@@ -9787,10 +9787,6 @@ static void ggml_vk_silu_back(ggml_backend_vk_context * ctx, vk_context& subctx,
97879787 ggml_vk_op_f32<vk_op_push_constants>(ctx, subctx, src0, src1, nullptr, nullptr, dst, GGML_OP_SILU_BACK, { (uint32_t)ggml_nelements(src0), 0, 0.0f, 0.0f });
97889788}
97899789
9790- static void ggml_vk_geglu_back(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst, bool dryrun = false) {
9791- ggml_vk_op_f32<vk_op_push_constants>(ctx, subctx, src0, src1, nullptr, dst, GGML_OP_GEGLU_BACK, { (uint32_t)ggml_nelements(src0), 0, 0.0f, 0.0f }, dryrun);
9792- }
9793-
97949790static void ggml_vk_norm(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, ggml_tensor * dst) {
97959791 float * op_params = (float *)dst->op_params;
97969792
@@ -10022,19 +10018,6 @@ static void ggml_vk_diag_mask_inf(ggml_backend_vk_context * ctx, vk_context& sub
1002210018 ggml_vk_op_f32<vk_op_diag_mask_push_constants>(ctx, subctx, src0, nullptr, nullptr, nullptr, dst, GGML_OP_DIAG_MASK_INF, { (uint32_t)src0->ne[0], (uint32_t)src0->ne[1], op_params[0] });
1002310019}
1002410020
10025- static void ggml_vk_cross_entropy_loss_back(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, const ggml_tensor * src2, ggml_tensor * dst, bool dryrun = false) {
10026- const int64_t nclasses = src1->ne[0];
10027- const int64_t nrows = ggml_nrows(src1);
10028-
10029- ggml_vk_op_f32<vk_op_push_constants>(ctx, subctx, src0, src1, src2, dst, GGML_OP_CROSS_ENTROPY_LOSS_BACK, {
10030- (uint32_t)nclasses,
10031- (uint32_t)nrows,
10032- 0.0f,
10033- 0.0f
10034- }, dryrun);
10035-
10036- }
10037-
1003810021static void ggml_vk_soft_max(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, const ggml_tensor * src2, ggml_tensor * dst) {
1003910022 float * op_params = (float *)dst->op_params;
1004010023
@@ -10190,24 +10173,6 @@ static void ggml_vk_count_equal(ggml_backend_vk_context * ctx, vk_context& subct
1019010173 ggml_vk_op_f32<vk_op_push_constants>(ctx, subctx, src0, src1, nullptr, nullptr, dst, GGML_OP_COUNT_EQUAL, { (uint32_t)ggml_nelements(src0), 0, 0.0f, 0.0f });
1019110174}
1019210175
10193- static void ggml_vk_out_prod(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst, bool dryrun = false) {
10194- const uint32_t src0_type_size = ggml_type_size(src0->type);
10195- const uint32_t src1_type_size = ggml_type_size(src1->type);
10196- const uint32_t dst_type_size = ggml_type_size(dst->type);
10197-
10198- const int64_t r2 = src1->ne[2] / src0->ne[2];
10199- const int64_t r3 = src1->ne[3] / src0->ne[3];
10200-
10201- ggml_vk_op_f32<vk_op_binary_push_constants>(ctx, subctx, src0, src1, nullptr, dst, GGML_OP_OUT_PROD, {
10202- (uint32_t)ggml_nelements(dst),
10203- (uint32_t)src0->ne[0], (uint32_t)src0->ne[1], (uint32_t)src0->ne[2],(uint32_t)src0->ne[3], (uint32_t)src0->nb[0] / src0_type_size, (uint32_t)src0->nb[1] / src0_type_size, (uint32_t)src0->nb[2] / src0_type_size, (uint32_t)src0->nb[3] / src0_type_size,
10204- (uint32_t)src1->ne[0], (uint32_t)src1->ne[1], (uint32_t)src1->ne[2],(uint32_t)src1->ne[3], (uint32_t)src1->nb[0] / src1_type_size, (uint32_t)src1->nb[1] / src1_type_size, (uint32_t)src1->nb[2] / src1_type_size, (uint32_t)src1->nb[3] / src1_type_size,
10205- (uint32_t) dst->ne[0], (uint32_t) dst->ne[1], (uint32_t) dst->ne[2],(uint32_t) dst->ne[3], (uint32_t) dst->nb[0] / dst_type_size, (uint32_t) dst->nb[1] / dst_type_size, (uint32_t) dst->nb[2] / dst_type_size, (uint32_t) dst->nb[3] / dst_type_size,
10206- 0,
10207- 0.0f, (float) r2, (int32_t) r3
10208- }, dryrun);
10209- }
10210-
1021110176static void ggml_vk_im2col(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst) {
1021210177 const int32_t s0 = dst->op_params[0];
1021310178 const int32_t s1 = dst->op_params[1];
0 commit comments