Skip to content

减少w4afp8 编译时间 #3287

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

Closed
wants to merge 1 commit into from
Closed

减少w4afp8 编译时间 #3287

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
21 changes: 11 additions & 10 deletions custom_ops/gpu_ops/w4afp8_gemm/w4afp8_gemm.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,16 +77,17 @@ void DisPatchW4AFp8Gemm(
max_tokens,
stream)
} else {
GEMM_SWITCH_FP16(
M, K, batch_size, token_padding_size, kBlockN, TailN,
weight,
input,
out,
weight_scale,
input_row_sum,
tokens,
max_tokens,
stream)
PD_THROW("Only supported dtype in ['FP16'].");
// GEMM_SWITCH_FP16(
// M, K, batch_size, token_padding_size, kBlockN, TailN,
// weight,
// input,
// out,
// weight_scale,
// input_row_sum,
// tokens,
// max_tokens,
// stream)
}
}

Expand Down
8 changes: 4 additions & 4 deletions custom_ops/utils/auto_gen_w4afp8_gemm_kernel.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -85,12 +85,12 @@

gemm_case = [
[8192, 3584, 8, 0], # eb45T ffn1
[8192, 3584, 8, 2048], # eb45T ffn1
[7168, 8192, 8, 0], # eb45T ffn2
[7168, 8192, 8, 2048], # eb45T ffn2
# [8192, 3584, 8, 2048], # eb45T ffn1
# [7168, 8192, 8, 0], # eb45T ffn2
# [7168, 8192, 8, 2048], # eb45T ffn2
]

dtype = ["BF16", "FP16"]
dtype = ["BF16"]


def get_cutlass_type(type):
Expand Down
Loading