ggml-cuda: Fix HIP build by adding define for __trap (#4569)

Regression of 1398823922
HIP doesn't have trap, only abort

README.md

@@ -982,6 +982,8 @@ docker run --gpus all -v /path/to/models:/models local/llama.cpp:light-cuda -m /
 - There are no strict rules for the code style, but try to follow the patterns in the code (indentation, spaces, etc.). Vertical alignment makes things more readable and easier to batch edit
 - Clean-up any trailing whitespaces, use 4 spaces for indentation, brackets on the same line, `void * ptr`, `int & a`
 - See [good first issues](https://github.com/ggerganov/llama.cpp/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22) for tasks suitable for first contributions
+- Tensors store data in row-major order. We refer to dimension 0 as columns, 1 as rows, 2 as matrices
+- Matrix multiplication is unconventional: [`z = ggml_mul_mat(ctx, x, y)`](https://github.com/ggerganov/llama.cpp/blob/880e352277fc017df4d5794f0c21c44e1eae2b84/ggml.h#L1058-L1064) means `zT = x @ yT`
 
 ### Docs
 

common/common.cpp

@@ -920,7 +920,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     printf("  -m FNAME, --model FNAME\n");
     printf("                        model path (default: %s)\n", params.model.c_str());
     printf("  -md FNAME, --model-draft FNAME\n");
-    printf("                        draft model for speculative decoding (default: %s)\n", params.model.c_str());
+    printf("                        draft model for speculative decoding\n");
     printf("  -ld LOGDIR, --logdir LOGDIR\n");
     printf("                        path under which to save YAML logs (no logging if unset)\n");
     printf("  --override-kv KEY=TYPE:VALUE\n");

ggml-cuda.cu

@@ -80,6 +80,7 @@
 #define cudaStreamWaitEvent(stream, event, flags) hipStreamWaitEvent(stream, event, flags)
 #define cudaStream_t hipStream_t
 #define cudaSuccess hipSuccess
+#define __trap abort
 #else
 #include <cuda_runtime.h>
 #include <cublas_v2.h>
@@ -512,6 +513,14 @@ static size_t g_scratch_offset = 0;
 
 static cublasHandle_t g_cublas_handles[GGML_CUDA_MAX_DEVICES] = {nullptr};
 
+[[noreturn]]
+static __device__ void bad_arch() {
+    printf("ERROR: ggml-cuda was compiled without support for the current GPU architecture.\n");
+    __trap();
+
+    (void) bad_arch; // suppress unused function warning
+}
+
 static __device__ __forceinline__ float warp_reduce_sum(float x) {
 #pragma unroll
     for (int mask = 16; mask > 0; mask >>= 1) {
@@ -1972,8 +1981,7 @@ template <int vdr> static __device__ __forceinline__ float vec_dot_q4_0_q8_1_imp
     // second part effectively subtracts 8 from each quant value
     return d4 * (sumi * ds8f.x - (8*vdr/QI4_0) * ds8f.y);
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2010,8 +2018,7 @@ template <int vdr> static __device__ __forceinline__ float vec_dot_q4_1_q8_1_imp
     // scale second part of sum by QI8_1/(vdr * QR4_1) to compensate for multiple threads adding it
     return sumi * d4d8 + m4s8 / (QI8_1 / (vdr * QR4_1));
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2046,8 +2053,7 @@ template <int vdr> static __device__ __forceinline__ float vec_dot_q5_0_q8_1_imp
     // second part effectively subtracts 16 from each quant value
     return d5 * (sumi * ds8f.x - (16*vdr/QI5_0) * ds8f.y);
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2092,8 +2098,7 @@ template <int vdr> static __device__ __forceinline__ float vec_dot_q5_1_q8_1_imp
     return sumi*d5d8 + m5s8 / (QI5_1 / vdr);
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2114,8 +2119,7 @@ template <int vdr> static __device__ __forceinline__ float vec_dot_q8_0_q8_1_imp
 
     return d8_0*d8_1 * sumi;
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2145,8 +2149,7 @@ template <int vdr> static __device__ __forceinline__ float vec_dot_q8_1_q8_1_imp
     // scale second part of sum by QI8_1/ vdr to compensate for multiple threads adding it
     return sumi*d8d8 + m8s8 / (QI8_1 / vdr);
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2181,8 +2184,7 @@ static __device__ __forceinline__ float vec_dot_q2_K_q8_1_impl_mmvq(
 
     return dm2f.x*sumf_d - dm2f.y*sumf_m;
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2219,8 +2221,7 @@ static __device__ __forceinline__ float vec_dot_q2_K_q8_1_impl_mmq(
 
     return d8 * (dm2f.x*sumi_d - dm2f.y*sumi_m);
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2260,8 +2261,7 @@ static __device__ __forceinline__ float vec_dot_q3_K_q8_1_impl_mmvq(
 
     return d3 * sumf;
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2286,8 +2286,7 @@ static __device__ __forceinline__ float vec_dot_q3_K_q8_1_impl_mmq(
 
     return d3*d8 * sumi;
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2320,8 +2319,7 @@ static __device__ __forceinline__ float vec_dot_q4_K_q8_1_impl_vmmq(
     return dm4f.x*sumf_d - dm4f.y*sumf_m;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2354,8 +2352,7 @@ static __device__ __forceinline__ float vec_dot_q4_K_q8_1_impl_mmq(
     return dm4f.x*sumf_d - dm4f.y*sumf_m;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2395,8 +2392,7 @@ static __device__ __forceinline__ float vec_dot_q5_K_q8_1_impl_vmmq(
     return dm5f.x*sumf_d - dm5f.y*sumf_m;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2429,8 +2425,7 @@ static __device__ __forceinline__ float vec_dot_q5_K_q8_1_impl_mmq(
     return dm4f.x*sumf_d - dm4f.y*sumf_m;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2460,8 +2455,7 @@ static __device__ __forceinline__ float vec_dot_q6_K_q8_1_impl_mmvq(
 
     return d*sumf;
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -2492,8 +2486,7 @@ static __device__ __forceinline__ float vec_dot_q6_K_q8_1_impl_mmq(
     return d6 * sumf_d;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 }
 
@@ -3359,8 +3352,7 @@ static __device__ __forceinline__ float vec_dot_q4_K_q8_1(
     return dall * sumf_d - dmin * sumf_m;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 
 #endif
@@ -3543,8 +3535,7 @@ static __device__ __forceinline__ float vec_dot_q5_K_q8_1(
     return d * sumf_d;
 
 #else
-    assert(false);
-    return 0.0f; // only to satisfy the compiler
+    bad_arch();
 #endif // __CUDA_ARCH__ >= MIN_CC_DP4A
 
 #endif
@@ -3954,7 +3945,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q4_0_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4023,7 +4014,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q4_1_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4090,7 +4081,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q5_0_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4157,7 +4148,7 @@ mul_mat_q5_1(
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q5_1_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4224,7 +4215,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q8_0_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4291,7 +4282,7 @@ mul_mat_q2_K(
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q2_K_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4360,7 +4351,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q3_K_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4429,7 +4420,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q4_K_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4496,7 +4487,7 @@ mul_mat_q5_K(
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q5_K_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -4565,7 +4556,7 @@ template <bool need_check> static __global__ void
         (vx, vy, dst, ncols_x, nrows_x, ncols_y, nrows_y, nrows_dst);
 #else
     (void) vec_dot_q6_K_q8_1_mul_mat;
-    assert(false);
+    bad_arch();
 #endif // __CUDA_ARCH__ >= CC_VOLTA
 }
 
@@ -6825,6 +6816,7 @@ static void ggml_cuda_op_get_rows(
             break;
         default:
             // TODO: k-quants
+            fprintf(stderr, "%s: unsupported type: %s\n", __func__, ggml_type_name(src0->type));
             GGML_ASSERT(false);
             break;
     }
@@ -8782,8 +8774,6 @@ static void ggml_cuda_mul_mat_id(const ggml_tensor * src0, const ggml_tensor * s
     // TODO: mmq/mmv support
 #endif
 
-    GGML_ASSERT(dst->backend == GGML_BACKEND_GPU);
-
     const int64_t nb11 = src1->nb[1];
     const int64_t nb1  =  dst->nb[1];
 
@@ -8812,13 +8802,21 @@ static void ggml_cuda_mul_mat_id(const ggml_tensor * src0, const ggml_tensor * s
     ggml_tensor src1_row = *src1;
     ggml_tensor dst_row = *dst;
 
+    src1_row.backend = GGML_BACKEND_GPU;
+    dst_row.backend  = GGML_BACKEND_GPU;
+
     src1_row.extra = &src1_row_extra;
     dst_row.extra = &dst_row_extra;
 
-    char * src1_original = (char *) src1_extra->data_device[g_main_device];
-    char * dst_original  = (char *)  dst_extra->data_device[g_main_device];
+    char * src1_original = src1->backend == GGML_BACKEND_CPU ?
+        (char *) src1->data : (char *) src1_extra->data_device[g_main_device];
+    char * dst_original  =  dst->backend == GGML_BACKEND_CPU ?
+        (char *)  dst->data : (char *)  dst_extra->data_device[g_main_device];
 
     if (src1->ne[1] == 1) {
+        GGML_ASSERT(src1->backend == GGML_BACKEND_GPU);
+        GGML_ASSERT(dst->backend  == GGML_BACKEND_GPU);
+
         for (int64_t i01 = 0; i01 < ids->ne[1]; i01++) {
             //int32_t row_id;
             //CUDA_CHECK(cudaMemcpyAsync(&row_id, ids_dev + i01*ids->nb[1] + id*ids->nb[0], sizeof(int32_t), cudaMemcpyDeviceToHost, g_cudaStreams[g_main_device][0]));
@@ -8846,6 +8844,11 @@ static void ggml_cuda_mul_mat_id(const ggml_tensor * src0, const ggml_tensor * s
         src1_row_extra.data_device[g_main_device] = src1_contiguous;
         dst_row_extra.data_device[g_main_device]  =  dst_contiguous;
 
+        const cudaMemcpyKind src1_kind = src1->backend == GGML_BACKEND_CPU ?
+            cudaMemcpyHostToDevice : cudaMemcpyDeviceToDevice;
+        const cudaMemcpyKind dst_kind  =  dst->backend == GGML_BACKEND_CPU ?
+            cudaMemcpyHostToDevice : cudaMemcpyDeviceToDevice;
+
         for (int32_t row_id = 0; row_id < n_as; ++row_id) {
             const struct ggml_tensor * src0_row = dst->src[row_id + 2];
 
@@ -8860,7 +8863,7 @@ static void ggml_cuda_mul_mat_id(const ggml_tensor * src0, const ggml_tensor * s
                 GGML_ASSERT(row_id >= 0 && row_id < n_as);
 
                 CUDA_CHECK(cudaMemcpyAsync(src1_contiguous + num_src1_rows*nb11, src1_original + i01*nb11,
-                                        nb11, cudaMemcpyDeviceToDevice, stream));
+                                        nb11, src1_kind, stream));
                 num_src1_rows++;
             }
 
@@ -8892,7 +8895,7 @@ static void ggml_cuda_mul_mat_id(const ggml_tensor * src0, const ggml_tensor * s
                 GGML_ASSERT(row_id >= 0 && row_id < n_as);
 
                 CUDA_CHECK(cudaMemcpyAsync(dst_original + i01*nb1, dst_contiguous + num_src1_rows*nb1,
-                                        nb1, cudaMemcpyDeviceToDevice, stream));
+                                        nb1, dst_kind, stream));
                 num_src1_rows++;
             }
         }
@@ -8900,6 +8903,10 @@ static void ggml_cuda_mul_mat_id(const ggml_tensor * src0, const ggml_tensor * s
         ggml_cuda_pool_free(src1_contiguous, as_src1);
         ggml_cuda_pool_free(dst_contiguous,  as_dst);
     }
+
+    if (dst->backend == GGML_BACKEND_CPU) {
+        CUDA_CHECK(cudaStreamSynchronize(stream));
+    }
 }
 
 static void ggml_cuda_scale(const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst) {
@@ -9091,7 +9098,7 @@ void ggml_cuda_transform_tensor(void * data, struct ggml_tensor * tensor) {
 }
 
 void ggml_cuda_free_data(struct ggml_tensor * tensor) {
-    if (!tensor || (tensor->backend != GGML_BACKEND_GPU && tensor->backend != GGML_BACKEND_GPU_SPLIT) ) {
+    if (!tensor || !tensor->extra || (tensor->backend != GGML_BACKEND_GPU && tensor->backend != GGML_BACKEND_GPU_SPLIT) ) {
         return;
     }
 
@@ -9298,7 +9305,7 @@ bool ggml_cuda_compute_forward(struct ggml_compute_params * params, struct ggml_
         || (tensor->src[0] != nullptr && (tensor->src[0]->backend == GGML_BACKEND_GPU || tensor->src[0]->backend == GGML_BACKEND_GPU_SPLIT))
         || (tensor->src[1] != nullptr && tensor->src[1]->backend == GGML_BACKEND_GPU);
 
-    if (!any_on_device && tensor->op != GGML_OP_MUL_MAT) {
+    if (!any_on_device && tensor->op != GGML_OP_MUL_MAT && tensor->op != GGML_OP_MUL_MAT_ID) {
         return false;
     }
 

gguf-py/gguf/vocab.py

@@ -84,7 +84,7 @@ class SpecialVocab:
         merges_file = path / 'merges.txt'
         if not merges_file.is_file():
             return False
-        with open(merges_file, 'r') as fp:
+        with open(merges_file, 'r', encoding = 'utf-8') as fp:
             first_line = next(fp, '').strip()
             if not first_line.startswith('#'):
                 fp.seek(0)

llama.cpp

@@ -2083,7 +2083,7 @@ struct llama_model_loader {
                     type_max   = meta->type;
                 }
 
-                LLAMA_LOG_INFO("%s: - tensor %4d: %32s %-8s [ %s ]\n", __func__, i, name, ggml_type_name(meta->type), llama_format_tensor_shape(meta).c_str());
+                // LLAMA_LOG_INFO("%s: - tensor %4d: %32s %-8s [ %s ]\n", __func__, i, name, ggml_type_name(meta->type), llama_format_tensor_shape(meta).c_str());
             }
 
             switch (type_max) {