Skip to content

[Joint matrix] Remove cpu-unsupported shapes/layouts in Matrix tests. #19475

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
wants to merge 6 commits into
base: sycl
Choose a base branch
from
Open

[Joint matrix] Remove cpu-unsupported shapes/layouts in Matrix tests. #19475

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
e2b05e8
Remove unsupported shapes/layouts
ggojska Jul 15, 2025
9be9ea5
Add Jira tracker. Add all shapes and layouts
ggojska Jul 17, 2025
c9de02b
Add test cases
ggojska Jul 18, 2025
62ac27f
fix indentations
ggojska Jul 18, 2025
e1ae6c6
fix indentations 2
ggojska Jul 18, 2025
abb9ad2
Formatter changes for comments
ggojska Jul 18, 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: 6 additions & 2 deletions sycl/test-e2e/Matrix/SG32/joint_matrix_bfloat16.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,9 +36,13 @@ int main() {
for (unsigned int i = 0; i < combinations.size(); i++) {
if (combinations[i].nsize == 0) { // Intel AMX
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::row_major, 1>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::ext_intel_packed, 2>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
layout::ext_intel_packed, 2>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::row_major, 1>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
layout::row_major, 1>();
break;
}

Expand Down
9 changes: 7 additions & 2 deletions sycl/test-e2e/Matrix/SG32/joint_matrix_half.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,9 +35,14 @@ int main() {
for (unsigned int i = 0; i < combinations.size(); i++) {
if (combinations[i].nsize == 0) { // Intel AMX
test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::row_major, 1>();
test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::ext_intel_packed, 2>();
test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
layout::ext_intel_packed, 2>();
// VNNI transform does not work for half yet (CMPLRLLVM-69129)
// test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
// layout::row_major, 1>();
// test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
// layout::row_major, 1>();
break;
}

Expand Down
8 changes: 6 additions & 2 deletions sycl/test-e2e/Matrix/joint_matrix_bfloat16.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,9 +29,13 @@ int main() {
for (unsigned int i = 0; i < combinations.size(); i++) {
if (combinations[i].nsize == 0) { // Intel AMX
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::row_major, 1>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::ext_intel_packed, 2>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
layout::ext_intel_packed, 2>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::row_major, 1>();
test<bfloat16, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
layout::row_major, 1>();
break;
}

Expand Down
9 changes: 7 additions & 2 deletions sycl/test-e2e/Matrix/joint_matrix_half.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,9 +26,14 @@ int main() {
for (unsigned int i = 0; i < combinations.size(); i++) {
if (combinations[i].nsize == 0) { // Intel AMX
test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::row_major, 1>();
test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
layout::ext_intel_packed, 2>();
test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
layout::ext_intel_packed, 2>();
// VNNI transform does not work for half yet (CMPLRLLVM-69129)
// test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 16,
// layout::row_major, 1>();
// test<half, float, float, /*TM*/ 16, /*TN*/ 16, /*TK*/ 32,
// layout::row_major, 1>();
break;
}

Expand Down
Loading