledMMLinearKernel only supports symmetric quantization.
ROCM_TRITON_SCALED_MM_SUPPORTED_INT8_MODEL = [
"nm-testing/tinyllama-oneshot-w8w8-test-static-shape-change",
"nm-testing/tinyllama-oneshot-w8-channel-a8-tensor",
"neuralmagic/Llama-3.2-1B-quantized.w8a8",
"nm-testing/tinyllama-oneshot-w8a8-dynamic-token-v2",
"nm-testing/tinyllama-oneshot-w8a8-channel-dynamic-token-v2",
]
@pytest.fixture(scope="function", autouse=True) @pytest.fixture(scope="function", autouse=True)
def use_v0_only(monkeypatch): def use_v0_only(monkeypatch):
@ -57,6 +74,11 @@ def use_v0_only(monkeypatch):
) )
def test_compressed_tensors_w8a8_static_setup(vllm_runner, model_args): def test_compressed_tensors_w8a8_static_setup(vllm_runner, model_args):
model_path, strategy, quant_type, shape_0, is_symmetric = model_args model_path, strategy, quant_type, shape_0, is_symmetric = model_args
if current_platform.is_rocm(
) and model_path not in ROCM_TRITON_SCALED_MM_SUPPORTED_INT8_MODEL:
pytest.skip(f"Skip model {model_path} as it is not support on ROCm.")
with vllm_runner(model_path, enforce_eager=True) as llm: with vllm_runner(model_path, enforce_eager=True) as llm:
def check_model(model): def check_model(model):
@ -123,6 +145,8 @@ def test_compressed_tensors_w8a8_static_setup(vllm_runner, model_args):
) )
@pytest.mark.parametrize("max_tokens", [32]) @pytest.mark.parametrize("max_tokens", [32])
@pytest.mark.parametrize("num_logprobs", [10]) @pytest.mark.parametrize("num_logprobs", [10])
@pytest.mark.parametrize(
"use_aiter", [True, False] if current_platform.is_rocm() else [False])
def test_compressed_tensors_w8a8_logprobs( def test_compressed_tensors_w8a8_logprobs(
hf_runner, hf_runner,
vllm_runner, vllm_runner,
@ -130,7 +154,21 @@ def test_compressed_tensors_w8a8_logprobs(
model_path, model_path,
max_tokens, max_tokens,
num_logprobs, num_logprobs,
use_aiter,
monkeypatch,
): ):
if current_platform.is_rocm(
) and model_path not in ROCM_TRITON_SCALED_MM_SUPPORTED_INT8_MODEL:
pytest.skip(f"Skip model {model_path} as it is not support on ROCm.")
if use_aiter:
if model_path not in ROCM_AITER_SUPPORTED_INT8_MODEL:
pytest.skip(
f"Skip model {model_path} as it is not support by aiter.")
# this will enable VLLM_ROCM_USE_AITER_LINEAR
monkeypatch.setenv("VLLM_ROCM_USE_AITER", "1")
dtype = "bfloat16" dtype = "bfloat16"
# skip language translation prompt for the static per tensor asym model # skip language translation prompt for the static per tensor asym model
@ -154,6 +192,9 @@ def test_compressed_tensors_w8a8_logprobs(
name_1="vllm", name_1="vllm",
) )
if current_platform.is_rocm():
torch.cuda.synchronize()
def test_compressed_tensors_no_enforce_eager(vllm_runner): def test_compressed_tensors_no_enforce_eager(vllm_runner):
model_path = "nm-testing/tinyllama-oneshot-w8w8-test-static-shape-change" model_path = "nm-testing/tinyllama-oneshot-w8w8-test-static-shape-change"
@ -177,8 +218,27 @@ def test_compressed_tensors_no_enforce_eager(vllm_runner):
), ),
], ],
) )
def test_compressed_tensors_w8a8_dynamic_per_token(vllm_runner, model_args): @pytest.mark.parametrize(
"use_aiter", [True, False] if current_platform.is_rocm() else [False])
def test_compressed_tensors_w8a8_dynamic_per_token(
vllm_runner,
model_args,
use_aiter,
monkeypatch,
):
model_path, strategy = model_args model_path, strategy = model_args
if current_platform.is_rocm(
) and model_path not in ROCM_TRITON_SCALED_MM_SUPPORTED_INT8_MODEL:
pytest.skip(f"Skip model {model_path} as it is not support on ROCm.")
if use_aiter:
if model_path not in ROCM_AITER_SUPPORTED_INT8_MODEL:
pytest.skip(
f"Skip model {model_path} as it is not support by aiter.")
# this will enable VLLM_ROCM_USE_AITER_LINEAR
monkeypatch.setenv("VLLM_ROCM_USE_AITER", "1")
with vllm_runner(model_path, dtype=torch.float16) as llm: with vllm_runner(model_path, dtype=torch.float16) as llm:
def check_model(model): def check_model(model):
@ -207,6 +267,8 @@ def test_compressed_tensors_w8a8_dynamic_per_token(vllm_runner, model_args):
("nm-testing/tinyllama-oneshot-w8a16-per-channel", "channel", None, 4), ("nm-testing/tinyllama-oneshot-w8a16-per-channel", "channel", None, 4),
], ],
) )
@pytest.mark.skipif(not current_platform.is_cuda(),
reason="The tests are skipped on non-CUDA platform.")
def test_compressed_tensors_wNa16(vllm_runner, wNa16_args): def test_compressed_tensors_wNa16(vllm_runner, wNa16_args):
model, strategy, group, pack_factor = wNa16_args model, strategy, group, pack_factor = wNa16_args
with vllm_runner(model) as llm: with vllm_runner(model) as llm:
@ -231,6 +293,8 @@ def test_compressed_tensors_wNa16(vllm_runner, wNa16_args):
assert output assert output
@pytest.mark.skipif(not current_platform.is_cuda(),
reason="This test is skipped on non-CUDA platform.")
def test_compressed_tensors_w4a16_marlin24(vllm_runner): def test_compressed_tensors_w4a16_marlin24(vllm_runner):
model_path = "nm-testing/llama7b-one-shot-2_4-w4a16-marlin24-t" model_path = "nm-testing/llama7b-one-shot-2_4-w4a16-marlin24-t"
with vllm_runner(model_path) as llm: with vllm_runner(model_path) as llm:
@ -271,7 +335,7 @@ def test_compressed_tensors_fp8(vllm_runner):
if isinstance(qkv_proj.scheme, CompressedTensorsW8A8Fp8): if isinstance(qkv_proj.scheme, CompressedTensorsW8A8Fp8):
assert len(qkv_proj.input_scale.shape) == 0 assert len(qkv_proj.input_scale.shape) == 0
assert qkv_proj.weight.dtype is torch.float8_e4m3fn assert qkv_proj.weight.dtype is current_platform.fp8_dtype()
assert qkv_proj.weight_scale.dtype is torch.float32 assert qkv_proj.weight_scale.dtype is torch.float32
assert len(qkv_proj.weight_scale.shape) == 0 assert len(qkv_proj.weight_scale.shape) == 0
@ -281,6 +345,8 @@ def test_compressed_tensors_fp8(vllm_runner):
assert output assert output
@pytest.mark.skipif(not current_platform.is_cuda(),
reason="This test is skipped on non-CUDA platform.")
def test_compressed_tensors_kv_cache(vllm_runner): def test_compressed_tensors_kv_cache(vllm_runner):
model_path = "nm-testing/TinyLlama-1.1B-compressed-tensors-kv-cache-scheme" model_path = "nm-testing/TinyLlama-1.1B-compressed-tensors-kv-cache-scheme"
with vllm_runner(model_path, kv_cache_dtype="fp8") as llm: with vllm_runner(model_path, kv_cache_dtype="fp8") as llm:
@ -309,7 +375,8 @@ def _test_2of4_quant_models(qkv_proj,
@pytest.mark.skipif( @pytest.mark.skipif(
not current_platform.has_device_capability(90), not current_platform.is_cuda()
or not current_platform.has_device_capability(90),
reason="Sparse FP8 is not yet supported on this GPU type.", reason="Sparse FP8 is not yet supported on this GPU type.",
) )
@pytest.mark.parametrize( @pytest.mark.parametrize(
@ -356,7 +423,8 @@ def test_compressed_tensors_2of4_quant_fp8(vllm_runner, args_2of4):
@pytest.mark.skipif( @pytest.mark.skipif(
not current_platform.has_device_capability(90), not current_platform.is_cuda()
or not current_platform.has_device_capability(90),
reason="Sparse FP8 is not yet supported on this GPU type.", reason="Sparse FP8 is not yet supported on this GPU type.",
) )
@pytest.mark.parametrize( @pytest.mark.parametrize(

View File

@ -75,6 +75,7 @@ if TYPE_CHECKING:
VLLM_DISABLED_KERNELS: list[str] = [] VLLM_DISABLED_KERNELS: list[str] = []
VLLM_USE_V1: bool = True VLLM_USE_V1: bool = True
VLLM_ROCM_USE_AITER: bool = False VLLM_ROCM_USE_AITER: bool = False
VLLM_ROCM_USE_AITER_LINEAR: bool = True
VLLM_ROCM_USE_AITER_MOE: bool = True VLLM_ROCM_USE_AITER_MOE: bool = True
VLLM_ROCM_USE_AITER_FP8_BLOCK_SCALED_MOE: bool = False VLLM_ROCM_USE_AITER_FP8_BLOCK_SCALED_MOE: bool = False
VLLM_ROCM_USE_AITER_RMSNORM: bool = True VLLM_ROCM_USE_AITER_RMSNORM: bool = True
@ -524,6 +525,13 @@ environment_variables: dict[str, Callable[[], Any]] = {
lambda: (os.getenv("VLLM_ROCM_USE_AITER", "False").lower() in lambda: (os.getenv("VLLM_ROCM_USE_AITER", "False").lower() in
("true", "1")), ("true", "1")),
# use aiter linear op if aiter ops are enabled
# The following list of related ops
# - scaled_mm (per-tensor / rowwise)
"VLLM_ROCM_USE_AITER_LINEAR":
lambda: (os.getenv("VLLM_ROCM_USE_AITER_LINEAR", "True").lower() in
("true", "1")),
# Whether to use aiter moe ops. # Whether to use aiter moe ops.
# By default is enabled. # By default is enabled.
"VLLM_ROCM_USE_AITER_MOE": "VLLM_ROCM_USE_AITER_MOE":

View File

@ -3,6 +3,8 @@
import os import os
from typing import Dict, List, Optional, Type from typing import Dict, List, Optional, Type
from vllm.model_executor.layers.quantization.kernels.scaled_mm.aiter import (
AiterScaledMMLinearKernel)
from vllm.model_executor.layers.quantization.kernels.scaled_mm.cutlass import ( from vllm.model_executor.layers.quantization.kernels.scaled_mm.cutlass import (
CutlassScaledMMLinearKernel) CutlassScaledMMLinearKernel)
from vllm.model_executor.layers.quantization.kernels.scaled_mm.ScaledMMLinearKernel import ( # noqa: E501 from vllm.model_executor.layers.quantization.kernels.scaled_mm.ScaledMMLinearKernel import ( # noqa: E501
@ -17,7 +19,7 @@ from vllm.platforms import PlatformEnum, current_platform
_POSSIBLE_KERNELS: Dict[PlatformEnum, List[Type[ScaledMMLinearKernel]]] = { _POSSIBLE_KERNELS: Dict[PlatformEnum, List[Type[ScaledMMLinearKernel]]] = {
PlatformEnum.CPU: [CutlassScaledMMLinearKernel], PlatformEnum.CPU: [CutlassScaledMMLinearKernel],
PlatformEnum.CUDA: [CutlassScaledMMLinearKernel], PlatformEnum.CUDA: [CutlassScaledMMLinearKernel],
PlatformEnum.ROCM: [TritonScaledMMLinearKernel], PlatformEnum.ROCM: [AiterScaledMMLinearKernel, TritonScaledMMLinearKernel],
PlatformEnum.TPU: [XLAScaledMMLinearKernel], PlatformEnum.TPU: [XLAScaledMMLinearKernel],
} }

View File

@ -0,0 +1,119 @@
# SPDX-License-Identifier: Apache-2.0
from typing import Optional, Tuple
import torch
import vllm.envs as envs
from vllm import _custom_ops as ops
from vllm.platforms import current_platform
from .cutlass import CutlassScaledMMLinearKernel
from .ScaledMMLinearKernel import ScaledMMLinearLayerConfig
class AiterScaledMMLinearKernel(CutlassScaledMMLinearKernel):
@classmethod
def get_min_capability(cls) -> int:
return 90
@classmethod
def can_implement(
cls, c: ScaledMMLinearLayerConfig) -> Tuple[bool, Optional[str]]:
if not current_platform.is_rocm():
return (
False,
"AiterScaledMMLinearKernel requires `aiter` which is not " +
"currently supported on non-ROCm platform.")
try:
import aiter # noqa: F401 # deliberately attempt to import aiter
except Exception:
return (
False,
"AiterScaledMMLinearKernel requires `aiter` which is not " +
"installed on ROCm.")
# Check if rocm_aiter_gemm_w8a8_scaled_mm is enabled
if not (
envs.VLLM_ROCM_USE_AITER_LINEAR \
and envs.VLLM_ROCM_USE_AITER
):
return (False, "AiterScaledMMLinearKernel is disabled. " +
"Enable by setting `VLLM_ROCM_USE_AITER=1` " +
"and `VLLM_ROCM_USE_AITER_LINEAR=1`. " +
"`VLLM_ROCM_USE_AITER_LINEAR` default is True.")
if not c.input_symmetric:
return (False,
"AiterScaledMMLinearKernel only supports symmetric " +
"quantization.")
return True, None
def process_weights_after_loading(self, layer: torch.nn.Module) -> None:
super().process_weights_after_loading(layer)
def apply_weights(self,
layer: torch.nn.Module,
x: torch.Tensor,
bias: Optional[torch.Tensor] = None) -> torch.Tensor:
"""
`AiterScaledMMLinearKernel` implements a fused version of
`output = torch.mm((scale_a * a), (scale_b * b)).to(out_dtype)`
where scale_a * a and scale_b * b are implemented using numpy-style
broadcasting.
Currently only support per-tensor-per-tensor GEMM
and per-token-per-channel GEMM through AITER
w8a8 scaled gemm. `AiterScaledMMLinearKernel` also does not support
ATIER block scaled GEMM and mix-precision GEMM.
"""
w_q, w_s, i_s, i_zp, azp_adj = self._get_weight_params(layer)
# ops.scaled_int8_quant supports both dynamic and static quant:
# * dynamic, i_s is None and x_s computed from x.
# * static, i_s is scalar and x_s is i_s.
symmetric = azp_adj is None
assert symmetric, ("AiterScaledMMLinearKernel only supports"
" symmetric quantization.")
x_q, x_s, x_zp = ops.scaled_int8_quant(x,
i_s,
i_zp,
symmetric=symmetric)
assert x_zp is None, ("AiterScaledMMLinearKernel only supports"
" symmetric quantization.")
out_dtype = x.dtype
assert (w_q.shape[0] % 16 == 0 and w_q.shape[1] % 16 == 0)
assert (out_dtype is torch.bfloat16 or out_dtype is torch.float16)
assert bias is None or bias.shape[0] == w_q.shape[
1] and bias.dtype == out_dtype
m = x_q.shape[0] # a
n = w_q.shape[1] # b
per_tensor_scale_a = (x_s.numel() == 1)
per_tensor_scale_b = (w_s.numel() == 1)
per_token_scale_a = (x_s.numel() == m)
per_channel_scale_b = (w_s.numel() == n)
# @TODO:
# Maybe broadcast the per-tensor-scale into per-channel-scale
# if one of the scale is a per-channel-scale.
# For now, it only supports:
# - per-tensor-per-tensor a8w8 scaled GEMM, and
# - per-token-per-channel a8w8 scaled GEMM
assert ((per_tensor_scale_a and per_tensor_scale_b)
or (per_token_scale_a and per_channel_scale_b)), (
"Currently only support per-tensor-per-tensor GEMM " +
" and per-token-per-channel GEMM through AITER"
" w8a8 scaled gemm. `AiterScaledMMLinearKernel` " +
"does not support AITER block scaled GEMM.")
from aiter import gemm_a8w8_CK
# gemm_a8w8_CK(a, b, scale_a, scale_b, bias) expects
# a to be [M, K]
# b to be [N, K]
# CutlassScaledMMLinearKernel prepare weight `w_q` in [K, N] format
return gemm_a8w8_CK(x_q, w_q.t(), x_s, w_s, bias).to(out_dtype)