diff --git a/ggml-metal.m b/ggml-metal.m index 2380c4310..bc881395a 100644 --- a/ggml-metal.m +++ b/ggml-metal.m @@ -238,14 +238,17 @@ struct ggml_metal_context * ggml_metal_init(int n_cb) { // load kernels { NSError * error = nil; -#define GGML_METAL_ADD_KERNEL(name) \ - ctx->function_##name = [ctx->library newFunctionWithName:@"kernel_"#name]; \ - ctx->pipeline_##name = [ctx->device newComputePipelineStateWithFunction:ctx->function_##name error:&error]; \ + + /* GGML_METAL_LOG_INFO("%s: loaded %-32s %16p | th_max = %4d | th_width = %4d\n", __func__, "kernel_"#name, (void *) ctx->pipeline_##name, \ (int) ctx->pipeline_##name.maxTotalThreadsPerThreadgroup, \ (int) ctx->pipeline_##name.threadExecutionWidth); \ + */ +#define GGML_METAL_ADD_KERNEL(name) \ + ctx->function_##name = [ctx->library newFunctionWithName:@"kernel_"#name]; \ + ctx->pipeline_##name = [ctx->device newComputePipelineStateWithFunction:ctx->function_##name error:&error]; \ if (error) { \ - GGML_METAL_LOG_ERROR("%s: error: load pipeline error: %s\n", __func__, [[error description] UTF8String]); \ + GGML_METAL_LOG_ERROR("%s: error: load pipeline error: %s\n", __func__, [[error description] UTF8String]); \ return NULL; \ } diff --git a/ggml.h b/ggml.h index 8c954904e..9d16c5a72 100644 --- a/ggml.h +++ b/ggml.h @@ -709,7 +709,7 @@ extern "C" { // Context tensor enumeration and lookup GGML_API struct ggml_tensor * ggml_get_first_tensor(struct ggml_context * ctx); GGML_API struct ggml_tensor * ggml_get_next_tensor (struct ggml_context * ctx, struct ggml_tensor * tensor); - GGML_API struct ggml_tensor * ggml_get_tensor(struct ggml_context * ctx, const char * name); + GGML_API struct ggml_tensor * ggml_get_tensor (struct ggml_context * ctx, const char * name); GGML_API struct ggml_tensor * ggml_set_zero(struct ggml_tensor * tensor); GGML_API struct ggml_tensor * ggml_set_i32 (struct ggml_tensor * tensor, int32_t value); diff --git a/llama.cpp b/llama.cpp index 7ee589298..ead1d421d 100644 --- a/llama.cpp +++ b/llama.cpp @@ -60,7 +60,9 @@ #include #include #include +#include #include +#include #include #include #include @@ -69,11 +71,10 @@ #include #include #include +#include #include #include #include -#include -#include #if defined(_MSC_VER) #pragma warning(disable: 4244 4267) // possible loss of data @@ -969,7 +970,7 @@ struct llama_mlock { typedef void (*offload_func_t)(struct ggml_tensor * tensor); -static void llama_nop(struct ggml_tensor * tensor) { // don't offload by default +static void ggml_offload_nop(struct ggml_tensor * tensor) { (void) tensor; } @@ -1113,13 +1114,13 @@ struct llama_layer { struct ggml_tensor * ffn_norm_b; // ff - struct ggml_tensor * w1; // ffn_gate - struct ggml_tensor * w2; // ffn_down - struct ggml_tensor * w3; // ffn_up + struct ggml_tensor * ffn_gate; // w1 + struct ggml_tensor * ffn_down; // w2 + struct ggml_tensor * ffn_up; // w3 // ff bias - struct ggml_tensor * b2; // ffn_down - struct ggml_tensor * b3; // ffn_up + struct ggml_tensor * ffn_down_b; // b2 + struct ggml_tensor * ffn_up_b; // b3 }; struct llama_kv_cell { @@ -1225,8 +1226,8 @@ struct llama_model { llama_hparams hparams = {}; llama_vocab vocab; - struct ggml_tensor * tok_embeddings; - struct ggml_tensor * pos_embeddings; + struct ggml_tensor * tok_embd; + struct ggml_tensor * pos_embd; struct ggml_tensor * tok_norm; struct ggml_tensor * tok_norm_b; @@ -2482,7 +2483,7 @@ static void llm_load_tensors( case LLM_ARCH_LLAMA: case LLM_ARCH_REFACT: { - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); // output { @@ -2536,21 +2537,21 @@ static void llm_load_tensors( layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); - layer.w1 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_GATE, "weight", i), {n_embd, n_ff}, backend_split); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_gate = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_GATE, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); if (backend == GGML_BACKEND_GPU) { vram_weights += - ggml_nbytes(layer.attn_norm) + ggml_nbytes(layer.wq) + ggml_nbytes(layer.wk) + - ggml_nbytes(layer.wv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.ffn_norm) + - ggml_nbytes(layer.w1) + ggml_nbytes(layer.w2) + ggml_nbytes(layer.w3); + ggml_nbytes(layer.attn_norm) + ggml_nbytes(layer.wq) + ggml_nbytes(layer.wk) + + ggml_nbytes(layer.wv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.ffn_norm) + + ggml_nbytes(layer.ffn_gate) + ggml_nbytes(layer.ffn_down) + ggml_nbytes(layer.ffn_up); } } } break; case LLM_ARCH_BAICHUAN: { - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); { ggml_backend_type backend_norm; ggml_backend_type backend_output; @@ -2602,15 +2603,15 @@ static void llm_load_tensors( layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); - layer.w1 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_GATE, "weight", i), {n_embd, n_ff}, backend_split); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_gate = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_GATE, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); if (backend == GGML_BACKEND_GPU) { vram_weights += - ggml_nbytes(layer.attn_norm) + ggml_nbytes(layer.wq) + ggml_nbytes(layer.wk) + - ggml_nbytes(layer.wv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.ffn_norm) + - ggml_nbytes(layer.w1) + ggml_nbytes(layer.w2) + ggml_nbytes(layer.w3); + ggml_nbytes(layer.attn_norm) + ggml_nbytes(layer.wq) + ggml_nbytes(layer.wk) + + ggml_nbytes(layer.wv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.ffn_norm) + + ggml_nbytes(layer.ffn_gate) + ggml_nbytes(layer.ffn_down) + ggml_nbytes(layer.ffn_up); } } } break; @@ -2618,7 +2619,7 @@ static void llm_load_tensors( { // TODO: CPU-only for now - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); // output { @@ -2681,21 +2682,21 @@ static void llm_load_tensors( layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, backend_split); layer.wo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, backend_split); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); if (backend == GGML_BACKEND_GPU) { vram_weights += ggml_nbytes(layer.attn_norm) + ggml_nbytes(layer.attn_norm_b) + ggml_nbytes(layer.wqkv) + ggml_nbytes(layer.wo) + - ggml_nbytes(layer.w2) + ggml_nbytes(layer.w3); + ggml_nbytes(layer.ffn_down) + ggml_nbytes(layer.ffn_up); } } } break; case LLM_ARCH_STARCODER: { - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); - model.pos_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_POS_EMBD, "weight"), {n_embd, hparams.n_ctx_train}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.pos_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_POS_EMBD, "weight"), {n_embd, hparams.n_ctx_train}, GGML_BACKEND_CPU); // output { @@ -2754,11 +2755,11 @@ static void llm_load_tensors( layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); layer.ffn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "bias", i), {n_embd}, backend); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split); - layer.b2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split); + layer.ffn_down_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); - layer.b3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff}, backend); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_up_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff}, backend); if (backend == GGML_BACKEND_GPU) { vram_weights += @@ -2766,14 +2767,14 @@ static void llm_load_tensors( ggml_nbytes(layer.wqkv) + ggml_nbytes(layer.bqkv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.bo) + ggml_nbytes(layer.ffn_norm) + ggml_nbytes(layer.ffn_norm_b) + - ggml_nbytes(layer.w2) + ggml_nbytes(layer.b2) + - ggml_nbytes(layer.w3) + ggml_nbytes(layer.b3); + ggml_nbytes(layer.ffn_down) + ggml_nbytes(layer.ffn_down_b) + + ggml_nbytes(layer.ffn_up) + ggml_nbytes(layer.ffn_up_b); } } } break; case LLM_ARCH_PERSIMMON: { - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); { ggml_backend_type backend_norm; @@ -2814,31 +2815,31 @@ static void llm_load_tensors( const ggml_backend_type backend = int(i) < i_gpu_start ? GGML_BACKEND_CPU : LLAMA_BACKEND_OFFLOAD; const ggml_backend_type backend_split = int(i) < i_gpu_start ? GGML_BACKEND_CPU : LLAMA_BACKEND_OFFLOAD_SPLIT; auto & layer = model.layers[i]; - layer.attn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd}, backend); - layer.attn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_NORM, "bias", i), {n_embd}, backend); - layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, backend_split); - layer.bqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "bias", i), {n_embd + 2*n_embd_gqa}, backend_split); - layer.wo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, backend_split); - layer.bo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "bias", i), {n_embd}, backend_split); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split); - layer.b2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend_split); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); - layer.b3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff}, backend_split); - layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); - layer.ffn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "bias", i), {n_embd}, backend); + layer.attn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd}, backend); + layer.attn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_NORM, "bias", i), {n_embd}, backend); + layer.wqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, backend_split); + layer.bqkv = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_QKV, "bias", i), {n_embd + 2*n_embd_gqa}, backend_split); + layer.wo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, backend_split); + layer.bo = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_OUT, "bias", i), {n_embd}, backend_split); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split); + layer.ffn_down_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend_split); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_up_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff}, backend_split); + layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); + layer.ffn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "bias", i), {n_embd}, backend); layer.attn_q_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_Q_NORM, "weight", i), {64}, backend); - layer.attn_q_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_Q_NORM, "bias", i), {64}, backend); + layer.attn_q_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_Q_NORM, "bias", i), {64}, backend); layer.attn_k_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_K_NORM, "weight", i), {64}, backend); - layer.attn_k_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_K_NORM, "bias", i), {64}, backend); + layer.attn_k_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_ATTN_K_NORM, "bias", i), {64}, backend); } } break; case LLM_ARCH_BLOOM: { // TODO: CPU-only for now - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); - model.tok_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD_NORM, "weight"), {n_embd}, GGML_BACKEND_CPU); - model.tok_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD_NORM, "bias"), {n_embd}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.tok_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD_NORM, "weight"), {n_embd}, GGML_BACKEND_CPU); + model.tok_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD_NORM, "bias"), {n_embd}, GGML_BACKEND_CPU); // output { @@ -2897,11 +2898,11 @@ static void llm_load_tensors( layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); layer.ffn_norm_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "bias", i), {n_embd}, backend); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split); - layer.b2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend_split); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), {n_ff, n_embd}, backend_split); + layer.ffn_down_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "bias", i), {n_embd}, backend_split); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); - layer.b3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff}, backend_split); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_up_b = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "bias", i), {n_ff}, backend_split); if (backend == GGML_BACKEND_GPU) { vram_weights += @@ -2909,14 +2910,14 @@ static void llm_load_tensors( ggml_nbytes(layer.wqkv) + ggml_nbytes(layer.bqkv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.bo) + ggml_nbytes(layer.ffn_norm) + ggml_nbytes(layer.ffn_norm_b) + - ggml_nbytes(layer.w3) + ggml_nbytes(layer.b3) + - ggml_nbytes(layer.w2) + ggml_nbytes(layer.b2); + ggml_nbytes(layer.ffn_up) + ggml_nbytes(layer.ffn_up_b) + + ggml_nbytes(layer.ffn_down) + ggml_nbytes(layer.ffn_down_b); } } } break; case LLM_ARCH_MPT: { - model.tok_embeddings = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); + model.tok_embd = ml.create_tensor(ctx, tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, GGML_BACKEND_CPU); // output { @@ -2967,8 +2968,8 @@ static void llm_load_tensors( layer.ffn_norm = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, backend); - layer.w2 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); - layer.w3 = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); + layer.ffn_down = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_DOWN, "weight", i), { n_ff, n_embd}, backend_split); + layer.ffn_up = ml.create_tensor(ctx, tn(LLM_TENSOR_FFN_UP, "weight", i), {n_embd, n_ff}, backend_split); if (backend == GGML_BACKEND_GPU) { vram_weights += @@ -2976,8 +2977,8 @@ static void llm_load_tensors( ggml_nbytes(layer.wqkv) + ggml_nbytes(layer.wo) + ggml_nbytes(layer.ffn_norm) + - ggml_nbytes(layer.w2) + - ggml_nbytes(layer.w3); + ggml_nbytes(layer.ffn_down) + + ggml_nbytes(layer.ffn_up); } } } break; @@ -3007,10 +3008,10 @@ static void llm_load_tensors( #ifdef GGML_USE_CUBLAS const int max_backend_supported_layers = hparams.n_layer + 3; - const int max_offloadable_layers = hparams.n_layer + 3; -#elif defined(GGML_USE_CLBLAST) + const int max_offloadable_layers = hparams.n_layer + 3; +#elif GGML_USE_CLBLAST const int max_backend_supported_layers = hparams.n_layer + 1; - const int max_offloadable_layers = hparams.n_layer + 1; + const int max_offloadable_layers = hparams.n_layer + 1; #endif // GGML_USE_CUBLAS LLAMA_LOG_INFO("%s: offloaded %d/%d layers to GPU\n", __func__, std::min(n_gpu_layers, max_offloadable_layers), max_backend_supported_layers); @@ -3089,9 +3090,359 @@ static bool llama_model_load( return true; } +using llm_build_cb = std::function; + +enum llm_rope_type { + LLM_ROPE, + LLM_ROPE_NEOX, + LLM_ROPE_GLM, +}; + +static struct ggml_tensor * llm_build_inp_embd( + struct ggml_context * ctx, + const llama_batch & batch, + struct ggml_tensor * tok_embd, + int64_t n_embd, + int32_t n_tokens, + const llm_build_cb & cb) { + struct ggml_tensor * inpL; + + if (batch.token) { + struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx, GGML_TYPE_I32, n_tokens); + cb(inp_tokens, "inp_tokens", -1); + + inpL = ggml_get_rows(ctx, tok_embd, inp_tokens); + } else { +#ifdef GGML_USE_MPI + GGML_ASSERT(false && "not implemented"); +#endif + + inpL = ggml_new_tensor_2d(ctx, GGML_TYPE_F32, n_embd, n_tokens); + } + + return inpL; +} + +// Persimmon: n_rot = n_embd_head/2 +// Other: n_rot = n_embd_head +static void llm_build_k_shift( + const llama_context & lctx, + struct ggml_context * ctx, + struct ggml_cgraph * graph, + int64_t n_rot, + llm_rope_type type, + const llm_build_cb & cb) { + const auto & model = lctx.model; + const auto & kv_self = lctx.kv_self; + const auto & cparams = lctx.cparams; + + const auto & hparams = model.hparams; + + const int64_t n_layer = hparams.n_layer; + const int64_t n_head_kv = hparams.n_head_kv; + const int64_t n_embd_gqa = hparams.n_embd_gqa(); + const int64_t n_embd_head = hparams.n_embd_head(); + + const int64_t n_ctx = lctx.cparams.n_ctx; + + const float freq_base = cparams.rope_freq_base; + const float freq_scale = cparams.rope_freq_scale; + + GGML_ASSERT(n_embd_head % n_rot == 0); + + struct ggml_tensor * K_shift = ggml_new_tensor_1d(ctx, GGML_TYPE_I32, n_ctx); + cb(K_shift, "K_shift", -1); + + int rope_type = 0; + + switch (type) { + case LLM_ROPE: rope_type = 0; break; + case LLM_ROPE_NEOX: rope_type = 2; break; + case LLM_ROPE_GLM: rope_type = 4; break; + } + + for (int il = 0; il < n_layer; ++il) { + struct ggml_tensor * tmp = + // we rotate only the first n_rot dimensions + ggml_rope_custom_inplace(ctx, + ggml_view_3d(ctx, kv_self.k, + n_rot, n_head_kv, n_ctx, + ggml_element_size(kv_self.k)*n_embd_head, + ggml_element_size(kv_self.k)*n_embd_gqa, + ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il), + K_shift, n_rot, rope_type, 0, freq_base, freq_scale); + cb(tmp, "K_shifted", il); + ggml_build_forward_expand(graph, tmp); + } +} + +static void llm_build_kv_store( + const llama_context & lctx, + struct ggml_context * ctx, + struct ggml_cgraph * graph, + struct ggml_tensor * k_cur, + struct ggml_tensor * v_cur, + int32_t n_tokens, + int32_t kv_head, + const llm_build_cb & cb, + int64_t il) { + const auto & model = lctx.model; + const auto & kv_self = lctx.kv_self; + const auto & cparams = lctx.cparams; + + const auto & hparams = model.hparams; + + const int64_t n_ctx = cparams.n_ctx; + const int64_t n_embd_gqa = hparams.n_embd_gqa(); + + // compute the transposed [n_tokens, n_embd] V matrix + struct ggml_tensor * v_cur_t = ggml_transpose(ctx, ggml_reshape_2d(ctx, v_cur, n_embd_gqa, n_tokens)); + //struct ggml_tensor * v_cur_t = ggml_transpose(ctx, v_cur); // TODO: reshape above is likely not needed + cb(v_cur_t, "v_cur_t", il); + + struct ggml_tensor * k_cache_view = ggml_view_1d(ctx, kv_self.k, n_tokens*n_embd_gqa, + (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); + cb(k_cache_view, "k_cache_view", il); + + struct ggml_tensor * v_cache_view = ggml_view_2d(ctx, kv_self.v, n_tokens, n_embd_gqa, + ( n_ctx)*ggml_element_size(kv_self.v), + (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); + cb(v_cache_view, "v_cache_view", il); + + // important: storing RoPE-ed version of K in the KV cache! + ggml_build_forward_expand(graph, ggml_cpy(ctx, k_cur, k_cache_view)); + ggml_build_forward_expand(graph, ggml_cpy(ctx, v_cur_t, v_cache_view)); +} + +enum llm_norm_type { + LLM_NORM, + LLM_NORM_RMS, +}; + +static struct ggml_tensor * llm_build_norm( + struct ggml_context * ctx, + struct ggml_tensor * cur, + struct ggml_tensor * mw, + struct ggml_tensor * mb, + llm_norm_type type, + float eps, + const llm_build_cb & cb, + int il) { + switch (type) { + case LLM_NORM: cur = ggml_norm (ctx, cur, eps); break; + case LLM_NORM_RMS: cur = ggml_rms_norm(ctx, cur, eps); break; + } + + if (mw || mb) { + cb(cur, "norm", il); + } + + if (mw) { + cur = ggml_mul(ctx, cur, mw); + if (mb) { + cb(cur, "norm_w", il); + } + } + + if (mb) { + cur = ggml_add(ctx, cur, mb); + } + + return cur; +} + +enum llm_ffn_op_type { + LLM_FFN_SILU, + LLM_FFN_GELU, + LLM_FFN_RELU, + LLM_FFN_RELU_SQR, +}; + +enum llm_ffn_gate_type { + LLM_FFN_SEQ, + LLM_FFN_PAR, // ffn_gate is parallel to ffn_up +}; + +static struct ggml_tensor * llm_build_ffn( + struct ggml_context * ctx, + struct ggml_tensor * cur, + struct ggml_tensor * up, + struct ggml_tensor * up_b, + struct ggml_tensor * gate, + struct ggml_tensor * gate_b, + struct ggml_tensor * down, + struct ggml_tensor * down_b, + llm_ffn_op_type type_op, + llm_ffn_gate_type type_gate, + const llm_build_cb & cb, + int il) { + struct ggml_tensor * tmp = ggml_mul_mat(ctx, up, cur); + cb(tmp, "ffn_up", il); + + if (up_b) { + tmp = ggml_add(ctx, tmp, up_b); + cb(tmp, "ffn_up_b", il); + } + + if (gate) { + switch (type_gate) { + case LLM_FFN_SEQ: + { + cur = ggml_mul_mat(ctx, gate, tmp); + cb(cur, "ffn_gate", il); + } break; + case LLM_FFN_PAR: + { + cur = ggml_mul_mat(ctx, gate, cur); + cb(cur, "ffn_gate", il); + } break; + } + + if (gate_b) { + cur = ggml_add(ctx, cur, gate_b); + cb(cur, "ffn_gate_b", il); + } + } else { + cur = tmp; + } + + switch (type_op) { + case LLM_FFN_SILU: + { + cur = ggml_silu(ctx, cur); + cb(cur, "ffn_silu", il); + } break; + case LLM_FFN_GELU: + { + cur = ggml_gelu(ctx, cur); + cb(cur, "ffn_gelu", il); + } break; + case LLM_FFN_RELU: + { + cur = ggml_relu(ctx, cur); + cb(cur, "ffn_relu", il); + } break; + case LLM_FFN_RELU_SQR: + { + cur = ggml_relu(ctx, cur); + cb(cur, "ffn_relu", il); + + cur = ggml_sqr(ctx, cur); + cb(cur, "ffn_sqr(relu)", il); + } break; + } + + if (type_gate == LLM_FFN_PAR) { + cur = ggml_mul(ctx, cur, tmp); + cb(cur, "ffn_gate_par", il); + } + + cur = ggml_mul_mat(ctx, down, cur); + if (down_b) { + cb(cur, "ffn_down", il); + } + + if (down_b) { + cur = ggml_add(ctx, cur, down_b); + } + + return cur; +} + +// if max_alibi_bias > 0 then apply ALiBi +static struct ggml_tensor * llm_build_kqv( + const llama_context & lctx, + struct ggml_context * ctx, + struct ggml_tensor * cur, + struct ggml_tensor * wo, + struct ggml_tensor * wo_b, + struct ggml_tensor * q_cur, + struct ggml_tensor * kq_scale, + struct ggml_tensor * kq_mask, + int32_t n_tokens, + int32_t n_kv, + float alibi_bias_max, + const llm_build_cb & cb, + int il) { + const auto & model = lctx.model; + const auto & kv_self = lctx.kv_self; + const auto & cparams = lctx.cparams; + + const auto & hparams = model.hparams; + + const int64_t n_ctx = cparams.n_ctx; + const int64_t n_embd = hparams.n_embd; + const int64_t n_head = hparams.n_head; + const int64_t n_head_kv = hparams.n_head_kv; + const int64_t n_embd_head = hparams.n_embd_head(); + const int64_t n_embd_gqa = hparams.n_embd_gqa(); + + struct ggml_tensor * q = ggml_permute(ctx, q_cur, 0, 2, 1, 3); + cb(q, "q", il); + + struct ggml_tensor * k = + ggml_view_3d(ctx, kv_self.k, + n_embd_head, n_kv, n_head_kv, + ggml_element_size(kv_self.k)*n_embd_gqa, + ggml_element_size(kv_self.k)*n_embd_head, + ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); + cb(k, "k", il); + + struct ggml_tensor * kq = ggml_mul_mat(ctx, k, q); + cb(kq, "kq", il); + + kq = ggml_scale(ctx, kq, kq_scale); + cb(kq, "kq_scaled", il); + + if (alibi_bias_max > 0.0f) { + // TODO: n_head or n_head_kv + // TODO: K-shift is likely not working + // TODO: change to ggml_add + kq = ggml_alibi(ctx, kq, /*n_past*/ 0, n_head, alibi_bias_max); + cb(kq, "kq_scaled_alibi", il); + } + + kq = ggml_add(ctx, kq, kq_mask); + cb(kq, "kq_masked", il); + + kq = ggml_soft_max(ctx, kq); + cb(kq, "kq_soft_max", il); + + // split cached v into n_head heads + struct ggml_tensor * v = + ggml_view_3d(ctx, kv_self.v, + n_kv, n_embd_head, n_head_kv, + ggml_element_size(kv_self.v)*n_ctx, + ggml_element_size(kv_self.v)*n_ctx*n_embd_head, + ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); + cb(v, "v", il); + + struct ggml_tensor * kqv = ggml_mul_mat(ctx, v, kq); + cb(kqv, "kqv", il); + + struct ggml_tensor * kqv_merged = ggml_permute(ctx, kqv, 0, 2, 1, 3); + cb(kqv_merged, "kqv_merged", il); + + cur = ggml_cont_2d(ctx, kqv_merged, n_embd, n_tokens); + cb(cur, "kqv_merged_cont", il); + + cur = ggml_mul_mat(ctx, wo, cur); + if (wo_b) { + cb(cur, "kqv_wo", il); + } + + if (wo_b) { + cur = ggml_add(ctx, cur, wo_b); + } + + return cur; +} + static struct ggml_cgraph * llm_build_llama( - llama_context & lctx, - const llama_batch & batch) { + llama_context & lctx, + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; const auto & cparams = lctx.cparams; @@ -3106,7 +3457,6 @@ static struct ggml_cgraph * llm_build_llama( const int64_t n_head = hparams.n_head; const int64_t n_head_kv = hparams.n_head_kv; const int64_t n_embd_head = hparams.n_embd_head(); - const int64_t n_embd_gqa = hparams.n_embd_gqa(); GGML_ASSERT(n_embd_head == hparams.n_rot); @@ -3114,13 +3464,11 @@ static struct ggml_cgraph * llm_build_llama( const float freq_scale = cparams.rope_freq_scale; const float norm_rms_eps = hparams.f_norm_rms_eps; - const int n_gpu_layers = model.n_gpu_layers; - const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; - const bool do_rope_shift = ggml_allocr_is_measure(lctx.alloc) || kv_self.has_shift; + const bool do_rope_shift = worst_case || kv_self.has_shift; //printf("n_kv = %d\n", n_kv); @@ -3139,314 +3487,81 @@ static struct ggml_cgraph * llm_build_llama( struct ggml_tensor * cur; struct ggml_tensor * inpL; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - - inpL = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - inpL = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inpL); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inpL->data, batch.embd, n_tokens * n_embd * ggml_element_size(inpL)); - } - } - - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - - // offload functions set the tensor output backend to GPU - // tensors are GPU-accelerated if any input or the output has been offloaded - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (n_gpu_layers > n_layer) { - offload_func_nr = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 1) { - offload_func_v = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 2) { - offload_func_kq = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + // inp_pos - contains the positions + struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + cb(inp_pos, "inp_pos", -1); // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd_head))); - } + cb(KQ_scale, "KQ_scale", -1); // KQ_mask (mask for 1 head, it will be broadcasted to all heads) struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - offload_func_kq(KQ_mask); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); - - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - // KQ_pos - contains the positions - struct ggml_tensor * KQ_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - offload_func_kq(KQ_pos); - ggml_set_name(KQ_pos, "KQ_pos"); - ggml_allocr_alloc(lctx.alloc, KQ_pos); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) KQ_pos->data; - for (int i = 0; i < n_tokens; ++i) { - data[i] = batch.pos[i]; - } - } + cb(KQ_mask, "KQ_mask", -1); // shift the entire K-cache if needed if (do_rope_shift) { - struct ggml_tensor * K_shift = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_ctx); - offload_func_kq(K_shift); - ggml_set_name(K_shift, "K_shift"); - ggml_allocr_alloc(lctx.alloc, K_shift); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) K_shift->data; - for (int i = 0; i < n_ctx; ++i) { - data[i] = kv_self.cells[i].delta; - } - } - - for (int il = 0; il < n_layer; ++il) { - struct ggml_tensor * tmp = - ggml_rope_custom_inplace(ctx0, - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_head_kv, n_ctx, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il), - K_shift, n_embd_head, 0, 0, freq_base, freq_scale); - offload_func_kq(tmp); - ggml_build_forward_expand(gf, tmp); - } + llm_build_k_shift(lctx, ctx0, gf, n_embd_head, LLM_ROPE, cb); } for (int il = 0; il < n_layer; ++il) { - ggml_format_name(inpL, "layer_inp_%d", il); - - offload_func_t offload_func = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (il >= i_gpu_start) { - offload_func = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS - struct ggml_tensor * inpSA = inpL; // norm - { - cur = ggml_rms_norm(ctx0, inpL, norm_rms_eps); - offload_func(cur); - ggml_set_name(cur, "rms_norm_0"); - - // cur = cur*attn_norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.layers[il].attn_norm); - offload_func(cur); - ggml_set_name(cur, "attention_norm_0"); - } + cur = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, il); + cb(cur, "attn_norm", il); // self-attention { // compute Q and K and RoPE them - struct ggml_tensor * tmpk = ggml_mul_mat(ctx0, model.layers[il].wk, cur); - offload_func_kq(tmpk); - ggml_set_name(tmpk, "tmpk"); + struct ggml_tensor * Qcur = ggml_mul_mat(ctx0, model.layers[il].wq, cur); + cb(Qcur, "Qcur", il); - struct ggml_tensor * tmpq = ggml_mul_mat(ctx0, model.layers[il].wq, cur); - offload_func_kq(tmpq); - ggml_set_name(tmpq, "tmpq"); + struct ggml_tensor * Kcur = ggml_mul_mat(ctx0, model.layers[il].wk, cur); + cb(Kcur, "Kcur", il); - struct ggml_tensor * Kcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpk, n_embd_head, n_head_kv, n_tokens), KQ_pos, n_embd_head, 0, 0, freq_base, freq_scale); - offload_func_kq(Kcur); - ggml_set_name(Kcur, "Kcur"); + struct ggml_tensor * Vcur = ggml_mul_mat(ctx0, model.layers[il].wv, cur); + cb(Vcur, "Vcur", il); - struct ggml_tensor * Qcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpq, n_embd_head, n_head, n_tokens), KQ_pos, n_embd_head, 0, 0, freq_base, freq_scale); - offload_func_kq(Qcur); - ggml_set_name(Qcur, "Qcur"); + Qcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens), inp_pos, n_embd_head, 0, 0, freq_base, freq_scale); + cb(Qcur, "Qcur", il); - // store key and value to memory - { - // compute the transposed [n_tokens, n_embd] V matrix + Kcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens), inp_pos, n_embd_head, 0, 0, freq_base, freq_scale); + cb(Kcur, "Kcur", il); - struct ggml_tensor * tmpv = ggml_mul_mat(ctx0, model.layers[il].wv, cur); - offload_func_v(tmpv); - ggml_set_name(tmpv, "tmpv"); + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, tmpv, n_embd_gqa, n_tokens)); - offload_func_v(Vcur); - ggml_set_name(Vcur, "Vcur"); - - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - offload_func_kq(k); - ggml_set_name(k, "k"); - - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - ggml_set_name(v, "v"); - - // important: storing RoPE-ed version of K in the KV cache! - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); - offload_func_kq(Q); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - offload_func_kq(K); - ggml_set_name(K, "K"); - - // K * Q - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - // KQ_scaled = KQ / sqrt(n_embd_head) - // KQ_scaled shape [n_kv, n_tokens, n_head, 1] - struct ggml_tensor * KQ_scaled = ggml_scale(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - // KQ_masked = mask_past(KQ_scaled) - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled, KQ_mask); - offload_func_kq(KQ_masked); - ggml_set_name(KQ_masked, "KQ_masked"); - - // KQ = soft_max(KQ_masked) - struct ggml_tensor * KQ_soft_max = ggml_soft_max(ctx0, KQ_masked); - offload_func_v(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - // split cached V into n_head heads - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - offload_func_v(V); - ggml_set_name(V, "V"); - -#if 1 - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); -#else - // make V contiguous in memory to speed up the matmul, however we waste time on the copy - // on M1 this is faster for the perplexity computation, but ~5% slower for the single-token generation - // is there a better way? - struct ggml_tensor * V_cont = ggml_cpy(ctx0, V, ggml_new_tensor_3d(ctx0, kv_self.v->type, n_ctx, n_embd_head, n_head)); - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V_cont, KQ_soft_max); -#endif - - // KQV_merged = KQV.permute(0, 2, 1, 3) - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - // cur = KQV_merged.contiguous().view(n_embd, n_tokens) - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); - - // projection (no bias) - cur = ggml_mul_mat(ctx0, - model.layers[il].wo, - cur); - offload_func(cur); - ggml_set_name(cur, "result_wo"); + cur = llm_build_kqv(lctx, ctx0, cur, + model.layers[il].wo, NULL, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, -1.0f, cb, il); + cb(cur, "kqv_out", il); } - struct ggml_tensor * inpFF = ggml_add(ctx0, cur, inpSA); - offload_func(inpFF); - ggml_set_name(inpFF, "inpFF"); + struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA); + cb(ffn_inp, "ffn_inp", il); // feed-forward network { - // norm - { - cur = ggml_rms_norm(ctx0, inpFF, norm_rms_eps); - offload_func(cur); - ggml_set_name(cur, "rms_norm_1"); + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, il); + cb(cur, "ffn_norm", il); - // cur = cur*ffn_norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.layers[il].ffn_norm); - offload_func(cur); - ggml_set_name(cur, "ffn_norm"); - } - - struct ggml_tensor * tmp = ggml_mul_mat(ctx0, - model.layers[il].w3, - cur); - offload_func(tmp); - ggml_set_name(tmp, "result_w3"); - - cur = ggml_mul_mat(ctx0, - model.layers[il].w1, - cur); - offload_func(cur); - ggml_set_name(cur, "result_w1"); - - // SILU activation - cur = ggml_silu(ctx0, cur); - offload_func(cur); - ggml_set_name(cur, "silu"); - - cur = ggml_mul(ctx0, cur, tmp); - offload_func(cur); - ggml_set_name(cur, "silu_x_result_w3"); - - cur = ggml_mul_mat(ctx0, - model.layers[il].w2, - cur); - offload_func(cur); - ggml_set_name(cur, "result_w2"); + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, NULL, + model.layers[il].ffn_gate, NULL, + model.layers[il].ffn_down, NULL, + LLM_FFN_SILU, LLM_FFN_PAR, cb, il); + cb(cur, "ffn_out", il); } - cur = ggml_add(ctx0, cur, inpFF); - offload_func(cur); - ggml_set_name(cur, "inpFF_+_result_w2"); + cur = ggml_add(ctx0, cur, ffn_inp); + cb(cur, "l_out", il); // input for next layer inpL = cur; @@ -3454,21 +3569,14 @@ static struct ggml_cgraph * llm_build_llama( cur = inpL; - // norm - { - cur = ggml_rms_norm(ctx0, cur, norm_rms_eps); - offload_func_nr(cur); - ggml_set_name(cur, "rms_norm_2"); - - // cur = cur*norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.output_norm); - // offload_func_nr(cur); // TODO CPU + GPU mirrored backend - ggml_set_name(cur, "result_norm"); - } + cur = llm_build_norm(ctx0, cur, + model.output_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, -1); + cb(cur, "result_norm", -1); // lm_head cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); ggml_build_forward_expand(gf, cur); @@ -3479,7 +3587,9 @@ static struct ggml_cgraph * llm_build_llama( static struct ggml_cgraph * llm_build_baichaun( llama_context & lctx, - const llama_batch & batch) { + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; const auto & cparams = lctx.cparams; @@ -3494,7 +3604,6 @@ static struct ggml_cgraph * llm_build_baichaun( const int64_t n_head = hparams.n_head; const int64_t n_head_kv = hparams.n_head_kv; const int64_t n_embd_head = hparams.n_embd_head(); - const int64_t n_embd_gqa = hparams.n_embd_gqa(); GGML_ASSERT(n_embd_head == hparams.n_rot); @@ -3502,13 +3611,11 @@ static struct ggml_cgraph * llm_build_baichaun( const float freq_scale = cparams.rope_freq_scale; const float norm_rms_eps = hparams.f_norm_rms_eps; - const int n_gpu_layers = model.n_gpu_layers; - const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; - const bool do_rope_shift = ggml_allocr_is_measure(lctx.alloc) || kv_self.has_shift; + const bool do_rope_shift = worst_case || kv_self.has_shift; auto & buf_compute = lctx.buf_compute; @@ -3525,331 +3632,91 @@ static struct ggml_cgraph * llm_build_baichaun( struct ggml_tensor * cur; struct ggml_tensor * inpL; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - - inpL = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - inpL = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inpL); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inpL->data, batch.embd, n_tokens * n_embd * ggml_element_size(inpL)); - } - } - - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - - // offload functions set the tensor output backend to GPU - // tensors are GPU-accelerated if any input or the output has been offloaded - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (n_gpu_layers > n_layer) { - offload_func_nr = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 1) { - offload_func_v = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 2) { - offload_func_kq = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + // inp_pos - contains the positions + struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + cb(inp_pos, "inp_pos", -1); // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd)/n_head)); - } + cb(KQ_scale, "KQ_scale", -1); // KQ_mask (mask for 1 head, it will be broadcasted to all heads) struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - offload_func_kq(KQ_mask); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); - - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - // KQ_pos - contains the positions - struct ggml_tensor * KQ_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - offload_func_kq(KQ_pos); - ggml_set_name(KQ_pos, "KQ_pos"); - ggml_allocr_alloc(lctx.alloc, KQ_pos); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) KQ_pos->data; - for (int i = 0; i < n_tokens; ++i) { - data[i] = batch.pos[i]; - } - } + cb(KQ_mask, "KQ_mask", -1); // shift the entire K-cache if needed if (do_rope_shift) { - struct ggml_tensor * K_shift = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_ctx); - offload_func_kq(K_shift); - ggml_set_name(K_shift, "K_shift"); - ggml_allocr_alloc(lctx.alloc, K_shift); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) K_shift->data; - for (int i = 0; i < n_ctx; ++i) { - data[i] = kv_self.cells[i].delta; - } - } - - for (int il = 0; il < n_layer; ++il) { - struct ggml_tensor * tmp = - ggml_rope_custom_inplace(ctx0, - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_head_kv, n_ctx, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il), - K_shift, n_embd_head, 0, 0, freq_base, freq_scale); - offload_func_kq(tmp); - ggml_build_forward_expand(gf, tmp); - } + llm_build_k_shift(lctx, ctx0, gf, n_embd_head, LLM_ROPE, cb); } for (int il = 0; il < n_layer; ++il) { - ggml_format_name(inpL, "layer_inp_%d", il); - - offload_func_t offload_func = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (il >= i_gpu_start) { - offload_func = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS - struct ggml_tensor * inpSA = inpL; - // norm - { - cur = ggml_rms_norm(ctx0, inpL, norm_rms_eps); - offload_func(cur); - ggml_set_name(cur, "rms_norm_0"); - - // cur = cur*attn_norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.layers[il].attn_norm); - offload_func(cur); - ggml_set_name(cur, "attention_norm_0"); - } + cur = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, il); + cb(cur, "attn_norm", il); // self-attention { - // compute Q and K and RoPE them - struct ggml_tensor * tmpk = ggml_mul_mat(ctx0, model.layers[il].wk, cur); - offload_func_kq(tmpk); - ggml_set_name(tmpk, "tmpk"); + struct ggml_tensor * Qcur = ggml_mul_mat(ctx0, model.layers[il].wq, cur); + cb(Qcur, "Qcur", il); - struct ggml_tensor * tmpq = ggml_mul_mat(ctx0, model.layers[il].wq, cur); - offload_func_kq(tmpq); - ggml_set_name(tmpq, "tmpq"); + struct ggml_tensor * Kcur = ggml_mul_mat(ctx0, model.layers[il].wk, cur); + cb(Kcur, "Kcur", il); + + struct ggml_tensor * Vcur = ggml_mul_mat(ctx0, model.layers[il].wv, cur); + cb(Vcur, "Vcur", il); - struct ggml_tensor * Kcur; - struct ggml_tensor * Qcur; switch (model.type) { case MODEL_7B: - Kcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpk, n_embd_head, n_head_kv, n_tokens), KQ_pos, n_embd_head, 0, 0, freq_base, freq_scale); - Qcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, tmpq, n_embd_head, n_head, n_tokens), KQ_pos, n_embd_head, 0, 0, freq_base, freq_scale); + Qcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens), inp_pos, n_embd_head, 0, 0, freq_base, freq_scale); + Kcur = ggml_rope_custom(ctx0, ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens), inp_pos, n_embd_head, 0, 0, freq_base, freq_scale); break; case MODEL_13B: - Kcur = ggml_reshape_3d(ctx0, tmpk, n_embd/n_head, n_head, n_tokens); - Qcur = ggml_reshape_3d(ctx0, tmpq, n_embd/n_head, n_head, n_tokens); + Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd/n_head, n_head, n_tokens); + Kcur = ggml_reshape_3d(ctx0, Kcur, n_embd/n_head, n_head, n_tokens); break; default: GGML_ASSERT(false); } + cb(Qcur, "Qcur", il); + cb(Kcur, "Kcur", il); - offload_func_kq(Kcur); - ggml_set_name(Kcur, "Kcur"); + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); - offload_func_kq(Qcur); - ggml_set_name(Qcur, "Qcur"); + // apply ALiBi for 13B model + const float alibi_bias_max = model.type == MODEL_13B ? 8.0f : -1.0f; - // store key and value to memory - { - // compute the transposed [n_tokens, n_embd] V matrix - - struct ggml_tensor * tmpv = ggml_mul_mat(ctx0, model.layers[il].wv, cur); - offload_func_v(tmpv); - ggml_set_name(tmpv, "tmpv"); - - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, tmpv, n_embd_gqa, n_tokens)); - offload_func_v(Vcur); - ggml_set_name(Vcur, "Vcur"); - - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - offload_func_kq(k); - ggml_set_name(k, "k"); - - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - ggml_set_name(v, "v"); - - // important: storing RoPE-ed version of K in the KV cache! - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); - offload_func_kq(Q); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - offload_func_kq(K); - ggml_set_name(K, "K"); - - // K * Q - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - // KQ_scaled = KQ / sqrt(n_embd_head) - // KQ_scaled shape [n_past + n_tokens, n_tokens, n_head, 1] - struct ggml_tensor * KQ_scaled = ggml_scale(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - struct ggml_tensor * KQ_masked; - struct ggml_tensor * KQ_scaled_alibi; - - switch (model.type) { - case MODEL_7B: - KQ_masked = ggml_add(ctx0, KQ_scaled, KQ_mask); - break; - case MODEL_13B: - // TODO: replace with ggml_add() - KQ_scaled_alibi = ggml_alibi(ctx0, KQ_scaled, /*n_past*/ 0, n_head, 8); - ggml_set_name(KQ_scaled_alibi, "KQ_scaled_alibi"); - KQ_masked = ggml_add(ctx0, KQ_scaled_alibi, KQ_mask); - break; - default: - GGML_ASSERT(false); - } - - // KQ = soft_max(KQ_masked) - struct ggml_tensor * KQ_soft_max = ggml_soft_max(ctx0, KQ_masked); - offload_func_v(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - // split cached V into n_head heads - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - offload_func_v(V); - ggml_set_name(V, "V"); - - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); - - // KQV_merged = KQV.permute(0, 2, 1, 3) - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - // cur = KQV_merged.contiguous().view(n_embd, n_tokens) - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); - - // projection (no bias) - cur = ggml_mul_mat(ctx0, - model.layers[il].wo, - cur); - offload_func(cur); - ggml_set_name(cur, "result_wo"); + cur = llm_build_kqv(lctx, ctx0, cur, + model.layers[il].wo, NULL, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, alibi_bias_max, cb, il); + cb(cur, "kqv_out", il); } - struct ggml_tensor * inpFF = ggml_add(ctx0, cur, inpSA); - offload_func(inpFF); - ggml_set_name(inpFF, "inpFF"); + struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA); + cb(ffn_inp, "ffn_inp", il); // feed-forward network { - // norm - { - cur = ggml_rms_norm(ctx0, inpFF, norm_rms_eps); - offload_func(cur); - ggml_set_name(cur, "rms_norm_1"); + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, il); + cb(cur, "ffn_norm", il); - // cur = cur*ffn_norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.layers[il].ffn_norm); - offload_func(cur); - ggml_set_name(cur, "ffn_norm"); - } - - struct ggml_tensor * tmp = ggml_mul_mat(ctx0, - model.layers[il].w3, - cur); - offload_func(tmp); - ggml_set_name(tmp, "result_w3"); - - cur = ggml_mul_mat(ctx0, - model.layers[il].w1, - cur); - offload_func(cur); - ggml_set_name(cur, "result_w1"); - - // SILU activation - cur = ggml_silu(ctx0, cur); - offload_func(cur); - ggml_set_name(cur, "silu"); - - cur = ggml_mul(ctx0, cur, tmp); - offload_func(cur); - ggml_set_name(cur, "silu_x_result_w3"); - - cur = ggml_mul_mat(ctx0, - model.layers[il].w2, - cur); - offload_func(cur); - ggml_set_name(cur, "result_w2"); + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, NULL, + model.layers[il].ffn_gate, NULL, + model.layers[il].ffn_down, NULL, + LLM_FFN_SILU, LLM_FFN_PAR, cb, il); + cb(cur, "ffn_out", il); } - cur = ggml_add(ctx0, cur, inpFF); - offload_func(cur); - ggml_set_name(cur, "inpFF_+_result_w2"); + cur = ggml_add(ctx0, cur, ffn_inp); + cb(cur, "l_out", il); // input for next layer inpL = cur; @@ -3857,366 +3724,14 @@ static struct ggml_cgraph * llm_build_baichaun( cur = inpL; - // norm - { - cur = ggml_rms_norm(ctx0, cur, norm_rms_eps); - offload_func_nr(cur); - ggml_set_name(cur, "rms_norm_2"); - - // cur = cur*norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.output_norm); - // offload_func_nr(cur); // TODO CPU + GPU mirrored backend - ggml_set_name(cur, "result_norm"); - } + cur = llm_build_norm(ctx0, cur, + model.output_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, -1); + cb(cur, "result_norm", -1); // lm_head cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); - - ggml_build_forward_expand(gf, cur); - - ggml_free(ctx0); - - return gf; -} - -static struct ggml_cgraph * llm_build_refact( - llama_context & lctx, - const llama_batch & batch) { - const auto & model = lctx.model; - const auto & hparams = model.hparams; - const auto & cparams = lctx.cparams; - - const auto & kv_self = lctx.kv_self; - - GGML_ASSERT(!!kv_self.ctx); - - const int64_t n_embd = hparams.n_embd; - const int64_t n_layer = hparams.n_layer; - const int64_t n_ctx = cparams.n_ctx; - const int64_t n_head = hparams.n_head; - const int64_t n_head_kv = hparams.n_head_kv; - const int64_t n_embd_head = hparams.n_embd_head(); - const int64_t n_embd_gqa = hparams.n_embd_gqa(); - - const float norm_rms_eps = hparams.f_norm_rms_eps; - - const int n_gpu_layers = model.n_gpu_layers; - - const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; - - // printf("n_kv = %d\n", n_kv); - - auto & buf_compute = lctx.buf_compute; - - struct ggml_init_params params = { - /*.mem_size =*/ buf_compute.size, - /*.mem_buffer =*/ buf_compute.data, - /*.no_alloc =*/ true, - }; - - struct ggml_context * ctx0 = ggml_init(params); - - ggml_cgraph * gf = ggml_new_graph(ctx0); - - struct ggml_tensor * cur; - struct ggml_tensor * inpL; - - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - - inpL = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - inpL = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inpL); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inpL->data, batch.embd, n_tokens * n_embd * ggml_element_size(inpL)); - } - } - - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - - // offload functions set the tensor output backend to GPU - // tensors are GPU-accelerated if any input or the output has been offloaded - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (n_gpu_layers > n_layer) { - offload_func_nr = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 1) { - offload_func_v = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 2) { - offload_func_kq = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS - - // KQ_scale - struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd_head))); - } - - // KQ_mask (mask for 1 head, it will be broadcasted to all heads) - struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - offload_func_kq(KQ_mask); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); - - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - for (int il = 0; il < n_layer; ++il) { - ggml_format_name(inpL, "layer_inp_%d", il); - - offload_func_t offload_func = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (il >= i_gpu_start) { - offload_func = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS - - struct ggml_tensor * inpSA = inpL; - - // norm - { - cur = ggml_rms_norm(ctx0, inpL, norm_rms_eps); - offload_func(cur); - ggml_set_name(cur, "rms_norm_0"); - - // cur = cur*attn_norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.layers[il].attn_norm); - offload_func(cur); - ggml_set_name(cur, "attention_norm_0"); - } - - // self-attention - { - // compute Q and K - struct ggml_tensor * tmpk = ggml_mul_mat(ctx0, model.layers[il].wk, cur); - offload_func_kq(tmpk); - ggml_set_name(tmpk, "tmpk"); - - struct ggml_tensor * tmpq = ggml_mul_mat(ctx0, model.layers[il].wq, cur); - offload_func_kq(tmpq); - ggml_set_name(tmpq, "tmpq"); - - struct ggml_tensor * Kcur = ggml_reshape_3d(ctx0, tmpk, n_embd_head, n_head_kv, n_tokens); - offload_func_kq(Kcur); - ggml_set_name(Kcur, "Kcur"); - - struct ggml_tensor * Qcur = ggml_reshape_3d(ctx0, tmpq, n_embd_head, n_head, n_tokens); - offload_func_kq(Qcur); - ggml_set_name(Qcur, "Qcur"); - - // store key and value to memory - { - // compute the transposed [n_tokens, n_embd] V matrix - - struct ggml_tensor * tmpv = ggml_mul_mat(ctx0, model.layers[il].wv, cur); - offload_func_v(tmpv); - ggml_set_name(tmpv, "tmpv"); - - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, tmpv, n_embd_gqa, n_tokens)); - offload_func_v(Vcur); - ggml_set_name(Vcur, "Vcur"); - - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - offload_func_kq(k); - ggml_set_name(k, "k"); - - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - ggml_set_name(v, "v"); - - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); - offload_func_kq(Q); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - offload_func_kq(K); - ggml_set_name(K, "K"); - - // K * Q - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - // KQ_scaled = KQ / sqrt(n_embd_head) - // KQ_scaled shape [n_kv, n_tokens, n_head, 1] - struct ggml_tensor * KQ_scaled = ggml_scale(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - // KQ_masked = mask_past(KQ_scaled) - struct ggml_tensor * KQ_scaled_alibi = ggml_alibi(ctx0, KQ_scaled, /*n_past*/ 0, n_head, 8); - ggml_set_name(KQ_scaled_alibi, "KQ_scaled_alibi"); - - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled_alibi, KQ_mask); - offload_func_kq(KQ_masked); - ggml_set_name(KQ_masked, "KQ_masked"); - - // KQ = soft_max(KQ_masked) - struct ggml_tensor * KQ_soft_max = ggml_soft_max(ctx0, KQ_masked); - offload_func_v(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - // split cached V into n_head heads - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - offload_func_v(V); - ggml_set_name(V, "V"); - -#if 1 - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); -#else - // make V contiguous in memory to speed up the matmul, however we waste time on the copy - // on M1 this is faster for the perplexity computation, but ~5% slower for the single-token generation - // is there a better way? - struct ggml_tensor * V_cont = ggml_cpy(ctx0, V, ggml_new_tensor_3d(ctx0, kv_self.v->type, n_ctx, n_embd_head, n_head)); - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V_cont, KQ_soft_max); -#endif - - // KQV_merged = KQV.permute(0, 2, 1, 3) - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - // cur = KQV_merged.contiguous().view(n_embd, n_tokens) - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); - - // projection (no bias) - cur = ggml_mul_mat(ctx0, - model.layers[il].wo, - cur); - offload_func(cur); - ggml_set_name(cur, "result_wo"); - } - - struct ggml_tensor * inpFF = ggml_add(ctx0, cur, inpSA); - offload_func(inpFF); - ggml_set_name(inpFF, "inpFF"); - - // feed-forward network - { - // norm - { - cur = ggml_rms_norm(ctx0, inpFF, norm_rms_eps); - offload_func(cur); - ggml_set_name(cur, "rms_norm_1"); - - // cur = cur*ffn_norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.layers[il].ffn_norm); - offload_func(cur); - ggml_set_name(cur, "ffn_norm"); - } - - struct ggml_tensor * tmp = ggml_mul_mat(ctx0, - model.layers[il].w3, - cur); - offload_func(tmp); - ggml_set_name(tmp, "result_w3"); - - cur = ggml_mul_mat(ctx0, - model.layers[il].w1, - cur); - offload_func(cur); - ggml_set_name(cur, "result_w1"); - - // SILU activation - cur = ggml_silu(ctx0, cur); - offload_func(cur); - ggml_set_name(cur, "silu"); - - cur = ggml_mul(ctx0, cur, tmp); - offload_func(cur); - ggml_set_name(cur, "silu_x_result_w3"); - - cur = ggml_mul_mat(ctx0, - model.layers[il].w2, - cur); - offload_func(cur); - ggml_set_name(cur, "result_w2"); - } - - cur = ggml_add(ctx0, cur, inpFF); - offload_func(cur); - ggml_set_name(cur, "inpFF_+_result_w2"); - - // input for next layer - inpL = cur; - } - - cur = inpL; - - // norm - { - cur = ggml_rms_norm(ctx0, cur, norm_rms_eps); - offload_func_nr(cur); - ggml_set_name(cur, "rms_norm_2"); - - // cur = cur*norm(broadcasted) - cur = ggml_mul(ctx0, cur, model.output_norm); - // offload_func_nr(cur); // TODO CPU + GPU mirrored backend - ggml_set_name(cur, "result_norm"); - } - - // lm_head - cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); ggml_build_forward_expand(gf, cur); @@ -4227,7 +3742,9 @@ static struct ggml_cgraph * llm_build_refact( static struct ggml_cgraph * llm_build_falcon( llama_context & lctx, - const llama_batch & batch) { + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; const auto & cparams = lctx.cparams; @@ -4250,13 +3767,11 @@ static struct ggml_cgraph * llm_build_falcon( const float freq_scale = cparams.rope_freq_scale; const float norm_eps = hparams.f_norm_eps; - const int n_gpu_layers = model.n_gpu_layers; - const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; - const bool do_rope_shift = ggml_allocr_is_measure(lctx.alloc) || kv_self.has_shift; + const bool do_rope_shift = worst_case || kv_self.has_shift; //printf("kv_head = %d, n_kv = %d, n_tokens = %d, n_ctx = %d, is_measure = %d, has_shift = %d\n", // kv_head, n_kv, n_tokens, n_ctx, ggml_allocr_is_measure(lctx.alloc), kv_self.has_shift); @@ -4276,294 +3791,94 @@ static struct ggml_cgraph * llm_build_falcon( struct ggml_tensor * cur; struct ggml_tensor * inpL; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - - inpL = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - inpL = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inpL); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inpL->data, batch.embd, n_tokens * n_embd * ggml_element_size(inpL)); - } - } - - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - - // offload functions set the tensor output backend to GPU - // tensors are GPU-accelerated if any input or the output has been offloaded - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (n_gpu_layers > n_layer) { - offload_func_nr = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 1) { - offload_func_v = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 2) { - offload_func_kq = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + // inp_pos - contains the positions + struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + cb(inp_pos, "inp_pos", -1); // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd)/n_head)); - } + cb(KQ_scale, "KQ_scale", -1); // KQ_mask (mask for 1 head, it will be broadcasted to all heads) struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - offload_func_kq(KQ_mask); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); - - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - // KQ_pos - contains the positions - struct ggml_tensor * KQ_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - offload_func_kq(KQ_pos); - ggml_set_name(KQ_pos, "KQ_pos"); - ggml_allocr_alloc(lctx.alloc, KQ_pos); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) KQ_pos->data; - for (int i = 0; i < n_tokens; ++i) { - data[i] = batch.pos[i]; - } - } + cb(KQ_mask, "KQ_mask", -1); // shift the entire K-cache if needed if (do_rope_shift) { - struct ggml_tensor * K_shift = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_ctx); - offload_func_kq(K_shift); - ggml_set_name(K_shift, "K_shift"); - ggml_allocr_alloc(lctx.alloc, K_shift); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) K_shift->data; - for (int i = 0; i < n_ctx; ++i) { - data[i] = kv_self.cells[i].delta; - } - } - - for (int il = 0; il < n_layer; ++il) { - struct ggml_tensor * tmp = - ggml_rope_custom_inplace(ctx0, - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_head_kv, n_ctx, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il), - K_shift, n_embd_head, 2, 0, freq_base, freq_scale); - offload_func_kq(tmp); - ggml_build_forward_expand(gf, tmp); - } + llm_build_k_shift(lctx, ctx0, gf, n_embd_head, LLM_ROPE_NEOX, cb); } for (int il = 0; il < n_layer; ++il) { struct ggml_tensor * attn_norm; - offload_func_t offload_func = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (il >= i_gpu_start) { - offload_func = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + attn_norm = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, + model.layers[il].attn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(attn_norm, "attn_norm", il); // self-attention - // TODO: refactor into common function (shared with LLaMA) { - attn_norm = ggml_norm(ctx0, inpL, norm_eps); - offload_func(attn_norm); - - attn_norm = ggml_add(ctx0, - ggml_mul(ctx0, attn_norm, model.layers[il].attn_norm), - model.layers[il].attn_norm_b); - offload_func(attn_norm->src[0]); - offload_func(attn_norm); - - if (model.layers[il].attn_norm_2) { // Falcon-40B - cur = ggml_norm(ctx0, inpL, norm_eps); - offload_func(cur); - - cur = ggml_add(ctx0, - ggml_mul(ctx0, cur, model.layers[il].attn_norm_2), - model.layers[il].attn_norm_2_b); - offload_func(cur->src[0]); - offload_func(cur); - } else { // Falcon 7B + if (model.layers[il].attn_norm_2) { + // Falcon-40B + cur = llm_build_norm(ctx0, attn_norm, + model.layers[il].attn_norm_2, + model.layers[il].attn_norm_2_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "attn_norm_2", il); + } else { cur = attn_norm; } - // compute QKV - cur = ggml_mul_mat(ctx0, model.layers[il].wqkv, cur); - offload_func_kq(cur); + cb(cur, "wqkv", il); - // Note that the strides for Kcur, Vcur are set up so that the - // resulting views are misaligned with the tensor's storage - // (by applying the K/V offset we shift the tensor's original - // view to stick out behind the viewed QKV tensor's allocated - // memory, so to say). This is ok because no actual accesses - // happen to that out-of-range memory, but it can require some - // trickery when trying to accurately dump these views for - // debugging. + struct ggml_tensor * Qcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd))); + struct ggml_tensor * Kcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd))); + struct ggml_tensor * Vcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd + n_embd_gqa))); - const size_t wsize = ggml_type_size(cur->type); + cb(Qcur, "Qcur", il); + cb(Kcur, "Kcur", il); + cb(Vcur, "Vcur", il); - // TODO: these 2 ggml_conts are technically not needed, but we add them until CUDA support for - // non-contiguous views is added for the rope operator - struct ggml_tensor * tmpq = ggml_cont(ctx0, ggml_view_3d( - ctx0, cur, n_embd_head, n_head, n_tokens, - wsize * n_embd_head, - wsize * n_embd_head * (n_head + 2 * n_head_kv), - 0)); - offload_func_kq(tmpq); - - struct ggml_tensor * tmpk = ggml_cont(ctx0, ggml_view_3d( - ctx0, cur, n_embd_head, n_head_kv, n_tokens, - wsize * n_embd_head, - wsize * n_embd_head * (n_head + 2 * n_head_kv), - wsize * n_embd_head * n_head)); - offload_func_kq(tmpk); - - struct ggml_tensor * tmpv = ggml_view_3d( - ctx0, cur, n_embd_head, n_head_kv, n_tokens, - wsize * n_embd_head, - wsize * n_embd_head * (n_head + 2 * n_head_kv), - wsize * n_embd_head * (n_head + n_head_kv)); - offload_func_v(tmpv); + Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens); + Kcur = ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens); // using mode = 2 for neox mode - struct ggml_tensor * Qcur = ggml_rope_custom(ctx0, tmpq, KQ_pos, n_embd_head, 2, 0, freq_base, freq_scale); - offload_func_kq(Qcur); - struct ggml_tensor * Kcur = ggml_rope_custom(ctx0, tmpk, KQ_pos, n_embd_head, 2, 0, freq_base, freq_scale); - offload_func_kq(Kcur); + Qcur = ggml_rope_custom(ctx0, Qcur, inp_pos, n_embd_head, 2, 0, freq_base, freq_scale); + cb(Qcur, "Qcur", il); - { - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, ggml_cont(ctx0, tmpv), n_embd_gqa, n_tokens)); - offload_func_v(Vcur); - offload_func_v(Vcur->src[0]->src[0]); - ggml_set_name(Vcur, "Vcur"); + Kcur = ggml_rope_custom(ctx0, Kcur, inp_pos, n_embd_head, 2, 0, freq_base, freq_scale); + cb(Kcur, "Kcur", il); - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - offload_func_kq(k); - ggml_set_name(k, "k"); + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); - offload_func_kq(Q); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - offload_func_kq(K); - ggml_set_name(K, "K"); - - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - struct ggml_tensor * KQ_scaled = ggml_scale(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled, KQ_mask); - offload_func_kq(KQ_masked); - ggml_set_name(KQ_masked, "KQ_masked"); - - struct ggml_tensor * KQ_soft_max = ggml_soft_max(ctx0, KQ_masked); - offload_func_v(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - offload_func_v(V); - ggml_set_name(V, "V"); - - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); - - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); - - cur = ggml_mul_mat(ctx0, model.layers[il].wo, cur); - offload_func(cur); - ggml_set_name(cur, "result_wo"); + cur = llm_build_kqv(lctx, ctx0, attn_norm, + model.layers[il].wo, NULL, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, -1.0f, cb, il); + cb(cur, "kqv_out", il); } - struct ggml_tensor * attn_out = cur; + struct ggml_tensor * ffn_inp = cur; // feed forward { - struct ggml_tensor * inpFF = attn_norm; - - cur = ggml_mul_mat(ctx0, model.layers[il].w3, inpFF); - offload_func(cur); - - cur = ggml_gelu(ctx0, cur); - offload_func(cur); - cur = ggml_mul_mat(ctx0, model.layers[il].w2, cur); - offload_func(cur); + cur = llm_build_ffn(ctx0, attn_norm, // !! use the attn norm, not the result + model.layers[il].ffn_up, NULL, + NULL, NULL, + model.layers[il].ffn_down, NULL, + LLM_FFN_GELU, LLM_FFN_SEQ, cb, il); + cb(cur, "ffn_out", il); } - cur = ggml_add(ctx0, cur, attn_out); - offload_func(cur); + cur = ggml_add(ctx0, cur, ffn_inp); + cb(cur, "l_out", il); + cur = ggml_add(ctx0, cur, inpL); - offload_func(cur); + cb(cur, "l_out", il); // input for next layer inpL = cur; @@ -4572,18 +3887,14 @@ static struct ggml_cgraph * llm_build_falcon( cur = inpL; // norm - { - cur = ggml_norm(ctx0, cur, norm_eps); - offload_func_nr(cur); - - cur = ggml_add(ctx0, - ggml_mul(ctx0, cur, model.output_norm), - model.output_norm_b); - ggml_set_name(cur, "result_norm"); - } + cur = llm_build_norm(ctx0, cur, + model.output_norm, + model.output_norm_b, + LLM_NORM, norm_eps, cb, -1); + cb(cur, "result_norm", -1); cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); ggml_build_forward_expand(gf, cur); @@ -4594,7 +3905,9 @@ static struct ggml_cgraph * llm_build_falcon( static struct ggml_cgraph * llm_build_starcoder( llama_context & lctx, - const llama_batch & batch) { + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; const auto & cparams = lctx.cparams; @@ -4607,7 +3920,6 @@ static struct ggml_cgraph * llm_build_starcoder( const int64_t n_layer = hparams.n_layer; const int64_t n_ctx = cparams.n_ctx; const int64_t n_head = hparams.n_head; - const int64_t n_head_kv = hparams.n_head_kv; const int64_t n_embd_head = hparams.n_embd_head(); const int64_t n_embd_gqa = hparams.n_embd_gqa(); @@ -4615,11 +3927,9 @@ static struct ggml_cgraph * llm_build_starcoder( const float norm_eps = hparams.f_norm_eps; - const int n_gpu_layers = model.n_gpu_layers; - const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; auto & buf_compute = lctx.buf_compute; @@ -4634,266 +3944,95 @@ static struct ggml_cgraph * llm_build_starcoder( ggml_cgraph * gf = ggml_new_graph(ctx0); struct ggml_tensor * cur; - struct ggml_tensor * token; - struct ggml_tensor * position; + struct ggml_tensor * pos; struct ggml_tensor * inpL; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - - token = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - token = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, token); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(token->data, batch.embd, n_tokens * n_embd * ggml_element_size(token)); - } - } - - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - - // offload functions set the tensor output backend to GPU - // tensors are GPU-accelerated if any input or the output has been offloaded - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (n_gpu_layers > n_layer) { - offload_func_nr = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 1) { - offload_func_v = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 2) { - offload_func_kq = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS - - { - // Compute position embeddings. - struct ggml_tensor * inp_positions = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - ggml_allocr_alloc(lctx.alloc, inp_positions); - if (!ggml_allocr_is_measure(lctx.alloc)) { - for (int i = 0; i < n_tokens; ++i) { - ((int32_t *) inp_positions->data)[i] = batch.pos[i]; - } - } - ggml_set_name(inp_positions, "inp_positions"); - - position = ggml_get_rows(ctx0, model.pos_embeddings, inp_positions); - } + // inp_pos - contains the positions + struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + cb(inp_pos, "inp_pos", -1); // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd)/n_head)); - } + cb(KQ_scale, "KQ_scale", -1); // KQ_mask (mask for 1 head, it will be broadcasted to all heads) struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - ggml_set_name(KQ_mask, "KQ_mask"); - offload_func_kq(KQ_mask); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); + cb(KQ_mask, "KQ_mask", -1); - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; + pos = ggml_get_rows(ctx0, model.pos_embd, inp_pos); + cb(pos, "pos_embd", -1); - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - inpL = ggml_add(ctx0, token, position); - ggml_set_name(inpL, "inpL"); + inpL = ggml_add(ctx0, inpL, pos); + cb(inpL, "inpL", -1); for (int il = 0; il < n_layer; ++il) { - offload_func_t offload_func = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (il >= i_gpu_start) { - offload_func = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + cur = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, + model.layers[il].attn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "attn_norm", il); + // self-attention { - // Norm - cur = ggml_norm(ctx0, inpL, norm_eps); - offload_func(cur); - - cur = ggml_add(ctx0, ggml_mul(ctx0, cur, model.layers[il].attn_norm), model.layers[il].attn_norm_b); - offload_func(cur); - } - - { - // Self Attention cur = ggml_mul_mat(ctx0, model.layers[il].wqkv, cur); - offload_func_kq(cur); + cb(cur, "wqkv", il); cur = ggml_add(ctx0, cur, model.layers[il].bqkv); - offload_func_kq(cur); + cb(cur, "bqkv", il); - struct ggml_tensor * tmpq = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd))); - struct ggml_tensor * tmpk = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd))); - struct ggml_tensor * tmpv = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd + n_embd_gqa))); + struct ggml_tensor * Qcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd))); + struct ggml_tensor * Kcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd))); + struct ggml_tensor * Vcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd + n_embd_gqa))); - ggml_set_name(tmpq, "tmpq"); - ggml_set_name(tmpk, "tmpk"); - ggml_set_name(tmpv, "tmpv"); + cb(Qcur, "Qcur", il); + cb(Kcur, "Kcur", il); + cb(Vcur, "Vcur", il); - offload_func_kq(tmpq); - offload_func_kq(tmpk); - offload_func_v (tmpv); + Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens); - struct ggml_tensor * Qcur = ggml_reshape_3d(ctx0, tmpq, n_embd_head, n_head, n_tokens); - struct ggml_tensor * Kcur = tmpk; + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); - { - struct ggml_tensor * Vcur = ggml_transpose(ctx0, tmpv); - offload_func_v(Vcur); - ggml_set_name(Vcur, "Vcur"); - - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - offload_func_kq(k); - ggml_set_name(k, "k"); - - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - ggml_set_name(v, "v"); - - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); - offload_func_kq(Q); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - offload_func_kq(K); - ggml_set_name(K, "K"); - - // K * Q - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - // KQ_scaled = KQ / sqrt(n_embd_head) - // KQ_scaled shape [n_past + n_tokens, n_tokens, n_head, 1] - struct ggml_tensor * KQ_scaled = ggml_scale_inplace(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - // KQ_masked = mask_past(KQ_scaled) - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled, KQ_mask); - offload_func_kq(KQ_masked); - ggml_set_name(KQ_masked, "KQ_masked"); - - // KQ = soft_max(KQ_masked) - struct ggml_tensor * KQ_soft_max = ggml_soft_max_inplace(ctx0, KQ_masked); - offload_func_v(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - // split cached V into n_head heads - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - ggml_set_name(V, "V"); - - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); - - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); + cur = llm_build_kqv(lctx, ctx0, cur, + model.layers[il].wo, model.layers[il].bo, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, -1.0f, cb, il); + cb(cur, "kqv_out", il); } - // Projection - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].wo, cur), model.layers[il].bo); - offload_func(cur); - - // Add the input - cur = ggml_add(ctx0, cur, inpL); - offload_func(cur); - - struct ggml_tensor * inpFF = cur; + // add the input + struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpL); + cb(ffn_inp, "ffn_inp", il); // FF { - // Norm - { - cur = ggml_norm(ctx0, inpFF, norm_eps); - offload_func_nr(cur); + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, + model.layers[il].ffn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "ffn_norm", il); - cur = ggml_add(ctx0, ggml_mul(ctx0, cur, model.layers[il].ffn_norm), model.layers[il].ffn_norm_b); - offload_func_nr(cur); - } - - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].w3, cur), model.layers[il].b3); - offload_func(cur); - - // GELU activation - cur = ggml_gelu(ctx0, cur); - offload_func(cur); - - // Projection - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].w2, cur), model.layers[il].b2); - offload_func(cur); + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, model.layers[il].ffn_up_b, + NULL, NULL, + model.layers[il].ffn_down, model.layers[il].ffn_down_b, + LLM_FFN_GELU, LLM_FFN_SEQ, cb, il); + cb(cur, "ffn_out", il); } - inpL = ggml_add(ctx0, cur, inpFF); - + inpL = ggml_add(ctx0, cur, ffn_inp); + cb(inpL, "l_out", il); } - // Output Norm - { - cur = ggml_norm(ctx0, inpL, norm_eps); - offload_func_nr(cur); - - cur = ggml_add(ctx0, ggml_mul(ctx0, cur, model.output_norm), model.output_norm_b); - ggml_set_name(cur, "result_norm"); - } + cur = llm_build_norm(ctx0, inpL, + model.output_norm, + model.output_norm_b, + LLM_NORM, norm_eps, cb, -1); + cb(cur, "result_norm", -1); cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); ggml_build_forward_expand(gf, cur); ggml_free(ctx0); @@ -4903,7 +4042,9 @@ static struct ggml_cgraph * llm_build_starcoder( static struct ggml_cgraph * llm_build_persimmon( llama_context & lctx, - const llama_batch & batch) { + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; @@ -4912,29 +4053,27 @@ static struct ggml_cgraph * llm_build_persimmon( GGML_ASSERT(!!kv_self.ctx); const auto & cparams = lctx.cparams; + const int64_t n_embd = hparams.n_embd; const int64_t n_layer = hparams.n_layer; const int64_t n_ctx = cparams.n_ctx; const int64_t n_head_kv = hparams.n_head_kv; const int64_t n_head = hparams.n_head; const int64_t n_embd_head = hparams.n_embd_head(); - const int64_t n_embd_gqa = hparams.n_embd_gqa(); - const size_t n_rot = n_embd_head / 2; + const int64_t n_rot = n_embd_head / 2; const float freq_base = cparams.rope_freq_base; const float freq_scale = cparams.rope_freq_scale; - const float norm_eps = hparams.f_norm_eps; - - const int n_gpu_layers = model.n_gpu_layers; - + const float norm_eps = hparams.f_norm_eps; const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; - const bool do_rope_shift = ggml_allocr_is_measure(lctx.alloc) || kv_self.has_shift; + const bool do_rope_shift = worst_case || kv_self.has_shift; auto & buf_compute = lctx.buf_compute; + struct ggml_init_params params = { /*.mem_size =*/ buf_compute.size, /*.mem_buffer =*/ buf_compute.data, @@ -4948,146 +4087,77 @@ static struct ggml_cgraph * llm_build_persimmon( struct ggml_tensor * cur; struct ggml_tensor * inpL; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "imp_embd", -1); + + struct ggml_tensor * inp_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); + cb(inp_pos, "inp_pos", -1); - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - inpL = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { - inpL = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - ggml_allocr_alloc(lctx.alloc, inpL); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inpL->data, batch.embd, n_tokens * n_embd * ggml_element_size(inpL)); - } - } - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd_head))); - } - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); + cb(KQ_scale, "KQ_scale", -1); + struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - offload_func_kq(KQ_mask); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); + cb(KQ_mask, "KQ_mask", -1); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - struct ggml_tensor * KQ_pos = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - offload_func_kq(KQ_pos); - ggml_set_name(KQ_pos, "KQ_pos"); - ggml_allocr_alloc(lctx.alloc, KQ_pos); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) KQ_pos->data; - for (int i = 0; i < n_tokens; ++i) { - data[i] = batch.pos[i]; - } - } if (do_rope_shift) { - struct ggml_tensor * K_shift = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_ctx); - offload_func_kq(K_shift); - ggml_set_name(K_shift, "K_shift"); - ggml_allocr_alloc(lctx.alloc, K_shift); - if (!ggml_allocr_is_measure(lctx.alloc)) { - int * data = (int *) K_shift->data; - for (int i = 0; i < n_ctx; ++i) { - data[i] = kv_self.cells[i].delta; - } - } - for (int il = 0; il < n_layer; ++il) { - struct ggml_tensor * tmp = - // we rotate only the first n_rot dimensions. - ggml_rope_custom_inplace(ctx0, - ggml_view_3d(ctx0, kv_self.k, - n_rot, n_head, n_ctx, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*(n_embd_head*n_ctx*il) - ), - K_shift, n_rot, 2, 0, freq_base, freq_scale); - offload_func_kq(tmp); - ggml_build_forward_expand(gf, tmp); - } + llm_build_k_shift(lctx, ctx0, gf, n_rot, LLM_ROPE_NEOX, cb); } - for (int il=0; il < n_layer; ++il) { + + for (int il = 0; il < n_layer; ++il) { struct ggml_tensor * residual = inpL; - offload_func_t offload_func = llama_nop; - { - cur = ggml_norm(ctx0, inpL, norm_eps); - offload_func(cur); - cur = ggml_mul(ctx0, cur, model.layers[il].attn_norm); - offload_func(cur); - cur = ggml_add(ctx0, cur, model.layers[il].attn_norm_b); - offload_func(cur); - ggml_format_name(cur, "input_layernorm_%d", il); - } + + cur = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, + model.layers[il].attn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "attn_norm", il); + // self attention { cur = ggml_mul_mat(ctx0, model.layers[il].wqkv, cur); - offload_func_kq(cur); + cb(cur, "wqkv", il); + cur = ggml_add(ctx0, cur, model.layers[il].bqkv); - offload_func_kq(cur); + cb(cur, "bqkv", il); // split qkv GGML_ASSERT(n_head_kv == n_head); - ggml_set_name(cur, format("qkv_%d", il).c_str()); + struct ggml_tensor * tmpqkv = ggml_reshape_4d(ctx0, cur, n_embd_head, 3, n_head, n_tokens); - offload_func_kq(tmpqkv); + cb(tmpqkv, "tmpqkv", il); + struct ggml_tensor * tmpqkv_perm = ggml_cont(ctx0, ggml_permute(ctx0, tmpqkv, 0, 3, 1, 2)); - offload_func_kq(tmpqkv_perm); - ggml_format_name(tmpqkv_perm, "tmpqkv_perm_%d", il); + cb(tmpqkv_perm, "tmpqkv", il); + struct ggml_tensor * tmpq = ggml_view_3d( ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens, ggml_element_size(tmpqkv_perm) * n_embd_head, ggml_element_size(tmpqkv_perm) * n_embd_head * n_head, 0 ); - offload_func_kq(tmpq); + cb(tmpq, "tmpq", il); + struct ggml_tensor * tmpk = ggml_view_3d( ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens, ggml_element_size(tmpqkv_perm) * n_embd_head, ggml_element_size(tmpqkv_perm) * n_embd_head * n_head, ggml_element_size(tmpqkv_perm) * n_embd_head * n_head * n_tokens ); - offload_func_kq(tmpk); - // Q/K Layernorm - tmpq = ggml_norm(ctx0, tmpq, norm_eps); - offload_func_kq(tmpq); - tmpq = ggml_mul(ctx0, tmpq, model.layers[il].attn_q_norm); - offload_func_kq(tmpq); - tmpq = ggml_add(ctx0, tmpq, model.layers[il].attn_q_norm_b); - offload_func_kq(tmpq); + cb(tmpk, "tmpk", il); - tmpk = ggml_norm(ctx0, tmpk, norm_eps); - offload_func_v(tmpk); - tmpk = ggml_mul(ctx0, tmpk, model.layers[il].attn_k_norm); - offload_func_v(tmpk); - tmpk = ggml_add(ctx0, tmpk, model.layers[il].attn_k_norm_b); - offload_func_v(tmpk); + // Q/K Layernorm + tmpq = llm_build_norm(ctx0, tmpq, + model.layers[il].attn_q_norm, + model.layers[il].attn_q_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(tmpq, "tmpq", il); + + tmpk = llm_build_norm(ctx0, tmpk, + model.layers[il].attn_k_norm, + model.layers[il].attn_k_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(tmpk, "tmpk", il); // RoPE the first n_rot of q/k, pass the other half, and concat. struct ggml_tensor * qrot = ggml_view_3d( @@ -5096,16 +4166,15 @@ static struct ggml_cgraph * llm_build_persimmon( ggml_element_size(tmpq) * n_embd_head * n_head, 0 ); - offload_func_kq(qrot); - ggml_format_name(qrot, "qrot_%d", il); + cb(qrot, "qrot", il); + struct ggml_tensor * krot = ggml_view_3d( ctx0, tmpk, n_rot, n_head, n_tokens, ggml_element_size(tmpk) * n_embd_head, ggml_element_size(tmpk) * n_embd_head * n_head, 0 ); - offload_func_kq(krot); - ggml_format_name(krot, "krot_%d", il); + cb(krot, "krot", il); // get the second half of tmpq, e.g tmpq[n_rot:, :, :] struct ggml_tensor * qpass = ggml_view_3d( @@ -5114,193 +4183,117 @@ static struct ggml_cgraph * llm_build_persimmon( ggml_element_size(tmpq) * n_embd_head * n_head, ggml_element_size(tmpq) * n_rot ); - offload_func_kq(qpass); - ggml_format_name(qpass, "qpass_%d", il); + cb(qpass, "qpass", il); + struct ggml_tensor * kpass = ggml_view_3d( ctx0, tmpk, n_rot, n_head, n_tokens, ggml_element_size(tmpk) * n_embd_head, ggml_element_size(tmpk) * n_embd_head * n_head, ggml_element_size(tmpk) * n_rot ); - offload_func_kq(kpass); - ggml_format_name(kpass, "kpass_%d", il); + cb(kpass, "kpass", il); - struct ggml_tensor * qrotated = ggml_rope_custom( - ctx0, qrot, KQ_pos, n_rot, 2, 0, freq_base, freq_scale + struct ggml_tensor * qrotated = ggml_rope_custom( + ctx0, qrot, inp_pos, n_rot, 2, 0, freq_base, freq_scale ); - offload_func_kq(qrotated); + cb(qrotated, "qrotated", il); + struct ggml_tensor * krotated = ggml_rope_custom( - ctx0, krot, KQ_pos, n_rot, 2, 0, freq_base, freq_scale + ctx0, krot, inp_pos, n_rot, 2, 0, freq_base, freq_scale ); - offload_func_kq(krotated); + cb(krotated, "krotated", il); + // ggml currently only supports concatenation on dim=2 // so we need to permute qrot, qpass, concat, then permute back. qrotated = ggml_cont(ctx0, ggml_permute(ctx0, qrotated, 2, 1, 0, 3)); - offload_func_kq(qrotated); + cb(qrotated, "qrotated", il); + krotated = ggml_cont(ctx0, ggml_permute(ctx0, krotated, 2, 1, 0, 3)); - offload_func_kq(krotated); + cb(krotated, "krotated", il); qpass = ggml_cont(ctx0, ggml_permute(ctx0, qpass, 2, 1, 0, 3)); - offload_func_kq(qpass); + cb(qpass, "qpass", il); + kpass = ggml_cont(ctx0, ggml_permute(ctx0, kpass, 2, 1, 0, 3)); - offload_func_kq(kpass); + cb(kpass, "kpass", il); struct ggml_tensor * Qcur = ggml_concat(ctx0, qrotated, qpass); - offload_func_kq(Qcur); + cb(Qcur, "Qcur", il); + struct ggml_tensor * Kcur = ggml_concat(ctx0, krotated, kpass); - offload_func_kq(Kcur); + cb(Kcur, "Kcur", il); struct ggml_tensor * Q = ggml_cont(ctx0, ggml_permute(ctx0, Qcur, 1, 2, 0, 3)); - offload_func_kq(Q); + cb(Q, "Q", il); Kcur = ggml_cont(ctx0, ggml_permute(ctx0, Kcur, 2, 1, 0, 3)); - offload_func_kq(Kcur); - { - struct ggml_tensor * tmpv = ggml_view_3d( - ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens, - ggml_element_size(tmpqkv_perm) * n_embd_head, - ggml_element_size(tmpqkv_perm) * n_embd_head * n_head, - ggml_element_size(tmpqkv_perm) * n_embd_head * n_head * n_tokens * 2 + cb(Kcur, "Kcur", il); + + struct ggml_tensor * Vcur = ggml_view_3d( + ctx0, tmpqkv_perm, n_embd_head, n_head, n_tokens, + ggml_element_size(tmpqkv_perm) * n_embd_head, + ggml_element_size(tmpqkv_perm) * n_embd_head * n_head, + ggml_element_size(tmpqkv_perm) * n_embd_head * n_head * n_tokens * 2 ); - offload_func_v(tmpv); - // store K, V in cache - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, tmpv, n_embd_gqa, n_tokens)); - offload_func_v(Vcur); - ggml_set_name(Vcur, "Vcur"); + cb(Vcur, "Vcur", il); - struct ggml_tensor * k = ggml_view_1d( - ctx0, kv_self.k, n_tokens*n_embd_gqa, - (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head) - ); - offload_func_kq(k); - ggml_set_name(k, "k"); + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - ggml_set_name(v, "v"); - - // important: storing RoPE-ed version of K in the KV cache! - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - struct ggml_tensor * K = ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - - offload_func_kq(K); - ggml_format_name(K, "K_%d", il); - - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - struct ggml_tensor * KQ_scaled = ggml_scale(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled, KQ_mask); - offload_func_kq(KQ_masked); - ggml_set_name(KQ_masked, "KQ_masked"); - - struct ggml_tensor * KQ_soft_max = ggml_soft_max_inplace(ctx0, KQ_masked); - offload_func_kq(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - offload_func_v(V); - ggml_set_name(V, "V"); - - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); - - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); - - cur = ggml_mul_mat(ctx0, model.layers[il].wo, cur); - offload_func(cur); - cur = ggml_add(ctx0, cur, model.layers[il].bo); - offload_func(cur); - ggml_set_name(cur, "result_wo"); + // TODO: not tested, could be broken + cur = llm_build_kqv(lctx, ctx0, Q, + model.layers[il].wo, model.layers[il].bo, + Q, KQ_scale, KQ_mask, n_tokens, n_kv, -1.0f, cb, il); + cb(cur, "kqv_out", il); } - struct ggml_tensor * inpFF = ggml_add(ctx0, residual, cur); - offload_func(inpFF); - ggml_set_name(inpFF, "inpFF"); + struct ggml_tensor * ffn_inp = ggml_add(ctx0, residual, cur); + cb(ffn_inp, "ffn_inp", il); + + // feed-forward network { - // MLP - { - // Norm - cur = ggml_norm(ctx0, inpFF, norm_eps); - offload_func(cur); - cur = ggml_add(ctx0, - ggml_mul(ctx0, cur, model.layers[il].ffn_norm), - model.layers[il].ffn_norm_b - ); - ggml_set_name(cur, "ffn_norm"); - offload_func(cur); - } - cur = ggml_mul_mat(ctx0, model.layers[il].w3, cur); - offload_func(cur); + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, + model.layers[il].ffn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "ffn_norm", il); - cur = ggml_add(ctx0, cur, model.layers[il].b3); - offload_func(cur); - ggml_set_name(cur, "result_ffn_up"); - - cur = ggml_sqr(ctx0, ggml_relu(ctx0, cur)); - ggml_set_name(cur, "result_ffn_act"); - offload_func(cur); - offload_func(cur->src[0]); - - cur = ggml_mul_mat(ctx0, model.layers[il].w2, cur); - offload_func(cur); - cur = ggml_add(ctx0, - cur, - model.layers[il].b2); - offload_func(cur); - ggml_set_name(cur, "outFF"); + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, model.layers[il].ffn_up_b, + NULL, NULL, + model.layers[il].ffn_down, model.layers[il].ffn_down_b, + LLM_FFN_RELU_SQR, LLM_FFN_SEQ, cb, il); + cb(cur, "ffn_out", il); } - cur = ggml_add(ctx0, cur, inpFF); - offload_func(cur); - ggml_set_name(cur, "inpFF_+_outFF"); + + cur = ggml_add(ctx0, cur, ffn_inp); + cb(cur, "l_out", il); + inpL = cur; } + cur = inpL; - { - cur = ggml_norm(ctx0, cur, norm_eps); - offload_func_nr(cur); - cur = ggml_mul(ctx0, cur, model.output_norm); - offload_func_nr(cur); - cur = ggml_add(ctx0, cur, model.output_norm_b); - // offload_func_nr(cur); + cur = llm_build_norm(ctx0, cur, + model.output_norm, + model.output_norm_b, + LLM_NORM, norm_eps, cb, -1); + cb(cur, "result_norm", -1); - ggml_set_name(cur, "result_norm"); - } cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); + ggml_build_forward_expand(gf, cur); + ggml_free(ctx0); + return gf; } -static struct ggml_cgraph * llm_build_bloom( +static struct ggml_cgraph * llm_build_refact( llama_context & lctx, - const llama_batch & batch) { + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; const auto & cparams = lctx.cparams; @@ -5315,6 +4308,133 @@ static struct ggml_cgraph * llm_build_bloom( const int64_t n_head = hparams.n_head; const int64_t n_head_kv = hparams.n_head_kv; const int64_t n_embd_head = hparams.n_embd_head(); + + const float norm_rms_eps = hparams.f_norm_rms_eps; + + const int32_t n_tokens = batch.n_tokens; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; + + auto & buf_compute = lctx.buf_compute; + + struct ggml_init_params params = { + /*.mem_size =*/ buf_compute.size, + /*.mem_buffer =*/ buf_compute.data, + /*.no_alloc =*/ true, + }; + + struct ggml_context * ctx0 = ggml_init(params); + + ggml_cgraph * gf = ggml_new_graph(ctx0); + + struct ggml_tensor * cur; + struct ggml_tensor * inpL; + + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); + + // KQ_scale + struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); + cb(KQ_scale, "KQ_scale", -1); + + // KQ_mask (mask for 1 head, it will be broadcasted to all heads) + struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); + cb(KQ_mask, "KQ_mask", -1); + + for (int il = 0; il < n_layer; ++il) { + struct ggml_tensor * inpSA = inpL; + + cur = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, il); + cb(cur, "attn_norm", il); + + // self-attention + { + struct ggml_tensor * Qcur = ggml_mul_mat(ctx0, model.layers[il].wq, cur); + cb(Qcur, "Qcur", il); + + struct ggml_tensor * Kcur = ggml_mul_mat(ctx0, model.layers[il].wk, cur); + cb(Kcur, "Kcur", il); + + struct ggml_tensor * Vcur = ggml_mul_mat(ctx0, model.layers[il].wv, cur); + cb(Vcur, "Vcur", il); + + Kcur = ggml_reshape_3d(ctx0, Kcur, n_embd_head, n_head_kv, n_tokens); + cb(Kcur, "Kcur", il); + + Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens); + cb(Qcur, "Qcur", il); + + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); + + cur = llm_build_kqv(lctx, ctx0, Qcur, + model.layers[il].wo, NULL, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, 8.0f, cb, il); + cb(cur, "kqv_out", il); + } + + struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpSA); + cb(ffn_inp, "ffn_inp", il); + + // feed-forward network + { + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, il); + cb(cur, "ffn_norm", il); + + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, NULL, + model.layers[il].ffn_gate, NULL, + model.layers[il].ffn_down, NULL, + LLM_FFN_SILU, LLM_FFN_PAR, cb, il); + cb(cur, "ffn_out", il); + } + + cur = ggml_add(ctx0, cur, ffn_inp); + cb(cur, "l_out", il); + + // input for next layer + inpL = cur; + } + + cur = inpL; + + cur = llm_build_norm(ctx0, cur, + model.output_norm, NULL, + LLM_NORM_RMS, norm_rms_eps, cb, -1); + cb(cur, "result_norm", -1); + + // lm_head + cur = ggml_mul_mat(ctx0, model.output, cur); + cb(cur, "result_output", -1); + + ggml_build_forward_expand(gf, cur); + + ggml_free(ctx0); + + return gf; +} + +static struct ggml_cgraph * llm_build_bloom( + llama_context & lctx, + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { + const auto & model = lctx.model; + const auto & hparams = model.hparams; + const auto & cparams = lctx.cparams; + + const auto & kv_self = lctx.kv_self; + + GGML_ASSERT(!!kv_self.ctx); + + const int64_t n_embd = hparams.n_embd; + const int64_t n_layer = hparams.n_layer; + const int64_t n_ctx = cparams.n_ctx; + const int64_t n_head = hparams.n_head; + const int64_t n_embd_head = hparams.n_embd_head(); const int64_t n_embd_gqa = hparams.n_embd_gqa(); GGML_ASSERT(n_embd_head == hparams.n_rot); @@ -5322,8 +4442,8 @@ static struct ggml_cgraph * llm_build_bloom( const float norm_eps = hparams.f_norm_eps; const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; auto & buf_compute = lctx.buf_compute; @@ -5340,198 +4460,90 @@ static struct ggml_cgraph * llm_build_bloom( ggml_cgraph * gf = ggml_new_graph(ctx0); struct ggml_tensor * cur; - struct ggml_tensor * token; struct ggml_tensor * inpL; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - } - ggml_set_name(inp_tokens, "inp_tokens"); - - token = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - token = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, token); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(token->data, batch.embd, n_tokens * n_embd * ggml_element_size(token)); - } - } + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd)/n_head)); - } + cb(KQ_scale, "KQ_scale", -1); // KQ_mask (mask for 1 head, it will be broadcasted to all heads) struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); + cb(KQ_mask, "KQ_mask", -1); - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } - - // norm - { - inpL = ggml_norm(ctx0, token, norm_eps); - inpL = ggml_add(ctx0, ggml_mul(ctx0, inpL, model.tok_norm), model.tok_norm_b); - } - - ggml_set_name(inpL, "inpL"); + inpL = llm_build_norm(ctx0, inpL, + model.tok_norm, + model.tok_norm_b, + LLM_NORM, norm_eps, cb, -1); + cb(inpL, "inp_norm", -1); for (int il = 0; il < n_layer; ++il) { + cur = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, + model.layers[il].attn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "attn_norm", il); + + // self-attention { - // Norm - cur = ggml_norm(ctx0, inpL, norm_eps); - cur = ggml_add(ctx0, ggml_mul(ctx0, cur, model.layers[il].attn_norm), model.layers[il].attn_norm_b); + cur = ggml_mul_mat(ctx0, model.layers[il].wqkv, cur); + cb(cur, "wqkv", il); + + cur = ggml_add(ctx0, cur, model.layers[il].bqkv); + cb(cur, "bqkv", il); + + struct ggml_tensor * Qcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd))); + struct ggml_tensor * Kcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd))); + struct ggml_tensor * Vcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd + n_embd_gqa))); + + cb(Qcur, "Qcur", il); + cb(Kcur, "Kcur", il); + cb(Vcur, "Vcur", il); + + Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens); + + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); + + cur = llm_build_kqv(lctx, ctx0, Qcur, + model.layers[il].wo, model.layers[il].bo, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, 8.0f, cb, il); + cb(cur, "kqv_out", il); } - { - // Self Attention - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].wqkv, cur), model.layers[il].bqkv); - - struct ggml_tensor * tmpq = ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*n_embd); - struct ggml_tensor * tmpk = ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], sizeof(float)*n_embd); - struct ggml_tensor * tmpv = ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], sizeof(float)*(n_embd + n_embd_gqa)); - - struct ggml_tensor * Qcur = tmpq; - struct ggml_tensor * Kcur = tmpk; - - // store key and value to memory - { - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, ggml_cont(ctx0, tmpv), n_embd_gqa, n_tokens)); - ggml_set_name(Vcur, "Vcur"); - - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - ggml_set_name(k, "k"); - - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = - ggml_permute(ctx0, - ggml_cpy(ctx0, - Qcur, - ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_embd_head, n_head, n_tokens)), - 0, 2, 1, 3); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - ggml_set_name(K, "K"); - - // K * Q - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - ggml_set_name(KQ, "KQ"); - - // KQ_scaled = KQ / sqrt(n_embd_head) - // KQ_scaled shape [n_past + n_tokens, n_tokens, n_head, 1] - struct ggml_tensor * KQ_scaled = ggml_scale_inplace(ctx0, KQ, KQ_scale); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - struct ggml_tensor * KQ_scaled_alibi = ggml_alibi(ctx0, KQ_scaled, /*n_past*/ kv_head, n_head, 8); - ggml_set_name(KQ_scaled_alibi, "KQ_scaled_alibi"); - - // KQ_masked = mask_past(KQ_scaled) - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled_alibi, KQ_mask); - ggml_set_name(KQ_masked, "KQ_masked"); - - // KQ = soft_max(KQ_masked) - struct ggml_tensor * KQ_soft_max = ggml_soft_max_inplace(ctx0, KQ_masked); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - // split cached V into n_head heads - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - ggml_set_name(V, "V"); - - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - ggml_set_name(KQV, "KQV"); - - // KQV_merged = KQV.permute(0, 2, 1, 3) - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - ggml_set_name(KQV_merged, "KQV_merged"); - - // cur = KQV_merged.contiguous().view(n_embd, n_tokens) - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - ggml_set_name(cur, "KQV_merged_contiguous"); - } - - // Projection - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].wo, cur), model.layers[il].bo); - // Add the input - cur = ggml_add(ctx0, cur, inpL); - - struct ggml_tensor * inpFF = cur; + struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpL); + cb(ffn_inp, "ffn_inp", il); // FF { - // Norm - { - cur = ggml_norm(ctx0, inpFF, norm_eps); - cur = ggml_add(ctx0, ggml_mul(ctx0, cur, model.layers[il].ffn_norm), model.layers[il].ffn_norm_b); - } + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, + model.layers[il].ffn_norm_b, + LLM_NORM, norm_eps, cb, il); + cb(cur, "ffn_norm", il); - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].w3, cur), model.layers[il].b3); - - // GELU activation - cur = ggml_gelu(ctx0, cur); - - // Projection - cur = ggml_add(ctx0, ggml_mul_mat(ctx0, model.layers[il].w2, cur), model.layers[il].b2); + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, model.layers[il].ffn_up_b, + NULL, NULL, + model.layers[il].ffn_down, model.layers[il].ffn_down_b, + LLM_FFN_GELU, LLM_FFN_SEQ, cb, il); + cb(cur, "ffn_out", il); } - inpL = ggml_add(ctx0, cur, inpFF); + inpL = ggml_add(ctx0, cur, ffn_inp); + cb(inpL, "l_out", il); } - // Output Norm - { - cur = ggml_norm(ctx0, inpL, norm_eps); - cur = ggml_add(ctx0, ggml_mul(ctx0, cur, model.output_norm), model.output_norm_b); - } - ggml_set_name(cur, "result_norm"); + cur = llm_build_norm(ctx0, inpL, + model.output_norm, + model.output_norm_b, + LLM_NORM, norm_eps, cb, -1); + cb(cur, "result_norm", -1); cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); ggml_build_forward_expand(gf, cur); @@ -5542,7 +4554,9 @@ static struct ggml_cgraph * llm_build_bloom( static struct ggml_cgraph * llm_build_mpt( llama_context & lctx, - const llama_batch & batch) { + const llama_batch & batch, + const llm_build_cb & cb, + bool worst_case) { const auto & model = lctx.model; const auto & hparams = model.hparams; const auto & cparams = lctx.cparams; @@ -5555,7 +4569,6 @@ static struct ggml_cgraph * llm_build_mpt( const int64_t n_layer = hparams.n_layer; const int64_t n_ctx = cparams.n_ctx; const int64_t n_head = hparams.n_head; - const int64_t n_head_kv = hparams.n_head_kv; const int64_t n_embd_head = hparams.n_embd_head(); const int64_t n_embd_gqa = hparams.n_embd_gqa(); @@ -5563,11 +4576,9 @@ static struct ggml_cgraph * llm_build_mpt( const float clamp_kqv = hparams.f_clamp_kqv; const float max_alibi_bias = hparams.f_max_alibi_bias; - const int n_gpu_layers = model.n_gpu_layers; - const int32_t n_tokens = batch.n_tokens; - const int32_t n_kv = ggml_allocr_is_measure(lctx.alloc) ? n_ctx : kv_self.n; - const int32_t kv_head = ggml_allocr_is_measure(lctx.alloc) ? n_ctx - n_tokens : kv_self.head; + const int32_t n_kv = worst_case ? n_ctx : kv_self.n; + const int32_t kv_head = worst_case ? n_ctx - n_tokens : kv_self.head; auto & buf_compute = lctx.buf_compute; @@ -5586,269 +4597,93 @@ static struct ggml_cgraph * llm_build_mpt( struct ggml_tensor * cur; struct ggml_tensor * inpL; - //int warmup = 0; - if (batch.token) { - struct ggml_tensor * inp_tokens = ggml_new_tensor_1d(ctx0, GGML_TYPE_I32, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inp_tokens); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inp_tokens->data, batch.token, n_tokens*ggml_element_size(inp_tokens)); - //warmup = ((uint32_t*) inp_tokens->data)[0] == 0; - } - - ggml_set_name(inp_tokens, "inp_tokens"); - - inpL = ggml_get_rows(ctx0, model.tok_embeddings, inp_tokens); - } else { -#ifdef GGML_USE_MPI - GGML_ASSERT(false && "not implemented"); -#endif - - inpL = ggml_new_tensor_2d(ctx0, GGML_TYPE_F32, n_embd, n_tokens); - - ggml_allocr_alloc(lctx.alloc, inpL); - if (!ggml_allocr_is_measure(lctx.alloc)) { - memcpy(inpL->data, batch.embd, n_tokens * n_embd * ggml_element_size(inpL)); - } - } - - const int i_gpu_start = n_layer - n_gpu_layers; - (void) i_gpu_start; - - // offload functions set the tensor output backend to GPU - // tensors are GPU-accelerated if any input or the output has been offloaded - offload_func_t offload_func_nr = llama_nop; // nr = non-repeating - offload_func_t offload_func_kq = llama_nop; - offload_func_t offload_func_v = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (n_gpu_layers > n_layer) { - offload_func_nr = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 1) { - offload_func_v = ggml_cuda_assign_buffers_no_alloc; - } - if (n_gpu_layers > n_layer + 2) { - offload_func_kq = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + inpL = llm_build_inp_embd(ctx0, batch, model.tok_embd, n_embd, n_tokens, cb); + cb(inpL, "inp_embd", -1); // KQ_scale struct ggml_tensor * KQ_scale = ggml_new_tensor_1d(ctx0, GGML_TYPE_F32, 1); - ggml_set_name(KQ_scale, "1/sqrt(n_embd_head)"); - ggml_allocr_alloc(lctx.alloc, KQ_scale); - if (!ggml_allocr_is_measure(lctx.alloc)) { - ggml_set_f32(KQ_scale, 1.0f/sqrtf(float(n_embd)/n_head)); - } + cb(KQ_scale, "KQ_scale", -1); // KQ_mask (mask for 1 head, it will be broadcasted to all heads) struct ggml_tensor * KQ_mask = ggml_new_tensor_3d(ctx0, GGML_TYPE_F32, n_kv, n_tokens, 1); - offload_func_kq(KQ_mask); - ggml_set_name(KQ_mask, "KQ_mask"); - ggml_allocr_alloc(lctx.alloc, KQ_mask); - if (!ggml_allocr_is_measure(lctx.alloc)) { - float * data = (float *) KQ_mask->data; - memset(data, 0, ggml_nbytes(KQ_mask)); - - for (int h = 0; h < 1; ++h) { - for (int j = 0; j < n_tokens; ++j) { - const llama_pos pos = batch.pos[j]; - const llama_seq_id seq_id = batch.seq_id[j][0]; - - for (int i = 0; i < n_kv; ++i) { - if (!kv_self.cells[i].has_seq_id(seq_id) || kv_self.cells[i].pos > pos) { - data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; - } - } - } - } - } + cb(KQ_mask, "KQ_mask", -1); for (int il = 0; il < n_layer; ++il) { struct ggml_tensor * attn_norm; - offload_func_t offload_func = llama_nop; - -#ifdef GGML_USE_CUBLAS - if (il >= i_gpu_start) { - offload_func = ggml_cuda_assign_buffers_no_alloc; - } -#endif // GGML_USE_CUBLAS + attn_norm = llm_build_norm(ctx0, inpL, + model.layers[il].attn_norm, + NULL, + LLM_NORM, norm_eps, cb, il); + cb(attn_norm, "attn_norm", il); // self-attention - // TODO: refactor into common function (shared with LLaMA) { - attn_norm = ggml_norm(ctx0, inpL, norm_eps); - offload_func(attn_norm); - - attn_norm = ggml_mul(ctx0, attn_norm, model.layers[il].attn_norm); - offload_func(attn_norm); - - if (1) { - cur = attn_norm; - } - - // compute QKV + cur = attn_norm; cur = ggml_mul_mat(ctx0, model.layers[il].wqkv, cur); - offload_func_kq(cur); + cb(cur, "wqkv", il); if (clamp_kqv > 0.0f) { cur = ggml_clamp(ctx0, cur, -clamp_kqv, clamp_kqv); - offload_func_kq(cur); + cb(cur, "wqkv_clamped", il); } - const size_t wsize = ggml_type_size(cur->type); + struct ggml_tensor * Qcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd, n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd))); + struct ggml_tensor * Kcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd))); + struct ggml_tensor * Vcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd + n_embd_gqa))); - struct ggml_tensor * Qcur = ggml_view_3d( - ctx0, cur, n_embd_head, n_head, n_tokens, - wsize * n_embd_head, - wsize * n_embd_head * (n_head + 2 * n_head_kv), - 0); - offload_func_kq(Qcur); + cb(Qcur, "Qcur", il); + cb(Kcur, "Kcur", il); + cb(Vcur, "Vcur", il); - struct ggml_tensor * Kcur = ggml_view_3d( - ctx0, cur, n_embd_head, n_head_kv, n_tokens, - wsize * n_embd_head, - wsize * n_embd_head * (n_head + 2 * n_head_kv), - wsize * n_embd_head * n_head); - offload_func_kq(Kcur); + Qcur = ggml_reshape_3d(ctx0, Qcur, n_embd_head, n_head, n_tokens); - struct ggml_tensor * tmpv = ggml_view_3d( - ctx0, cur, n_embd_head, n_head_kv, n_tokens, - wsize * n_embd_head, - wsize * n_embd_head * (n_head + 2 * n_head_kv), - wsize * n_embd_head * (n_head + n_head_kv)); - offload_func_kq(Kcur); + llm_build_kv_store(lctx, ctx0, gf, Kcur, Vcur, n_tokens, kv_head, cb, il); - ggml_set_name(Qcur, "Qcur"); - ggml_set_name(Kcur, "Kcur"); - - { - struct ggml_tensor * Vcur = ggml_transpose(ctx0, ggml_reshape_2d(ctx0, ggml_cont(ctx0, tmpv), n_embd_gqa, n_tokens)); - offload_func_v(Vcur); - offload_func_v(Vcur->src[0]->src[0]); - ggml_set_name(Vcur, "Vcur"); - - struct ggml_tensor * k = ggml_view_1d(ctx0, kv_self.k, n_tokens*n_embd_gqa, (ggml_element_size(kv_self.k)*n_embd_gqa)*(il*n_ctx + kv_head)); - offload_func_kq(k); - ggml_set_name(k, "k"); - - struct ggml_tensor * v = ggml_view_2d(ctx0, kv_self.v, n_tokens, n_embd_gqa, - ( n_ctx)*ggml_element_size(kv_self.v), - (il*n_ctx)*ggml_element_size(kv_self.v)*n_embd_gqa + kv_head*ggml_element_size(kv_self.v)); - offload_func_v(v); - - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Kcur, k)); - ggml_build_forward_expand(gf, ggml_cpy(ctx0, Vcur, v)); - } - - struct ggml_tensor * Q = ggml_permute(ctx0, Qcur, 0, 2, 1, 3); - offload_func_kq(Q); - ggml_set_name(Q, "Q"); - - struct ggml_tensor * K = - ggml_view_3d(ctx0, kv_self.k, - n_embd_head, n_kv, n_head_kv, - ggml_element_size(kv_self.k)*n_embd_gqa, - ggml_element_size(kv_self.k)*n_embd_head, - ggml_element_size(kv_self.k)*n_embd_gqa*n_ctx*il); - offload_func_kq(K); - ggml_set_name(K, "K"); - - struct ggml_tensor * KQ = ggml_mul_mat(ctx0, K, Q); - offload_func_kq(KQ); - ggml_set_name(KQ, "KQ"); - - struct ggml_tensor * KQ_scaled = ggml_scale(ctx0, KQ, KQ_scale); - offload_func_kq(KQ_scaled); - ggml_set_name(KQ_scaled, "KQ_scaled"); - - // TODO: replace with ggml_add() - struct ggml_tensor * KQ_scaled_alibi = - ggml_alibi(ctx0, KQ_scaled, 0, n_head, max_alibi_bias); - offload_func_kq(KQ_scaled_alibi); - ggml_set_name(KQ_scaled_alibi, "KQ_scaled_alibi"); - - struct ggml_tensor * KQ_masked = ggml_add(ctx0, KQ_scaled_alibi, KQ_mask); - offload_func_kq(KQ_masked); - ggml_set_name(KQ_masked, "KQ_masked"); - - struct ggml_tensor * KQ_soft_max = ggml_soft_max(ctx0, KQ_masked); - offload_func_v(KQ_soft_max); - ggml_set_name(KQ_soft_max, "KQ_soft_max"); - - struct ggml_tensor * V = - ggml_view_3d(ctx0, kv_self.v, - n_kv, n_embd_head, n_head_kv, - ggml_element_size(kv_self.v)*n_ctx, - ggml_element_size(kv_self.v)*n_ctx*n_embd_head, - ggml_element_size(kv_self.v)*n_ctx*n_embd_gqa*il); - offload_func_v(V); - ggml_set_name(V, "V"); - - struct ggml_tensor * KQV = ggml_mul_mat(ctx0, V, KQ_soft_max); - offload_func_v(KQV); - ggml_set_name(KQV, "KQV"); - - struct ggml_tensor * KQV_merged = ggml_permute(ctx0, KQV, 0, 2, 1, 3); - offload_func_v(KQV_merged); - ggml_set_name(KQV_merged, "KQV_merged"); - - cur = ggml_cont_2d(ctx0, KQV_merged, n_embd, n_tokens); - offload_func_v(cur); - ggml_set_name(cur, "KQV_merged_contiguous"); - - cur = ggml_mul_mat(ctx0, model.layers[il].wo, cur); - offload_func(cur); - ggml_set_name(cur, "result_wo"); + cur = llm_build_kqv(lctx, ctx0, Qcur, + model.layers[il].wo, NULL, + Qcur, KQ_scale, KQ_mask, n_tokens, n_kv, max_alibi_bias, cb, il); + cb(cur, "kqv_out", il); } // Add the input - cur = ggml_add(ctx0, cur, inpL); - offload_func(cur); - - struct ggml_tensor * attn_out = cur; + struct ggml_tensor * ffn_inp = ggml_add(ctx0, cur, inpL); + cb(ffn_inp, "ffn_inp", il); // feed forward { - // Norm - { - cur = ggml_norm(ctx0, attn_out, norm_eps); - offload_func(cur); + cur = llm_build_norm(ctx0, ffn_inp, + model.layers[il].ffn_norm, + NULL, + LLM_NORM, norm_eps, cb, il); + cb(cur, "ffn_norm", il); - cur = ggml_mul(ctx0, cur, model.layers[il].ffn_norm); - offload_func(cur); - } - - cur = ggml_mul_mat(ctx0, model.layers[il].w3, cur); - offload_func(cur); - - cur = ggml_gelu(ctx0, cur); - offload_func(cur); - cur = ggml_mul_mat(ctx0, model.layers[il].w2, cur); - offload_func(cur); + cur = llm_build_ffn(ctx0, cur, + model.layers[il].ffn_up, NULL, + NULL, NULL, + model.layers[il].ffn_down, NULL, + LLM_FFN_GELU, LLM_FFN_SEQ, cb, il); + cb(cur, "ffn_out", il); } - cur = ggml_add(ctx0, cur, attn_out); - offload_func(cur); + cur = ggml_add(ctx0, cur, ffn_inp); + cb(cur, "l_out", il); + // input for next layer inpL = cur; } cur = inpL; - // norm - { - cur = ggml_norm(ctx0, cur, norm_eps); - offload_func_nr(cur); - - cur = ggml_mul(ctx0, cur, model.output_norm); - ggml_set_name(cur, "result_norm"); - } + cur = llm_build_norm(ctx0, cur, + model.output_norm, + NULL, + LLM_NORM, norm_eps, cb, -1); + cb(cur, "result_norm", -1); cur = ggml_mul_mat(ctx0, model.output, cur); - ggml_set_name(cur, "result_output"); + cb(cur, "result_output", -1); ggml_build_forward_expand(gf, cur); @@ -5857,50 +4692,494 @@ static struct ggml_cgraph * llm_build_mpt( return gf; } +// +// tensor offloading helpers +// +// TODO: will be removed with backend v2 + +enum llm_offload_func_e { + OFFLOAD_FUNC_NOP, + OFFLOAD_FUNC, + OFFLOAD_FUNC_KQ, + OFFLOAD_FUNC_V, + OFFLOAD_FUNC_NR, + OFFLOAD_FUNC_EMB, + OFFLOAD_FUNC_OUT, +}; + +// TODO: will be removed with backend v2 +struct llm_offload_trie { + struct node { + ~node() { + for (int i = 0; i < 256; ++i) { + if (children[i]) { + delete children[i]; + } + } + } + + node * children[256] = { nullptr }; + llm_offload_func_e func = OFFLOAD_FUNC_NOP; + }; + + llm_offload_trie() { + root = new node; + } + + llm_offload_trie(const std::unordered_map & map) { + root = new node; + + for (const auto & kv : map) { + add(kv.first, kv.second); + } + } + + ~llm_offload_trie() { + delete root; + } + + void add(const char * name, llm_offload_func_e func) { + node * cur = root; + + for (int i = 0; ; ++i) { + const uint8_t c = name[i]; + + if (!c) { + break; + } + + if (!cur->children[c]) { + cur->children[c] = new node; + } + + cur = cur->children[c]; + } + + cur->func = func; + } + + llm_offload_func_e find(const char * name) const { + const node * cur = root; + + for (int i = 0; ; ++i) { + const uint8_t c = name[i]; + + if (!c) { + break; + } + + if (!cur->children[c]) { + return OFFLOAD_FUNC_NOP; + } + + cur = cur->children[c]; + } + + return cur->func; + } + + node * root = nullptr; +}; + +// TODO: will be removed with backend v2 +static const std::unordered_map k_offload_map = { + //{ "inp_tokens", OFFLOAD_FUNC_NR }, // TODO: missing K-quants get_rows kernel + //{ "inp_embd", OFFLOAD_FUNC_NR }, // TODO: missing K-quants get_rows kernel + { "pos_embd", OFFLOAD_FUNC_NR }, + + { "inp_pos", OFFLOAD_FUNC_KQ }, // this is often used for KQ ops (e.g. rope) + { "KQ_scale", OFFLOAD_FUNC_KQ }, + { "KQ_mask", OFFLOAD_FUNC_KQ }, + { "K_shift", OFFLOAD_FUNC_KQ }, + { "K_shifted", OFFLOAD_FUNC_KQ }, + + { "inp_norm", OFFLOAD_FUNC_NR }, + { "inp_norm_w", OFFLOAD_FUNC_NR }, + { "inp_norm_wb", OFFLOAD_FUNC_NR }, + + { "norm", OFFLOAD_FUNC }, + { "norm_w", OFFLOAD_FUNC }, + { "norm_wb", OFFLOAD_FUNC }, + + { "attn_norm", OFFLOAD_FUNC }, + { "attn_norm_2", OFFLOAD_FUNC }, + + { "wqkv", OFFLOAD_FUNC_KQ }, + { "bqkv", OFFLOAD_FUNC_KQ }, + { "wqkv_clamped", OFFLOAD_FUNC_KQ }, + + { "tmpk", OFFLOAD_FUNC_KQ }, + { "tmpq", OFFLOAD_FUNC_KQ }, + { "tmpv", OFFLOAD_FUNC_V }, + { "Kcur", OFFLOAD_FUNC_KQ }, + { "Qcur", OFFLOAD_FUNC_KQ }, + { "Vcur", OFFLOAD_FUNC_V }, + + { "krot", OFFLOAD_FUNC_KQ }, + { "qrot", OFFLOAD_FUNC_KQ }, + { "kpass", OFFLOAD_FUNC_KQ }, + { "qpass", OFFLOAD_FUNC_KQ }, + { "krotated", OFFLOAD_FUNC_KQ }, + { "qrotated", OFFLOAD_FUNC_KQ }, + + { "q", OFFLOAD_FUNC_KQ }, + { "k", OFFLOAD_FUNC_KQ }, + { "kq", OFFLOAD_FUNC_KQ }, + { "kq_scaled", OFFLOAD_FUNC_KQ }, + { "kq_scaled_alibi", OFFLOAD_FUNC_KQ }, + { "kq_masked", OFFLOAD_FUNC_KQ }, + { "kq_soft_max", OFFLOAD_FUNC_V }, + { "v", OFFLOAD_FUNC_V }, + { "kqv", OFFLOAD_FUNC_V }, + { "kqv_merged", OFFLOAD_FUNC_V }, + { "kqv_merged_cont", OFFLOAD_FUNC_V }, + { "kqv_wo", OFFLOAD_FUNC_V }, + { "kqv_out", OFFLOAD_FUNC_V }, + + { "ffn_inp", OFFLOAD_FUNC }, + { "ffn_norm", OFFLOAD_FUNC }, + + { "ffn_up", OFFLOAD_FUNC }, + { "ffn_up_b", OFFLOAD_FUNC }, + { "ffn_gate", OFFLOAD_FUNC }, + { "ffn_gate_b", OFFLOAD_FUNC }, + { "ffn_gate_par", OFFLOAD_FUNC }, + { "ffn_down", OFFLOAD_FUNC }, + { "ffn_down_b", OFFLOAD_FUNC }, + { "ffn_out", OFFLOAD_FUNC }, + + { "ffn_silu", OFFLOAD_FUNC }, + { "ffn_gelu", OFFLOAD_FUNC }, + { "ffn_relu", OFFLOAD_FUNC }, + { "ffn_sqr(relu)", OFFLOAD_FUNC }, + + { "l_out", OFFLOAD_FUNC }, + + { "result_norm", OFFLOAD_FUNC_EMB }, + { "result_output", OFFLOAD_FUNC_OUT }, +}; + +static llm_offload_trie k_offload_func_trie(k_offload_map); + static struct ggml_cgraph * llama_build_graph( llama_context & lctx, const llama_batch & batch) { const auto & model = lctx.model; + // check if we should build the worst-case graph (for memory measurement) + const bool worst_case = ggml_allocr_is_measure(lctx.alloc); + + // keep track of the input that has already been allocated + bool alloc_inp_tokens = false; + bool alloc_inp_embd = false; + bool alloc_inp_pos = false; + bool alloc_inp_KQ_scale = false; + bool alloc_inp_KQ_mask = false; + bool alloc_inp_K_shift = false; + +#ifdef GGML_USE_CUBLAS + const bool do_offload = true; +#else + const bool do_offload = true; // TODO: set to false after finishing refactoring +#endif + + int n_non_view = 0; // number of non-view tensors that have been processed by the callback + + // this callback allows us to apply custom logic to each tensor (e.g. ggml-alloc, offloading, etc.) + // TODO: will be removed with backend v2 + llm_build_cb cb = [&](struct ggml_tensor * cur, const char * name, int il) { + if (il >= 0) { + ggml_format_name(cur, "%s-%d", name, il); + } else { + ggml_set_name(cur, name); + } + + // + // allocate input tensors and set input data + // + // TODO: will be removed with backend v2 + + if (!alloc_inp_tokens && strcmp(name, "inp_tokens") == 0) { + ggml_allocr_alloc(lctx.alloc, cur); + + if (!ggml_allocr_is_measure(lctx.alloc) && batch.token) { + const int64_t n_tokens = cur->ne[0]; + + memcpy(cur->data, batch.token, n_tokens*ggml_element_size(cur)); + } + + alloc_inp_tokens = true; + } + + if (!alloc_inp_embd && strcmp(name, "inp_embd") == 0) { + ggml_allocr_alloc(lctx.alloc, cur); + + if (!ggml_allocr_is_measure(lctx.alloc) && batch.embd) { + const int64_t n_embd = cur->ne[0]; + const int64_t n_tokens = cur->ne[1]; + + memcpy(cur->data, batch.embd, n_tokens*n_embd*ggml_element_size(cur)); + } + + alloc_inp_embd = true; + } + + if (!alloc_inp_pos && strcmp(name, "inp_pos") == 0) { + ggml_allocr_alloc(lctx.alloc, cur); + + if (!ggml_allocr_is_measure(lctx.alloc) && batch.pos) { + const int64_t n_tokens = cur->ne[0]; + + int32_t * data = (int32_t *) cur->data; + + for (int i = 0; i < n_tokens; ++i) { + data[i] = batch.pos[i]; + } + } + + alloc_inp_pos = true; + } + + if (!alloc_inp_KQ_scale && strcmp(name, "KQ_scale") == 0) { + ggml_allocr_alloc(lctx.alloc, cur); + + if (!ggml_allocr_is_measure(lctx.alloc)) { + const int64_t n_embd_head = model.hparams.n_embd_head(); + ggml_set_f32(cur, 1.0f/sqrtf(float(n_embd_head))); + } + + alloc_inp_KQ_scale = true; + } + + if (!alloc_inp_KQ_mask && strcmp(name, "KQ_mask") == 0) { + ggml_allocr_alloc(lctx.alloc, cur); + + if (!ggml_allocr_is_measure(lctx.alloc)) { + const int64_t n_kv = cur->ne[0]; + const int64_t n_tokens = cur->ne[1]; + + float * data = (float *) cur->data; + memset(data, 0, ggml_nbytes(cur)); + + for (int h = 0; h < 1; ++h) { + for (int j = 0; j < n_tokens; ++j) { + const llama_pos pos = batch.pos[j]; + const llama_seq_id seq_id = batch.seq_id[j][0]; + + for (int i = 0; i < n_kv; ++i) { + if (!lctx.kv_self.cells[i].has_seq_id(seq_id) || lctx.kv_self.cells[i].pos > pos) { + data[h*(n_kv*n_tokens) + j*n_kv + i] = -INFINITY; + } + } + } + } + } + + alloc_inp_KQ_mask = true; + } + + if (!alloc_inp_K_shift && strcmp(name, "K_shift") == 0) { + ggml_allocr_alloc(lctx.alloc, cur); + + if (!ggml_allocr_is_measure(lctx.alloc)) { + const int64_t n_ctx = cur->ne[0]; + + int32_t * data = (int32_t *) cur->data; + + for (int i = 0; i < n_ctx; ++i) { + data[i] = lctx.kv_self.cells[i].delta; + } + } + + alloc_inp_K_shift = true; + } + + // view tensors are not processed further + if (cur->view_src != nullptr) { + return; + } + + if (cur->op != GGML_OP_NONE) { + n_non_view++; + } + + // + // offload layers + // + // TODO: will be removed with backend v2 + +//#define LLAMA_OFFLOAD_DEBUG + + if (!do_offload) { + return; + } + + const int n_layer = model.hparams.n_layer; + + const int n_gpu_layers = model.n_gpu_layers; + const int i_gpu_start = n_layer - n_gpu_layers; + + // should we offload the final norm? yes if we are not computing embeddings + const bool offload_emb = lctx.embedding.empty(); + + static const std::unordered_map> k_offload_func_name = { + { OFFLOAD_FUNC_NOP, "CPU" }, + { OFFLOAD_FUNC_OUT, "CPU" }, +#ifdef GGML_USE_CUBLAS + { OFFLOAD_FUNC, "GPU (CUDA)" }, + { OFFLOAD_FUNC_KQ, "GPU (CUDA) KQ" }, + { OFFLOAD_FUNC_V, "GPU (CUDA) V" }, + { OFFLOAD_FUNC_NR, "GPU (CUDA) NR" }, + { OFFLOAD_FUNC_EMB, "GPU (CUDA) EMB" }, +#else + { OFFLOAD_FUNC, "CPU" }, + { OFFLOAD_FUNC_KQ, "CPU" }, + { OFFLOAD_FUNC_V, "CPU" }, + { OFFLOAD_FUNC_NR, "CPU" }, + { OFFLOAD_FUNC_EMB, "CPU" }, +#endif // GGML_USE_CUBLAS + }; + + // check the global map for what offload function to use for this tensor + llm_offload_func_e func_e = k_offload_func_trie.find(name); + + if (func_e == OFFLOAD_FUNC_NOP) { +#ifdef LLAMA_OFFLOAD_DEBUG + // if a tensor hasn't been offloaded, we warn the user + if (worst_case) { + LLAMA_LOG_WARN("%s: %32s: not offloaded (ref: %s)\n", __func__, + cur->name, "https://github.com/ggerganov/llama.cpp/pull/3837"); + } +#endif + + return; + } + + // count the number of layers and respect the provided n_gpu_layers + switch (func_e) { + case OFFLOAD_FUNC_NOP: + case OFFLOAD_FUNC_OUT: + break; + case OFFLOAD_FUNC: + if (n_gpu_layers < n_layer) { + if (il < i_gpu_start) { + func_e = OFFLOAD_FUNC_NOP; + } + } + break; + case OFFLOAD_FUNC_NR: + if (n_gpu_layers <= n_layer + 0) { + func_e = OFFLOAD_FUNC_NOP; + } + break; + case OFFLOAD_FUNC_V: + if (n_gpu_layers <= n_layer + 1) { + func_e = OFFLOAD_FUNC_NOP; + } + break; + case OFFLOAD_FUNC_KQ: + if (n_gpu_layers <= n_layer + 2) { + func_e = OFFLOAD_FUNC_NOP; + } + break; + case OFFLOAD_FUNC_EMB: + if (!offload_emb || n_gpu_layers < n_layer) { + func_e = OFFLOAD_FUNC_NOP; + } + break; + default: GGML_ASSERT(false); + } + + offload_func_t func = ggml_offload_nop; + + // this is needed for compatibility with Metal for example +#ifdef GGML_USE_CUBLAS + static offload_func_t ggml_offload_gpu = ggml_cuda_assign_buffers_no_alloc; +#else + static offload_func_t ggml_offload_gpu = ggml_offload_nop; +#endif + + switch (func_e) { + case OFFLOAD_FUNC_NOP: + case OFFLOAD_FUNC_OUT: func = ggml_offload_nop; break; + case OFFLOAD_FUNC: + case OFFLOAD_FUNC_KQ: + case OFFLOAD_FUNC_V: + case OFFLOAD_FUNC_NR: + case OFFLOAD_FUNC_EMB: func = ggml_offload_gpu; break; + default: GGML_ASSERT(false); + } + + // apply offload function to the tensor + func(cur); + +#ifdef LLAMA_OFFLOAD_DEBUG + if (worst_case) { + LLAMA_LOG_INFO("%s: %32s: %s\n", __func__, cur->name, k_offload_func_name.at(func_e).c_str()); + } +#endif + }; + struct ggml_cgraph * result = NULL; switch (model.arch) { case LLM_ARCH_LLAMA: { - result = llm_build_llama(lctx, batch); + result = llm_build_llama(lctx, batch, cb, worst_case); } break; case LLM_ARCH_BAICHUAN: { - result = llm_build_baichaun(lctx, batch); + result = llm_build_baichaun(lctx, batch, cb, worst_case); } break; case LLM_ARCH_FALCON: { - result = llm_build_falcon(lctx, batch); + result = llm_build_falcon(lctx, batch, cb, worst_case); } break; case LLM_ARCH_STARCODER: { - result = llm_build_starcoder(lctx, batch); + result = llm_build_starcoder(lctx, batch, cb, worst_case); } break; case LLM_ARCH_PERSIMMON: { - result = llm_build_persimmon(lctx, batch); + result = llm_build_persimmon(lctx, batch, cb, worst_case); } break; case LLM_ARCH_REFACT: { - result = llm_build_refact(lctx, batch); + result = llm_build_refact(lctx, batch, cb, worst_case); } break; case LLM_ARCH_BLOOM: { - result = llm_build_bloom(lctx, batch); + result = llm_build_bloom(lctx, batch, cb, worst_case); } break; case LLM_ARCH_MPT: { - result = llm_build_mpt(lctx, batch); + result = llm_build_mpt(lctx, batch, cb, worst_case); } break; default: GGML_ASSERT(false); } + if (worst_case) { + int n_non_view_total = 0; + + for (int i = 0; i < result->n_nodes; ++i) { + if (result->nodes[i]->view_src == nullptr) { + n_non_view_total++; + } + } + + LLAMA_LOG_INFO("%s: non-view tensors processed: %d/%d\n", __func__, n_non_view, n_non_view_total); + + if (n_non_view != n_non_view_total) { + LLAMA_LOG_WARN("%s: ****************************************************************\n", __func__); + LLAMA_LOG_WARN("%s: not all non-view tensors have been processed with a callback\n", __func__); + LLAMA_LOG_WARN("%s: this can indicate an inefficiency in the graph implementation\n", __func__); + LLAMA_LOG_WARN("%s: build with LLAMA_OFFLOAD_DEBUG for more info\n", __func__); + LLAMA_LOG_WARN("%s: ref: https://github.com/ggerganov/llama.cpp/pull/3837\n", __func__); + LLAMA_LOG_WARN("%s: ****************************************************************\n", __func__); + } + } + return result; } @@ -6043,11 +5322,13 @@ static int llama_decode_internal( } // If all tensors can be run on the GPU then using more than 1 thread is detrimental. - const bool full_offload_supported = model.arch == LLM_ARCH_LLAMA || + const bool full_offload_supported = + model.arch == LLM_ARCH_LLAMA || model.arch == LLM_ARCH_BAICHUAN || - model.arch == LLM_ARCH_FALCON || - model.arch == LLM_ARCH_REFACT || + model.arch == LLM_ARCH_FALCON || + model.arch == LLM_ARCH_REFACT || model.arch == LLM_ARCH_MPT; + const bool fully_offloaded = model.n_gpu_layers >= (int) hparams.n_layer + 3; if (ggml_cpu_has_cublas() && full_offload_supported && fully_offloaded) { n_threads = 1; @@ -6102,6 +5383,8 @@ static int llama_decode_internal( //} // extract logits + // TODO: do not compute and extract logits if only embeddings are needed + // need to update the graphs to skip "result_output" { auto & logits_out = lctx.logits; @@ -8713,8 +7996,8 @@ static int llama_apply_lora_from_file_internal( ggml_tensor * dest_t = model_tensors[base_name]; - offload_func_t offload_func = llama_nop; - offload_func_t offload_func_force_inplace = llama_nop; + offload_func_t offload_func = ggml_offload_nop; + offload_func_t offload_func_force_inplace = ggml_offload_nop; #ifdef GGML_USE_CUBLAS if (dest_t->backend == GGML_BACKEND_GPU || dest_t->backend == GGML_BACKEND_GPU_SPLIT) {