Skip to content

【Quant】update e8m0 for stack_quant #77163

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

Open
liuruyan wants to merge 2 commits into
base: develop
Choose a base branch
from
Open

【Quant】update e8m0 for stack_quant #77163

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a82f5fe
add stack_quant e8m0
liuruyan Dec 31, 2025
e7a42a7
fix ci
liuruyan Jan 3, 2026
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
40 changes: 36 additions & 4 deletions paddle/phi/infermeta/fusion.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4287,35 +4287,67 @@ std::tuple<int64_t, int64_t, int64_t> FusedStackQuantCommonCheck(
}

void FusedStackTransposeQuantInferMeta(const std::vector<const MetaTensor*>& x,
bool using_pow2_scaling,
bool using_ue8m0_scale,
bool output_scale_transpose,
MetaTensor* out,
MetaTensor* scale) {
int64_t N, M, K;
std::tie(N, M, K) = FusedStackQuantCommonCheck(x);

std::vector<int64_t> out_shape = {N * K, M};
std::vector<int64_t> scale_shape = {N * K / 128, M / 128};
std::vector<int64_t> scale_shape;
if (using_ue8m0_scale) {
if (output_scale_transpose) {
scale_shape = {M / 128 / 4, N * K};
} else {
scale_shape = {N * K, M / 128 / 4};
}
} else {
if (output_scale_transpose) {
scale_shape = {M / 128, N * K / 128};
} else {
scale_shape = {N * K / 128, M / 128};
}
}
out->set_dims(common::make_ddim(out_shape));
scale->set_dims(common::make_ddim(scale_shape));
out->set_dtype(DataType::FLOAT8_E4M3FN);
scale->set_dtype(DataType::FLOAT32);
scale->set_dtype(using_ue8m0_scale ? DataType::INT32 : DataType::FLOAT32);
out->share_lod(*x.at(0));
scale->share_lod(*x.at(0));
out->set_layout(x.at(0)->layout());
scale->set_layout(x.at(0)->layout());
}

void FusedStackQuantInferMeta(const std::vector<const MetaTensor*>& x,
bool using_pow2_scaling,
bool using_ue8m0_scale,
bool output_scale_transpose,
MetaTensor* out,
MetaTensor* scale) {
int64_t N, M, K;
std::tie(N, M, K) = FusedStackQuantCommonCheck(x);

std::vector<int64_t> out_shape = {N * M, K};
std::vector<int64_t> scale_shape = {N * M / 128, K / 128};
std::vector<int64_t> scale_shape;
if (using_ue8m0_scale) {
if (output_scale_transpose) {
scale_shape = {K / 128 / 4, N * M};
} else {
scale_shape = {N * M, K / 128 / 4};
}
} else {
if (output_scale_transpose) {
scale_shape = {K / 128, N * M / 128};
} else {
scale_shape = {N * M / 128, K / 128};
}
}
out->set_dims(common::make_ddim(out_shape));
scale->set_dims(common::make_ddim(scale_shape));
out->set_dtype(DataType::FLOAT8_E4M3FN);
scale->set_dtype(DataType::FLOAT32);
scale->set_dtype(using_ue8m0_scale ? DataType::INT32 : DataType::FLOAT32);
out->share_lod(*x.at(0));
scale->share_lod(*x.at(0));
out->set_layout(x.at(0)->layout());
Expand Down
6 changes: 6 additions & 0 deletions paddle/phi/infermeta/fusion.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -943,11 +943,17 @@ PADDLE_API void FusionSeqExpandConcatFCInferMeta(

PADDLE_API void FusedStackTransposeQuantInferMeta(
const std::vector<const MetaTensor*>& x,
bool using_pow2_scaling,
bool using_ue8m0_scale,
bool output_scale_transpose,
MetaTensor* out,
MetaTensor* scale);

PADDLE_API void FusedStackQuantInferMeta(
const std::vector<const MetaTensor*>& x,
bool using_pow2_scaling,
bool using_ue8m0_scale,
bool output_scale_transpose,
MetaTensor* out,
MetaTensor* scale);

Expand Down
Loading
Loading