upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-07-23 11:10:55 -05:00
Code Issues Packages Projects Releases Wiki Activity

fix compiler warning

Browse Source
This commit is contained in:
Ruben Ortlam
2026-07-17 08:08:07 +02:00
parent 0e8431faf9
commit 3989780be6
1 changed files with 7 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
ggml/src/ggml-vulkan/ggml-vulkan.cpp
View File

@@ -4363,7 +4363,7 @@ static void ggml_vk_load_shaders(vk_device& device, vk_pipeline requested) {
#if defined(VK_NV_cooperative_matrix2) && defined(GGML_VULKAN_COOPMAT2_GLSLC_SUPPORT)
if (device->coopmat2) {
auto const &ggml_vk_mul_mm_cm2_spec = [&](std::vector<uint32_t> spec, bool aligned, bool mul_mat_id, uint32_t type = UINT32_MAX) {
auto const &ggml_vk_mul_mm_cm2_spec = [&](std::vector<uint32_t> spec, bool aligned, uint32_t type = UINT32_MAX) {
spec.push_back(aligned ? 1u : 0u); // ALIGNED
spec.push_back(device->subgroup_size); // subgroup_size
if (type != UINT32_MAX) {
@@ -4375,8 +4375,7 @@ static void ggml_vk_load_shaders(vk_device& device, vk_pipeline requested) {
std::vector<vk_tile_config> tc_mmq_k = {{s_warptile_mmq_k, s_mmq_wg_denoms_k, s_align}, {m_warptile_mmq_k, m_mmq_wg_denoms_k, m_align}, {l_warptile_mmq_k, l_mmq_wg_denoms_k, l_align}};
std::vector<vk_tile_config> tc_mmqid = {{s_warptile_mmqid, s_mmqid_wg_denoms, s_align}, {m_warptile_mmqid, m_mmqid_wg_denoms, m_align}, {l_warptile_mmqid, l_mmqid_wg_denoms, l_align}};
spec_fn_t cm2_spec = [&](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a, false); };
spec_fn_t cm2_spec_id = [&](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a, true); };
spec_fn_t cm2_spec = [&](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a); };
// F16 x F16
create_mm_pipelines({GGML_TYPE_F16, GGML_TYPE_F16, false, true}, tc_mm, "matmul_f16_f16acc", matmul_f16_f16acc_cm2_len, matmul_f16_f16acc_cm2_data, sizeof(vk_mat_mat_push_constants), 3, cm2_spec, true);
@@ -4388,7 +4387,7 @@ static void ggml_vk_load_shaders(vk_device& device, vk_pipeline requested) {
#endif
for (const auto type : quant_types) {
auto& tc = (type >= GGML_TYPE_Q2_K && type <= GGML_TYPE_Q6_K) ? tc_mmq_k : tc_mmq;
spec_fn_t qs = [&, type](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a, false, (uint32_t)type); };
spec_fn_t qs = [&, type](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a, (uint32_t)type); };
#if defined(GGML_VULKAN_FLOAT_E2M1_GLSLC_SUPPORT) && defined(GGML_VULKAN_FLOAT_E4M3_GLSLC_SUPPORT)
if (device->ocp_fp4 && (type == GGML_TYPE_MXFP4 || type == GGML_TYPE_NVFP4)) {
create_mm_pipelines({type, GGML_TYPE_F16, false, true}, tc, "matmul_quant_f16_ocp_f16acc", matmul_quant_f16_ocp_f16acc_cm2_len, matmul_quant_f16_ocp_f16acc_cm2_data, sizeof(vk_mat_mat_push_constants), 3, qs, true);
@@ -4404,15 +4403,15 @@ static void ggml_vk_load_shaders(vk_device& device, vk_pipeline requested) {
GGML_ASSERT(device->subgroup_ballot);
// ID: F16
create_mm_pipelines({GGML_TYPE_F16, GGML_TYPE_F16, true, true}, tc_mm, "matmul_id_subgroup_f16_f16acc", matmul_id_subgroup_f16_f16acc_cm2_len, matmul_id_subgroup_f16_f16acc_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, cm2_spec_id, true);
create_mm_pipelines({GGML_TYPE_F16, GGML_TYPE_F16, true, false}, tc_mm, "matmul_id_subgroup_f16", matmul_id_subgroup_f16_cm2_len, matmul_id_subgroup_f16_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, cm2_spec_id, true);
create_mm_pipelines({GGML_TYPE_F16, GGML_TYPE_F16, true, true}, tc_mm, "matmul_id_subgroup_f16_f16acc", matmul_id_subgroup_f16_f16acc_cm2_len, matmul_id_subgroup_f16_f16acc_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, cm2_spec, true);
create_mm_pipelines({GGML_TYPE_F16, GGML_TYPE_F16, true, false}, tc_mm, "matmul_id_subgroup_f16", matmul_id_subgroup_f16_cm2_len, matmul_id_subgroup_f16_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, cm2_spec, true);
#if defined(GGML_VULKAN_BFLOAT16_GLSLC_SUPPORT)
if (device->coopmat_bf16_support) {
create_mm_pipelines({GGML_TYPE_BF16, GGML_TYPE_BF16, true, false}, tc_mm, "matmul_id_subgroup_bf16", matmul_id_subgroup_bf16_cm2_len, matmul_id_subgroup_bf16_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, cm2_spec_id, true);
create_mm_pipelines({GGML_TYPE_BF16, GGML_TYPE_BF16, true, false}, tc_mm, "matmul_id_subgroup_bf16", matmul_id_subgroup_bf16_cm2_len, matmul_id_subgroup_bf16_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, cm2_spec, true);
}
#endif
for (const auto type : quant_types) {
spec_fn_t qs_id = [&, type](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a, true, (uint32_t)type); };
spec_fn_t qs_id = [&, type](const std::vector<uint32_t>& wt, bool a) { return ggml_vk_mul_mm_cm2_spec(wt, a, (uint32_t)type); };
#if defined(GGML_VULKAN_FLOAT_E2M1_GLSLC_SUPPORT) && defined(GGML_VULKAN_FLOAT_E4M3_GLSLC_SUPPORT)
if (device->ocp_fp4 && (type == GGML_TYPE_MXFP4 || type == GGML_TYPE_NVFP4)) {
create_mm_pipelines({type, GGML_TYPE_F16, true, true}, tc_mmqid, "matmul_id_subgroup_quant_f16_ocp_f16acc", matmul_id_subgroup_quant_f16_ocp_f16acc_cm2_len, matmul_id_subgroup_quant_f16_ocp_f16acc_cm2_data, sizeof(vk_mat_mat_id_push_constants), 5, qs_id, true);