Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix: Build error seen on Power Architecture #10421

Merged
merged 24 commits into from
Nov 19, 2024
Merged
Changes from 1 commit
Commits
Show all changes
24 commits
Select commit Hold shift + click to select a range
0ce09ab
Fix: Build error seen on Power Architecture
Nov 18, 2024
201a6da
[Model][LoRA]LoRA support added for glm-4v (#10418)
B-201 Nov 18, 2024
9120161
[Model] Remove transformers attention porting in VITs (#10414)
Isotr0py Nov 18, 2024
d7b14ce
[Doc] Update doc for LoRA support in GLM-4V (#10425)
B-201 Nov 18, 2024
06de800
[5/N][torch.compile] torch.jit.script --> torch.compile (#10406)
youkaichao Nov 18, 2024
d3a6317
[Doc] Add documentation for Structured Outputs (#9943)
ismael-dm Nov 18, 2024
b2a1685
Fix open_collective value in FUNDING.yml (#10426)
andrew Nov 18, 2024
4c610a5
[Model][Bugfix] Support TP for PixtralHF ViT (#10405)
mgoin Nov 18, 2024
96d56ba
[Hardware][XPU] AWQ/GPTQ support for xpu backend (#10107)
yma11 Nov 18, 2024
21da67d
[Kernel] Explicitly specify other value in tl.load calls (#9014)
angusYuhao Nov 18, 2024
2b855b1
[Kernel] Initial Machete W4A8 support + Refactors (#9855)
LucasWilkinson Nov 18, 2024
887d326
[3/N][torch.compile] consolidate custom op logging (#10399)
youkaichao Nov 18, 2024
db5dddb
[ci][bugfix] fix kernel tests (#10431)
youkaichao Nov 18, 2024
53e3a96
[misc] partial prefix & random input generation benchmark (#9929)
rickyyx Nov 18, 2024
ec45058
[ci/build] Have dependabot ignore all patch update (#10436)
khluu Nov 19, 2024
cce69dc
[Bugfix]Fix Phi-3 BNB online quantization (#10417)
jeejeelee Nov 19, 2024
2ce7cd4
[Platform][Refactor] Extract func `get_default_attn_backend` to `Plat…
MengqingCao Nov 19, 2024
6372003
Add openai.beta.chat.completions.parse example to structured_outputs.…
mgoin Nov 19, 2024
c0482f6
[Bugfix] Guard for negative counter metrics to prevent crash (#10430)
tjohnson31415 Nov 19, 2024
3fcfe67
[Misc] Avoid misleading warning messages (#10438)
jeejeelee Nov 19, 2024
392acf9
[Doc] Add the start of an arch overview page (#10368)
russellb Nov 19, 2024
30deada
[misc][plugin] improve plugin loading (#10443)
youkaichao Nov 19, 2024
cd96cde
Fix for clang-format (3.11)
mikejuliet13 Nov 19, 2024
6b053cb
Merge branch 'vllm-project:main' into vllm-power-issue
mikejuliet13 Nov 19, 2024
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
[ci][bugfix] fix kernel tests (#10431)
Signed-off-by: youkaichao <youkaichao@gmail.com>
Signed-off-by: Manjul Mohan <manjul.mohan@ibm.com>
  • Loading branch information
youkaichao authored and mikejuliet13 committed Nov 19, 2024
commit db5dddb050c6d7804e0156839e8794f8fbf032d2
23 changes: 13 additions & 10 deletions vllm/plugins/__init__.py
Original file line number Diff line number Diff line change
Expand Up @@ -6,9 +6,6 @@

if TYPE_CHECKING:
from vllm.config import CompilationConfig, VllmConfig
else:
CompilationConfig = None
VllmConfig = None

logger = logging.getLogger(__name__)

Expand Down Expand Up @@ -50,23 +47,23 @@ def load_general_plugins():
logger.exception("Failed to load plugin %s", plugin.name)


_compilation_config: Optional[CompilationConfig] = None
_compilation_config: Optional["CompilationConfig"] = None


def set_compilation_config(config: Optional[CompilationConfig]):
def set_compilation_config(config: Optional["CompilationConfig"]):
global _compilation_config
_compilation_config = config


def get_compilation_config() -> Optional[CompilationConfig]:
def get_compilation_config() -> Optional["CompilationConfig"]:
return _compilation_config


_current_vllm_config: Optional[VllmConfig] = None
_current_vllm_config: Optional["VllmConfig"] = None


@contextmanager
def set_current_vllm_config(vllm_config: VllmConfig):
def set_current_vllm_config(vllm_config: "VllmConfig"):
"""
Temporarily set the current VLLM config.
Used during model initialization.
Expand All @@ -87,6 +84,12 @@ def set_current_vllm_config(vllm_config: VllmConfig):
_current_vllm_config = old_vllm_config


def get_current_vllm_config() -> VllmConfig:
assert _current_vllm_config is not None, "Current VLLM config is not set."
def get_current_vllm_config() -> "VllmConfig":
if _current_vllm_config is None:
# in ci, usually when we test custom ops/modules directly,
# we don't set the vllm config. In that case, we set a default
# config.
logger.warning("Current VLLM config is not set.")
from vllm.config import VllmConfig
return VllmConfig()
return _current_vllm_config