Skip to content

[release/2.7] [SWDEV-543214] Reland #2416 Fix warps runtime #2421

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.

Sign up for GitHub

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

Merged
merged 3 commits into from
Jul 30, 2025
Merged

[release/2.7] [SWDEV-543214] Reland #2416 Fix warps runtime #2421

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
9cd09fd
Revert "[SWDEV-543214] Set max_numwarps based on warp_size instead of…
jataylo Jul 25, 2025
74321c2
Set max_numwarps based on warp_size instead of hardcoding
jataylo Jul 28, 2025
bd7dc14
Merge branch 'release/2.7' into fix_warps_runtime
jithunnair-amd Jul 30, 2025
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
2 changes: 2 additions & 0 deletions torch/_inductor/runtime/coordinate_descent_tuner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
import itertools
import logging
from typing import Callable, Optional, TYPE_CHECKING
from functools import lru_cache

from .hints import TRITON_MAX_BLOCK
from .runtime_utils import red_text, triton_config_to_hashable
Expand Down Expand Up @@ -60,6 +61,7 @@ def get_config_max(self, prefix: str) -> int:
size_hint = self.size_hints.get(prefix) if self.size_hints is not None else None
return min(max_block, size_hint) if size_hint is not None else max_block

@lru_cache(maxsize=1)
def get_warpsmax(self):
# Currently, CUDA has a maximum of 1024 threads, so 32 is the max
# number of warps.
Expand Down