Skip to content

[ARM CPU] SVE support for Sgemm kernel #26027

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
akote123 wants to merge 11 commits into
base: main
Choose a base branch
from
+1,115 −113
Open

[ARM CPU] SVE support for Sgemm kernel #26027

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
d762023
ADD support for SVE sgemm
akote123 Sep 12, 2025
de79d1a
Updates to fix issue in Gr2
akote123 Sep 15, 2025
c508e83
Cleanup
akote123 Sep 23, 2025
d997270
Fix Compilation errors
akote123 Sep 23, 2025
5dd2161
Fix Accuracy Errors
akote123 Sep 24, 2025
11f7267
Fix CI Failure
akote123 Sep 30, 2025
9729a1a
Code formating
akote123 Nov 18, 2025
f12f872
Update
akote123 Nov 18, 2025
0699ab8
Merge branch 'main' into aruna/sgemm_sve
akote123 Nov 18, 2025
db2bfe9
code formating
akote123 Nov 18, 2025
2e2c6db
code formating
akote123 Nov 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
2 changes: 2 additions & 0 deletions cmake/onnxruntime_mlas.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -467,6 +467,8 @@ else()
list(APPEND mlas_platform_srcs ${MLAS_SRC_DIR}/sve/mlasi_sve.h)
list(APPEND mlas_platform_srcs ${MLAS_SRC_DIR}/sve/elementwise_sve.cpp)
set_source_files_properties(${MLAS_SRC_DIR}/sve/elementwise_sve.cpp PROPERTIES COMPILE_FLAGS " -march=armv8.2-a+sve+fp16 ")
list(APPEND mlas_platform_srcs ${MLAS_SRC_DIR}/sve/sgemm_sve.cpp)
set_source_files_properties(${MLAS_SRC_DIR}/sve/sgemm_sve.cpp PROPERTIES COMPILE_FLAGS "-march=armv8.2-a+sve -O3 -ffast-math -funroll-loops")
list(APPEND mlas_private_compile_definitions MLAS_USE_SVE)
endif()

Expand Down
106 changes: 94 additions & 12 deletions onnxruntime/core/mlas/lib/sgemm.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,10 @@ Module Name:

#include "mlasi.h"

#if defined(MLAS_USE_SVE)
#include "sve/mlasi_sve.h"
#endif

//
// Define the number of rows from matrix A to transpose to a local buffer.
//
Expand Down Expand Up @@ -259,8 +263,15 @@ Return Value:

do {

#if defined(MLAS_NEON_INTRINSICS)
vst4q_f32(D, vld4q_f32(b));
#if defined(MLAS_USE_SVE) && defined(MLAS_NEON_INTRINSICS)
if (MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()) {
SVE_LOAD_STORE(D, b);
}
else{
vst4q_f32(D, vld4q_f32(b));
}
#elif defined(MLAS_NEON_INTRINSICS)
vst4q_f32(D, vld4q_f32(b));
#else
MLAS_FLOAT32X4 t0 = MlasLoadFloat32x4(&b[0]);
MLAS_FLOAT32X4 t1 = MlasLoadFloat32x4(&b[4]);
Expand Down Expand Up @@ -303,8 +314,14 @@ Return Value:
float* d = D;
const float* b = B;

#if defined(MLAS_NEON_INTRINSICS)
vst4q_f32(d, ZeroFloat32x4x4);
#if defined(MLAS_USE_SVE) && defined(MLAS_NEON_INTRINSICS)
if (MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()) {
Copy link
Member

@hariharans29 hariharans29 Oct 1, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Same comment as before - what happens when MLAS_USE_SVE is false and MLAS_NEON_INTRINSICS is true and HasArmSve() evaluates to true ?

SVE_ZERO_INITIALIZE(d);
} else {
vst4q_f32(d, ZeroFloat32x4x4);
}
#elif defined(MLAS_NEON_INTRINSICS)
vst4q_f32(d, ZeroFloat32x4x4);
#else
MlasStoreAlignedFloat32x4(d, ZeroFloat32x4);
MlasStoreAlignedFloat32x4(d + 4, ZeroFloat32x4);
Expand Down Expand Up @@ -486,6 +503,21 @@ Return Value:
x -= 4;
}

#elif defined(MLAS_USE_SVE)
if(MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()) {
SVE_TRANSPOSE(D,b,ldb,x);
}
else
{
while (x >= 4) {

MlasSgemmTransposePackBNx4<16>(&D[0], &b[0], ldb);

D += 16 * 4;
b += 4;
x -= 4;
}
}
#else

while (x >= 4) {
Expand Down Expand Up @@ -564,8 +596,15 @@ Return Value:
const float* b = B;

if ((CountY & 8) != 0) {

MlasSgemmTransposePackBNx4<8>(&d[0], &b[0], ldb);
#if defined(MLAS_USE_SVE)
if (MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()){
MlasSveTransposePackBNx4<8>(&d[0], &b[0], ldb);}
else{
MlasSgemmTransposePackBNx4<8>(&d[0], &b[0], ldb);
}
#else
MlasSgemmTransposePackBNx4<8>(&d[0], &b[0], ldb);
#endif

d += 8;
b += ldb * 8;
Expand All @@ -584,7 +623,15 @@ Return Value:

if ((CountY & 4) != 0) {

MlasSgemmTransposePackBNx4<4>(&d[0], &b[0], ldb);
#if defined(MLAS_USE_SVE)
if (MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()){
MlasSveTransposePackBNx4<4>(&d[0], &b[0], ldb);}
else{
MlasSgemmTransposePackBNx4<4>(&d[0], &b[0], ldb);
}
#else
MlasSgemmTransposePackBNx4<4>(&d[0], &b[0], ldb);
#endif

d += 4;
b += ldb * 4;
Expand Down Expand Up @@ -631,7 +678,19 @@ Return Value:

if ((CountY & 1) != 0) {

#if defined(MLAS_NEON_INTRINSICS)
#if defined(MLAS_USE_SVE) && defined(MLAS_NEON_INTRINSICS)
if(MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve())
{
SCATTER_STORE(&d[0],&b[0]);
}
else{
MLAS_FLOAT32X4 t0 = MlasLoadFloat32x4(&b[0]);

MlasStoreLaneFloat32x4<0>(&d[0], t0);
MlasStoreLaneFloat32x4<1>(&d[16], t0);
MlasStoreLaneFloat32x4<2>(&d[32], t0);
MlasStoreLaneFloat32x4<3>(&d[48], t0);}
#elif defined(MLAS_NEON_INTRINSICS)
MLAS_FLOAT32X4 t0 = MlasLoadFloat32x4(&b[0]);

MlasStoreLaneFloat32x4<0>(&d[0], t0);
Expand Down Expand Up @@ -1004,8 +1063,7 @@ Return Value:
#endif

MLAS_FORCEINLINE
float*
MlasSgemmKernelLoop(
float* MlasSgemmKernelLoop(
const float* A,
const float* B,
float* C,
Expand Down Expand Up @@ -1059,18 +1117,41 @@ Return Value:
{
while (CountM > 0) {

size_t RowsHandled;
size_t RowsHandled = 0;

#if (defined(MLAS_TARGET_AMD64_IX86) || defined(MLAS_TARGET_POWER) || defined(MLAS_TARGET_S390X) || defined(MLAS_TARGET_LARCH64)) && !defined(FORCE_GENERIC_ALGORITHMS)
RowsHandled = GetMlasPlatform().GemmFloatKernel(A, B, C, CountK, CountM, CountN, lda, ldc, alpha, ZeroMode);

#else

if (ZeroMode) {

#if defined(MLAS_USE_SVE)
if (MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()) {
RowsHandled = MlasSgemmKernelZero_sve(A, B, C, CountK, CountM, CountN, lda, ldc, alpha);
} else {
RowsHandled = MlasSgemmKernelZero(A, B, C, CountK, CountM, CountN, lda, ldc, alpha);
}
#else
RowsHandled = MlasSgemmKernelZero(A, B, C, CountK, CountM, CountN, lda, ldc, alpha);
#endif

} else {

#if defined(MLAS_USE_SVE)
if (MLAS_CPUIDINFO::GetCPUIDInfo().HasArmSve()) {
RowsHandled = MlasSgemmKernelAdd_sve(A, B, C, CountK, CountM, CountN, lda, ldc, alpha);
} else {
RowsHandled = MlasSgemmKernelAdd(A, B, C, CountK, CountM, CountN, lda, ldc, alpha);
}
#else
RowsHandled = MlasSgemmKernelAdd(A, B, C, CountK, CountM, CountN, lda, ldc, alpha);
}
#endif

}

#endif // platform check

C += ldc * RowsHandled;
A += lda * RowsHandled;
CountM -= RowsHandled;
Expand All @@ -1079,6 +1160,7 @@ Return Value:
return C;
}


void
MlasSgemmOperation(
CBLAS_TRANSPOSE TransA,
Expand Down
Loading
Loading