Revert "[V1] Scatter and gather placeholders in the model runner" (#16075)

This commit is contained in:
Roger Wang
2025-04-04 14:50:57 -07:00
committed by GitHub
parent f5722a5052
commit af51d80fa1
42 changed files with 942 additions and 496 deletions
+29 -17
View File
@@ -3,7 +3,7 @@
import pytest
import torch
from vllm.multimodal.inputs import MultiModalKwargs, PlaceholderRange
from vllm.multimodal.inputs import MultiModalKwargs
from vllm.sampling_params import SamplingParams
from vllm.utils import sha256
# disable yapf here as it formats differently than isort such that both fail
@@ -158,10 +158,13 @@ def test_generate_block_hash_extra_keys():
request = make_request(
request_id=0,
prompt_token_ids=[_ for _ in range(20)],
mm_positions=[
PlaceholderRange(offset=0, length=5),
PlaceholderRange(offset=10, length=5),
],
mm_positions=[{
"offset": 0,
"length": 5
}, {
"offset": 10,
"length": 5
}],
mm_hashes=["hash1", "hash2"],
)
@@ -219,10 +222,13 @@ def test_hash_request_tokens(hash_fn):
request = make_request(
request_id=0,
prompt_token_ids=[_ for _ in range(6)],
mm_positions=[
PlaceholderRange(offset=0, length=3),
PlaceholderRange(offset=3, length=3),
],
mm_positions=[{
"offset": 0,
"length": 3
}, {
"offset": 3,
"length": 3
}],
mm_hashes=["hash1", "hash2"],
)
@@ -247,19 +253,25 @@ def test_hash_tokens_different_mm_input(hash_fn):
request1 = make_request(
request_id=0,
prompt_token_ids=[_ for _ in range(6)],
mm_positions=[
PlaceholderRange(offset=0, length=3),
PlaceholderRange(offset=3, length=3),
],
mm_positions=[{
"offset": 0,
"length": 3
}, {
"offset": 3,
"length": 3
}],
mm_hashes=["hash1", "hash2"],
)
request2 = make_request(
request_id=1,
prompt_token_ids=[_ for _ in range(6)],
mm_positions=[
PlaceholderRange(offset=0, length=3),
PlaceholderRange(offset=3, length=3),
],
mm_positions=[{
"offset": 0,
"length": 3
}, {
"offset": 3,
"length": 3
}],
mm_hashes=["hash3", "hash2"],
)
block_size = 3