tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-07-08 12:47:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

batched-bench : fix oob write

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-07-04 13:08:31 +03:00
parent b123d89445
commit ab8443d0ea
2 changed files with 6 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
examples/parallel/parallel.cpp
View File

@ -224,6 +224,7 @@ int main(int argc, char ** argv) {
auto & client = clients[i]; auto & client = clients[i];
client.id = i; client.id = i;
client.smpl = common_sampler_init(model, params.sampling); client.smpl = common_sampler_init(model, params.sampling);
//params.sampling.seed++;
} }
std::vector<llama_token> tokens_system; std::vector<llama_token> tokens_system;
@ -345,7 +346,7 @@ int main(int argc, char ** argv) {
client.n_decoded = 0; client.n_decoded = 0;
client.i_batch = batch.n_tokens - 1; client.i_batch = batch.n_tokens - 1;
LOG_INF("\033[31mClient %3d, seq %4d, junk = %4d, started decoding ...\033[0m\n", client.id, client.seq_id, n_junk_cur); LOG_INF("\033[31mClient %3d, seq %4d, junk = %4d, prompt = %d, started decoding ...\033[0m\n", client.id, client.seq_id, n_junk_cur, client.n_prompt);
g_seq_id += 1; g_seq_id += 1;

5
tools/batched-bench/batched-bench.cpp
View File

@ -130,7 +130,10 @@ int main(int argc, char ** argv) {
common_batch_add(batch, 0, i, { j }, false); common_batch_add(batch, 0, i, { j }, false);
} }
} }
batch.logits[batch.n_tokens - 1] = true;
if (batch.n_tokens > 0) {
batch.logits[batch.n_tokens - 1] = true;
}
const auto t_pp_start = ggml_time_us(); const auto t_pp_start = ggml_time_us();