DarthAffe/stable-diffusion.cpp
1
0 Fork 0
mirror of https://github.com/leejet/stable-diffusion.cpp.git synced 2026-11-11 03:24:30 +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/qwen_image.hpp
+21 -18
View File
@@ -4,6 +4,7 @@
#include <memory>
#include "core/ggml_extend_backend.h"
#include "core/ggml_tensor_utils.h"
#include "core/pe_cache.h"
#include "core/util.h"
#include "model/common/block.hpp"
@@ -561,10 +562,11 @@ namespace Qwen {
};
struct QwenImageRunner : public DiffusionModelRunner {
sd::PECache pe_cache_;
public:
QwenImageConfig config;
QwenImageModel qwen_image;
std::vector<float> pe_vec;
std::vector<float> modulate_index_vec;
std::vector<int32_t> additional_t_cond_vec;
SDVersion version;
@@ -635,23 +637,24 @@ namespace Qwen {
ref_index_mode = Rope::RefIndexMode::DECREASE;
}
pe_vec = finish_rope_pe(Rope::gen_qwen_image_pe(time_len,
static_cast<int>(x->ne[1]),
static_cast<int>(x->ne[0]),
config.patch_size,
batch_size,
static_cast<int>(context->ne[1]),
ref_latents,
ref_index_mode,
config.theta,
config.axes_dim));
int pos_len = static_cast<int>(pe_vec.size() / config.axes_dim_sum / 2);
// LOG_VERBOSE("pos_len %d", pos_len);
auto pe = ggml_new_tensor_4d(compute_ctx, GGML_TYPE_F32, 2, 2, config.axes_dim_sum / 2, pos_len);
// pe->data = pe_vec.data();
// print_ggml_tensor(pe, true, "pe");
// pe->data = nullptr;
set_backend_tensor_data(pe, pe_vec.data());
sd::PECacheKey pe_key("qwen_image.pe");
pe_key.add_shape(x).add_shape(context).add(config.theta).add(config.axes_dim).add(config.axes_dim_sum);
pe_key.add(circular_x_enabled).add(circular_y_enabled).add(time_len).add(batch_size).add(config.patch_size);
pe_key.add_shapes(ref_latents).add(ref_index_mode);
auto pe = make_input(pe_cache_.get_or_create(pe_key, [&]() {
auto pe_vec = finish_rope_pe(Rope::gen_qwen_image_pe(time_len,
static_cast<int>(x->ne[1]),
static_cast<int>(x->ne[0]),
config.patch_size,
batch_size,
static_cast<int>(context->ne[1]),
ref_latents,
ref_index_mode,
config.theta,
config.axes_dim));
int pos_len = static_cast<int>(pe_vec.size() / config.axes_dim_sum / 2);
return sd::Tensor<float>({2, 2, config.axes_dim_sum / 2, pos_len}, std::move(pe_vec));
}));
ggml_tensor* modulate_index = nullptr;
if (config.zero_cond_t) {