Skip to content

server : fix off-by-1 in server_tokens::size_up_to_pos() #20279

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ggerganov merged 2 commits into from
Mar 9, 2026
Merged

server : fix off-by-1 in server_tokens::size_up_to_pos() #20279

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b5c75c8
server : fix off-by-1 in server_tokens::size_up_to_pos()
ggerganov Mar 9, 2026
eef2328
cont : fix typo [no ci]
ggerganov Mar 9, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions tools/server/server-common.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -276,7 +276,7 @@ llama_pos server_tokens::pos_next(int64_t n_tokens) const {

size_t server_tokens::size_up_to_pos(llama_pos max_pos) const {
if (!has_mtmd) {
return std::min((size_t)(max_pos + 1), tokens.size());
return std::min((size_t)max_pos, tokens.size());
}

size_t idx = 0;
Expand All @@ -296,7 +296,7 @@ size_t server_tokens::size_up_to_pos(llama_pos max_pos) const {
idx++;
}

if (pos > max_pos) {
if (pos >= max_pos) {
break;
}
}
Expand Down
2 changes: 1 addition & 1 deletion tools/server/server-common.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -170,7 +170,7 @@ struct server_tokens {
// the next position after n_tokens. if n_tokens < 0, return the next position after all tokens.
llama_pos pos_next(int64_t n_tokens = -1) const;

// number of tokens with position <= max_pos
// number of tokens with position < max_pos
size_t size_up_to_pos(llama_pos max_pos) const;

const mtmd::input_chunk_ptr & find_chunk(size_t idx) const;
Expand Down
8 changes: 4 additions & 4 deletions tools/server/server-context.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -570,7 +570,7 @@ struct server_context_impl {
std::vector<server_slot> slots;

int slots_debug = 0;
int n_empty_consequtive = 0;
int n_empty_consecutive = 0;

std::unique_ptr<server_prompt_cache> prompt_cache;

Expand Down Expand Up @@ -2372,7 +2372,7 @@ struct server_context_impl {
} else {
pos_next = std::min(pos_next, std::max(it->pos_min + 1, it->pos_max));
n_past = std::min(slot.prompt.tokens.size_up_to_pos(pos_next), (size_t) it->n_tokens);
SLT_WRN(slot, "restored context checkpoint (pos_min = %d, pos_max = %d, n_tokens = %" PRId64 ", size = %.3f MiB)\n", it->pos_min, it->pos_max, it->n_tokens, (float) checkpoint_size / 1024 / 1024);
SLT_WRN(slot, "restored context checkpoint (pos_min = %d, pos_max = %d, n_tokens = %" PRId64 ", n_past = %d, size = %.3f MiB)\n", it->pos_min, it->pos_max, it->n_tokens, n_past, (float) checkpoint_size / 1024 / 1024);
}
}

Expand Down Expand Up @@ -2630,11 +2630,11 @@ struct server_context_impl {
if (batch.n_tokens == 0) {
SRV_WRN("%s", "no tokens to decode\n");

if (++n_empty_consequtive > 3) {
if (++n_empty_consecutive > 3) {
GGML_ABORT("fatal error - please provide logs and repro in %s\n", "https://github.com/ggml-org/llama.cpp/pull/20277");
}
} else {
n_empty_consequtive = 0;
n_empty_consecutive = 0;
}

int32_t i_next = 0;
Expand Down