public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-10-18 04:35:33 +02:00
Code Issues Packages Projects Releases Wiki Activity

talk-llama : sync llama.cpp

Browse Source
This commit is contained in:
Georgi Gerganov
2025-12-31 17:52:09 +02:00
parent 54fa8216ea
commit 7359ac94d5
23 changed files with 1125 additions and 198 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/talk-llama/llama-context.cpp
+21 -20
View File
@@ -294,8 +294,8 @@ llama_context::llama_context(
// enabling pipeline parallelism in the scheduler increases memory usage, so it is only done when necessary
bool pipeline_parallel =
model.n_devices() > 1 &&
model.params.n_gpu_layers > (int) model.hparams.n_layer &&
model.params.split_mode == LLAMA_SPLIT_MODE_LAYER &&
model.n_gpu_layers() > model.hparams.n_layer &&
model.split_mode() == LLAMA_SPLIT_MODE_LAYER &&
cparams.offload_kqv &&
!model.has_tensor_overrides();
@@ -459,23 +459,22 @@ llama_context::llama_context(
}
llama_context::~llama_context() {
// FIXME this currently results in a use-after-free bug if the model is freed before the context
// if (!model.hparams.no_alloc) {
// for (size_t i = 0; i < backend_ptrs.size(); ++i) {
// ggml_backend_t backend = backend_ptrs[i];
// ggml_backend_buffer_type_t buft = backend_buft[i];
if (!model.hparams.no_alloc) {
for (size_t i = 0; i < backend_ptrs.size(); ++i) {
ggml_backend_t backend = backend_ptrs[i];
ggml_backend_buffer_type_t buft = backend_buft[i];
// const size_t size_exp = backend_buf_exp_size[i];
// const size_t size_act = ggml_backend_sched_get_buffer_size(sched.get(), backend);
// if (size_exp == size_act) {
// LLAMA_LOG_DEBUG("%s: %10s compute buffer size is %8.4f MiB, matches expectation of %8.4f MiB\n",
// __func__, ggml_backend_buft_name(buft), size_act / (1024.0*1024.0), size_exp / (1024.0*1024.0));
// } else {
// LLAMA_LOG_WARN("%s: %10s compute buffer size of %8.4f MiB, does not match expectation of %8.4f MiB\n",
// __func__, ggml_backend_buft_name(buft), size_act / (1024.0*1024.0), size_exp / (1024.0*1024.0));
// }
// }
// }
const size_t size_exp = backend_buf_exp_size[i];
const size_t size_act = ggml_backend_sched_get_buffer_size(sched.get(), backend);
if (size_exp == size_act) {
LLAMA_LOG_DEBUG("%s: %10s compute buffer size is %8.4f MiB, matches expectation of %8.4f MiB\n",
__func__, ggml_backend_buft_name(buft), size_act / (1024.0*1024.0), size_exp / (1024.0*1024.0));
} else {
LLAMA_LOG_WARN("%s: %10s compute buffer size of %8.4f MiB, does not match expectation of %8.4f MiB\n",
__func__, ggml_backend_buft_name(buft), size_act / (1024.0*1024.0), size_exp / (1024.0*1024.0));
}
}
}
ggml_opt_free(opt_ctx);
}
@@ -1443,7 +1442,9 @@ uint32_t llama_context::graph_max_nodes(uint32_t n_tokens) const {
if (model.arch == LLM_ARCH_QWEN3NEXT) {
return std::max<uint32_t>(n_tokens * 40, 32u * model.n_tensors());
}
return std::max<uint32_t>(1024u, 8u*model.n_tensors());
uint32_t res = std::max<uint32_t>(1024u, 8u*model.n_tensors());
res += model.n_lora_nodes;
return res;
}
llm_graph_result * llama_context::get_gf_res_reserve() const {
@@ -1571,7 +1572,7 @@ llm_graph_cb llama_context::graph_get_cb() const {
// norm may be automatically assigned to the backend of the previous layer, increasing data transfer between backends
// FIXME: fix in ggml_backend_sched
const bool full_offload = model.params.n_gpu_layers > (int) model.hparams.n_layer;
const bool full_offload = model.n_gpu_layers() > model.hparams.n_layer;
if (ubatch.n_tokens < 32 || full_offload) {
if (il != -1 && strcmp(name, "norm") == 0) {
const auto & dev_layer = model.dev_layer(il);