mikhail/llama.cpp-mtp-turboquant
1
0
Fork 0
Code Issues Pull Requests Actions 32 Packages Projects Releases Wiki Activity

logs : reduce (#23021)

Browse Source 
* logs : reduce

* args : fix envs

* server : fix build

* common : print verbosity level at start

* server : clean-up logs

* server : print prompt processing timings + sampling params

* minor : whitespaces
This commit is contained in:
Georgi Gerganov
2026-05-14 13:05:52 +03:00
committed by GitHub
parent 81b0d882ae
commit 67b2b7f2f2
17 changed files with 207 additions and 147 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-task.cpp
+6 -6
View File
@@ -1988,7 +1988,7 @@ server_prompt * server_prompt_cache::alloc(const server_prompt & prompt, size_t
const int cur_lcp_len = it->tokens.get_common_prefix(prompt.tokens);
if (cur_lcp_len == (int) prompt.tokens.size()) {
SRV_WRN("%s", " - prompt is already in the cache, skipping\n");
SRV_INF("%s", " - prompt is already in the cache, skipping\n");
return nullptr;
}
}
@@ -2043,7 +2043,7 @@ bool server_prompt_cache::load(server_prompt & prompt, const server_tokens & tok
float f_keep_best = prompt.tokens.size() > 0 ? float(lcp_best) / prompt.tokens.size() : -1.0f; // empty slot: any cache entry wins
float sim_best = float(lcp_best) / tokens_new.size();
SRV_WRN(" - looking for better prompt, base f_keep = %.3f, sim = %.3f\n", f_keep_best, sim_best);
SRV_INF(" - looking for better prompt, base f_keep = %.3f, sim = %.3f\n", f_keep_best, sim_best);
auto it_best = states.end();
@@ -2068,7 +2068,7 @@ bool server_prompt_cache::load(server_prompt & prompt, const server_tokens & tok
}
if (it_best != states.end()) {
SRV_WRN(" - found better prompt with f_keep = %.3f, sim = %.3f\n", f_keep_best, sim_best);
SRV_INF(" - found better prompt with f_keep = %.3f, sim = %.3f\n", f_keep_best, sim_best);
{
auto & data = it_best->data.main;
@@ -2076,7 +2076,7 @@ bool server_prompt_cache::load(server_prompt & prompt, const server_tokens & tok
const size_t size = data.size();
const size_t n = llama_state_seq_set_data_ext(ctx_tgt, data.data(), size, id_slot, 0);
if (n != size) {
SRV_WRN("failed to restore state with size %zu\n", size);
SRV_ERR("failed to restore state with size %zu\n", size);
return false;
}
@@ -2145,11 +2145,11 @@ void server_prompt_cache::update() {
}
}
SRV_WRN(" - cache state: %zu prompts, %.3f MiB (limits: %.3f MiB, %zu tokens, %zu est)\n",
SRV_INF(" - cache state: %zu prompts, %.3f MiB (limits: %.3f MiB, %zu tokens, %zu est)\n",
states.size(), size() / (1024.0 * 1024.0), limit_size / (1024.0 * 1024.0), limit_tokens, limit_tokens_cur);
for (const auto & state : states) {
SRV_WRN(" - prompt %p: %7d tokens, checkpoints: %2zu, %9.3f MiB\n",
SRV_INF(" - prompt %p: %7d tokens, checkpoints: %2zu, %9.3f MiB\n",
(const void *)&state, state.n_tokens(), state.checkpoints.size(), state.size() / (1024.0 * 1024.0));
}
}