Skip to content

[NPU]: Adjust MAX_FUSED_SIZE when using fused_linear_cross_entropy #985

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
lancerts merged 3 commits into from
Dec 23, 2025
Merged

[NPU]: Adjust MAX_FUSED_SIZE when using fused_linear_cross_entropy #985

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
da0caef
Adjust MAX_FUSED_SIZE when using fused_linear_cross_entropy
zheliuyu Dec 23, 2025
b63611f
Merge branch 'main' into main
lancerts Dec 23, 2025
f68c24f
Merge branch 'main' into main
lancerts Dec 23, 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
8 changes: 7 additions & 1 deletion src/liger_kernel/ops/cross_entropy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -289,7 +289,13 @@ def liger_cross_entropy_kernel(
# The hard limit of TRITON_MAX_TENSOR_NUMEL is 1048576 https://github.com/triton-lang/triton/blob/ba42a5c68fd0505f8c42f4202d53be0f8d9a5fe0/python/triton/language/core.py#L19
# However, setting limit as 65536 as in LayerNorm tutorial is faster because of less register spilling
# The optimal maximum block size depends on your hardware, your kernel, and your dtype
MAX_FUSED_SIZE = 4096 if infer_device() == "xpu" else 65536 // 2 # the best size we found by manually tuning
# the best size we found by manually tuning on xpu and npu.
if infer_device() == "xpu":
MAX_FUSED_SIZE = 4096
elif infer_device() == "npu":
MAX_FUSED_SIZE = 2048
else:
MAX_FUSED_SIZE = 65536 // 2


def cross_entropy_forward(
Expand Down
3 changes: 2 additions & 1 deletion src/liger_kernel/ops/fused_linear_cross_entropy.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,11 +6,12 @@
from liger_kernel.ops.utils import amp_custom_fwd
from liger_kernel.ops.utils import element_mul_kernel
from liger_kernel.ops.utils import is_hip
from liger_kernel.utils import infer_device

# The hard limit of TRITON_MAX_TENSOR_NUMEL is 1048576 https://github.com/triton-lang/triton/blob/ba42a5c68fd0505f8c42f4202d53be0f8d9a5fe0/python/triton/language/core.py#L19
# However, setting limit as 65536 as in LayerNorm tutorial is faster because of less register spilling
# The optimal maximum block size depends on your hardware, your kernel, and your dtype
MAX_FUSED_SIZE = 65536 // 2
MAX_FUSED_SIZE = 2048 if infer_device() == "npu" else 65536 // 2


def fused_linear_cross_entropy_forward(
Expand Down
Loading