tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-02 05:15:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

apply various in places

Browse Source
This commit is contained in:
Xuan Son Nguyen
2025-03-01 20:42:18 +01:00
parent 1d6ba97789
commit 46596caf6d
12 changed files with 142 additions and 133 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/llama-bench/llama-bench.cpp
View File

@ -1444,7 +1444,8 @@ static void test_prompt(llama_context * ctx, int n_prompt, int n_batch, int n_th
for (int i = 1; i < n_tokens; i++) {
tokens[i] = std::rand() % n_vocab;
}
llama_decode(ctx, llama_batch_get_one(tokens.data(), n_tokens));
llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(tokens.data(), n_tokens, 0, 0));
llama_decode_ext(ctx, batch.get());
n_processed += n_tokens;
}
@ -1461,7 +1462,8 @@ static void test_gen(llama_context * ctx, int n_gen, int n_threads) {
llama_token token = llama_vocab_get_add_bos(vocab) ? llama_vocab_bos(vocab) : std::rand() % n_vocab;
for (int i = 0; i < n_gen; i++) {
llama_decode(ctx, llama_batch_get_one(&token, 1));
llama_batch_ext_ptr batch(llama_batch_ext_init_from_text(&token, 1, 0, 0));
llama_decode_ext(ctx, batch.get());
llama_synchronize(ctx);
token = std::rand() % n_vocab;
}