DarthAffe/stable-diffusion.cpp
1
0 Fork 0
mirror of https://github.com/leejet/stable-diffusion.cpp.git synced 2026-11-11 19:36:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

perf: cache CPU position embeddings across DiT runners (#2211)

Browse Source
This commit is contained in:
leejet authored and GitHub committed 2026-10-25 05:18:30 +08:00
1 parent a85db96aa5
commit ea209dbfcf
31 files changed
+855 -610
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

src/model/diffusion/ideogram4.hpp
+10 -4
View File
@@ -8,6 +8,7 @@
#include <memory>
#include <string>
#include <vector>
#include "core/pe_cache.h"
#include "core/ggml_extend.h"
#include "core/ggml_graph_cut.h"
@@ -422,6 +423,8 @@ namespace Ideogram4 {
};
class Ideogram4Runner : public DiffusionModelRunner {
sd::PECache pe_cache_;
protected:
bool should_use_uncond_model(const DiffusionParams& diffusion_params) const {
return has_uncond_model &&
@@ -436,7 +439,6 @@ namespace Ideogram4 {
Ideogram4Transformer uncond_model;
bool has_uncond_model = false;
std::string uncond_prefix;
std::vector<float> pe_vec;
std::vector<int32_t> image_indicator_vec;
Ideogram4Runner(ggml_backend_t backend,
@@ -496,15 +498,19 @@ namespace Ideogram4 {
int64_t head_dim = config.emb_dim / config.num_heads;
auto runner_ctx = get_context();
pe_vec = finish_rope_pe(gen_ideogram4_pe(static_cast<int>(grid_h),
sd::PECacheKey pe_key("ideogram4.pe");
pe_key.add(grid_h).add(grid_w).add(x->ne[3]).add(context_len).add(head_dim).add(pos_len).add(config.rope_theta);
pe_key.add(config.mrope_section).add(circular_x_enabled).add(circular_y_enabled);
auto pe = make_input(pe_cache_.get_or_create(pe_key, [&]() {
auto pe_vec = finish_rope_pe(gen_ideogram4_pe(static_cast<int>(grid_h),
static_cast<int>(grid_w),
static_cast<int>(x->ne[3]),
static_cast<int>(context_len),
static_cast<int>(head_dim),
static_cast<int>(config.rope_theta),
config.mrope_section));
auto pe = ggml_new_tensor_4d(compute_ctx, GGML_TYPE_F32, 2, 2, head_dim / 2, pos_len);
set_backend_tensor_data(pe, pe_vec.data());
return sd::Tensor<float>({2, 2, head_dim / 2, pos_len}, std::move(pe_vec));
}));
image_indicator_vec.assign(static_cast<size_t>(pos_len), 1);
for (int64_t i = 0; i < context_len; ++i) {