Skip to content

fix subgroup ballot testing for non-uniform workgroups #2709

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
karolherbst wants to merge 2 commits into
base: main
Choose a base branch
from
Open

fix subgroup ballot testing for non-uniform workgroups #2709

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
167700b
subgroups: fix set_last_workgroup_params for uniform subgroup sizes
karolherbst May 31, 2026
74237f8
fix subgroup ballot broadcast cases for non-uniform workgroups
karolherbst Jun 1, 2026
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
4 changes: 3 additions & 1 deletion test_conformance/subgroups/subhelpers.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -215,8 +215,10 @@ void set_last_workgroup_params(int non_uniform_size, int &number_of_subgroups,
int subgroup_size, int &workgroup_size,
int &last_subgroup_size)
{
number_of_subgroups = 1 + non_uniform_size / subgroup_size;
number_of_subgroups =
(subgroup_size + non_uniform_size - 1) / subgroup_size;
last_subgroup_size = non_uniform_size % subgroup_size;
if (last_subgroup_size == 0) last_subgroup_size = subgroup_size;
workgroup_size = non_uniform_size;
}

Expand Down
22 changes: 20 additions & 2 deletions test_conformance/subgroups/test_subgroup_ballot.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -766,7 +766,16 @@ __kernel void test_sub_group_non_uniform_broadcast(const __global Type *in, __gl
int gid = get_global_id(0);
XY(xy,gid);
Type x = in[gid];
if (xy[gid].x < (get_sub_group_size() >> 1)) {

uint sub_group_size = get_sub_group_size();
// If we are at the edge, calculate our own sub_group_size as it's implementation defined otherwise.
if (get_local_size(0) != get_enqueued_local_size(0) && get_sub_group_id() == get_num_sub_groups() - 1) {
uint new_sub_group_size = get_local_size(0) % sub_group_size;
if (new_sub_group_size != 0)
sub_group_size = new_sub_group_size;
}

if (xy[gid].x < (sub_group_size >> 1)) {
out[gid] = sub_group_non_uniform_broadcast(x, xy[gid].z);
} else {
out[gid] = sub_group_non_uniform_broadcast(x, xy[gid].w);
Expand All @@ -778,7 +787,16 @@ __kernel void test_sub_group_broadcast_first(const __global Type *in, __global i
int gid = get_global_id(0);
XY(xy,gid);
Type x = in[gid];
if (xy[gid].x < (get_sub_group_size() >> 1)) {

uint sub_group_size = get_sub_group_size();
// If we are at the edge, calculate our own sub_group_size as it's implementation defined otherwise.
if (get_local_size(0) != get_enqueued_local_size(0) && get_sub_group_id() == get_num_sub_groups() - 1) {
uint new_sub_group_size = get_local_size(0) % sub_group_size;
if (new_sub_group_size != 0)
sub_group_size = new_sub_group_size;
}

if (xy[gid].x < (sub_group_size >> 1)) {
out[gid] = sub_group_broadcast_first(x);;
} else {
out[gid] = sub_group_broadcast_first(x);;
Expand Down
Loading