Skip to content

[SYCL][Matrix] Make joint_matrix_mad return A*B+C's result instead of… #3586

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 1 commit into from
Apr 23, 2021
Merged

[SYCL][Matrix] Make joint_matrix_mad return A*B+C's result instead of… #3586

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
15 changes: 8 additions & 7 deletions sycl/include/CL/sycl/ONEAPI/matrix/matrix-amx.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -405,26 +405,27 @@ inline __SYCL_ALWAYS_INLINE typename std::enable_if<
(LayoutA == matrix_layout::row_major) &&
(LayoutB == matrix_layout::packed_b) &&
(LayoutC == matrix_layout::row_major),
void>::type
joint_matrix<Group, T2, NumRowsC, NumColsC, LayoutC>>::type
joint_matrix_mad(Group sg,
joint_matrix<Group, T1, NumRowsA, NumColsA, LayoutA> &jmA,
joint_matrix<Group, T1, NumRowsB, NumColsB, LayoutB> &jmB,
joint_matrix<Group, T2, NumRowsC, NumColsC, LayoutC> &jmC) {
joint_matrix<Group, T2, NumRowsC, NumColsC, LayoutC> res(jmC);
constexpr size_t epd = detail::elems_per_dword<T1>::value;
// If A is large and C is small, in joint_matrix_load, we do memcpy for A, and
// we do tileload for C whose shape is not tile_size*tile_size*4. In
// joint_matrix_mad, we do tileload for A and shape is tile_size*tile_size*4.
// So we need to reshape C before we do dpbssd.
bool Cshouldreload = jmC.isSmall && !jmA.isSmall && !jmB.isSmall;
bool Cshouldreload = res.isSmall && !jmA.isSmall && !jmB.isSmall;
bool Ashouldreload = jmA.isSmall && !jmB.isSmall;
bool Bshouldreload = jmB.isSmall && !jmA.isSmall;

for (int m = 0; m < jmC.trows; ++m) {
for (int n = 0; n < jmC.tcols; ++n) {
for (int m = 0; m < res.trows; ++m) {
for (int n = 0; n < res.tcols; ++n) {
detail::submatrix<T2> sub_c;

// AMX: 8 register tiles : 1k byte size, SMmaxxSKmax =16x64
submatrix_load(sub_c, jmC, m * tile_size, n * tile_size, jmC.stride,
submatrix_load(sub_c, res, m * tile_size, n * tile_size, res.stride,
matrix_layout::row_major, Cshouldreload);
for (int k = 0; k < jmA.tcols; ++k) { // K->int8_t
detail::submatrix<T1> sub_a;
Expand All @@ -436,11 +437,11 @@ joint_matrix_mad(Group sg,
jmB.stride, matrix_layout::packed_b, Bshouldreload);
submatrix_mad(sub_a, sub_b, sub_c);
}
submatrix_store(sub_c, jmC, m * tile_size, n * tile_size, jmC.stride,
submatrix_store(sub_c, res, m * tile_size, n * tile_size, res.stride,
matrix_layout::row_major, Cshouldreload);
}
}
return;
return res;
}

} // namespace experimental::matrix
Expand Down
2 changes: 1 addition & 1 deletion sycl/test/matrix/matrix-amx-bf16-test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,7 +83,7 @@ void matrix_multiply(big_matrix<T1, NUM_ROWS_C, NUM_COLS_C> &C, big_matrix<T2, N
accB.get_pointer() +
(k * TK / 2) * (N * 2) + sg_starty * TN * 2,
N * 2, matrix_layout::packed_b);
joint_matrix_mad(sg, sub_a, sub_b, sub_c);
sub_c = joint_matrix_mad(sg, sub_a, sub_b, sub_c);
}
joint_matrix_store(sg, sub_c,
accC.get_pointer() + (sg_startx * TM) * N +
Expand Down
2 changes: 1 addition & 1 deletion sycl/test/matrix/matrix-amx-int8-test.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -83,7 +83,7 @@ void matrix_multiply(big_matrix<T1, NUM_ROWS_C, NUM_COLS_C> &C, big_matrix<T2, N
accB.get_pointer() +
(k * TK / 4) * (N * 4) + sg_starty * TN * 4,
N * 4, matrix_layout::packed_b);
joint_matrix_mad(sg, sub_a, sub_b, sub_c);
sub_c = joint_matrix_mad(sg, sub_a, sub_b, sub_c);
}
joint_matrix_store(sg, sub_c,
accC.get_pointer() + (sg_startx * TM) * N +
Expand Down