Fix llama-android.cpp for error - "common/common.h not found" (#8145)

- Path seems to be wrong for the common.h header file in llama-android.cpp file. Fixing the path so the Android Build doesn't fail with the error "There is no file common/common.h"

.github/workflows/build.yml

@@ -10,10 +10,10 @@ on:
   push:
     branches:
       - master
-    paths: ['.github/workflows/**', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp', '**/*.cu', '**/*.swift', '**/*.m']
+    paths: ['.github/workflows/build.yml', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp', '**/*.cu', '**/*.cuh', '**/*.swift', '**/*.m', '**/*.metal']
   pull_request:
     types: [opened, synchronize, reopened]
-    paths: ['.github/workflows/build.yml', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp', '**/*.cu', '**/*.cuh', '**/*.swift', '**/*.m']
+    paths: ['.github/workflows/build.yml', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp', '**/*.cu', '**/*.cuh', '**/*.swift', '**/*.m', '**/*.metal']
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref && github.ref || github.run_id }}

.github/workflows/docker.yml

@@ -14,6 +14,7 @@ on:
   push:
     branches:
       - master
+    paths: ['.github/workflows/docker.yml', '.devops/*.Dockerfile', '**/CMakeLists.txt', '**/Makefile', '**/*.h', '**/*.hpp', '**/*.c', '**/*.cpp', '**/*.cu', '**/*.cuh', '**/*.swift', '**/*.m', '**/*.metal']
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref && github.ref || github.run_id }}

CMakeLists.txt

@@ -80,6 +80,7 @@ set(GGML_SANITIZE_UNDEFINED ${LLAMA_SANITIZE_UNDEFINED})
 set(GGML_ALL_WARNINGS       ${LLAMA_ALL_WARNINGS})
 set(GGML_FATAL_WARNINGS     ${LLAMA_FATAL_WARNINGS})
 set(GGML_LLAMAFILE          ON)
+set(GGML_CUDA_USE_GRAPHS    ON)
 
 # transition helpers
 function (llama_option_depr TYPE OLD NEW)

examples/llama.android/llama/src/main/cpp/llama-android.cpp

@@ -5,7 +5,7 @@
 #include <string>
 #include <unistd.h>
 #include "llama.h"
-#include "common/common.h"
+#include "common.h"
 
 // Write C++ code here.
 //

examples/llava/clip.cpp

@@ -1121,20 +1121,20 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
             }
             if (n < 32)
                 hparams.image_grid_pinpoints[n] = 0;
-        } catch (std::runtime_error & e) {
+        } catch (std::runtime_error & /*e*/) {
             hparams.image_grid_pinpoints[0]=0;
         }
 
         try {
             int idx = get_key_idx(ctx, KEY_MM_PATCH_MERGE_TYPE);
             strcpy(hparams.mm_patch_merge_type, gguf_get_val_str(ctx, idx));
-        } catch (std::runtime_error & e) {
+        } catch (std::runtime_error & /*e*/) {
             strcpy(hparams.mm_patch_merge_type, "flat");
         }
 
         try {
             hparams.image_crop_resolution = get_u32(ctx, KEY_IMAGE_CROP_RESOLUTION); // llava-1.6
-        } catch(const std::exception& e) {
+        } catch(const std::exception& /*e*/) {
             hparams.image_crop_resolution = hparams.image_size;
         }
 
@@ -1173,7 +1173,7 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
         try {
             vision_model.class_embedding  = get_tensor(new_clip->ctx_data, TN_CLASS_EMBD);
             new_clip->has_class_embedding = true;
-        } catch (const std::exception& e) {
+        } catch (const std::exception& /*e*/) {
             new_clip->has_class_embedding = false;
         }
 
@@ -1181,7 +1181,7 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
             vision_model.pre_ln_w  = get_tensor(new_clip->ctx_data, format(TN_LN_PRE, "v", "weight"));
             vision_model.pre_ln_b  = get_tensor(new_clip->ctx_data, format(TN_LN_PRE, "v", "bias"));
             new_clip->has_pre_norm = true;
-        } catch (std::exception & e) {
+        } catch (std::exception & /*e*/) {
             new_clip->has_pre_norm = false;
         }
 
@@ -1189,21 +1189,21 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
             vision_model.post_ln_w  = get_tensor(new_clip->ctx_data, format(TN_LN_POST, "v", "weight"));
             vision_model.post_ln_b  = get_tensor(new_clip->ctx_data, format(TN_LN_POST, "v", "bias"));
             new_clip->has_post_norm = true;
-        } catch (std::exception & e) {
+        } catch (std::exception & /*e*/) {
             new_clip->has_post_norm = false;
         }
 
         try {
             vision_model.patch_bias = get_tensor(new_clip->ctx_data, TN_PATCH_BIAS);
             new_clip->has_patch_bias = true;
-        } catch (std::exception & e) {
+        } catch (std::exception & /*e*/) {
             new_clip->has_patch_bias = false;
         }
 
         try {
             vision_model.patch_embeddings    = get_tensor(new_clip->ctx_data, TN_PATCH_EMBD);
             vision_model.position_embeddings = get_tensor(new_clip->ctx_data, format(TN_POS_EMBD, "v"));
-        } catch(const std::exception& e) {
+        } catch(const std::exception& /*e*/) {
             LOG_TEE("%s: failed to load vision model tensors\n", __func__);
         }
 
@@ -1215,26 +1215,26 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
                 // Yi-type llava
                 vision_model.mm_1_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 1, "weight"));
                 vision_model.mm_1_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 1, "bias"));
-            } catch (std::runtime_error & e) {  }
+            } catch (std::runtime_error & /*e*/) { }
             try {
                 // missing in Yi-type llava
                 vision_model.mm_2_w              = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 2, "weight"));
                 vision_model.mm_2_b              = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 2, "bias"));
-            } catch (std::runtime_error & e) {  }
+            } catch (std::runtime_error & /*e*/) { }
             try {
                 // Yi-type llava
                 vision_model.mm_3_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 3, "weight"));
                 vision_model.mm_3_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 3, "bias"));
-            } catch (std::runtime_error & e) {  }
+            } catch (std::runtime_error & /*e*/) { }
             try {
                 // Yi-type llava
                 vision_model.mm_4_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 4, "weight"));
                 vision_model.mm_4_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 4, "bias"));
-            } catch (std::runtime_error & e) {  }
+            } catch (std::runtime_error & /*e*/) { }
             try {
                 vision_model.image_newline = get_tensor(new_clip->ctx_data, TN_IMAGE_NEWLINE);
                 // LOG_TEE("%s: image_newline tensor (llava-1.6) found\n", __func__);
-            } catch (std::runtime_error & e) {  }
+            } catch (std::runtime_error & /*e*/) { }
         } else if (new_clip->proj_type == PROJECTOR_TYPE_LDP) {
             // MobileVLM projection
             vision_model.mm_model_mlp_1_w               = get_tensor(new_clip->ctx_data, format(TN_MVLM_PROJ_MLP, 1, "weight"));

ggml/CMakeLists.txt

@@ -109,6 +109,7 @@ option(GGML_LLAMAFILE                       "ggml: use ggml SGEMM"
 option(GGML_CUDA                            "ggml: use CUDA"                                  OFF)
 option(GGML_CUDA_FORCE_DMMV                 "ggml: use dmmv instead of mmvq CUDA kernels"     OFF)
 option(GGML_CUDA_FORCE_MMQ                  "ggml: use mmq kernels instead of cuBLAS"         OFF)
+option(GGML_CUDA_FORCE_CUBLAS               "ggml: always use cuBLAS instead of mmq kernels"  OFF)
 set   (GGML_CUDA_DMMV_X   "32" CACHE STRING "ggml: x stride for dmmv CUDA kernels")
 set   (GGML_CUDA_MMV_Y     "1" CACHE STRING "ggml: y block size for mmv CUDA kernels")
 option(GGML_CUDA_F16                        "ggml: use 16 bit floats for some calculations"   OFF)
@@ -119,6 +120,7 @@ set   (GGML_CUDA_PEER_MAX_BATCH_SIZE "128" CACHE STRING
 option(GGML_CUDA_NO_PEER_COPY               "ggml: do not use peer to peer copies"            OFF)
 option(GGML_CUDA_NO_VMM                     "ggml: do not try to use CUDA VMM"                OFF)
 option(GGML_CUDA_FA_ALL_QUANTS              "ggml: compile all quants for FlashAttention"     OFF)
+option(GGML_CUDA_USE_GRAPHS                 "ggml: use CUDA graphs (llama.cpp only)"          OFF)
 
 option(GGML_CURL                            "ggml: use libcurl to download model from an URL" OFF)
 option(GGML_HIPBLAS                         "ggml: use hipBLAS"                               OFF)

ggml/src/CMakeLists.txt

@@ -295,12 +295,15 @@ if (GGML_CUDA)
 
         list(APPEND GGML_CDEF_PUBLIC GGML_USE_CUDA)
 
-        add_compile_definitions(GGML_CUDA_USE_GRAPHS)
         add_compile_definitions(GGML_CUDA_DMMV_X=${GGML_CUDA_DMMV_X})
         add_compile_definitions(GGML_CUDA_MMV_Y=${GGML_CUDA_MMV_Y})
         add_compile_definitions(K_QUANTS_PER_ITERATION=${GGML_CUDA_KQUANTS_ITER})
         add_compile_definitions(GGML_CUDA_PEER_MAX_BATCH_SIZE=${GGML_CUDA_PEER_MAX_BATCH_SIZE})
 
+        if (GGML_CUDA_USE_GRAPHS)
+            add_compile_definitions(GGML_CUDA_USE_GRAPHS)
+        endif()
+
         if (GGML_CUDA_FORCE_DMMV)
             add_compile_definitions(GGML_CUDA_FORCE_DMMV)
         endif()

scripts/sync-ggml-am.sh

@@ -136,42 +136,41 @@ if [ -f $SRC_LLAMA/ggml-src.patch ]; then
     # LICENSE                -> LICENSE
     # scripts/gen-authors.sh -> scripts/gen-authors.sh
 
-    cat ggml-src.patch | sed \
-        -e 's/CMakeLists.txt/ggml\/CMakeLists.txt/g' \
-        -e 's/src\/CMakeLists.txt/ggml\/src\/CMakeLists.txt/g' \
-        -e 's/cmake\/FindSIMD.cmake/ggml\/cmake\/FindSIMD.cmake/g' \
-        -e 's/src\/ggml\.c/ggml/src/ggml.c/g' \
-        -e 's/src\/ggml-alloc\.c/ggml/src/ggml-alloc.c/g' \
-        -e 's/src\/ggml-backend-impl\.h/ggml/src/ggml-backend-impl.h/g' \
-        -e 's/src\/ggml-backend\.c/ggml/src/ggml-backend.c/g' \
-        -e 's/src\/ggml-common\.h/ggml/src/ggml-common.h/g' \
-        -e 's/src\/ggml-cuda\//ggml-cuda\//g' \
-        -e 's/src\/ggml-cuda\.cu/ggml/src/ggml-cuda.cu/g' \
-        -e 's/src\/ggml-impl\.h/ggml/src/ggml-impl.h/g' \
-        -e 's/src\/ggml-kompute\.cpp/ggml/src/ggml-kompute.cpp/g' \
-        -e 's/src\/ggml-metal\.m/ggml/src/ggml-metal.m/g' \
-        -e 's/src\/ggml-quants\.c/ggml/src/ggml-quants.c/g' \
-        -e 's/src\/ggml-quants\.h/ggml/src/ggml-quants.h/g' \
-        -e 's/src\/ggml-rpc\.cpp/ggml/src/ggml-rpc.cpp/g' \
-        -e 's/src\/ggml-sycl\.cpp/ggml/src/ggml-sycl.cpp/g' \
-        -e 's/src\/ggml-vulkan\.cpp/ggml/src/ggml-vulkan.cpp/g' \
-        -e 's/include\/ggml\.h/ggml/include/ggml.h/g' \
-        -e 's/include\/ggml-alloc\.h/ggml/include/ggml-alloc.h/g' \
-        -e 's/include\/ggml-backend\.h/ggml/include/ggml-backend.h/g' \
-        -e 's/include\/ggml-blas\.h/ggml/include/ggml-blas.h/g' \
-        -e 's/include\/ggml-cuda\.h/ggml/include/ggml-cuda.h/g' \
-        -e 's/include\/ggml-kompute\.h/ggml/include/ggml-kompute.h/g' \
-        -e 's/include\/ggml-metal\.h/ggml/include/ggml-metal.h/g' \
-        -e 's/include\/ggml-rpc\.h/ggml/include/ggml-rpc.h/g' \
-        -e 's/include\/ggml-sycl\.h/ggml/include/ggml-sycl.h/g' \
-        -e 's/include\/ggml-vulkan\.h/ggml/include/ggml-vulkan.h/g' \
-        -e 's/tests\/test-opt\.cpp/tests\/test-opt.cpp/g' \
-        -e 's/tests\/test-grad0\.cpp/tests\/test-grad0.cpp/g' \
-        -e 's/tests\/test-quantize-fns\.cpp/tests\/test-quantize-fns.cpp/g' \
-        -e 's/tests\/test-quantize-perf\.cpp/tests\/test-quantize-perf.cpp/g' \
-        -e 's/tests\/test-backend-ops\.cpp/tests\/test-backend-ops.cpp/g' \
-        -e 's/LICENSE/LICENSE/g' \
-        -e 's/scripts\/gen-authors\.sh/scripts\/gen-authors.sh/g' \
+    cat ggml-src.patch | sed -E \
+        -e 's/([[:space:]]|[ab]\/)CMakeLists.txt/\1ggml\/CMakeLists.txt/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/CMakeLists.txt/\1ggml\/src\/CMakeLists.txt/g' \
+        -e 's/([[:space:]]|[ab]\/)cmake\/FindSIMD.cmake/\1ggml\/cmake\/FindSIMD.cmake/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml\.c/\1ggml\/src\/ggml.c/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-alloc\.c/\1ggml\/src\/ggml-alloc.c/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-backend-impl\.h/\1ggml\/src\/ggml-backend-impl.h/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-backend\.c/\1ggml\/src\/ggml-backend.c/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-common\.h/\1ggml\/src\/ggml-common.h/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-cuda\//\1ggml\/src\/ggml-cuda\//g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-cuda\.cu/\1ggml\/src\/ggml-cuda.cu/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-impl\.h/\1ggml\/src\/ggml-impl.h/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-kompute\.cpp/\1ggml\/src\/ggml-kompute.cpp/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-metal\.m/\1ggml\/src\/ggml-metal.m/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-quants\.c/\1ggml\/src\/ggml-quants.c/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-quants\.h/\1ggml\/src\/ggml-quants.h/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-rpc\.cpp/\1ggml\/src\/ggml-rpc.cpp/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-sycl\.cpp/\1ggml\/src\/ggml-sycl.cpp/g' \
+        -e 's/([[:space:]]|[ab]\/)src\/ggml-vulkan\.cpp/\1ggml\/src\/ggml-vulkan.cpp/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml\.h/\1ggml\/include\/ggml.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-alloc\.h/\1ggml\/include\/ggml-alloc.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-backend\.h/\1ggml\/include\/ggml-backend.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-blas\.h/\1ggml\/include\/ggml-blas.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-cuda\.h/\1ggml\/include\/ggml-cuda.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-kompute\.h/\1ggml\/include\/ggml-kompute.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-metal\.h/\1ggml\/include\/ggml-metal.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-rpc\.h/\1ggml\/include\/ggml-rpc.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-sycl\.h/\1ggml\/include\/ggml-sycl.h/g' \
+        -e 's/([[:space:]]|[ab]\/)include\/ggml-vulkan\.h/\1ggml\/include\/ggml-vulkan.h/g' \
+        -e 's/([[:space:]]|[ab]\/)examples\/common\.h/examples\/common.h/g' \
+        -e 's/([[:space:]]|[ab]\/)examples\/common\.cpp/examples\/common.cpp/g' \
+        -e 's/([[:space:]]|[ab]\/)examples\/common-ggml\.h/examples\/common-ggml.h/g' \
+        -e 's/([[:space:]]|[ab]\/)examples\/common-ggml\.cpp/examples\/common-ggml.cpp/g' \
+        -e 's/([[:space:]]|[ab]\/)LICENSE/LICENSE/g' \
+        -e 's/([[:space:]]|[ab]\/)scripts\/gen-authors\.sh/scripts\/gen-authors.sh/g' \
         > ggml-src.patch.tmp
     mv ggml-src.patch.tmp ggml-src.patch