Skip to content

[UR][L0] Optimize calculateKernelWorkDimensions function #20357

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

Draft
wants to merge 1 commit into
base: sycl
Choose a base branch
from
Draft

[UR][L0] Optimize calculateKernelWorkDimensions function #20357

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
21 changes: 11 additions & 10 deletions unified-runtime/source/adapters/level_zero/helpers/kernel_helpers.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,36 +96,37 @@ ur_result_t calculateKernelWorkDimensions(
// New variable needed because GlobalWorkSize parameter might not be of size
// 3
size_t GlobalWorkSize3D[3]{1, 1, 1};
std::copy(GlobalWorkSize, GlobalWorkSize + WorkDim, GlobalWorkSize3D);

if (LocalWorkSize) {
WG[0] = ur_cast<uint32_t>(LocalWorkSize[0]);
WG[1] = WorkDim >= 2 ? ur_cast<uint32_t>(LocalWorkSize[1]) : 1;
WG[2] = WorkDim == 3 ? ur_cast<uint32_t>(LocalWorkSize[2]) : 1;
} else {
std::copy(GlobalWorkSize, GlobalWorkSize + WorkDim, GlobalWorkSize3D);
UR_CALL(getSuggestedLocalWorkSize(Device, Kernel, GlobalWorkSize3D, WG));
}
const size_t *GlobalWorkSizePtr = LocalWorkSize ? GlobalWorkSize : GlobalWorkSize3D;

// TODO: assert if sizes do not fit into 32-bit?
switch (WorkDim) {
case 3:
ZeThreadGroupDimensions.groupCountX =
ur_cast<uint32_t>(GlobalWorkSize3D[0] / WG[0]);
ur_cast<uint32_t>(GlobalWorkSizePtr[0] / WG[0]);
ZeThreadGroupDimensions.groupCountY =
ur_cast<uint32_t>(GlobalWorkSize3D[1] / WG[1]);
ur_cast<uint32_t>(GlobalWorkSizePtr[1] / WG[1]);
ZeThreadGroupDimensions.groupCountZ =
ur_cast<uint32_t>(GlobalWorkSize3D[2] / WG[2]);
ur_cast<uint32_t>(GlobalWorkSizePtr[2] / WG[2]);
break;
case 2:
ZeThreadGroupDimensions.groupCountX =
ur_cast<uint32_t>(GlobalWorkSize3D[0] / WG[0]);
ur_cast<uint32_t>(GlobalWorkSizePtr[0] / WG[0]);
ZeThreadGroupDimensions.groupCountY =
ur_cast<uint32_t>(GlobalWorkSize3D[1] / WG[1]);
ur_cast<uint32_t>(GlobalWorkSizePtr[1] / WG[1]);
WG[2] = 1;
break;
case 1:
ZeThreadGroupDimensions.groupCountX =
ur_cast<uint32_t>(GlobalWorkSize3D[0] / WG[0]);
ur_cast<uint32_t>(GlobalWorkSizePtr[0] / WG[0]);
WG[1] = WG[2] = 1;
break;

Expand All @@ -135,19 +136,19 @@ ur_result_t calculateKernelWorkDimensions(
}

// Error handling for non-uniform group size case
if (GlobalWorkSize3D[0] !=
if (GlobalWorkSizePtr[0] !=
size_t(ZeThreadGroupDimensions.groupCountX) * WG[0]) {
UR_LOG(ERR, "calculateKernelWorkDimensions: invalid work_dim. The range "
"is not a multiple of the group size in the 1st dimension");
return UR_RESULT_ERROR_INVALID_WORK_GROUP_SIZE;
}
if (GlobalWorkSize3D[1] !=
if (WorkDim >= 2 && GlobalWorkSizePtr[1] !=
size_t(ZeThreadGroupDimensions.groupCountY) * WG[1]) {
UR_LOG(ERR, "calculateKernelWorkDimensions: invalid work_dim. The range "
"is not a multiple of the group size in the 2nd dimension");
return UR_RESULT_ERROR_INVALID_WORK_GROUP_SIZE;
}
if (GlobalWorkSize3D[2] !=
if (WorkDim == 3 && GlobalWorkSizePtr[2] !=
size_t(ZeThreadGroupDimensions.groupCountZ) * WG[2]) {
UR_LOG(ERR, "calculateKernelWorkDimensions: invalid work_dim. The range "
"is not a multiple of the group size in the 3rd dimension");
Expand Down
Loading