sync : ggml (VMM, sync-ggml-am, dotprod ARM fixes, CUDA fixes) (#1691)

* scripts : add sync-ggml-am.sh

* sync : ggml (VMM, ARM dot prod fix, etc.)

* build : fix CUDA build

* ggml : fix some mul mat cases + add tests for src1 F16

ggerganov/ggml@dbd0295

CMakeLists.txt

@@ -218,11 +218,17 @@ if (WHISPER_CUBLAS)
         add_compile_definitions(GGML_USE_CUBLAS)
 
         if (WHISPER_STATIC)
-            set(WHISPER_EXTRA_LIBS ${WHISPER_EXTRA_LIBS} CUDA::cudart_static CUDA::cublas_static CUDA::cublasLt_static)
+            if (WIN32)
+                # As of 12.3.1 CUDA Tookit for Windows does not offer a static cublas library
+                set(WHISPER_EXTRA_LIBS ${WHISPER_EXTRA_LIBS} CUDA::cudart_static CUDA::cublas CUDA::cublasLt)
+            else ()
+                set(WHISPER_EXTRA_LIBS ${WHISPER_EXTRA_LIBS} CUDA::cudart_static CUDA::cublas_static CUDA::cublasLt_static)
+            endif()
         else()
             set(WHISPER_EXTRA_LIBS ${WHISPER_EXTRA_LIBS} CUDA::cudart CUDA::cublas CUDA::cublasLt)
         endif()
 
+        set(WHISPER_EXTRA_LIBS ${WHISPER_EXTRA_LIBS} CUDA::cuda_driver)
     else()
         message(FATAL_ERROR "cuBLAS not found")
     endif()

Makefile

@@ -206,7 +206,7 @@ ifdef WHISPER_CUBLAS
 
 	CFLAGS      += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/$(UNAME_M)-linux/include
 	CXXFLAGS    += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/$(UNAME_M)-linux/include
-	LDFLAGS     += -lcublas -lculibos -lcudart -lcublasLt -lpthread -ldl -lrt -L/usr/local/cuda/lib64 -L/opt/cuda/lib64 -L$(CUDA_PATH)/targets/$(UNAME_M)-linux/lib
+	LDFLAGS     += -lcuda -lcublas -lculibos -lcudart -lcublasLt -lpthread -ldl -lrt -L/usr/local/cuda/lib64 -L/opt/cuda/lib64 -L$(CUDA_PATH)/targets/$(UNAME_M)-linux/lib
 	WHISPER_OBJ += ggml-cuda.o
 	NVCC        = nvcc
 	NVCCFLAGS   = --forward-unknown-to-host-compiler -arch=$(CUDA_ARCH_FLAG)

extra/sync-ggml-am.sh

@@ -0,0 +1,138 @@
+#!/bin/bash
+#
+# Synchronize ggml changes to whisper.cpp
+#
+# Usage:
+#
+#   $ cd /path/to/whisper.cpp
+#   $ ./extra/sync-ggml-am.sh
+#
+
+set -e
+
+sd=$(dirname $0)
+cd $sd/../
+
+SRC_WHISPER=$(pwd)
+SRC_GGML=$(cd ../ggml; pwd)
+
+if [ ! -d $SRC_GGML ]; then
+    echo "ggml not found at $SRC_GGML"
+    exit 1
+fi
+
+lc=$(cat $SRC_WHISPER/extra/sync-ggml.last)
+echo "Syncing ggml changes since commit $lc"
+
+cd $SRC_GGML
+
+git log --oneline $lc..HEAD
+
+git format-patch $lc --stdout -- \
+    include/ggml/ggml*.h \
+    src/ggml*.h \
+    src/ggml*.c \
+    src/ggml*.cpp \
+    src/ggml*.m \
+    src/ggml*.metal \
+    src/ggml*.cu \
+    tests/test-opt.cpp \
+    tests/test-grad0.cpp \
+    tests/test-quantize-fns.cpp \
+    tests/test-quantize-perf.cpp \
+    tests/test-backend-ops.cpp \
+    > $SRC_WHISPER/ggml-src.patch
+
+# delete files if empty
+if [ ! -s $SRC_WHISPER/ggml-src.patch ]; then
+    rm -v $SRC_WHISPER/ggml-src.patch
+fi
+
+cd $SRC_WHISPER
+
+if [ -f $SRC_WHISPER/ggml-src.patch ]; then
+    # replace PR numbers
+    #
+    # Subject: some text (#1234)
+    # Subject: some text (ggml/1234)
+    cat ggml-src.patch | sed -e 's/^Subject: \(.*\) (#\([0-9]*\))/Subject: \1 (ggml\/\2)/' > ggml-src.patch.tmp
+    mv ggml-src.patch.tmp ggml-src.patch
+
+    cat ggml-src.patch | sed -e 's/^\(.*\) (#\([0-9]*\))$/\1 (ggml\/\2)/' > ggml-src.patch.tmp
+    mv ggml-src.patch.tmp ggml-src.patch
+
+    # replace filenames:
+    #
+    # src/ggml.c                  -> ggml.c
+    # src/ggml-alloc.c            -> ggml-alloc.c
+    # src/ggml-backend-impl.h     -> ggml-backend-impl.h
+    # src/ggml-backend.c          -> ggml-backend.c
+    # src/ggml-cuda.cu            -> ggml-cuda.cu
+    # src/ggml-cuda.h             -> ggml-cuda.h
+    # src/ggml-impl.h             -> ggml-impl.h
+    # src/ggml-metal.h            -> ggml-metal.h
+    # src/ggml-metal.m            -> ggml-metal.m
+    # src/ggml-metal.metal        -> ggml-metal.metal
+    # src/ggml-mpi.h              -> ggml-mpi.h
+    # src/ggml-mpi.c              -> ggml-mpi.c
+    # src/ggml-opencl.cpp         -> ggml-opencl.cpp
+    # src/ggml-opencl.h           -> ggml-opencl.h
+    # src/ggml-quants.c           -> ggml-quants.c
+    # src/ggml-quants.h           -> ggml-quants.h
+    # include/ggml/ggml.h         -> ggml.h
+    # include/ggml/ggml-alloc.h   -> ggml-alloc.h
+    # include/ggml/ggml-backend.h -> ggml-backend.h
+    #
+    # examples/common.h           -> examples/common.h
+    # examples/common.cpp         -> examples/common.cpp
+    # examples/common-ggml.h      -> examples/common-ggml.h
+    # examples/common-ggml.cpp    -> examples/common-ggml.cpp
+    #
+    # examples/whisper/whisper.h    -> whisper.h
+    # examples/whisper/whisper.cpp  -> whisper.cpp
+    # examples/whisper/main.cpp     -> examples/main/main.cpp
+    # examples/whisper/quantize.cpp -> examples/quantize/quantize.cpp
+
+    cat ggml-src.patch | sed \
+        -e 's/src\/ggml\.c/ggml.c/g' \
+        -e 's/src\/ggml-alloc\.c/ggml-alloc.c/g' \
+        -e 's/src\/ggml-backend-impl\.h/ggml-backend-impl.h/g' \
+        -e 's/src\/ggml-backend\.c/ggml-backend.c/g' \
+        -e 's/src\/ggml-cuda\.cu/ggml-cuda.cu/g' \
+        -e 's/src\/ggml-cuda\.h/ggml-cuda.h/g' \
+        -e 's/src\/ggml-impl\.h/ggml-impl.h/g' \
+        -e 's/src\/ggml-metal\.h/ggml-metal.h/g' \
+        -e 's/src\/ggml-metal\.m/ggml-metal.m/g' \
+        -e 's/src\/ggml-metal\.metal/ggml-metal.metal/g' \
+        -e 's/src\/ggml-mpi\.h/ggml-mpi.h/g' \
+        -e 's/src\/ggml-mpi\.c/ggml-mpi.c/g' \
+        -e 's/src\/ggml-opencl\.cpp/ggml-opencl.cpp/g' \
+        -e 's/src\/ggml-opencl\.h/ggml-opencl.h/g' \
+        -e 's/src\/ggml-quants\.c/ggml-quants.c/g' \
+        -e 's/src\/ggml-quants\.h/ggml-quants.h/g' \
+        -e 's/include\/ggml\/ggml\.h/ggml.h/g' \
+        -e 's/include\/ggml\/ggml-alloc\.h/ggml-alloc.h/g' \
+        -e 's/include\/ggml\/ggml-backend\.h/ggml-backend.h/g' \
+        -e 's/examples\/common\.h/examples\/common.h/g' \
+        -e 's/examples\/common\.cpp/examples\/common.cpp/g' \
+        -e 's/examples\/common-ggml\.h/examples\/common-ggml.h/g' \
+        -e 's/examples\/common-ggml\.cpp/examples\/common-ggml.cpp/g' \
+        -e 's/examples\/whisper\/whisper\.h/whisper.h/g' \
+        -e 's/examples\/whisper\/whisper\.cpp/whisper.cpp/g' \
+        -e 's/examples\/whisper\/main\.cpp/examples\/main\/main.cpp/g' \
+        -e 's/examples\/whisper\/quantize\.cpp/examples\/quantize\/quantize.cpp/g' \
+        > ggml-src.patch.tmp
+    mv ggml-src.patch.tmp ggml-src.patch
+
+    git am ggml-src.patch
+
+    rm -v $SRC_WHISPER/ggml-src.patch
+fi
+
+# update last commit
+cd $SRC_GGML
+git log -1 --format=%H > $SRC_WHISPER/extra/sync-ggml.last
+
+echo "Done"
+
+exit 0

extra/sync-ggml.last

@@ -0,0 +1 @@
+1467a4eb71bdb5ac316d248a7f3f26cdadc56b68

ggml-backend.c

@@ -297,7 +297,7 @@ static void ggml_backend_registry_init(void) {
 void ggml_backend_register(const char * name, ggml_backend_init_fn init_fn, ggml_backend_buffer_type_t default_buffer_type, void * user_data) {
     GGML_ASSERT(ggml_backend_registry_count < GGML_MAX_BACKENDS_REG);
 
-    int id = ggml_backend_registry_count;
+    size_t id = ggml_backend_registry_count;
 
     ggml_backend_registry[id] = (struct ggml_backend_reg) {
         /* .name                = */ {0},
@@ -330,6 +330,8 @@ size_t ggml_backend_reg_find_by_name(const char * name) {
             return i;
         }
     }
+
+    // not found
     return SIZE_MAX;
 }
 
@@ -340,15 +342,15 @@ ggml_backend_t ggml_backend_reg_init_backend_from_str(const char * backend_str)
     const char * params = strchr(backend_str, ':');
     char backend_name[128];
     if (params == NULL) {
-        strcpy(backend_name, backend_str);
+        snprintf(backend_name, sizeof(backend_name), "%s", backend_str);
         params = "";
     } else {
-        strncpy(backend_name, backend_str, params - backend_str);
-        backend_name[params - backend_str] = '\0';
+        snprintf(backend_name, sizeof(backend_name), "%.*s", (int)(params - backend_str), backend_str);
         params++;
     }
 
     size_t backend_i = ggml_backend_reg_find_by_name(backend_name);
+
     if (backend_i == SIZE_MAX) {
         fprintf(stderr, "%s: backend %s not found\n", __func__, backend_name);
         return NULL;
@@ -396,18 +398,12 @@ static void ggml_backend_cpu_buffer_free_buffer(ggml_backend_buffer_t buffer) {
 }
 
 static void ggml_backend_cpu_buffer_set_tensor(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor, const void * data, size_t offset, size_t size) {
-    GGML_ASSERT(offset + size <= ggml_nbytes(tensor) && "tensor write out of bounds");
-    GGML_ASSERT(tensor->data != NULL && "tensor not allocated");
-
     memcpy((char *)tensor->data + offset, data, size);
 
     GGML_UNUSED(buffer);
 }
 
 static void ggml_backend_cpu_buffer_get_tensor(ggml_backend_buffer_t buffer, const struct ggml_tensor * tensor, void * data, size_t offset, size_t size) {
-    GGML_ASSERT(offset + size <= ggml_nbytes(tensor) && "tensor read out of bounds");
-    GGML_ASSERT(tensor->data != NULL && "tensor not allocated");
-
     memcpy(data, (const char *)tensor->data + offset, size);
 
     GGML_UNUSED(buffer);
@@ -618,10 +614,14 @@ static void ggml_backend_cpu_graph_compute(ggml_backend_t backend, struct ggml_c
 }
 
 static bool ggml_backend_cpu_supports_op(ggml_backend_t backend, const struct ggml_tensor * op) {
-    return true;
+    switch (op->op) {
+        case GGML_OP_MUL_MAT:
+            return op->src[1]->type == GGML_TYPE_F32 || op->src[1]->type == ggml_internal_get_type_traits(op->src[0]->type).vec_dot_type;
+        default:
+            return true;
+    }
 
     GGML_UNUSED(backend);
-    GGML_UNUSED(op);
 }
 
 static struct ggml_backend_i cpu_backend_i = {