EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Update test_flashinfer (#560)

Browse Source
This commit is contained in:
Liangsheng Yin
2024-06-24 15:23:57 +08:00
committed by GitHub
parent 1fa15099d8
commit 05471f2103
1 changed files with 82 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files

113
test/srt/test_flashinfer.py
View File

@@ -1,18 +1,24 @@
import flashinfer
import pytest import pytest
import torch import torch
from flashinfer import (
BatchDecodeWithPagedKVCacheWrapper,
BatchPrefillWithPagedKVCacheWrapper,
)
from flashinfer.decode import _grouped_size_compiled_for_decode_kernels
from sglang.srt.layers.extend_attention import extend_attention_fwd from sglang.srt.layers.extend_attention import extend_attention_fwd, redundant_attention
from sglang.srt.layers.token_attention import token_attention_fwd from sglang.srt.layers.token_attention import token_attention_fwd
flashinfer_prefill_wrapper = None
flashinfer_decode_wrapper = None
@pytest.mark.parametrize("batch_size", [12, 37, 67]) @pytest.mark.parametrize("batch_size", [12, 37, 67])
@pytest.mark.parametrize("kv_len", [54, 97]) @pytest.mark.parametrize("kv_len", [54, 97])
@pytest.mark.parametrize("qo_len", [37, 17]) @pytest.mark.parametrize("qo_len", [37, 17])
@pytest.mark.parametrize("num_kv_heads", [4]) @pytest.mark.parametrize("num_kv_heads", [4])
@pytest.mark.parametrize("num_qo_heads", [4, 32]) @pytest.mark.parametrize("num_qo_heads", [32, 4])
@pytest.mark.parametrize("head_dim", [128]) @pytest.mark.parametrize("head_dim", [128])
@pytest.mark.parametrize("use_wrapper", [True, False])
def test_batch_prefill_with_paged_kv_cache( def test_batch_prefill_with_paged_kv_cache(
batch_size, batch_size,
kv_len, kv_len,
@@ -20,12 +26,13 @@ def test_batch_prefill_with_paged_kv_cache(
num_kv_heads, num_kv_heads,
num_qo_heads, num_qo_heads,
head_dim, head_dim,
use_wrapper,
): ):
init_flashinfer(num_qo_heads, num_kv_heads)
q = torch.randn(batch_size * qo_len, num_qo_heads, head_dim).to(0).half() q = torch.randn(batch_size * qo_len, num_qo_heads, head_dim).to(0).half()
q_indptr = torch.arange(0, batch_size + 1).to(0).int() * qo_len qo_indptr = torch.arange(0, batch_size + 1).to(0).int() * qo_len
total_tokens = kv_len * batch_size total_tokens = kv_len * batch_size
kv_data = torch.randn(total_tokens, 2, num_kv_heads, 1, head_dim).to(0).half() kv_data = torch.randn(total_tokens, 2, num_kv_heads, head_dim).to(0).half()
kv_indptr = torch.arange(0, batch_size + 1).to(0).int() * kv_len kv_indptr = torch.arange(0, batch_size + 1).to(0).int() * kv_len
kv_indices = torch.arange(0, total_tokens).to(0).int() kv_indices = torch.arange(0, total_tokens).to(0).int()
kv_last_page_len = torch.full((batch_size,), 1, dtype=torch.int32).to(0) kv_last_page_len = torch.full((batch_size,), 1, dtype=torch.int32).to(0)
@@ -70,21 +77,44 @@ def test_batch_prefill_with_paged_kv_cache(
max_len_extend, max_len_extend,
) )
if use_wrapper: o_redundant = torch.empty_like(q)
wrapper = flashinfer.BatchPrefillWithPagedKVCacheWrapper() b_start_loc = torch.zeros((batch_size,), dtype=torch.int32).to(0)
wrapper.begin_forward(q_indptr, batch_size, num_qo_heads, num_kv_heads) b_start_loc[1:] = torch.cumsum(b_seq_len[:-1], dim=0)
o = wrapper.forward( b_seq_len_prefix = b_seq_len - b_seq_len_extend
q, q_indptr, kv_data, kv_indptr, kv_indices, kv_last_page_len
) redundant_attention(
else: q,
o = flashinfer.batch_prefill_with_paged_kv_cache( k_extend,
q, v_extend,
q_indptr, o_redundant,
kv_data, k_buffer,
kv_indptr, v_buffer,
kv_indices, req_to_token,
kv_last_page_len, b_req_idx,
) b_start_loc,
b_seq_len,
b_seq_len_prefix,
max_len_in_batch,
)
print("Mean: ", torch.mean(torch.abs(o_redundant - o_triton)))
print("Max: ", torch.max(torch.abs(o_redundant - o_triton)))
assert torch.allclose(o_redundant, o_triton, rtol=1e-2, atol=1e-3)
flashinfer_prefill_wrapper.end_forward()
flashinfer_prefill_wrapper.begin_forward(
qo_indptr,
kv_indptr,
kv_indices,
kv_last_page_len,
num_qo_heads,
num_kv_heads,
head_dim,
1,
)
o = flashinfer_prefill_wrapper.forward(
q.contiguous().view(-1, num_qo_heads, head_dim), kv_data
)
print("Mean: ", torch.mean(torch.abs(o - o_triton))) print("Mean: ", torch.mean(torch.abs(o - o_triton)))
print("Max: ", torch.max(torch.abs(o - o_triton))) print("Max: ", torch.max(torch.abs(o - o_triton)))
@@ -105,10 +135,11 @@ def test_batch_decode_with_paged_kv_cache(
): ):
# note(lsyin): when pytest, the number of heads cannot change, because triton kernel has a cache # note(lsyin): when pytest, the number of heads cannot change, because triton kernel has a cache
# to test different shape of decode, change the parameters in the __main__, and run decode only once # to test different shape of decode, change the parameters in the __main__, and run decode only once
init_flashinfer(num_qo_heads, num_kv_heads)
q = torch.randn(batch_size, num_qo_heads, head_dim).to(0).half() q = torch.randn(batch_size, num_qo_heads, head_dim).to(0).half()
total_tokens = kv_len * batch_size total_tokens = kv_len * batch_size
kv_data = torch.randn(total_tokens, 2, num_kv_heads, 1, head_dim).to(0).half() kv_data = torch.randn(total_tokens, 2, num_kv_heads, head_dim).to(0).half()
kv_indptr = torch.arange(0, batch_size + 1).to(0).int() * kv_len kv_indptr = torch.arange(0, batch_size + 1).to(0).int() * kv_len
kv_indices = torch.arange(0, total_tokens).to(0).int() kv_indices = torch.arange(0, total_tokens).to(0).int()
kv_last_page_len = torch.full((batch_size,), 1, dtype=torch.int32).to(0) kv_last_page_len = torch.full((batch_size,), 1, dtype=torch.int32).to(0)
@@ -139,26 +170,46 @@ def test_batch_decode_with_paged_kv_cache(
total_tokens, total_tokens,
) )
wrapper = flashinfer.BatchDecodeWithPagedKVCacheWrapper() flashinfer_decode_wrapper.end_forward()
wrapper.begin_forward( flashinfer_decode_wrapper.begin_forward(
kv_indptr, kv_indptr,
kv_indices,
kv_last_page_len, kv_last_page_len,
batch_size,
num_qo_heads, num_qo_heads,
num_kv_heads, num_kv_heads,
head_dim, head_dim,
1, 1,
"NONE", pos_encoding_mode="NONE",
"float16", data_type="float16",
)
o = flashinfer_decode_wrapper.forward(
q.contiguous().view(-1, num_qo_heads, head_dim), kv_data
) )
o = wrapper.forward(q, kv_data, kv_indptr, kv_indices, kv_last_page_len)
print("Mean: ", torch.mean(torch.abs(o - o_triton))) print("Mean: ", torch.mean(torch.abs(o - o_triton)))
print("Max: ", torch.max(torch.abs(o - o_triton))) print("Max: ", torch.max(torch.abs(o - o_triton)))
assert torch.allclose(o, o_triton, rtol=1e-2, atol=2e-3) assert torch.allclose(o, o_triton, rtol=1e-2, atol=2e-3)
def init_flashinfer(num_attention_heads, num_kv_heads):
if not _grouped_size_compiled_for_decode_kernels(num_attention_heads, num_kv_heads):
use_tensor_cores = True
else:
use_tensor_cores = False
workspace_buffer = torch.empty(128 * 1024 * 1024, dtype=torch.int8, device="cuda")
global flashinfer_prefill_wrapper, flashinfer_decode_wrapper
flashinfer_prefill_wrapper = BatchPrefillWithPagedKVCacheWrapper(
workspace_buffer, "NHD"
)
flashinfer_decode_wrapper = BatchDecodeWithPagedKVCacheWrapper(
workspace_buffer, "NHD", use_tensor_cores=use_tensor_cores
)
if __name__ == "__main__": if __name__ == "__main__":
test_batch_prefill_with_paged_kv_cache(12, 54, 37, 8, 8, 128, False) test_batch_prefill_with_paged_kv_cache(12, 54, 37, 8, 8, 128)
test_batch_prefill_with_paged_kv_cache(37, 1111, 456, 32, 32, 128, True) test_batch_prefill_with_paged_kv_cache(37, 1111, 456, 32, 32, 128)
test_batch_decode_with_paged_kv_cache(12, 54, 4, 32, 128) test_batch_decode_with_paged_kv_cache(12, 54, 4, 32, 128)