* installing *source* package 'sd2R' ...
** this is package 'sd2R' version '0.2.1'
** package 'sd2R' successfully unpacked and MD5 sums checked
** using staged installation
* Downloading clip_merges.hpp...
      % Total    % Received % Xferd  Average Speed  Time    Time    Time   Current
                                 Dload  Upload  Total   Spent   Left   Speed

  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0

  0      0   0      0   0      0      0      0                              0
100  2.50M 100  2.50M   0      0  6.25M      0                              0
100  2.50M 100  2.50M   0      0  6.24M      0                              0
100  2.50M 100  2.50M   0      0  6.23M      0                              0
* clip_merges.hpp: downloaded successfully
* Downloading t5.hpp...
      % Total    % Received % Xferd  Average Speed  Time    Time    Time   Current
                                 Dload  Upload  Total   Spent   Left   Speed

  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0

  0      0   0      0   0      0      0      0                              0
100  6.93M 100  6.93M   0      0 16.44M      0                              0
100  6.93M 100  6.93M   0      0 16.42M      0                              0
100  6.93M 100  6.93M   0      0 16.41M      0                              0
* t5.hpp: downloaded successfully
* Downloading umt5.hpp...
      % Total    % Received % Xferd  Average Speed  Time    Time    Time   Current
                                 Dload  Upload  Total   Spent   Left   Speed

  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0

  0      0   0      0   0      0      0      0                              0
 17 43.61M  17  7.52M   0      0  6.02M      0   00:07   00:01   00:06  7.52M
100 43.61M 100 43.61M   0      0 24.43M      0   00:01   00:01          7.52M
100 43.61M 100 43.61M   0      0 24.42M      0   00:01   00:01          7.52M
100 43.61M 100 43.61M   0      0 24.42M      0   00:01   00:01          7.52M
* umt5.hpp: downloaded successfully
* Downloading qwen_merges.hpp...
      % Total    % Received % Xferd  Average Speed  Time    Time    Time   Current
                                 Dload  Upload  Total   Spent   Left   Speed

  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0
  0      0   0      0   0      0      0      0                              0

  0      0   0      0   0      0      0      0                              0
100  7.97M 100  7.97M   0      0 23.41M      0                              0
100  7.97M 100  7.97M   0      0 23.38M      0                              0
100  7.97M 100  7.97M   0      0 23.35M      0                              0
* qwen_merges.hpp: downloaded successfully
* ggmlR lib: D:/RCompile/CRANpkg/lib/4.6/ggmlR/lib
* ggmlR include: D:/RCompile/CRANpkg/lib/4.6/ggmlR/include
* OpenMP: enabled (C++: -fopenmp)
* Vulkan: not found (building without GPU support)
* Meta backend: enabled (ggml_backend_meta_device found in libggml.a)
Generating src/Makevars.win...
* Wrote src/Makevars.win
Configuration complete.
** libs
specified C++17
using C++ compiler: 'g++.exe (GCC) 14.3.0'
using C++17
make[1]: Entering directory '/d/temp/2026_06_20_01_50_00_3901/RtmpSYiQ2A/R.INSTALL71386a833739/sd2R/src'
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c RcppExports.cpp -o RcppExports.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd2R_interface.cpp -o sd2R_interface.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/stable-diffusion.cpp -o sd/stable-diffusion.o
In file included from sd/anima.hpp:9,
                 from sd/stable-diffusion.cpp:15:
sd/common_block.hpp: In member function 'ggml_tensor* CrossAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*)':
sd/common_block.hpp:318:17: warning: unused variable 'n_context' [-Wunused-variable]
  318 |         int64_t n_context = context->ne[1];
      |                 ^~~~~~~~~
sd/common_block.hpp:319:17: warning: unused variable 'inner_dim' [-Wunused-variable]
  319 |         int64_t inner_dim = d_head * n_head;
      |                 ^~~~~~~~~
sd/common_block.hpp: At global scope:
sd/common_block.hpp:126:26: warning: 'virtual ggml_tensor* ResBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*)' was hidden [-Woverloaded-virtual=]
  126 |     virtual ggml_tensor* forward(GGMLRunnerContext* ctx, ggml_tensor* x, ggml_tensor* emb = nullptr) {
      |                          ^~~~~~~
sd/common_block.hpp:569:18: note:   by 'ggml_tensor* VideoResBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int)'
  569 |     ggml_tensor* forward(GGMLRunnerContext* ctx,
      |                  ^~~~~~~
In file included from sd/flux.hpp:10,
                 from sd/anima.hpp:11:
sd/rope.hpp: In function 'std::vector<std::vector<float> > Rope::rope(const std::vector<float>&, int, float, const std::vector<int>&)':
sd/rope.hpp:66:34: warning: comparison of integer expressions of different signedness: 'size_t' {aka 'long long unsigned int'} and 'int' [-Wsign-compare]
   66 |             for (size_t j = 0; j < half_dim; ++j) {
      |                                ~~^~~~~~~~~~
sd/rope.hpp:89:27: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
   89 |         for (int i = 0; i < pos_size; ++i) {
      |                         ~~^~~~~~~~~~
sd/rope.hpp: In function 'std::vector<std::vector<float> > Rope::gen_flux_img_ids(int, int, int, int, int, int, int, int, bool)':
sd/rope.hpp:157:31: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<std::vector<float> >::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  157 |             for (int j = 0; j < img_ids.size(); ++j) {
      |                             ~~^~~~~~~~~~~~~~~~
sd/rope.hpp: In function 'std::vector<std::vector<float> > Rope::concat_ids(const std::vector<std::vector<float> >&, const std::vector<std::vector<float> >&, int)':
sd/rope.hpp:171:31: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  171 |             for (int j = 0; j < a_len; ++j) {
      |                             ~~^~~~~~~
sd/rope.hpp:174:31: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  174 |             for (int j = 0; j < b_len; ++j) {
      |                             ~~^~~~~~~
sd/rope.hpp: In function 'std::vector<float> Rope::embed_nd(const std::vector<std::vector<float> >&, int, const std::vector<float>&, const std::vector<int>&, const std::vector<std::vector<int> >&, EmbedNDLayout)':
sd/rope.hpp:202:41: warning: comparison of integer expressions of different signedness: 'size_t' {aka 'long long unsigned int'} and 'int' [-Wsign-compare]
  202 |             if (!wrap_dims.empty() && i < (int)wrap_dims.size()) {
      |                                       ~~^~~~~~~~~~~~~~~~~~~~~~~
sd/rope.hpp:212:35: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  212 |                 for (int j = 0; j < pos_len; ++j) {
      |                                 ~~^~~~~~~~~
In file included from sd/conditioner.hpp:8,
                 from sd/stable-diffusion.cpp:17:
sd/clip.hpp: In member function 'ggml_tensor* CLIPTextModelRunner::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, size_t, bool, int)':
sd/clip.hpp:508:16: warning: unused variable 'N' [-Wunused-variable]
  508 |         size_t N       = input_ids->ne[1];
      |                ^
sd/clip.hpp:509:16: warning: unused variable 'n_token' [-Wunused-variable]
  509 |         size_t n_token = input_ids->ne[0];
      |                ^~~~~~~
In file included from sd/conditioner.hpp:9:
sd/llm.hpp: In function 'ggml_tensor* LLM::splice_image_embeds(GGMLRunnerContext*, ggml_tensor*, const std::vector<std::pair<int, ggml_tensor*> >&)':
sd/llm.hpp:352:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<std::pair<int, ggml_tensor*> >::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  352 |         for (int i = 0; i < image_embeds.size(); i++) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~
sd/llm.hpp: In member function 'ggml_tensor* LLM::VisionAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)':
sd/llm.hpp:574:21: warning: unused variable 'n_token' [-Wunused-variable]
  574 |             int64_t n_token = x->ne[1];
      |                     ^~~~~~~
sd/llm.hpp:575:21: warning: unused variable 'N' [-Wunused-variable]
  575 |             int64_t N       = x->ne[2];
      |                     ^
sd/llm.hpp: In member function 'ggml_tensor* LLM::TextModel::forward_embeds(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, std::set<int>, ggml_tensor*, bool)':
sd/llm.hpp:1143:35: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<ggml_tensor*>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 1143 |                 for (int i = 1; i < intermediate_outputs.size(); i++) {
      |                                 ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/llm.hpp:1151:35: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<ggml_tensor*>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 1151 |                 for (int i = 1; i < intermediate_outputs.size(); i++) {
      |                                 ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/llm.hpp: In static member function 'static ggml_tensor* LLM::LLMRunner::encode_image_common(GGMLRunner*, ggml_context*, GGMLRunnerContext*, ggml_tensor*, const LLM::LLMVisionParams&, std::shared_ptr<LLM::VisionModel>, std::vector<int>&, std::vector<int>&, std::vector<float>&, std::vector<float>&, std::array<std::vector<int>, 4>&, std::array<std::vector<float>, 4>&)':
sd/llm.hpp:1420:47: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 1420 |             for (int seq_index = 0; seq_index < seqlens.size(); seq_index++) {
      |                                     ~~~~~~~~~~^~~~~~~~~~~~~~~~
In file included from D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-alloc.h:3,
                 from sd/ggml_extend.hpp:26,
                 from sd/stable-diffusion.cpp:5:
sd/llm.hpp:1429:65: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<float>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 1429 |                         GGML_ASSERT((i * (grid_h * grid_w) + j) < window_mask_vec.size());
      |                                     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml.h:285:30: note: in definition of macro 'GGML_ASSERT'
  285 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
      |                              ^
sd/llm.hpp: In member function 'int64_t LLM::LLMRunner::get_num_image_tokens(int64_t, int64_t, int64_t)':
sd/llm.hpp:1748:21: warning: unused variable 'llm_grid_h' [-Wunused-variable]
 1748 |             int64_t llm_grid_h = grid_h / params.vision.spatial_merge_size;
      |                     ^~~~~~~~~~
sd/llm.hpp:1749:21: warning: unused variable 'llm_grid_w' [-Wunused-variable]
 1749 |             int64_t llm_grid_w = grid_w / params.vision.spatial_merge_size;
      |                     ^~~~~~~~~~
sd/llm.hpp: In member function 'void LLM::LLMEmbedder::test()':
sd/llm.hpp:1936:37: warning: unused variable 'weights' [-Wunused-variable]
 1936 |                 std::vector<float>& weights = std::get<1>(tokens_and_weights);
      |                                     ^~~~~~~
sd/llm.hpp:1980:37: warning: unused variable 'weights' [-Wunused-variable]
 1980 |                 std::vector<float>& weights = std::get<1>(tokens_and_weights);
      |                                     ^~~~~~~
sd/llm.hpp:2005:37: warning: unused variable 'weights' [-Wunused-variable]
 2005 |                 std::vector<float>& weights = std::get<1>(tokens_and_weights);
      |                                     ^~~~~~~
sd/llm.hpp:2030:37: warning: unused variable 'weights' [-Wunused-variable]
 2030 |                 std::vector<float>& weights = std::get<1>(tokens_and_weights);
      |                                     ^~~~~~~
In file included from sd/conditioner.hpp:10:
sd/t5.hpp: In member function 'ggml_tensor* T5Runner::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)':
sd/t5.hpp:349:16: warning: unused variable 'N' [-Wunused-variable]
  349 |         size_t N       = input_ids->ne[1];
      |                ^
sd/t5.hpp:350:16: warning: unused variable 'n_token' [-Wunused-variable]
  350 |         size_t n_token = input_ids->ne[0];
      |                ^~~~~~~
sd/t5.hpp: In member function 'void T5Embedder::test()':
sd/t5.hpp:538:33: warning: unused variable 'weights' [-Wunused-variable]
  538 |             std::vector<float>& weights = std::get<1>(tokens_and_weights);
      |                                 ^~~~~~~
sd/conditioner.hpp: In member function 'std::tuple<std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<bool, std::allocator<bool> > > FrozenCLIPEmbedderWithCustomWords::tokenize_with_trigger_token(std::string, int, int32_t)':
sd/conditioner.hpp:409:58: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  409 |                 for (int i = class_token_index[0] + 1; i < clean_input_ids.size(); i++)
      |                                                        ~~^~~~~~~~~~~~~~~~~~~~~~~~
sd/conditioner.hpp:424:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  424 |         for (int i = 0; i < tokens.size(); i++) {
      |                         ~~^~~~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'SDCondition FrozenCLIPEmbedderWithCustomWords::get_learned_condition_common(int, std::vector<int>&, std::vector<float>&, int, int, int, bool)':
sd/conditioner.hpp:528:43: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  528 |         for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
      |                                 ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'SDCondition SD3CLIPEmbedder::get_learned_condition_common(int, std::vector<std::pair<std::vector<int>, std::vector<float> > >, int, bool)':
sd/conditioner.hpp:986:43: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  986 |         for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
      |                                 ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'SDCondition FluxCLIPEmbedder::get_learned_condition_common(int, std::vector<std::pair<std::vector<int>, std::vector<float> > >, int, bool)':
sd/conditioner.hpp:1338:43: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
 1338 |         for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
      |                                 ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'std::tuple<std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<float, std::allocator<float> > > T5CLIPEmbedder::tokenize(std::string, size_t, size_t)':
sd/conditioner.hpp:1519:14: warning: variable 'on_new_token_cb' set but not used [-Wunused-but-set-variable]
 1519 |         auto on_new_token_cb = [&](std::string& str, std::vector<int32_t>& bpe_tokens) -> bool {
      |              ^~~~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'SDCondition T5CLIPEmbedder::get_learned_condition_common(int, std::tuple<std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<float, std::allocator<float> > >, int, bool)':
sd/conditioner.hpp:1580:43: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
 1580 |         for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
      |                                 ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'std::tuple<std::vector<int, std::allocator<int> >, std::vector<float, std::allocator<float> >, std::vector<float, std::allocator<float> > > LLMEmbedder::tokenize(std::string, const std::pair<int, int>&, size_t, size_t, bool)':
sd/conditioner.hpp:1860:35: warning: comparison of integer expressions of different signedness: 'const int' and 'std::__cxx11::basic_string<char>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 1860 |             if (attn_range.second < text.size()) {
      |                 ~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function 'virtual SDCondition LLMEmbedder::get_learned_condition(int, const ConditionerParams&)':
sd/conditioner.hpp:1990:35: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<sd::Tensor<float> >::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 1990 |                 for (int i = 0; i < conditioner_params.ref_images->size(); i++) {
      |                                 ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/conditioner.hpp:2061:35: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<sd::Tensor<float> >::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 2061 |                 for (int i = 0; i < conditioner_params.ref_images->size(); i++) {
      |                                 ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/conditioner.hpp:2162:35: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<sd::Tensor<float> >::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 2162 |                 for (int i = 0; i < conditioner_params.ref_images->size() - 1; i++) {
      |                                 ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/conditioner.hpp:2261:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<std::__cxx11::basic_string<char> >::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
 2261 |         for (int i = 0; i < extra_prompts.size(); i++) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~~
In file included from sd/stable-diffusion.cpp:18:
sd/control.hpp: In constructor 'ControlNetBlock::ControlNetBlock(SDVersion)':
sd/control.hpp:111:27: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  111 |         for (int i = 0; i < len_mults; i++) {
      |                         ~~^~~~~~~~~~~
sd/control.hpp:136:19: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  136 |             if (i != len_mults - 1) {
      |                 ~~^~~~~~~~~~~~~~~~
sd/control.hpp: In member function 'std::vector<ggml_tensor*> ControlNetBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*)':
sd/control.hpp:267:27: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  267 |         for (int i = 0; i < len_mults; i++) {
      |                         ~~^~~~~~~~~~~
sd/control.hpp:282:19: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  282 |             if (i != len_mults - 1) {
      |                 ~~^~~~~~~~~~~~~~~~
sd/control.hpp:268:17: warning: unused variable 'mult' [-Wunused-variable]
  268 |             int mult = channel_mult[i];
      |                 ^~~~
sd/control.hpp: In member function 'void ControlNet::alloc_control_ctx(std::vector<ggml_tensor*>)':
sd/control.hpp:347:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<ggml_tensor*>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  347 |         for (int i = 0; i < outs.size() - 1; i++) {
      |                         ~~^~~~~~~~~~~~~~~~~
sd/control.hpp: In member function 'ggml_cgraph* ControlNet::build_graph(const sd::Tensor<float>&, const sd::Tensor<float>&, const sd::Tensor<float>&, const sd::Tensor<float>&, const sd::Tensor<float>&)':
sd/control.hpp:417:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<ggml_tensor*>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  417 |         for (int i = 0; i < outs.size() - 1; i++) {
      |                         ~~^~~~~~~~~~~~~~~~~
In file included from sd/stable-diffusion.cpp:19:
sd/denoiser.hpp: In function 'sd::Tensor<float> sample_tcd(denoise_cb_t, sd::Tensor<float>, const std::vector<float>&, std::shared_ptr<RNG>, float)':
sd/denoiser.hpp:1726:15: warning: unused variable 'alpha_prod_t' [-Wunused-variable]
 1726 |         float alpha_prod_t      = 1.0f / (sigma * sigma + 1.0f);
      |               ^~~~~~~~~~~~
In file included from sd/ernie_image.hpp:10,
                 from sd/stable-diffusion.cpp:21:
sd/qwen_image.hpp: In member function 'ggml_tensor* Qwen::QwenImageModel::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<ggml_tensor*>, ggml_tensor*)':
sd/qwen_image.hpp:456:21: warning: unused variable 'C' [-Wunused-variable]
  456 |             int64_t C = x->ne[2];
      |                     ^
sd/qwen_image.hpp:457:21: warning: unused variable 'N' [-Wunused-variable]
  457 |             int64_t N = x->ne[3];
      |                     ^
In file included from sd/ltx_vae.hpp:14,
                 from sd/stable-diffusion.cpp:30:
sd/wan.hpp: In constructor 'WAN::Encoder3d::Encoder3d(int64_t, int64_t, std::vector<int>, int, std::vector<bool>, bool)':
sd/wan.hpp:627:31: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<long long int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  627 |             for (int i = 0; i < dims.size() - 1; i++) {
      |                             ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:631:42: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<bool>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  631 |                     bool t_down_flag = i < temperal_downsample.size() ? temperal_downsample[i] : false;
      |                                        ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:636:92: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  636 |                                                                                          i != dim_mult.size() - 1));
      |                                                                                          ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:646:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  646 |                     if (i != dim_mult.size() - 1) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp: In member function 'ggml_tensor* WAN::Encoder3d::forward(GGMLRunnerContext*, ggml_tensor*, int64_t, std::vector<ggml_tensor*>&, int&, int)':
sd/wan.hpp:704:31: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<long long int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  704 |             for (int i = 0; i < dims.size() - 1; i++) {
      |                             ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:716:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  716 |                     if (i != dim_mult.size() - 1) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp: In constructor 'WAN::Decoder3d::Decoder3d(int64_t, int64_t, std::vector<int>, int, std::vector<bool>, bool)':
sd/wan.hpp:796:31: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<long long int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  796 |             for (int i = 0; i < dims.size() - 1; i++) {
      |                             ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:800:40: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<bool>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  800 |                     bool t_up_flag = i < temperal_upsample.size() ? temperal_upsample[i] : false;
      |                                      ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:805:88: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  805 |                                                                                      i != dim_mult.size() - 1));
      |                                                                                      ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:818:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  818 |                     if (i != dim_mult.size() - 1) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp: In member function 'ggml_tensor* WAN::Decoder3d::forward(GGMLRunnerContext*, ggml_tensor*, int64_t, std::vector<ggml_tensor*>&, int&, int)':
sd/wan.hpp:884:31: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<long long int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  884 |             for (int i = 0; i < dims.size() - 1; i++) {
      |                             ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:896:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<int>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  896 |                     if (i != dim_mult.size() - 1) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp: In static member function 'static void WAN::WanVAERunner::load_from_file_and_test(const std::string&)':
sd/wan.hpp:1332:23: warning: unused variable 'model_data_type' [-Wunused-variable]
 1332 |             ggml_type model_data_type         = GGML_TYPE_F16;
      |                       ^~~~~~~~~~~~~~~
sd/wan.hpp: At global scope:
sd/wan.hpp:1389:30: warning: 'virtual ggml_tensor* WAN::WanSelfAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)' was hidden [-Woverloaded-virtual=]
 1389 |         virtual ggml_tensor* forward(GGMLRunnerContext* ctx,
      |                              ^~~~~~~
sd/wan.hpp:1430:30: note:   by 'virtual ggml_tensor* WAN::WanCrossAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int64_t)'
 1430 |         virtual ggml_tensor* forward(GGMLRunnerContext* ctx,
      |                              ^~~~~~~
sd/wan.hpp: In member function 'virtual ggml_tensor* WAN::WanT2VCrossAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int64_t)':
sd/wan.hpp:1451:21: warning: unused variable 'N' [-Wunused-variable]
 1451 |             int64_t N       = x->ne[2];
      |                     ^
sd/wan.hpp:1452:21: warning: unused variable 'n_token' [-Wunused-variable]
 1452 |             int64_t n_token = x->ne[1];
      |                     ^~~~~~~
sd/wan.hpp: In member function 'virtual ggml_tensor* WAN::WanI2VCrossAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int64_t)':
sd/wan.hpp:1512:21: warning: unused variable 'n_token' [-Wunused-variable]
 1512 |             int64_t n_token         = x->ne[1];
      |                     ^~~~~~~
sd/wan.hpp: In member function 'ggml_tensor* WAN::Wan::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, float, int64_t)':
sd/wan.hpp:2064:21: warning: unused variable 'C' [-Wunused-variable]
 2064 |             int64_t C = x->ne[3];
      |                     ^
In file included from sd/stable-diffusion.cpp:32:
sd/mmdit.hpp: In constructor 'MMDiT::MMDiT(const String2TensorStorage&)':
sd/mmdit.hpp:682:17: warning: unused variable 'default_out_channels' [-Wunused-variable]
  682 |         int64_t default_out_channels = in_channels;
      |                 ^~~~~~~~~~~~~~~~~~~~
sd/mmdit.hpp: In static member function 'static void MMDiTRunner::load_from_file_and_test(const std::string&)':
sd/mmdit.hpp:951:19: warning: unused variable 'model_data_type' [-Wunused-variable]
  951 |         ggml_type model_data_type          = GGML_TYPE_F16;
      |                   ^~~~~~~~~~~~~~~
In file included from sd/stable-diffusion.cpp:33:
sd/pid.hpp: In member function 'ggml_cgraph* Pid::PiDRunner::build_graph(const sd::Tensor<float>&, const sd::Tensor<float>&, const sd::Tensor<float>&, const sd::Tensor<float>&, const sd::Tensor<float>&)':
sd/pid.hpp:739:21: warning: unused variable 'B' [-Wunused-variable]
  739 |             int64_t B  = x->ne[3];
      |                     ^
In file included from sd/stable-diffusion.cpp:34:
sd/pmid.hpp: In member function 'ggml_tensor* PerceiverAttention::reshape_tensor(ggml_context*, ggml_tensor*, int)':
sd/pmid.hpp:87:17: warning: variable 'ne' set but not used [-Wunused-but-set-variable]
   87 |         int64_t ne[4];
      |                 ^~
sd/pmid.hpp: In member function 'ggml_cgraph* PhotoMakerIDEncoder::build_graph(const sd::Tensor<float>&, const sd::Tensor<float>&, std::vector<bool>&, const sd::Tensor<float>&)':
sd/pmid.hpp:474:27: warning: comparison of integer expressions of different signedness: 'int' and 'std::vector<bool>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  474 |         for (int i = 0; i < class_tokens_mask.size(); i++) {
      |                         ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from sd/sample-cache.h:6,
                 from sd/stable-diffusion.cpp:36:
sd/cache_dit.hpp: In member function 'void TaylorSeerState::update_derivatives(const float*, size_t, int)':
sd/cache_dit.hpp:78:13: warning: unused variable 'order' [-Wunused-variable]
   78 |         int order = n_derivatives + 1;
      |             ^~~~~
sd/common_block.hpp: At global scope:
sd/common_block.hpp:470:26: warning: 'virtual ggml_tensor* SpatialTransformer::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*)' was hidden [-Woverloaded-virtual=]
  470 |     virtual ggml_tensor* forward(GGMLRunnerContext* ctx,
      |                          ^~~~~~~
In file included from sd/stable-diffusion.cpp:38:
sd/unet.hpp:64:18: note:   by 'ggml_tensor* SpatialVideoTransformer::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int)'
   64 |     ggml_tensor* forward(GGMLRunnerContext* ctx,
      |                  ^~~~~~~
sd/unet.hpp: In constructor 'SpatialVideoTransformer::SpatialVideoTransformer(int64_t, int64_t, int64_t, int64_t, int64_t, bool, int64_t, int)':
sd/unet.hpp:44:17: warning: unused variable 'time_mix_inner_dim' [-Wunused-variable]
   44 |         int64_t time_mix_inner_dim = time_mix_d_head * n_time_mix_heads;  // equal to inner_dim
      |                 ^~~~~~~~~~~~~~~~~~
sd/unet.hpp: In constructor 'UnetModelBlock::UnetModelBlock(SDVersion, const String2TensorStorage&)':
sd/unet.hpp:277:27: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  277 |         for (int i = 0; i < len_mults; i++) {
      |                         ~~^~~~~~~~~~~
sd/unet.hpp:310:19: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  310 |             if (i != len_mults - 1) {
      |                 ~~^~~~~~~~~~~~~~~~
sd/unet.hpp: In member function 'ggml_tensor* UnetModelBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int, std::vector<ggml_tensor*>, float)':
sd/unet.hpp:501:27: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  501 |         for (int i = 0; i < len_mults; i++) {
      |                         ~~^~~~~~~~~~~
sd/unet.hpp:517:19: warning: comparison of integer expressions of different signedness: 'int' and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  517 |             if (i != len_mults - 1) {
      |                 ~~^~~~~~~~~~~~~~~~
sd/unet.hpp:502:17: warning: unused variable 'mult' [-Wunused-variable]
  502 |             int mult = channel_mult[i];
      |                 ^~~~
In file included from sd/stable-diffusion.cpp:42:
sd/z_image.hpp: In member function 'ggml_tensor* ZImage::JointAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)':
sd/z_image.hpp:47:21: warning: unused variable 'n_token' [-Wunused-variable]
   47 |             int64_t n_token = x->ne[1];
      |                     ^~~~~~~
sd/z_image.hpp:48:21: warning: unused variable 'N' [-Wunused-variable]
   48 |             int64_t N       = x->ne[2];
      |                     ^
sd/z_image.hpp: In member function 'ggml_tensor* ZImage::ZImageModel::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<ggml_tensor*>)':
sd/z_image.hpp:442:21: warning: unused variable 'C' [-Wunused-variable]
  442 |             int64_t C = x->ne[2];
      |                     ^
sd/z_image.hpp:443:21: warning: unused variable 'N' [-Wunused-variable]
  443 |             int64_t N = x->ne[3];
      |                     ^
sd/stable-diffusion.cpp: In member function 'void StableDiffusionGGML::preview_image(int, const sd::Tensor<float>&, SDVersion, preview_t, std::function<void(int, int, sd_image_t*, bool, void*)>, void*, bool)':
sd/stable-diffusion.cpp:1792:34: warning: comparison of integer expressions of different signedness: 'int' and 'uint32_t' {aka 'unsigned int'} [-Wsign-compare]
 1792 |         auto _latents = channels != dim ? is_video ? sd::ops::slice(latents, 3, 0, channels)
      |                         ~~~~~~~~~^~~~~~
sd/ggml_extend.hpp: In instantiation of 'std::optional<sd::Tensor<T> > GGMLRunner::compute_with_graph_cuts(ggml_cgraph*, const GraphCutPlan&, int, bool, bool) [with T = float; GraphCutPlan = sd::ggml_graph_cut::Plan]':
sd/ggml_extend.hpp:3250:50:   required from 'std::optional<sd::Tensor<T> > GGMLRunner::compute(get_graph_cb_t, int, bool, bool) [with T = float; get_graph_cb_t = std::function<ggml_cgraph*()>]'
 3250 |                 return compute_with_graph_cuts<T>(gf,
      |                        ~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
 3251 |                                                   plan,
      |                                                   ~~~~~
 3252 |                                                   n_threads,
      |                                                   ~~~~~~~~~~
 3253 |                                                   free_compute_buffer_immediately,
      |                                                   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 3254 |                                                   no_return);
      |                                                   ~~~~~~~~~~
sd/flux.hpp:1486:85:   required from here
 1486 |             auto result = restore_trailing_singleton_dims(GGMLRunner::compute<float>(get_graph, n_threads, false), x.dim());
      |                                                           ~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/ggml_extend.hpp:2814:21: warning: unused variable 't_segment_begin' [-Wunused-variable]
 2814 |             int64_t t_segment_begin = ggml_time_ms();
      |                     ^~~~~~~~~~~~~~~
sd/stable-diffusion.cpp: At global scope:
sd/stable-diffusion.cpp:5476:29: warning: 'ggml_context* sd_lowlevel_work_ctx(size_t)' defined but not used [-Wunused-function]
 5476 | static struct ggml_context* sd_lowlevel_work_ctx(size_t mem_size) {
      |                             ^~~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:5427:28: warning: 'ggml_tensor* sd_tensor_to_ggml(ggml_context*, const sd_tensor*)' defined but not used [-Wunused-function]
 5427 | static struct ggml_tensor* sd_tensor_to_ggml(struct ggml_context* ctx, const sd_tensor* t) {
      |                            ^~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:5405:18: warning: 'sd_tensor ggml_to_sd_tensor(const ggml_tensor*)' defined but not used [-Wunused-function]
 5405 | static sd_tensor ggml_to_sd_tensor(const struct ggml_tensor* g) {
      |                  ^~~~~~~~~~~~~~~~~
sd/wan.hpp: In member function 'ggml_tensor* WAN::WanVAE::decode(GGMLRunnerContext*, ggml_tensor*, int64_t)':
sd/wan.hpp:1094:33: warning: 'out' may be used uninitialized [-Wmaybe-uninitialized]
 1094 |                 out = unpatchify(ctx->ggml_ctx, out, 2, b);
      |                       ~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:1081:26: note: 'out' was declared here
 1081 |             ggml_tensor* out;
      |                          ^~~
sd/wan.hpp: In member function 'ggml_tensor* WAN::WanVAE::encode(GGMLRunnerContext*, ggml_tensor*, int64_t)':
sd/wan.hpp:1060:37: warning: 'out' may be used uninitialized [-Wmaybe-uninitialized]
 1060 |             out     = conv1->forward(ctx, out);
      |                       ~~~~~~~~~~~~~~^~~~~~~~~~
sd/wan.hpp:1048:26: note: 'out' was declared here
 1048 |             ggml_tensor* out;
      |                          ^~~
sd/wan.hpp: In constructor 'WAN::Encoder3d::Encoder3d(int64_t, int64_t, std::vector<int>, int, std::vector<bool>, bool)':
sd/wan.hpp:654:95: warning: 'out_dim' may be used uninitialized [-Wmaybe-uninitialized]
  654 |             blocks["middle.0"] = std::shared_ptr<GGMLBlock>(new ResidualBlock(out_dim, out_dim));
      |                                                                                               ^
sd/wan.hpp:626:21: note: 'out_dim' was declared here
  626 |             int64_t out_dim;
      |                     ^~~~~~~
In constructor 'WAN::RMS_norm::RMS_norm(int64_t)',
    inlined from 'WAN::Decoder3d::Decoder3d(int64_t, int64_t, std::vector<int>, int, std::vector<bool>, bool)' at sd/wan.hpp:827:79:
sd/wan.hpp:102:15: warning: 'out_dim' may be used uninitialized [-Wmaybe-uninitialized]
  102 |             : dim(dim) {}
      |               ^~~~~~~~
sd/wan.hpp: In constructor 'WAN::Decoder3d::Decoder3d(int64_t, int64_t, std::vector<int>, int, std::vector<bool>, bool)':
sd/wan.hpp:795:21: note: 'out_dim' was declared here
  795 |             int64_t out_dim;
      |                     ^~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/model.cpp -o sd/model.o
sd/model.cpp: In function 'uint16_t f8_e4m3_to_f16(uint8_t)':
sd/model.cpp:119:31: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]
  119 |     return ggml_fp32_to_fp16(*reinterpret_cast<const float*>(&result));
      |                               ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-backend.h:3,
                 from sd/model.h:11,
                 from sd/model.cpp:16:
sd/model.cpp: In lambda function:
sd/model.cpp:1063:65: warning: comparison of integer expressions of different signedness: 'size_t' {aka 'long long unsigned int'} and 'int64_t' {aka 'long long int'} [-Wsign-compare]
 1063 |                             GGML_ASSERT(ggml_nbytes(dst_tensor) == tensor_storage.nbytes());
      |                                         ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml.h:285:30: note: in definition of macro 'GGML_ASSERT'
  285 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
      |                              ^
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/model_io/gguf_io.cpp -o sd/model_io/gguf_io.o
sd/model_io/gguf_io.cpp: In function 'bool read_gguf_file(const std::string&, std::vector<TensorStorage>&, std::string*)':
sd/model_io/gguf_io.cpp:81:32: warning: comparison of integer expressions of different signedness: 'size_t' {aka 'long long unsigned int'} and 'int64_t' {aka 'long long int'} [-Wsign-compare]
   81 |         if (ggml_nbytes(dummy) != tensor_storage.nbytes()) {
      |             ~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/model_io/pickle_io.cpp -o sd/model_io/pickle_io.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/model_io/safetensors_io.cpp -o sd/model_io/safetensors_io.o
sd/model_io/safetensors_io.cpp: In function 'bool read_safetensors_file(const std::string&, std::vector<TensorStorage>&, std::string*)':
sd/model_io/safetensors_io.cpp:197:55: warning: comparison of integer expressions of different signedness: 'int64_t' {aka 'long long int'} and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  197 |             tensor_size_ok = (tensor_storage.nbytes() == tensor_data_size * 2);
      |                               ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
sd/model_io/safetensors_io.cpp:201:55: warning: comparison of integer expressions of different signedness: 'int64_t' {aka 'long long int'} and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  201 |             tensor_size_ok = (tensor_storage.nbytes() == tensor_data_size * 2);
      |                               ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
sd/model_io/safetensors_io.cpp:205:59: warning: comparison of integer expressions of different signedness: 'int64_t' {aka 'long long int'} and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  205 |             tensor_size_ok = (tensor_storage.nbytes() * 2 == tensor_data_size);
      |                               ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
sd/model_io/safetensors_io.cpp:209:59: warning: comparison of integer expressions of different signedness: 'int64_t' {aka 'long long int'} and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  209 |             tensor_size_ok = (tensor_storage.nbytes() * 2 == tensor_data_size);
      |                               ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
sd/model_io/safetensors_io.cpp:211:55: warning: comparison of integer expressions of different signedness: 'int64_t' {aka 'long long int'} and 'size_t' {aka 'long long unsigned int'} [-Wsign-compare]
  211 |             tensor_size_ok = (tensor_storage.nbytes() == tensor_data_size);
      |                               ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/model_io/torch_legacy_io.cpp -o sd/model_io/torch_legacy_io.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/model_io/torch_zip_io.cpp -o sd/model_io/torch_zip_io.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/name_conversion.cpp -o sd/name_conversion.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/bpe_tokenizer.cpp -o sd/tokenizers/bpe_tokenizer.o
sd/tokenizers/bpe_tokenizer.cpp: In member function 'virtual std::vector<int> BPETokenizer::encode(const std::string&, on_new_token_cb_t)':
sd/tokenizers/bpe_tokenizer.cpp:165:35: warning: comparison of integer expressions of different signedness: 'int' and 'std::__cxx11::basic_string<char>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  165 |                 for (int i = 0; i < token_str.length(); i++) {
      |                                 ~~^~~~~~~~~~~~~~~~~~~~
sd/tokenizers/bpe_tokenizer.cpp:181:43: warning: comparison of integer expressions of different signedness: 'int' and 'std::__cxx11::basic_string<char>::size_type' {aka 'long long unsigned int'} [-Wsign-compare]
  181 |                         for (int i = 0; i < utf8_token_str.length(); i++) {
      |                                         ~~^~~~~~~~~~~~~~~~~~~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/clip_tokenizer.cpp -o sd/tokenizers/clip_tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/gemma_tokenizer.cpp -o sd/tokenizers/gemma_tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/gpt_oss_tokenizer.cpp -o sd/tokenizers/gpt_oss_tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/mistral_tokenizer.cpp -o sd/tokenizers/mistral_tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/qwen2_tokenizer.cpp -o sd/tokenizers/qwen2_tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/t5_unigram_tokenizer.cpp -o sd/tokenizers/t5_unigram_tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/tokenizer.cpp -o sd/tokenizers/tokenizer.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/tokenize_util.cpp -o sd/tokenizers/tokenize_util.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/tokenizers/vocab/vocab.cpp -o sd/tokenizers/vocab/vocab.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/upscaler.cpp -o sd/upscaler.o
In file included from sd/esrgan.hpp:4,
                 from sd/upscaler.h:4,
                 from sd/upscaler.cpp:1:
sd/ggml_extend.hpp: In instantiation of 'std::optional<sd::Tensor<T> > GGMLRunner::compute_with_graph_cuts(ggml_cgraph*, const GraphCutPlan&, int, bool, bool) [with T = float; GraphCutPlan = sd::ggml_graph_cut::Plan]':
sd/ggml_extend.hpp:3250:50:   required from 'std::optional<sd::Tensor<T> > GGMLRunner::compute(get_graph_cb_t, int, bool, bool) [with T = float; get_graph_cb_t = std::function<ggml_cgraph*()>]'
 3250 |                 return compute_with_graph_cuts<T>(gf,
      |                        ~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~
 3251 |                                                   plan,
      |                                                   ~~~~~
 3252 |                                                   n_threads,
      |                                                   ~~~~~~~~~~
 3253 |                                                   free_compute_buffer_immediately,
      |                                                   ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 3254 |                                                   no_return);
      |                                                   ~~~~~~~~~~
sd/esrgan.hpp:370:84:   required from here
  370 |         auto result    = restore_trailing_singleton_dims(GGMLRunner::compute<float>(get_graph, n_threads, false), x.dim());
      |                                                          ~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/ggml_extend.hpp:2814:21: warning: unused variable 't_segment_begin' [-Wunused-variable]
 2814 |             int64_t t_segment_begin = ggml_time_ms();
      |                     ^~~~~~~~~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/util.cpp -o sd/util.o
sd/util.cpp:84:9: warning: "NOMINMAX" redefined
   84 | #define NOMINMAX
      |         ^~~~~~~~
In file included from D:/rtools45/x86_64-w64-mingw32.static.posix/lib/gcc/x86_64-w64-mingw32.static.posix/14.3.0/include/c++/x86_64-w64-mingw32.static.posix/bits/c++config.h:683,
                 from D:/rtools45/x86_64-w64-mingw32.static.posix/lib/gcc/x86_64-w64-mingw32.static.posix/14.3.0/include/c++/cstdlib:41,
                 from D:/rtools45/x86_64-w64-mingw32.static.posix/lib/gcc/x86_64-w64-mingw32.static.posix/14.3.0/include/c++/stdlib.h:36,
                 from D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/r_ggml_compat.h:43,
                 from <command-line>:
D:/rtools45/x86_64-w64-mingw32.static.posix/lib/gcc/x86_64-w64-mingw32.static.posix/14.3.0/include/c++/x86_64-w64-mingw32.static.posix/bits/os_defines.h:45:9: note: this is the location of the previous definition
   45 | #define NOMINMAX 1
      |         ^~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -DSDCPP_BUILD_VERSION=master_2026_06_03 -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/version.cpp -o sd/version.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/ggml_extend_backend.cpp -o sd/ggml_extend_backend.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/ggml_graph_cut.cpp -o sd/ggml_graph_cut.o
sd/ggml_graph_cut.cpp: In function 'sd::ggml_graph_cut::Plan sd::ggml_graph_cut::resolve_plan(ggml_backend_t, ggml_cgraph*, PlanCache*, size_t, const std::unordered_set<const ggml_tensor*>&, const char*)':
sd/ggml_graph_cut.cpp:721:17: warning: unused variable 't_prepare_begin' [-Wunused-variable]
  721 |         int64_t t_prepare_begin = ggml_time_ms();
      |                 ^~~~~~~~~~~~~~~
In file included from sd/ggml_graph_cut.cpp:16:
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h: At global scope:
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:300:15: warning: 'size_t ggml_hash_find_or_insert(ggml_hash_set*, ggml_tensor*)' defined but not used [-Wunused-function]
  300 | static size_t ggml_hash_find_or_insert(struct ggml_hash_set * hash_set, struct ggml_tensor * key) {
      |               ^~~~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:279:15: warning: 'size_t ggml_hash_insert(ggml_hash_set*, ggml_tensor*)' defined but not used [-Wunused-function]
  279 | static size_t ggml_hash_insert(struct ggml_hash_set * hash_set, struct ggml_tensor * key) {
      |               ^~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:274:13: warning: 'bool ggml_hash_contains(const ggml_hash_set*, ggml_tensor*)' defined but not used [-Wunused-function]
  274 | static bool ggml_hash_contains(const struct ggml_hash_set * hash_set, struct ggml_tensor * key) {
      |             ^~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:205:15: warning: 'size_t ggml_bitset_size(size_t)' defined but not used [-Wunused-function]
  205 | static size_t ggml_bitset_size(size_t n) {
      |               ^~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:168:13: warning: 'void ggml_set_op_params_f32(ggml_tensor*, uint32_t, float)' defined but not used [-Wunused-function]
  168 | static void ggml_set_op_params_f32(struct ggml_tensor * tensor, uint32_t i, float value) {
      |             ^~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:163:13: warning: 'void ggml_set_op_params_i32(ggml_tensor*, uint32_t, int32_t)' defined but not used [-Wunused-function]
  163 | static void ggml_set_op_params_i32(struct ggml_tensor * tensor, uint32_t i, int32_t value) {
      |             ^~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:158:14: warning: 'float ggml_get_op_params_f32(const ggml_tensor*, uint32_t)' defined but not used [-Wunused-function]
  158 | static float ggml_get_op_params_f32(const struct ggml_tensor * tensor, uint32_t i) {
      |              ^~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:153:16: warning: 'int32_t ggml_get_op_params_i32(const ggml_tensor*, uint32_t)' defined but not used [-Wunused-function]
  153 | static int32_t ggml_get_op_params_i32(const struct ggml_tensor * tensor, uint32_t i) {
      |                ^~~~~~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:147:13: warning: 'void ggml_set_op_params(ggml_tensor*, const void*, size_t)' defined but not used [-Wunused-function]
  147 | static void ggml_set_op_params(struct ggml_tensor * tensor, const void * params, size_t params_size) {
      |             ^~~~~~~~~~~~~~~~~~
D:/RCompile/CRANpkg/lib/4.6/ggmlR/include/ggml-impl.h:90:13: warning: 'bool ggml_op_is_empty(ggml_op)' defined but not used [-Wunused-function]
   90 | static bool ggml_op_is_empty(enum ggml_op op) {
      |             ^~~~~~~~~~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/layer_registry.cpp -o sd/layer_registry.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/guidance.cpp -o sd/guidance.o
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -include r_ggml_compat.h -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/sample-cache.cpp -o sd/sample-cache.o
In file included from sd/sample-cache.h:6,
                 from sd/sample-cache.cpp:1:
sd/cache_dit.hpp: In member function 'void TaylorSeerState::update_derivatives(const float*, size_t, int)':
sd/cache_dit.hpp:78:13: warning: unused variable 'order' [-Wunused-variable]
   78 |         int order = n_derivatives + 1;
      |             ^~~~~
In file included from sd/sample-cache.h:7:
sd/denoiser.hpp: In function 'sd::Tensor<float> sample_tcd(denoise_cb_t, sd::Tensor<float>, const std::vector<float>&, std::shared_ptr<RNG>, float)':
sd/denoiser.hpp:1726:15: warning: unused variable 'alpha_prod_t' [-Wunused-variable]
 1726 |         float alpha_prod_t      = 1.0f / (sigma * sigma + 1.0f);
      |               ^~~~~~~~~~~~
g++  -std=gnu++17 -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"   -fopenmp   -O2 -Wall  -mfpmath=sse -msse2 -mstackrealign    -c sd/convert.cpp -o sd/convert.o
gcc  -I"D:/RCompile/recent/R-4.6.0/include" -DNDEBUG -Isd -Isd/thirdparty -Isd/tokenizers -Isd/tokenizers/vocab -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP  -DSD2R_HAVE_META_BACKEND=1 -D__USE_MINGW_ANSI_STDIO=1 -I'D:/RCompile/CRANpkg/lib/4.6/Rcpp/include' -I'D:/RCompile/CRANpkg/lib/4.6/ggmlR/include'   -I"d:/rtools45/x86_64-w64-mingw32.static.posix/include"      -pedantic -Wstrict-prototypes -O2 -Wall -std=gnu2x  -mfpmath=sse -msse2 -mstackrealign   -c sd/thirdparty/zip.c -o sd/thirdparty/zip.o
g++ -shared -s -static-libgcc -o sd2R.dll tmp.def RcppExports.o sd2R_interface.o sd/stable-diffusion.o sd/model.o sd/model_io/gguf_io.o sd/model_io/pickle_io.o sd/model_io/safetensors_io.o sd/model_io/torch_legacy_io.o sd/model_io/torch_zip_io.o sd/name_conversion.o sd/tokenizers/bpe_tokenizer.o sd/tokenizers/clip_tokenizer.o sd/tokenizers/gemma_tokenizer.o sd/tokenizers/gpt_oss_tokenizer.o sd/tokenizers/mistral_tokenizer.o sd/tokenizers/qwen2_tokenizer.o sd/tokenizers/t5_unigram_tokenizer.o sd/tokenizers/tokenizer.o sd/tokenizers/tokenize_util.o sd/tokenizers/vocab/vocab.o sd/upscaler.o sd/util.o sd/version.o sd/ggml_extend_backend.o sd/ggml_graph_cut.o sd/layer_registry.o sd/guidance.o sd/sample-cache.o sd/convert.o sd/thirdparty/zip.o -LD:/RCompile/CRANpkg/lib/4.6/ggmlR/lib -lggml -fopenmp -lpthread -lm -static-libstdc++ -static-libgcc -Ld:/rtools45/x86_64-w64-mingw32.static.posix/lib/x64 -Ld:/rtools45/x86_64-w64-mingw32.static.posix/lib -LD:/RCompile/recent/R-4.6.0/bin/x64 -lR
make[1]: Leaving directory '/d/temp/2026_06_20_01_50_00_3901/RtmpSYiQ2A/R.INSTALL71386a833739/sd2R/src'
make[1]: Entering directory '/d/temp/2026_06_20_01_50_00_3901/RtmpSYiQ2A/R.INSTALL71386a833739/sd2R/src'
make[1]: Leaving directory '/d/temp/2026_06_20_01_50_00_3901/RtmpSYiQ2A/R.INSTALL71386a833739/sd2R/src'
installing to d:/Rcompile/CRANpkg/lib/4.6/00LOCK-sd2R/00new/sd2R/libs/x64
** R
** inst
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* MD5 sums
packaged installation of 'sd2R' as sd2R_0.2.1.zip
* DONE (sd2R)