Skip to content

vulkan: Allow non-pow2 n_experts in topk_moe #17872

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
jeffbolznv wants to merge 1 commit into
base: master
Choose a base branch
from
+27 −11
Open

vulkan: Allow non-pow2 n_experts in topk_moe #17872

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
vulkan: Allow non-pow2 n_experts in topk_moe
  • Loading branch information
@jeffbolznv
jeffbolznv committed Dec 8, 2025
commit 9fee3f56baa75f35c2daad47e68f415568c99ccc
22 changes: 14 additions & 8 deletions ggml/src/ggml-vulkan/ggml-vulkan.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -757,7 +757,8 @@ struct vk_device_struct {

vk_pipeline pipeline_flash_attn_split_k_reduce;

vk_pipeline pipeline_topk_moe[num_topk_moe_pipelines][TOPK_MOE_COUNT];
// [2] is for whether to take n_experts from spec constant (0) or push constant (1)
vk_pipeline pipeline_topk_moe[num_topk_moe_pipelines][TOPK_MOE_COUNT][2];

std::vector<vk_pipeline_ref> all_pipelines;

Expand Down Expand Up @@ -1149,6 +1150,7 @@ static_assert(sizeof(vk_op_multi_add_push_constants) <= 256);

struct vk_op_topk_moe_push_constants {
uint32_t n_rows;
uint32_t n_experts_push;
uint32_t n_expert_used;
float clamp_min;
float clamp_max;
Expand Down Expand Up @@ -4204,10 +4206,12 @@ static void ggml_vk_load_shaders(vk_device& device) {
ggml_vk_create_pipeline(device, device->pipeline_conv2d_dw_whcn_f16_f32, "conv2d_dw_whcn_f16_f32", conv2d_dw_whcn_f16_f32_len, conv2d_dw_whcn_f16_f32_data, "main", 3, sizeof(vk_op_conv2d_dw_push_constants), {512, 1, 1}, {}, 1);
ggml_vk_create_pipeline(device, device->pipeline_conv2d_dw_cwhn_f16_f32, "conv2d_dw_cwhn_f16_f32", conv2d_dw_cwhn_f16_f32_len, conv2d_dw_cwhn_f16_f32_data, "main", 3, sizeof(vk_op_conv2d_dw_push_constants), {512, 1, 1}, {}, 1);

for (uint32_t i = 0; i < num_topk_moe_pipelines; ++i) {
ggml_vk_create_pipeline2(device, device->pipeline_topk_moe[i][TOPK_MOE_EARLY_SOFTMAX], "topk_moe_f32_early_softmax_"+std::to_string(i), topk_moe_f32_len, topk_moe_f32_data, "main", 3, sizeof(vk_op_topk_moe_push_constants), {1, 1, 1}, {device->subgroup_size, 1u<<i, 0, 0}, 1, true, true, device->subgroup_size);
ggml_vk_create_pipeline2(device, device->pipeline_topk_moe[i][TOPK_MOE_EARLY_SOFTMAX_NORM], "topk_moe_f32_early_softmax_norm"+std::to_string(i), topk_moe_f32_len, topk_moe_f32_data, "main", 3, sizeof(vk_op_topk_moe_push_constants), {1, 1, 1}, {device->subgroup_size, 1u<<i, 1, 0}, 1, true, true, device->subgroup_size);
ggml_vk_create_pipeline2(device, device->pipeline_topk_moe[i][TOPK_MOE_LATE_SOFTMAX], "topk_moe_f32_late_softmax"+std::to_string(i), topk_moe_f32_len, topk_moe_f32_data, "main", 3, sizeof(vk_op_topk_moe_push_constants), {1, 1, 1}, {device->subgroup_size, 1u<<i, 0, 1}, 1, true, true, device->subgroup_size);
for (uint32_t use_push = 0; use_push < 2; ++use_push) {
for (uint32_t i = 0; i < num_topk_moe_pipelines; ++i) {
ggml_vk_create_pipeline2(device, device->pipeline_topk_moe[i][TOPK_MOE_EARLY_SOFTMAX][use_push], "topk_moe_f32_early_softmax_"+std::to_string(i), topk_moe_f32_len, topk_moe_f32_data, "main", 3, sizeof(vk_op_topk_moe_push_constants), {1, 1, 1}, {device->subgroup_size, 1u<<i, 0, 0, use_push}, 1, true, true, device->subgroup_size);
ggml_vk_create_pipeline2(device, device->pipeline_topk_moe[i][TOPK_MOE_EARLY_SOFTMAX_NORM][use_push], "topk_moe_f32_early_softmax_norm"+std::to_string(i), topk_moe_f32_len, topk_moe_f32_data, "main", 3, sizeof(vk_op_topk_moe_push_constants), {1, 1, 1}, {device->subgroup_size, 1u<<i, 1, 0, use_push}, 1, true, true, device->subgroup_size);
ggml_vk_create_pipeline2(device, device->pipeline_topk_moe[i][TOPK_MOE_LATE_SOFTMAX][use_push], "topk_moe_f32_late_softmax"+std::to_string(i), topk_moe_f32_len, topk_moe_f32_data, "main", 3, sizeof(vk_op_topk_moe_push_constants), {1, 1, 1}, {device->subgroup_size, 1u<<i, 0, 1, use_push}, 1, true, true, device->subgroup_size);
}
}

for (auto &c : compiles) {
Expand Down Expand Up @@ -8554,7 +8558,9 @@ static vk_pipeline ggml_vk_op_get_pipeline(ggml_backend_vk_context * ctx, const
uint32_t idx = (uint32_t)ceilf(log2f(float(dst->ne[0])));
GGML_ASSERT(idx < num_topk_moe_pipelines);
topk_moe_mode mode = ggml_vk_num_additional_ops_to_topk_moe_mode(ctx->num_additional_fused_ops);
return ctx->device->pipeline_topk_moe[idx][mode];
// use n_experts from push constant if it's not equal to the power of two spec constant
bool use_push = dst->ne[0] != (1u << idx);
return ctx->device->pipeline_topk_moe[idx][mode][use_push];
}

if (src0->type == GGML_TYPE_F32 && (src1 == nullptr || src1->type == GGML_TYPE_F32) && dst->type == GGML_TYPE_F32) {
Expand Down Expand Up @@ -10158,6 +10164,7 @@ static void ggml_vk_topk_moe(ggml_backend_vk_context * ctx, vk_context& subctx,

vk_op_topk_moe_push_constants pc {};
pc.n_rows = n_rows;
pc.n_experts_push = n_experts;
pc.n_expert_used = n_expert_used;
if (mode == TOPK_MOE_EARLY_SOFTMAX_NORM) {
ggml_tensor * clamp = cgraph->nodes[node_idx + 7];
Expand Down Expand Up @@ -12832,8 +12839,7 @@ static bool ggml_vk_can_fuse_topk_moe(ggml_backend_vk_context * ctx, const struc
}

const int n_expert = softmax->ne[0];
// n_expert must be a power of 2
if (!is_pow2(n_expert) || n_expert > (1 << (num_topk_moe_pipelines-1))) {
if (n_expert > (1 << (num_topk_moe_pipelines-1))) {
return false;
}

Expand Down
12 changes: 9 additions & 3 deletions ggml/src/ggml-vulkan/vulkan-shaders/topk_moe.comp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@
layout (push_constant) uniform parameter
{
uint n_rows;
uint n_experts_push;
uint n_expert_used;
float clamp_min;
float clamp_max;
Expand All @@ -18,11 +19,16 @@ layout (push_constant) uniform parameter
layout(local_size_x_id = 0, local_size_y = 4, local_size_z = 1) in;

layout(constant_id = 0) const uint WARP_SIZE = 32;
layout(constant_id = 1) const uint n_experts = 512;
layout(constant_id = 1) const uint n_experts_spec = 512;
layout(constant_id = 2) const bool with_norm = true;
layout(constant_id = 3) const bool late_softmax = false;
layout(constant_id = 4) const bool nexperts_use_push = false;

const uint experts_per_thread = (n_experts > WARP_SIZE) ? n_experts / WARP_SIZE : 1;
uint n_experts = nexperts_use_push ? n_experts_push : n_experts_spec;

#define CEIL_DIV(a, b) (((a) + (b) - 1) / (b))

const uint experts_per_thread = CEIL_DIV(n_experts_spec, WARP_SIZE);

layout (binding = 0, std430) readonly buffer Logits {float logits[];};
layout (binding = 1, std430) writeonly buffer Weights {float weights[];};
Expand Down Expand Up @@ -94,7 +100,7 @@ void main() {
}

if (!late_softmax) {
softmax_warp_inplace(wt, n_experts, lane, false);
softmax_warp_inplace(wt, n_experts, lane, nexperts_use_push);
}

// at this point, each thread holds a portion of softmax,
Expand Down
4 changes: 4 additions & 0 deletions tests/test-backend-ops.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7927,8 +7927,12 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {

for (bool with_norm : {false, true}) {
test_cases.emplace_back(new test_topk_moe({8, 22, 1, 1}, 4, with_norm));
test_cases.emplace_back(new test_topk_moe({31, 22, 1, 1}, 8, with_norm));
test_cases.emplace_back(new test_topk_moe({32, 22, 1, 1}, 8, with_norm));
test_cases.emplace_back(new test_topk_moe({40, 22, 1, 1}, 8, with_norm));
test_cases.emplace_back(new test_topk_moe({71, 22, 1, 1}, 8, with_norm));
test_cases.emplace_back(new test_topk_moe({128, 1, 1, 1}, 128, with_norm));
test_cases.emplace_back(new test_topk_moe({129, 1, 1, 1}, 128, with_norm));
}

test_cases.emplace_back(new test_topk_moe({ 8, 22, 1, 1 }, 4, /*with_norm*/ false, /*delayed_softmax*/ true));
Expand Down
Loading