Skip to content

Align matmul grad with torch v2.9.1 #76922

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
wanghuancoder merged 7 commits into from
Dec 23, 2025
Merged

Align matmul grad with torch v2.9.1 #76922

Changes from 1 commit
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
206f82c
combine batch with aggregate dim, save reduce, align with torch
A-nnonymous Dec 15, 2025
a1b3234
fix miscs
A-nnonymous Dec 15, 2025
d7942f6
fix grad
A-nnonymous Dec 18, 2025
6efe644
add flag control
A-nnonymous Dec 18, 2025
8c5d13b
restrict new logic
A-nnonymous Dec 18, 2025
496c07b
fix dcu\cpu
A-nnonymous Dec 19, 2025
70ee1ba
fix high_grad
A-nnonymous Dec 22, 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
Prev Previous commit
Next Next commit
add flag control
  • Loading branch information
@A-nnonymous
A-nnonymous committed Dec 18, 2025
commit 6efe64473d63ca43b135ddd6d6cb4ef0bb322c7c
35 changes: 24 additions & 11 deletions paddle/phi/kernels/impl/matmul_grad_kernel_impl.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@ limitations under the License. */
#if defined(__NVCC__) || defined(__HIPCC__)
#include "paddle/phi/kernels/gpu/reduce.h"
#endif
COMMON_DECLARE_bool(use_legacy_gemm);

namespace phi {

Expand Down Expand Up @@ -225,15 +226,25 @@ void CalcInputGrad(const Context& dev_ctx,
DenseTensor a_processed = a, b_processed = b;
bool trans_a_processed = trans_a, trans_b_processed = trans_b;
if (need_combine) {
a_processed = is_fold_init_dims_a
? FoldInitDims(a)
: FoldBatchIntoAggregation<Context, T>(dev_ctx, a);
b_processed = is_fold_init_dims_b
? FoldInitDims(b)
: FoldBatchIntoAggregation<Context, T>(dev_ctx, b);
// Once we try to combine aggregation dimension to batch dimension,
trans_a_processed = is_fold_init_dims_a ? trans_a : !trans_a;
trans_b_processed = is_fold_init_dims_b ? trans_b : !trans_b;
if (!FLAGS_use_legacy_gemm) {
a_processed = is_fold_init_dims_a
? FoldInitDims(a)
: FoldBatchIntoAggregation<Context, T>(dev_ctx, a);
b_processed = is_fold_init_dims_b
? FoldInitDims(b)
: FoldBatchIntoAggregation<Context, T>(dev_ctx, b);
// Once we try to combine aggregation dimension to batch dimension,
// we need to flip the transpose flag
trans_a_processed = is_fold_init_dims_a ? trans_a : !trans_a;
trans_b_processed = is_fold_init_dims_b ? trans_b : !trans_b;
} else {
a_processed = is_fold_init_dims_a
? FoldInitDims(a)
: FoldHeadAndLastDims<Context, T>(dev_ctx, a);
b_processed = is_fold_init_dims_b
? FoldInitDims(b)
: FoldHeadAndLastDims<Context, T>(dev_ctx, b);
}
}
std::vector<std::int64_t> a_dims = common::vectorize(a_processed.dims());
std::vector<std::int64_t> b_dims = common::vectorize(b_processed.dims());
Expand Down Expand Up @@ -485,7 +496,8 @@ void MatmulGradKernel(const Context& dev_ctx,
VLOG(3)
<< "matmul grad case: transpose_x = false && transpose_y = false";
if (dx) {
if (is_x_been_broadcasted && x_ndim == 3 && ndim == 3) {
if (!FLAGS_use_legacy_gemm && is_x_been_broadcasted && x_ndim == 3 &&
ndim == 3) {
// Once x been broadcasted, we introduce a new aggregate dim
// original: [B, M, N] x [B, K, N]' -> [B, M, K] -(reduceB)-> [M, K]
// new: [BN, M] x [BN, K] -> [M, K]
Expand Down Expand Up @@ -535,7 +547,8 @@ void MatmulGradKernel(const Context& dev_ctx,
} // if is_x_been_broadcasted
} // if dx
if (dy) {
if (is_y_been_broadcasted && y_ndim == 3 && ndim == 3) {
if (!FLAGS_use_legacy_gemm && is_y_been_broadcasted && y_ndim == 3 &&
ndim == 3) {
// Once y been broadcasted, we introduce a new aggregate dim
// original: [B, M, K] x [B, M, N] -> [B, K, N] -(reduceB)-> [K, N]
// new: [BM, K]' x [BM, N] -> [K, N]
Expand Down
Loading