CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-31 08:51:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

adapt common

Browse Source
This commit is contained in:
Xuan Son Nguyen
2025-03-01 12:12:52 +01:00
parent a1b1dea33b
commit f0ffd81130
2 changed files with 18 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
common/common.cpp
View File

@@ -1047,7 +1047,8 @@ struct common_init_result common_init_from_params(common_params & params) {
}
if (llama_model_has_encoder(model)) {
llama_encode(lctx, llama_batch_get_one(tmp.data(), tmp.size()));
llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(tmp.data(), tmp.size(), 0, 0));
llama_encode_ext(lctx, batch.get());
llama_token decoder_start_token_id = llama_model_decoder_start_token(model);
if (decoder_start_token_id == LLAMA_TOKEN_NULL) {
decoder_start_token_id = bos;
@@ -1056,7 +1057,8 @@ struct common_init_result common_init_from_params(common_params & params) {
tmp.push_back(decoder_start_token_id);
}
if (llama_model_has_decoder(model)) {
llama_decode(lctx, llama_batch_get_one(tmp.data(), std::min(tmp.size(), (size_t) params.n_batch)));
llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(tmp.data(), std::min(tmp.size(), (size_t) params.n_batch), 0, 0));
llama_encode_ext(lctx, batch.get());
}
llama_kv_cache_clear(lctx);
llama_synchronize(lctx);