* installing *source* package ‘sd2R’ ...
** this is package ‘sd2R’ version ‘0.1.7’
** package ‘sd2R’ successfully unpacked and MD5 sums checked
** using staged installation
* Downloading vocab.hpp...
* vocab.hpp: downloaded successfully
* Downloading vocab_mistral.hpp...
* vocab_mistral.hpp: downloaded successfully
* Downloading vocab_qwen.hpp...
* vocab_qwen.hpp: downloaded successfully
* Downloading vocab_umt5.hpp...
* vocab_umt5.hpp: downloaded successfully
* ggmlR lib: /data/gannet/ripley/R/test-dev/ggmlR/lib
* ggmlR include: /data/gannet/ripley/R/test-dev/ggmlR/include
* OpenMP: enabled (C: -fopenmp, C++: -fopenmp)
* Vulkan: enabled (detected via ggmlR and pkg-config)
* Wrote src/Makevars
* Patching sd/ sources for R compatibility...
patched: rng_mt19937.hpp
patched: ucache.hpp
patched: util.h
patched: lora.hpp
patched: tokenize_util.h
patched: version.cpp
patched: z_image.hpp
patched: easycache.hpp
patched: qwen_image.hpp
patched: denoiser.hpp
patched: diffusion_model.hpp
patched: rng.hpp
patched: vocab_mistral.hpp
patched: pmid.hpp
patched: wan.hpp
patched: mmdit.hpp
patched: vae.hpp
patched: vocab_qwen.hpp
patched: gguf_reader.hpp
patched: tae.hpp
patched: rng_philox.hpp
patched: stable-diffusion.cpp
patched: common.hpp
patched: name_conversion.cpp
patched: stable-diffusion.h
patched: preprocessing.hpp
patched: name_conversion.h
patched: latent-preview.h
patched: model.h
patched: esrgan.hpp
patched: vocab.hpp
patched: control.hpp
patched: ordered_map.hpp
patched: upscaler.cpp
patched: vocab_umt5.hpp
patched: ltxv.hpp
patched: unet.hpp
patched: conditioner.hpp
patched: flux.hpp
patched: clip.hpp
patched: tokenize_util.cpp
patched: cache_dit.hpp
patched: rope.hpp
patched: util.cpp (unused variable size2)
patched: util.cpp (sign-compare)
patched: ggml_extend.hpp (sign-compare N)
patched: ggml_extend.hpp (unused variable param)
patched: model.h (unused function)
patched: thirdparty/json.hpp.inc (literal operator spacing)
patched: util.h (VA_OPT)
* Done.
** libs
specified C++17
using C++ compiler: ‘g++ (GCC) 15.2.1 20260123 (Red Hat 15.2.1-7)’
using C++17
make[1]: Entering directory '/data/gannet/ripley/R/packages/tests-devel/sd2R/src'
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c RcppExports.cpp -o RcppExports.o
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd2R_interface.cpp -o sd2R_interface.o
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/stable-diffusion.cpp -o sd/stable-diffusion.o
In file included from /data/gannet/ripley/R/test-dev/ggmlR/include/ggml-alloc.h:3,
from sd/ggml_extend.hpp:27,
from sd/stable-diffusion.cpp:1:
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:204:13: warning: ‘GGML_ATTRIBUTE_FORMAT’ redefined
204 | # define GGML_ATTRIBUTE_FORMAT(...) __attribute__((format(printf, __VA_ARGS__)))
| ^~~~~~~~~~~~~~~~~~~~~
In file included from <command-line>:
/data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h:53:9: note: this is the location of the previous definition
53 | #define GGML_ATTRIBUTE_FORMAT(...)
| ^~~~~~~~~~~~~~~~~~~~~
sd/ggml_extend.hpp: In function ‘void ggml_ext_tensor_iter(ggml_tensor*, const std::function<void(ggml_tensor*, long int)>&)’:
sd/ggml_extend.hpp:294:13: warning: unused variable ‘n0’ [-Wunused-variable]
294 | int64_t n0 = tensor->ne[0];
| ^~
sd/ggml_extend.hpp:295:13: warning: unused variable ‘n1’ [-Wunused-variable]
295 | int64_t n1 = tensor->ne[1];
| ^~
sd/ggml_extend.hpp:296:13: warning: unused variable ‘n2’ [-Wunused-variable]
296 | int64_t n2 = tensor->ne[2];
| ^~
sd/ggml_extend.hpp:297:13: warning: unused variable ‘n3’ [-Wunused-variable]
297 | int64_t n3 = tensor->ne[3];
| ^~
sd/ggml_extend.hpp: In function ‘ggml_tensor* load_tensor_from_file(ggml_context*, const std::string&)’:
sd/ggml_extend.hpp:348:18: warning: unused variable ‘bpe’ [-Wunused-variable]
348 | const size_t bpe = ggml_type_size(ggml_type(ttype));
| ^~~
sd/ggml_extend.hpp: In function ‘ggml_tensor* ggml_ext_conv_3d(ggml_context*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int64_t, int, int, int, int, int, int, int, int, int)’:
sd/ggml_extend.hpp:1154:13: warning: unused variable ‘OC’ [-Wunused-variable]
1154 | int64_t OC = w->ne[3] / IC;
| ^~
sd/ggml_extend.hpp:1155:13: warning: unused variable ‘N’ [-Wunused-variable]
1155 | int64_t N = x->ne[3] / IC;
| ^
In file included from sd/stable-diffusion.cpp:10:
sd/cache_dit.hpp: In member function ‘void TaylorSeerState::update_derivatives(const float*, size_t, int)’:
sd/cache_dit.hpp:76:13: warning: unused variable ‘order’ [-Wunused-variable]
76 | int order = n_derivatives + 1;
| ^~~~~
In file included from sd/conditioner.hpp:4,
from sd/stable-diffusion.cpp:11:
sd/clip.hpp: In static member function ‘static std::set<std::pair<std::__cxx11::basic_string<char32_t>, std::__cxx11::basic_string<char32_t> > > CLIPTokenizer::get_pairs(const std::vector<std::__cxx11::basic_string<char32_t> >&)’:
sd/clip.hpp:89:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
89 | for (int i = 1; i < subwords.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/clip.hpp: In member function ‘std::u32string CLIPTokenizer::bpe(const std::u32string&)’:
sd/clip.hpp:200:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::__cxx11::basic_string<char32_t>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
200 | for (int i = 0; i < token.size() - 1; i++) {
| ~~^~~~~~~~~~~~~~~~~~
sd/clip.hpp:235:22: warning: comparison of integer expressions of different signedness: ‘int32_t’ {aka ‘int’} and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
235 | while (i < word.size()) {
| ~~^~~~~~~~~~~~~
sd/clip.hpp:262:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
262 | for (int i = 0; i < word.size(); i++) {
| ~~^~~~~~~~~~~~~
sd/clip.hpp:264:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
264 | if (i != word.size() - 1) {
| ~~^~~~~~~~~~~~~~~~~~
sd/clip.hpp: In member function ‘void CLIPTokenizer::pad_tokens(std::vector<int>&, std::vector<float>&, size_t, bool)’:
sd/clip.hpp:310:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
310 | for (int i = 1; i < length; i++) {
| ~~^~~~~~~~
sd/clip.hpp:311:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
311 | if (token_idx >= tokens.size()) {
| ~~~~~~~~~~^~~~~~~~~~~~~~~~
sd/clip.hpp: In member function ‘std::vector<int> CLIPTokenizer::encode(std::string, on_new_token_cb_t)’:
sd/clip.hpp:424:35: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::__cxx11::basic_string<char>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
424 | for (int i = 0; i < token_str.length(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~
sd/clip.hpp: In member function ‘ggml_tensor* CLIPTextModelRunner::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, size_t, bool, int)’:
sd/clip.hpp:949:16: warning: unused variable ‘N’ [-Wunused-variable]
949 | size_t N = input_ids->ne[1];
| ^
sd/clip.hpp:950:16: warning: unused variable ‘n_token’ [-Wunused-variable]
950 | size_t n_token = input_ids->ne[0];
| ^~~~~~~
In file included from sd/llm.hpp:23,
from sd/conditioner.hpp:5:
sd/rope.hpp: In function ‘std::vector<std::vector<float> > Rope::rope(const std::vector<float>&, int, int, const std::vector<int>&)’:
sd/rope.hpp:61:34: warning: comparison of integer expressions of different signedness: ‘size_t’ {aka ‘long unsigned int’} and ‘int’ [-Wsign-compare]
61 | for (size_t j = 0; j < half_dim; ++j) {
| ~~^~~~~~~~~~
sd/rope.hpp:84:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
84 | 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:144:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::vector<float> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
144 | 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:158:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
158 | for (int j = 0; j < a_len; ++j) {
| ~~^~~~~~~
sd/rope.hpp:161:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
161 | 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, int, const std::vector<int>&, const std::vector<std::vector<int> >&)’:
sd/rope.hpp:188:41: warning: comparison of integer expressions of different signedness: ‘size_t’ {aka ‘long unsigned int’} and ‘int’ [-Wsign-compare]
188 | if (!wrap_dims.empty() && i < (int)wrap_dims.size()) {
| ~~^~~~~~~~~~~~~~~~~~~~~~~
sd/rope.hpp:194:35: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
194 | for (int j = 0; j < pos_len; ++j) {
| ~~^~~~~~~~~
sd/rope.hpp:195:39: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<float>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
195 | for (int k = 0; k < rope_emb[0].size(); ++k) {
| ~~^~~~~~~~~~~~~~~~~~~~
sd/rope.hpp: In function ‘std::vector<std::vector<float> > Rope::gen_qwen_image_ids(int, int, int, int, int, const std::vector<ggml_tensor*>&, bool)’:
sd/rope.hpp:351:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<float>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
351 | for (int j = 0; j < txt_ids.size(); ++j) {
| ~~^~~~~~~~~~~~~~~~
sd/rope.hpp: In function ‘std::vector<std::vector<float> > Rope::gen_vid_ids(int, int, int, int, int, int, int, int, int, int)’:
sd/rope.hpp:460:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::vector<float> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
460 | for (int j = 0; j < vid_ids.size(); ++j) {
| ~~^~~~~~~~~~~~~~~~
sd/llm.hpp: In static member function ‘static std::set<std::pair<std::__cxx11::basic_string<char32_t>, std::__cxx11::basic_string<char32_t> > > LLM::BPETokenizer::get_pairs(const std::vector<std::__cxx11::basic_string<char32_t> >&)’:
sd/llm.hpp:79:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
79 | for (int i = 1; i < subwords.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/llm.hpp: In member function ‘std::u32string LLM::BPETokenizer::bpe(const std::u32string&)’:
sd/llm.hpp:103:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::__cxx11::basic_string<char32_t>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
103 | for (int i = 0; i < token.size(); i++) {
| ~~^~~~~~~~~~~~~~
sd/llm.hpp:137:26: warning: comparison of integer expressions of different signedness: ‘int32_t’ {aka ‘int’} and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
137 | while (i < word.size()) {
| ~~^~~~~~~~~~~~~
sd/llm.hpp:164:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
164 | for (int i = 0; i < word.size(); i++) {
| ~~^~~~~~~~~~~~~
sd/llm.hpp:166:23: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::__cxx11::basic_string<char32_t> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
166 | if (i != word.size() - 1) {
| ~~^~~~~~~~~~~~~~~~~~
sd/llm.hpp: In member function ‘std::vector<int> LLM::BPETokenizer::encode(std::string, on_new_token_cb_t)’:
sd/llm.hpp:236:39: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::__cxx11::basic_string<char>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
236 | for (int i = 0; i < token_str.length(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~
sd/llm.hpp: In member function ‘ggml_tensor* LLM::VisionAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)’:
sd/llm.hpp:678:21: warning: unused variable ‘n_token’ [-Wunused-variable]
678 | int64_t n_token = x->ne[1];
| ^~~~~~~
sd/llm.hpp:679:21: warning: unused variable ‘N’ [-Wunused-variable]
679 | int64_t N = x->ne[2];
| ^
sd/llm.hpp: In member function ‘ggml_tensor* LLM::TextModel::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<std::pair<int, ggml_tensor*> >, std::set<int>)’:
sd/llm.hpp:966:35: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::pair<int, ggml_tensor*> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
966 | for (int i = 0; i < image_embeds.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~~
sd/llm.hpp:1007:35: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
1007 | for (int i = 1; i < intermediate_outputs.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/llm.hpp: In member function ‘int64_t LLM::LLMRunner::get_num_image_tokens(int64_t, int64_t, int64_t)’:
sd/llm.hpp:1261:21: warning: unused variable ‘llm_grid_h’ [-Wunused-variable]
1261 | int64_t llm_grid_h = grid_h / params.vision.spatial_merge_size;
| ^~~~~~~~~~
sd/llm.hpp:1262:21: warning: unused variable ‘llm_grid_w’ [-Wunused-variable]
1262 | int64_t llm_grid_w = grid_w / params.vision.spatial_merge_size;
| ^~~~~~~~~~
sd/llm.hpp: In member function ‘ggml_cgraph* LLM::LLMRunner::build_encode_image_graph(ggml_tensor*)’:
sd/llm.hpp:1358:47: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
1358 | for (int seq_index = 0; seq_index < seqlens.size(); seq_index++) {
| ~~~~~~~~~~^~~~~~~~~~~~~~~~
sd/llm.hpp:1368:65: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<float>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
1368 | GGML_ASSERT((i * (grid_h * grid_w) + j) < window_mask_vec.size());
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/llm.hpp:1299:17: warning: unused variable ‘grid_t’ [-Wunused-variable]
1299 | int grid_t = 1;
| ^~~~~~
sd/llm.hpp:1355:17: warning: unused variable ‘seq_window_size’ [-Wunused-variable]
1355 | int seq_window_size = (vit_merger_window_size * params.vision.spatial_merge_size) * (vit_merger_window_size * params.vision.spatial_merge_size);
| ^~~~~~~~~~~~~~~
sd/llm.hpp: In member function ‘void LLM::LLMEmbedder::test()’:
sd/llm.hpp:1547:37: warning: unused variable ‘weights’ [-Wunused-variable]
1547 | std::vector<float>& weights = std::get<1>(tokens_and_weights);
| ^~~~~~~
sd/llm.hpp:1595:37: warning: unused variable ‘weights’ [-Wunused-variable]
1595 | std::vector<float>& weights = std::get<1>(tokens_and_weights);
| ^~~~~~~
sd/llm.hpp:1626:37: warning: unused variable ‘weights’ [-Wunused-variable]
1626 | std::vector<float>& weights = std::get<1>(tokens_and_weights);
| ^~~~~~~
sd/llm.hpp:1657:37: warning: unused variable ‘weights’ [-Wunused-variable]
1657 | std::vector<float>& weights = std::get<1>(tokens_and_weights);
| ^~~~~~~
In file included from sd/conditioner.hpp:6:
sd/t5.hpp: In member function ‘EncodeResult T5UniGramTokenizer::EncodeOptimized(const std::string&) const’:
sd/t5.hpp:289:28: warning: comparison of integer expressions of different signedness: ‘std::size_t’ {aka ‘long unsigned int’} and ‘const int’ [-Wsign-compare]
289 | while (key_pos < size) {
| ~~~~~~~~^~~~~~
sd/t5.hpp:312:52: warning: comparison of integer expressions of different signedness: ‘const long unsigned int’ and ‘const int’ [-Wsign-compare]
312 | if (!has_single_node && length == mblen) {
| ~~~~~~~^~~~~~~~
sd/t5.hpp: In constructor ‘T5UniGramTokenizer::T5UniGramTokenizer(bool)’:
sd/t5.hpp:356:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<std::pair<std::__cxx11::basic_string<char>, float> >::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
356 | for (int i = 0; i < piece_score_pairs.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
sd/t5.hpp: In member function ‘void T5UniGramTokenizer::pad_tokens(std::vector<int>&, std::vector<float>&, std::vector<float>*, size_t, bool)’:
sd/t5.hpp:410:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
410 | for (int i = 0; i < length; i++) {
| ~~^~~~~~~~
sd/t5.hpp:411:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
411 | if (token_idx >= orig_token_num) {
| ~~~~~~~~~~^~~~~~~~~~~~~~~~~
sd/t5.hpp: In member function ‘ggml_tensor* T5Runner::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)’:
sd/t5.hpp:789:16: warning: unused variable ‘N’ [-Wunused-variable]
789 | size_t N = input_ids->ne[1];
| ^
sd/t5.hpp:790:16: warning: unused variable ‘n_token’ [-Wunused-variable]
790 | size_t n_token = input_ids->ne[0];
| ^~~~~~~
sd/t5.hpp: In member function ‘void T5Embedder::test()’:
sd/t5.hpp:980:33: warning: unused variable ‘weights’ [-Wunused-variable]
980 | 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, size_t, bool)’:
sd/conditioner.hpp:310:58: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
310 | for (int i = class_token_index[0] + 1; i < clean_input_ids.size(); i++)
| ~~^~~~~~~~~~~~~~~~~~~~~~~~
sd/conditioner.hpp:325:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
325 | for (int i = 0; i < tokens.size(); i++) {
| ~~^~~~~~~~~~~~~~~
sd/conditioner.hpp: In member function ‘SDCondition FrozenCLIPEmbedderWithCustomWords::get_learned_condition_common(ggml_context*, int, std::vector<int>&, std::vector<float>&, int, int, int, int, bool)’:
sd/conditioner.hpp:439:43: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
439 | for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
| ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function ‘SDCondition SD3CLIPEmbedder::get_learned_condition_common(ggml_context*, int, std::vector<std::pair<std::vector<int>, std::vector<float> > >, int, bool)’:
sd/conditioner.hpp:900:43: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
900 | for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
| ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp: In member function ‘SDCondition FluxCLIPEmbedder::get_learned_condition_common(ggml_context*, int, std::vector<std::pair<std::vector<int>, std::vector<float> > >, int, bool)’:
sd/conditioner.hpp:1282:43: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1282 | 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, bool)’:
sd/conditioner.hpp:1470:14: warning: variable ‘on_new_token_cb’ set but not used [-Wunused-but-set-variable]
1470 | 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(ggml_context*, 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:1533:43: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1533 | for (int chunk_idx = 0; chunk_idx < chunk_count; chunk_idx++) {
| ~~~~~~~~~~^~~~~~~~~~~~~
sd/conditioner.hpp:1526:29: warning: unused variable ‘pooled’ [-Wunused-variable]
1526 | struct ggml_tensor* pooled = nullptr;
| ^~~~~~
sd/conditioner.hpp: In member function ‘virtual SDCondition LLMEmbedder::get_learned_condition(ggml_context*, int, const ConditionerParams&)’:
sd/conditioner.hpp:1730:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<sd_image_t*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
1730 | for (int i = 0; i < conditioner_params.ref_images.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/conditioner.hpp:1823:31: warning: comparison of integer expressions of different signedness: ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} and ‘int’ [-Wsign-compare]
1823 | if (tokens.size() < max_length) {
| ~~~~~~~~~~~~~~^~~~~~~~~~~~
In file included from sd/control.hpp:4,
from sd/stable-diffusion.cpp:12:
sd/common.hpp: In member function ‘ggml_tensor* CrossAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*)’:
sd/common.hpp:311:17: warning: unused variable ‘n’ [-Wunused-variable]
311 | int64_t n = x->ne[2];
| ^
sd/common.hpp:312:17: warning: unused variable ‘n_token’ [-Wunused-variable]
312 | int64_t n_token = x->ne[1];
| ^~~~~~~
sd/common.hpp:313:17: warning: unused variable ‘n_context’ [-Wunused-variable]
313 | int64_t n_context = context->ne[1];
| ^~~~~~~~~
sd/common.hpp:314:17: warning: unused variable ‘inner_dim’ [-Wunused-variable]
314 | int64_t inner_dim = d_head * n_head;
| ^~~~~~~~~
sd/common.hpp: At global scope:
sd/common.hpp:124:33: warning: ‘virtual ggml_tensor* ResBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*)’ was hidden [-Woverloaded-virtual=]
124 | virtual struct ggml_tensor* forward(GGMLRunnerContext* ctx, struct ggml_tensor* x, struct ggml_tensor* emb = nullptr) {
| ^~~~~~~
sd/common.hpp:558:25: note: by ‘ggml_tensor* VideoResBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int)’
558 | struct ggml_tensor* forward(GGMLRunnerContext* ctx,
| ^~~~~~~
sd/control.hpp: In constructor ‘ControlNetBlock::ControlNetBlock(SDVersion)’:
sd/control.hpp:112:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
112 | for (int i = 0; i < len_mults; i++) {
| ~~^~~~~~~~~~~
sd/control.hpp:137:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
137 | 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:268:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
268 | for (int i = 0; i < len_mults; i++) {
| ~~^~~~~~~~~~~
sd/control.hpp:283:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
283 | if (i != len_mults - 1) {
| ~~^~~~~~~~~~~~~~~~
sd/control.hpp:269:17: warning: unused variable ‘mult’ [-Wunused-variable]
269 | int mult = channel_mult[i];
| ^~~~
sd/control.hpp: In member function ‘void ControlNet::alloc_control_ctx(std::vector<ggml_tensor*>)’:
sd/control.hpp:346:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
346 | for (int i = 0; i < outs.size() - 1; i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/control.hpp: In member function ‘ggml_cgraph* ControlNet::build_graph(ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*)’:
sd/control.hpp:410:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
410 | for (int i = 0; i < outs.size() - 1; i++) {
| ~~^~~~~~~~~~~~~~~~~
In file included from sd/stable-diffusion.cpp:13:
sd/denoiser.hpp: In function ‘bool sample_k_diffusion(sample_method_t, denoise_cb_t, ggml_context*, ggml_tensor*, std::vector<float>, std::shared_ptr<RNG>, float)’:
sd/denoiser.hpp:726:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
726 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:782:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
782 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:818:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
818 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:849:28: warning: unused variable ‘vec_d2’ [-Wunused-variable]
849 | float* vec_d2 = (float*)d->data;
| ^~~~~~
sd/denoiser.hpp:874:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
874 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:932:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
932 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1004:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1004 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1046:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1046 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1092:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1092 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1153:41: warning: comparison of integer expressions of different signedness: ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} and ‘int’ [-Wsign-compare]
1153 | if (buffer_model.size() == max_order - 1) {
| ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~
sd/denoiser.hpp:1170:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1170 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1206:31: warning: unused variable ‘h_n_2’ [-Wunused-variable]
1206 | float h_n_2 = (i > 1) ? (sigmas[i - 1] - sigmas[i - 2]) : h_n_1;
| ^~~~~
sd/denoiser.hpp:1217:31: warning: unused variable ‘h_n_3’ [-Wunused-variable]
1217 | float h_n_3 = (i > 2) ? (sigmas[i - 2] - sigmas[i - 3]) : h_n_2;
| ^~~~~
sd/denoiser.hpp:1229:41: warning: comparison of integer expressions of different signedness: ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} and ‘int’ [-Wsign-compare]
1229 | if (buffer_model.size() == max_order - 1) {
| ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~
sd/denoiser.hpp:1243:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1243 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1241:33: warning: unused variable ‘d’ [-Wunused-variable]
1241 | struct ggml_tensor* d = ggml_dup_tensor(work_ctx, x);
| ^
sd/denoiser.hpp:1313:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1313 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1500:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1500 | for (int i = 0; i < steps; i++) {
| ~~^~~~~~~
sd/denoiser.hpp:1506:39: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1506 | int prev_timestep = i >= steps - 1 ? 0 : TIMESTEPS - 1 - (TIMESTEPS / original_steps) * (int)floor((i + 1) * ((float)original_steps / steps));
| ~~^~~~~~~~~~~~
sd/denoiser.hpp:1616:34: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
1616 | if (eta > 0 && i != steps - 1) {
| ~~^~~~~~~~~~~~
In file included from sd/diffusion_model.hpp:4,
from sd/stable-diffusion.cpp:14:
sd/flux.hpp: In member function ‘std::vector<Flux::ModulationOut> Flux::DoubleStreamBlock::get_distil_img_mod(GGMLRunnerContext*, ggml_tensor*)’:
sd/flux.hpp:298:23: warning: unused variable ‘double_blocks_count’ [-Wunused-variable]
298 | const int double_blocks_count = 19;
| ^~~~~~~~~~~~~~~~~~~
sd/flux.hpp: In member function ‘ggml_tensor* Flux::Flux::forward_orig(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<int>)’:
sd/flux.hpp:933:25: warning: unused variable ‘mod_index_length’ [-Wunused-variable]
933 | int64_t mod_index_length = 344;
| ^~~~~~~~~~~~~~~~
sd/flux.hpp: In member function ‘ggml_cgraph* Flux::FluxRunner::build_graph(ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<ggml_tensor*>, bool, std::vector<int>)’:
sd/flux.hpp:1458:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
1458 | for (int i = 0; i < ref_latents.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~
In file included from sd/diffusion_model.hpp:5:
sd/mmdit.hpp: In constructor ‘MMDiT::MMDiT(const String2TensorStorage&)’:
sd/mmdit.hpp:681:17: warning: unused variable ‘default_out_channels’ [-Wunused-variable]
681 | 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:947:19: warning: unused variable ‘model_data_type’ [-Wunused-variable]
947 | ggml_type model_data_type = GGML_TYPE_F16;
| ^~~~~~~~~~~~~~~
In file included from sd/diffusion_model.hpp:6:
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:512:21: warning: unused variable ‘C’ [-Wunused-variable]
512 | int64_t C = x->ne[2];
| ^
sd/qwen_image.hpp:513:21: warning: unused variable ‘N’ [-Wunused-variable]
513 | int64_t N = x->ne[3];
| ^
sd/qwen_image.hpp: In member function ‘ggml_cgraph* Qwen::QwenImageRunner::build_graph(ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<ggml_tensor*>, bool)’:
sd/qwen_image.hpp:611:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
611 | for (int i = 0; i < ref_latents.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~
sd/common.hpp: At global scope:
sd/common.hpp:459:33: warning: ‘virtual ggml_tensor* SpatialTransformer::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*)’ was hidden [-Woverloaded-virtual=]
459 | virtual struct ggml_tensor* forward(GGMLRunnerContext* ctx,
| ^~~~~~~
In file included from sd/diffusion_model.hpp:7:
sd/unet.hpp:64:25: note: by ‘ggml_tensor* SpatialVideoTransformer::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int)’
64 | struct 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:273:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
273 | for (int i = 0; i < len_mults; i++) {
| ~~^~~~~~~~~~~
sd/unet.hpp:306:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
306 | 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:495:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
495 | for (int i = 0; i < len_mults; i++) {
| ~~^~~~~~~~~~~
sd/unet.hpp:510:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
510 | if (i != len_mults - 1) {
| ~~^~~~~~~~~~~~~~~~
sd/unet.hpp:496:17: warning: unused variable ‘mult’ [-Wunused-variable]
496 | int mult = channel_mult[i];
| ^~~~
sd/unet.hpp: In member function ‘ggml_cgraph* UNetModelRunner::build_graph(ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int, std::vector<ggml_tensor*>, float)’:
sd/unet.hpp:633:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
633 | for (int i = 0; i < controls.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~
In file included from sd/wan.hpp:12,
from sd/diffusion_model.hpp:8:
sd/vae.hpp: In constructor ‘Encoder::Encoder(int, std::vector<int>, int, int, int, bool, bool)’:
sd/vae.hpp:295:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
295 | for (int i = 0; i < num_resolutions; i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/vae.hpp:307:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
307 | if (i != num_resolutions - 1) {
| ~~^~~~~~~~~~~~~~~~~~~~~~
sd/vae.hpp: In member function ‘virtual ggml_tensor* Encoder::forward(GGMLRunnerContext*, ggml_tensor*)’:
sd/vae.hpp:335:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
335 | for (int i = 0; i < num_resolutions; i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/vae.hpp:342:19: warning: comparison of integer expressions of different signedness: ‘int’ and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
342 | if (i != num_resolutions - 1) {
| ~~^~~~~~~~~~~~~~~~~~~~~~
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 int>::size_type’ {aka ‘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 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 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 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:703:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<long int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
703 | for (int i = 0; i < dims.size() - 1; i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:715:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
715 | 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:793:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<long int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
793 | for (int i = 0; i < dims.size() - 1; i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:797:40: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<bool>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
797 | bool t_up_flag = i < temperal_upsample.size() ? temperal_upsample[i] : false;
| ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:802:88: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
802 | i != dim_mult.size() - 1));
| ~~^~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp:815:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
815 | 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:879:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<long int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
879 | for (int i = 0; i < dims.size() - 1; i++) {
| ~~^~~~~~~~~~~~~~~~~
sd/wan.hpp:891:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<int>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
891 | 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:1264:23: warning: unused variable ‘model_data_type’ [-Wunused-variable]
1264 | ggml_type model_data_type = GGML_TYPE_F16;
| ^~~~~~~~~~~~~~~
sd/wan.hpp: At global scope:
sd/wan.hpp:1318:37: warning: ‘virtual ggml_tensor* WAN::WanSelfAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)’ was hidden [-Woverloaded-virtual=]
1318 | virtual struct ggml_tensor* forward(GGMLRunnerContext* ctx,
| ^~~~~~~
sd/wan.hpp:1359:37: note: by ‘virtual ggml_tensor* WAN::WanCrossAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, int64_t)’
1359 | virtual struct 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:1380:21: warning: unused variable ‘N’ [-Wunused-variable]
1380 | int64_t N = x->ne[2];
| ^
sd/wan.hpp:1381:21: warning: unused variable ‘n_token’ [-Wunused-variable]
1381 | 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:1441:21: warning: unused variable ‘n_token’ [-Wunused-variable]
1441 | int64_t n_token = x->ne[1];
| ^~~~~~~
sd/wan.hpp: At global scope:
sd/wan.hpp:1534:37: warning: ‘virtual ggml_tensor* WAN::WanAttentionBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int64_t)’ was hidden [-Woverloaded-virtual=]
1534 | virtual struct ggml_tensor* forward(GGMLRunnerContext* ctx,
| ^~~~~~~
sd/wan.hpp:1609:47: note: by ‘std::pair<ggml_tensor*, ggml_tensor*> WAN::VaceWanAttentionBlock::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int64_t)’
1609 | std::pair<ggml_tensor*, ggml_tensor*> forward(GGMLRunnerContext* ctx,
| ^~~~~~~
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:1982:21: warning: unused variable ‘C’ [-Wunused-variable]
1982 | int64_t C = x->ne[3];
| ^
In file included from sd/diffusion_model.hpp:9:
sd/z_image.hpp: In member function ‘ggml_tensor* ZImage::JointAttention::forward(GGMLRunnerContext*, ggml_tensor*, ggml_tensor*, ggml_tensor*)’:
sd/z_image.hpp:50:21: warning: unused variable ‘n_token’ [-Wunused-variable]
50 | int64_t n_token = x->ne[1];
| ^~~~~~~
sd/z_image.hpp:51:21: warning: unused variable ‘N’ [-Wunused-variable]
51 | 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:495:21: warning: unused variable ‘C’ [-Wunused-variable]
495 | int64_t C = x->ne[2];
| ^
sd/z_image.hpp:496:21: warning: unused variable ‘N’ [-Wunused-variable]
496 | int64_t N = x->ne[3];
| ^
sd/z_image.hpp: In member function ‘ggml_cgraph* ZImage::ZImageRunner::build_graph(ggml_tensor*, ggml_tensor*, ggml_tensor*, std::vector<ggml_tensor*>, bool)’:
sd/z_image.hpp:564:31: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<ggml_tensor*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
564 | for (int i = 0; i < ref_latents.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~
In file included from sd/stable-diffusion.cpp:18:
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(ggml_tensor*, ggml_tensor*, std::vector<bool>&, ggml_tensor*)’:
sd/pmid.hpp:475:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<bool>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
475 | for (int i = 0; i < class_tokens_mask.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from sd/stable-diffusion.cpp:23:
sd/latent-preview.h: In function ‘void preview_latent_video(uint8_t*, ggml_tensor*, const float (*)[3], const float*, int)’:
sd/latent-preview.h:167:12: warning: unused variable ‘buffer_head’ [-Wunused-variable]
167 | size_t buffer_head = 0;
| ^~~~~~~~~~~
sd/stable-diffusion.cpp: In member function ‘bool StableDiffusionGGML::init(const sd_ctx_params_t*)’:
sd/stable-diffusion.cpp:597:22: warning: unused variable ‘is_chroma’ [-Wunused-variable]
597 | bool is_chroma = false;
| ^~~~~~~~~
sd/stable-diffusion.cpp: In member function ‘SDCondition StableDiffusionGGML::get_svd_condition(ggml_context*, sd_image_t, int, int, int, int, float, bool)’:
sd/stable-diffusion.cpp:1492:27: warning: comparison of integer expressions of different signedness: ‘int’ and ‘uint32_t’ {aka ‘unsigned int’} [-Wsign-compare]
1492 | if (width != init_image.width || height != init_image.height) {
| ~~~~~~^~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:1492:57: warning: comparison of integer expressions of different signedness: ‘int’ and ‘uint32_t’ {aka ‘unsigned int’} [-Wsign-compare]
1492 | if (width != init_image.width || height != init_image.height) {
| ~~~~~~~^~~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp: In function ‘sd_image_t* generate_image(sd_ctx_t*, const sd_img_gen_params_t*)’:
sd/stable-diffusion.cpp:3804:26: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<float>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
3804 | if (sample_steps != sigmas.size() - 1) {
| ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:3826:19: warning: comparison of integer expressions of different signedness: ‘size_t’ {aka ‘long unsigned int’} and ‘int’ [-Wsign-compare]
3826 | if (t_enc == sample_steps)
| ~~~~~~^~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:3991:23: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<sd_image_t*>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
3991 | for (int i = 0; i < ref_images.size(); i++) {
| ~~^~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp: In function ‘sd_image_t* generate_video(sd_ctx_t*, const sd_vid_gen_params_t*, int*)’:
sd/stable-diffusion.cpp:4125:25: warning: comparison of integer expressions of different signedness: ‘int’ and ‘std::vector<float>::size_type’ {aka ‘long unsigned int’} [-Wsign-compare]
4125 | if (total_steps != sigmas.size() - 1) {
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/model.cpp -o sd/model.o
In file included from sd/gguf_reader.hpp:9,
from sd/model.cpp:18:
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:204:13: warning: ‘GGML_ATTRIBUTE_FORMAT’ redefined
204 | # define GGML_ATTRIBUTE_FORMAT(...) __attribute__((format(printf, __VA_ARGS__)))
| ^~~~~~~~~~~~~~~~~~~~~
In file included from <command-line>:
/data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h:53:9: note: this is the location of the previous definition
53 | #define GGML_ATTRIBUTE_FORMAT(...)
| ^~~~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp: At global scope:
sd/stable-diffusion.cpp:175:13: warning: ‘void accumulate_tile(const ggml_tensor*, float*, float*, const float*, int, int, int, int, int, int)’ defined but not used [-Wunused-function]
175 | static void accumulate_tile(const ggml_tensor* tile_denoised,
| ^~~~~~~~~~~~~~~
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/name_conversion.cpp -o sd/name_conversion.o
In file included from /data/gannet/ripley/R/test-dev/ggmlR/include/ggml-backend.h:3,
from sd/model.h:14,
from sd/name_conversion.h:6,
from sd/name_conversion.cpp:4:
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:204:13: warning: ‘GGML_ATTRIBUTE_FORMAT’ redefined
204 | # define GGML_ATTRIBUTE_FORMAT(...) __attribute__((format(printf, __VA_ARGS__)))
| ^~~~~~~~~~~~~~~~~~~~~
In file included from <command-line>:
/data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h:53:9: note: this is the location of the previous definition
53 | #define GGML_ATTRIBUTE_FORMAT(...)
| ^~~~~~~~~~~~~~~~~~~~~
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/tokenize_util.cpp -o sd/tokenize_util.o
sd/model.cpp: In function ‘uint16_t f8_e4m3_to_f16(uint8_t)’:
sd/model.cpp:168:31: warning: dereferencing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing]
168 | return ggml_fp32_to_fp16(*reinterpret_cast<const float*>(&result));
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/model.cpp: In member function ‘bool ModelLoader::init_from_gguf_file(const std::string&, const std::string&)’:
sd/model.cpp:475:40: warning: comparison of integer expressions of different signedness: ‘size_t’ {aka ‘long unsigned int’} and ‘int64_t’ {aka ‘long int’} [-Wsign-compare]
475 | GGML_ASSERT(ggml_nbytes(dummy) == tensor_storage.nbytes());
| ~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/model.cpp:460:12: warning: unused variable ‘total_size’ [-Wunused-variable]
460 | size_t total_size = 0;
| ^~~~~~~~~~
sd/model.cpp: In member function ‘bool ModelLoader::init_from_safetensors_file(const std::string&, const std::string&)’:
sd/model.cpp:624:49: warning: comparison of integer expressions of different signedness: ‘int64_t’ {aka ‘long int’} and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
624 | GGML_ASSERT(tensor_storage.nbytes() == tensor_data_size * 2);
| ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/model.cpp:628:49: warning: comparison of integer expressions of different signedness: ‘int64_t’ {aka ‘long int’} and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
628 | GGML_ASSERT(tensor_storage.nbytes() == tensor_data_size * 2);
| ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/model.cpp:632:53: warning: comparison of integer expressions of different signedness: ‘int64_t’ {aka ‘long int’} and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
632 | GGML_ASSERT(tensor_storage.nbytes() * 2 == tensor_data_size);
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/model.cpp:636:53: warning: comparison of integer expressions of different signedness: ‘int64_t’ {aka ‘long int’} and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
636 | GGML_ASSERT(tensor_storage.nbytes() * 2 == tensor_data_size);
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/model.cpp:638:49: warning: comparison of integer expressions of different signedness: ‘int64_t’ {aka ‘long int’} and ‘size_t’ {aka ‘long unsigned int’} [-Wsign-compare]
638 | GGML_ASSERT(tensor_storage.nbytes() == tensor_data_size);
| ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
sd/model.cpp: In lambda function:
sd/model.cpp:1533:65: warning: comparison of integer expressions of different signedness: ‘size_t’ {aka ‘long unsigned int’} and ‘int64_t’ {aka ‘long int’} [-Wsign-compare]
1533 | GGML_ASSERT(ggml_nbytes(dst_tensor) == tensor_storage.nbytes());
| ~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:283:30: note: in definition of macro ‘GGML_ASSERT’
283 | #define GGML_ASSERT(x) if (!(x)) GGML_ABORT("GGML_ASSERT(%s) failed", #x)
| ^
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/upscaler.cpp -o sd/upscaler.o
In file included from /data/gannet/ripley/R/test-dev/ggmlR/include/ggml-alloc.h:3,
from sd/ggml_extend.hpp:27,
from sd/esrgan.hpp:4,
from sd/upscaler.cpp:1:
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:204:13: warning: ‘GGML_ATTRIBUTE_FORMAT’ redefined
204 | # define GGML_ATTRIBUTE_FORMAT(...) __attribute__((format(printf, __VA_ARGS__)))
| ^~~~~~~~~~~~~~~~~~~~~
In file included from <command-line>:
/data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h:53:9: note: this is the location of the previous definition
53 | #define GGML_ATTRIBUTE_FORMAT(...)
| ^~~~~~~~~~~~~~~~~~~~~
sd/ggml_extend.hpp: In function ‘void ggml_ext_tensor_iter(ggml_tensor*, const std::function<void(ggml_tensor*, long int)>&)’:
sd/ggml_extend.hpp:294:13: warning: unused variable ‘n0’ [-Wunused-variable]
294 | int64_t n0 = tensor->ne[0];
| ^~
sd/ggml_extend.hpp:295:13: warning: unused variable ‘n1’ [-Wunused-variable]
295 | int64_t n1 = tensor->ne[1];
| ^~
sd/ggml_extend.hpp:296:13: warning: unused variable ‘n2’ [-Wunused-variable]
296 | int64_t n2 = tensor->ne[2];
| ^~
sd/ggml_extend.hpp:297:13: warning: unused variable ‘n3’ [-Wunused-variable]
297 | int64_t n3 = tensor->ne[3];
| ^~
sd/ggml_extend.hpp: In function ‘ggml_tensor* load_tensor_from_file(ggml_context*, const std::string&)’:
sd/ggml_extend.hpp:348:18: warning: unused variable ‘bpe’ [-Wunused-variable]
348 | const size_t bpe = ggml_type_size(ggml_type(ttype));
| ^~~
sd/ggml_extend.hpp: In function ‘ggml_tensor* ggml_ext_conv_3d(ggml_context*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int64_t, int, int, int, int, int, int, int, int, int)’:
sd/ggml_extend.hpp:1154:13: warning: unused variable ‘OC’ [-Wunused-variable]
1154 | int64_t OC = w->ne[3] / IC;
| ^~
sd/ggml_extend.hpp:1155:13: warning: unused variable ‘N’ [-Wunused-variable]
1155 | int64_t N = x->ne[3] / IC;
| ^
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/util.cpp -o sd/util.o
In file included from /data/gannet/ripley/R/test-dev/ggmlR/include/ggml-alloc.h:3,
from sd/ggml_extend.hpp:27,
from sd/preprocessing.hpp:4,
from sd/util.cpp:16:
/data/gannet/ripley/R/test-dev/ggmlR/include/ggml.h:204:13: warning: ‘GGML_ATTRIBUTE_FORMAT’ redefined
204 | # define GGML_ATTRIBUTE_FORMAT(...) __attribute__((format(printf, __VA_ARGS__)))
| ^~~~~~~~~~~~~~~~~~~~~
In file included from <command-line>:
/data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h:53:9: note: this is the location of the previous definition
53 | #define GGML_ATTRIBUTE_FORMAT(...)
| ^~~~~~~~~~~~~~~~~~~~~
g++ -std=gnu++17 -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -include /data/gannet/ripley/R/test-dev/ggmlR/include/r_ggml_compat.h -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wno-ignored-attributes -Wno-parentheses -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -c sd/version.cpp -o sd/version.o
gcc -I"/data/gannet/ripley/R/R-devel/include" -DNDEBUG -Isd -Isd/thirdparty -DGGML_USE_CPU -D_GNU_SOURCE -DGGML_MAX_NAME=128 -DGGML_R_PACKAGE -DGGML_USE_OPENMP -DSD_USE_VULKAN -I'/data/gannet/ripley/R/test-dev/Rcpp/include' -I'/data/gannet/ripley/R/test-dev/ggmlR/include' -I/usr/local/include -fopenmp -fpic -g -O2 -Wall -pedantic -mtune=native -Wp,-D_FORTIFY_SOURCE=3 -fexceptions -fstack-protector-strong -fstack-clash-protection -fcf-protection -Werror=implicit-function-declaration -Wstrict-prototypes -c sd/thirdparty/zip.c -o sd/thirdparty/zip.o
sd/ggml_extend.hpp: In function ‘void ggml_ext_tensor_iter(ggml_tensor*, const std::function<void(ggml_tensor*, long int)>&)’:
sd/ggml_extend.hpp:294:13: warning: unused variable ‘n0’ [-Wunused-variable]
294 | int64_t n0 = tensor->ne[0];
| ^~
sd/ggml_extend.hpp:295:13: warning: unused variable ‘n1’ [-Wunused-variable]
295 | int64_t n1 = tensor->ne[1];
| ^~
sd/ggml_extend.hpp:296:13: warning: unused variable ‘n2’ [-Wunused-variable]
296 | int64_t n2 = tensor->ne[2];
| ^~
sd/ggml_extend.hpp:297:13: warning: unused variable ‘n3’ [-Wunused-variable]
297 | int64_t n3 = tensor->ne[3];
| ^~
sd/ggml_extend.hpp: In function ‘ggml_tensor* load_tensor_from_file(ggml_context*, const std::string&)’:
sd/ggml_extend.hpp:348:18: warning: unused variable ‘bpe’ [-Wunused-variable]
348 | const size_t bpe = ggml_type_size(ggml_type(ttype));
| ^~~
sd/ggml_extend.hpp: In function ‘ggml_tensor* ggml_ext_conv_3d(ggml_context*, ggml_tensor*, ggml_tensor*, ggml_tensor*, int64_t, int, int, int, int, int, int, int, int, int)’:
sd/ggml_extend.hpp:1154:13: warning: unused variable ‘OC’ [-Wunused-variable]
1154 | int64_t OC = w->ne[3] / IC;
| ^~
sd/ggml_extend.hpp:1155:13: warning: unused variable ‘N’ [-Wunused-variable]
1155 | int64_t N = x->ne[3] / IC;
| ^
sd/stable-diffusion.cpp: In member function ‘void StableDiffusionGGML::process_latent_in(ggml_tensor*)’:
sd/stable-diffusion.cpp:2669:50: warning: ‘mean’ may be used uninitialized [-Wmaybe-uninitialized]
2669 | value = (value - mean) * scale_factor / std_;
| ~~~~~~~^~~~~~~
sd/stable-diffusion.cpp:2654:19: note: ‘mean’ was declared here
2654 | float mean;
| ^~~~
sd/stable-diffusion.cpp:2669:41: warning: ‘std_’ may be used uninitialized [-Wmaybe-uninitialized]
2669 | value = (value - mean) * scale_factor / std_;
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:2655:19: note: ‘std_’ was declared here
2655 | float std_;
| ^~~~
sd/stable-diffusion.cpp: In member function ‘void StableDiffusionGGML::process_latent_out(ggml_tensor*)’:
sd/stable-diffusion.cpp:2708:41: warning: ‘mean’ may be used uninitialized [-Wmaybe-uninitialized]
2708 | value = value * std_ / scale_factor + mean;
| ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/stable-diffusion.cpp:2693:19: note: ‘mean’ was declared here
2693 | float mean;
| ^~~~
sd/stable-diffusion.cpp:2708:49: warning: ‘std_’ may be used uninitialized [-Wmaybe-uninitialized]
2708 | value = value * std_ / scale_factor + mean;
| ~~~~~~^~~~~~
sd/stable-diffusion.cpp:2694:19: note: ‘std_’ was declared here
2694 | float std_;
| ^~~~
sd/wan.hpp: In member function ‘ggml_tensor* WAN::WanVAE::encode(GGMLRunnerContext*, ggml_tensor*, int64_t)’:
sd/wan.hpp:1053:37: warning: ‘out’ may be used uninitialized [-Wmaybe-uninitialized]
1053 | out = conv1->forward(ctx, out);
| ~~~~~~~~~~~~~~^~~~~~~~~~
sd/wan.hpp:1041:33: note: ‘out’ was declared here
1041 | struct ggml_tensor* out;
| ^~~
In member function ‘ggml_tensor* WAN::WanVAE::unpatchify(ggml_context*, ggml_tensor*, int64_t, int64_t)’,
inlined from ‘ggml_tensor* WAN::WanVAE::decode(GGMLRunnerContext*, ggml_tensor*, int64_t)’ at sd/wan.hpp:1085:33:
sd/wan.hpp:1013:32: warning: ‘out’ may be used uninitialized [-Wmaybe-uninitialized]
1013 | x = ggml_reshape_4d(ctx, x, w * h, f, q * r, c * b); // [b*c, r*q, f, h*w]
| ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sd/wan.hpp: In member function ‘ggml_tensor* WAN::WanVAE::decode(GGMLRunnerContext*, ggml_tensor*, int64_t)’:
sd/wan.hpp:1072:33: note: ‘out’ was declared here
1072 | struct 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:824:79:
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:792:21: note: ‘out_dim’ was declared here
792 | int64_t out_dim;
| ^~~~~~~
g++ -std=gnu++17 -shared -L/usr/local/lib64 -o sd2R.so RcppExports.o sd2R_interface.o sd/stable-diffusion.o sd/model.o sd/name_conversion.o sd/tokenize_util.o sd/upscaler.o sd/util.o sd/version.o sd/thirdparty/zip.o -L/data/gannet/ripley/R/test-dev/ggmlR/lib -lggml -fopenmp -lvulkan -lpthread -lm
make[1]: Leaving directory '/data/gannet/ripley/R/packages/tests-devel/sd2R/src'
make[1]: Entering directory '/data/gannet/ripley/R/packages/tests-devel/sd2R/src'
make[1]: Leaving directory '/data/gannet/ripley/R/packages/tests-devel/sd2R/src'
installing to /data/gannet/ripley/R/packages/tests-devel/sd2R.Rcheck/00LOCK-sd2R/00new/sd2R/libs
** 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
Error: package or namespace load failed for ‘sd2R’ in dyn.load(file, DLLpath = DLLpath, ...):
unable to load shared object '/data/gannet/ripley/R/packages/tests-devel/sd2R.Rcheck/00LOCK-sd2R/00new/sd2R/libs/sd2R.so':
/data/gannet/ripley/R/packages/tests-devel/sd2R.Rcheck/00LOCK-sd2R/00new/sd2R/libs/sd2R.so: undefined symbol: ggml_backend_vk_init
Error: loading failed
Execution halted
ERROR: loading failed
* removing ‘/data/gannet/ripley/R/packages/tests-devel/sd2R.Rcheck/sd2R’