From ad8c522bafd9ffc14b45c374ff146a281cf9d6d9 Mon Sep 17 00:00:00 2001 From: slaren Date: Thu, 5 Sep 2024 03:29:01 +0200 Subject: [PATCH] fix defrag with quantized KV --- ggml/src/ggml-backend.c | 10 +++++++++- ggml/src/ggml-cuda.cu | 14 ++++++++++++-- ggml/src/ggml-cuda/cpy.cu | 30 ++++++++++++++++++------------ 3 files changed, 39 insertions(+), 15 deletions(-) diff --git a/ggml/src/ggml-backend.c b/ggml/src/ggml-backend.c index 5b877db3566e76..4d4b7d5e03fe92 100644 --- a/ggml/src/ggml-backend.c +++ b/ggml/src/ggml-backend.c @@ -1163,6 +1163,11 @@ static int ggml_backend_sched_backend_id_from_cur(ggml_backend_sched_t sched, st SET_CAUSE(tensor, "1.vsrc"); return cur_backend_id; } + if (tensor->view_src->buffer != NULL) { + // since the tensor is pre-allocated, it cannot be moved to another backend + // TODO: support moving pre-allocated tensors + GGML_ABORT("pre-allocatod tensor in a backend that cannot run the operation"); + } } // graph input @@ -1644,7 +1649,7 @@ static void ggml_backend_sched_split_graph(ggml_backend_sched_t sched, struct gg sched->prev_leaf_backend_ids = tmp; } - int graph_size = graph->n_nodes + sched->n_splits*GGML_SCHED_MAX_SPLIT_INPUTS*2; + int graph_size = MAX(graph->n_nodes, graph->n_leafs) + sched->n_splits*GGML_SCHED_MAX_SPLIT_INPUTS*2*sched->n_copies; if (sched->graph.size < graph_size) { sched->graph.size = graph_size; sched->graph.nodes = realloc(sched->graph.nodes, graph_size * sizeof(struct ggml_tensor *)); @@ -1696,6 +1701,7 @@ static void ggml_backend_sched_split_graph(ggml_backend_sched_t sched, struct gg for (int c = 0; c < sched->n_copies; c++) { struct ggml_tensor * input_cpy = tensor_id_copy(id, backend_id, c); sched->leaf_backend_ids[graph_copy->n_leafs] = backend_id; + assert(graph_copy->size > graph_copy->n_leafs); graph_copy->leafs[graph_copy->n_leafs++] = input_cpy; } } @@ -1709,6 +1715,7 @@ static void ggml_backend_sched_split_graph(ggml_backend_sched_t sched, struct gg for (int c = 0; c < sched->n_copies; c++) { struct ggml_tensor * input_cpy = tensor_id_copy(id, backend_id, c); sched->leaf_backend_ids[graph_copy->n_leafs] = backend_id; + assert(graph_copy->size > graph_copy->n_leafs); graph_copy->leafs[graph_copy->n_leafs++] = input_cpy; } } @@ -1719,6 +1726,7 @@ static void ggml_backend_sched_split_graph(ggml_backend_sched_t sched, struct gg for (int i = 0; i < graph->n_leafs; i++) { struct ggml_tensor * leaf = graph->leafs[i]; sched->leaf_backend_ids[graph_copy->n_leafs] = tensor_backend_id(leaf); + assert(graph_copy->size > graph_copy->n_leafs); graph_copy->leafs[graph_copy->n_leafs++] = leaf; } } diff --git a/ggml/src/ggml-cuda.cu b/ggml/src/ggml-cuda.cu index 8a844b02a27a5a..d33988d0277f6f 100644 --- a/ggml/src/ggml-cuda.cu +++ b/ggml/src/ggml-cuda.cu @@ -2572,8 +2572,15 @@ GGML_CALL static enum ggml_status ggml_backend_cuda_graph_compute(ggml_backend_t cuda_ctx->cuda_graph->updated_kernel_arg.push_back((char **) &(node->src[1]->data)); // store a pointer to each copy op CUDA kernel to identify it later void * ptr = ggml_cuda_cpy_fn(node->src[0], node->src[1]); - if (std::find(ggml_cuda_cpy_fn_ptrs.begin(), ggml_cuda_cpy_fn_ptrs.end(), ptr) == ggml_cuda_cpy_fn_ptrs.end()) { - ggml_cuda_cpy_fn_ptrs.push_back(ptr); + if (!ptr) { + use_cuda_graph = false; +#ifndef NDEBUG + GGML_CUDA_LOG_WARN("%s: disabling CUDA graphs due to unsupported copy op\n", __func__); +#endif + } else { + if (std::find(ggml_cuda_cpy_fn_ptrs.begin(), ggml_cuda_cpy_fn_ptrs.end(), ptr) == ggml_cuda_cpy_fn_ptrs.end()) { + ggml_cuda_cpy_fn_ptrs.push_back(ptr); + } } } @@ -2842,6 +2849,9 @@ GGML_CALL static bool ggml_backend_cuda_supports_op(ggml_backend_t backend, cons if (src0_type == GGML_TYPE_F16 && src1_type == GGML_TYPE_F32) { return true; } + if (src0_type == src1_type && ggml_is_contiguous(op->src[0]) && ggml_is_contiguous(op->src[1])) { + return true; + } return false; } break; case GGML_OP_DUP: diff --git a/ggml/src/ggml-cuda/cpy.cu b/ggml/src/ggml-cuda/cpy.cu index aad34bfe5b32b4..5a714019e5f3bb 100644 --- a/ggml/src/ggml-cuda/cpy.cu +++ b/ggml/src/ggml-cuda/cpy.cu @@ -428,7 +428,10 @@ void ggml_cuda_cpy(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, gg char * src0_ddc = (char *) src0->data; char * src1_ddc = (char *) src1->data; - if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F32) { + if (src0->type == src1->type && ggml_is_contiguous(src0) && ggml_is_contiguous(src1)) { + GGML_ASSERT(ggml_nbytes(src0) == ggml_nbytes(src1)); + CUDA_CHECK(cudaMemcpyAsync(src1_ddc, src0_ddc, ggml_nbytes(src0), cudaMemcpyDeviceToDevice, main_stream)); + } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F32) { ggml_cpy_f32_f32_cuda (src0_ddc, src1_ddc, ne, ne00, ne01, ne02, nb00, nb01, nb02, nb03, ne10, ne11, ne12, nb10, nb11, nb12, nb13, main_stream); } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F16) { ggml_cpy_f32_f16_cuda (src0_ddc, src1_ddc, ne, ne00, ne01, ne02, nb00, nb01, nb02, nb03, ne10, ne11, ne12, nb10, nb11, nb12, nb13, main_stream); @@ -451,6 +454,7 @@ void ggml_cuda_cpy(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, gg } else { fprintf(stderr, "%s: unsupported type combination (%s to %s)\n", __func__, ggml_type_name(src0->type), ggml_type_name(src1->type)); + fprintf(stderr, "contiguous: %d %d\n", ggml_is_contiguous(src0), ggml_is_contiguous(src1)); GGML_ABORT("fatal error"); } } @@ -461,26 +465,28 @@ void ggml_cuda_dup(ggml_backend_cuda_context & ctx, ggml_tensor * dst) { } void* ggml_cuda_cpy_fn(const ggml_tensor * src0, ggml_tensor * src1) { - if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F32) { - return (void*) cpy_f32_f16; + if (src0->type == src1->type && ggml_is_contiguous(src0) && ggml_is_contiguous(src1)) { + return nullptr; + } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F32) { + return (void*) cpy_f32_f16; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F16) { - return (void*) cpy_f32_f16; + return (void*) cpy_f32_f16; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_Q8_0) { - return (void*) cpy_f32_q; + return (void*) cpy_f32_q; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_Q4_0) { - return (void*) cpy_f32_q; + return (void*) cpy_f32_q; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_Q4_1) { - return (void*) cpy_f32_q; + return (void*) cpy_f32_q; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_Q5_0) { - return (void*) cpy_f32_q; + return (void*) cpy_f32_q; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_IQ4_NL) { - return (void*) cpy_f32_q; + return (void*) cpy_f32_q; } else if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_Q5_1) { - return (void*) cpy_f32_q; + return (void*) cpy_f32_q; } else if (src0->type == GGML_TYPE_F16 && src1->type == GGML_TYPE_F16) { - return (void*) cpy_f32_f16; + return (void*) cpy_f32_f16; } else if (src0->type == GGML_TYPE_F16 && src1->type == GGML_TYPE_F32) { - return (void*) cpy_f32_f16; + return (void*) cpy_f32_f16; } else { fprintf(stderr, "%s: unsupported type combination (%s to %s)\n", __func__, ggml_type_name(src0->type), ggml_type_name(src1->type));