CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-09 10:17:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

server : disable checkpoints with mtmd (#17045)

Browse Source
This commit is contained in:
Georgi Gerganov
2025-11-06 12:09:29 +02:00
committed by GitHub
parent 4882f0ff78
commit b7f9010d24
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tools/server/server.cpp
View File

@@ -3832,7 +3832,9 @@ struct server_context {
// the largest pos_min required for a checkpoint to be useful // the largest pos_min required for a checkpoint to be useful
const auto pos_min_thold = std::max(0, n_past - n_swa); const auto pos_min_thold = std::max(0, n_past - n_swa);
if (n_past > 0 && n_past < slot.prompt.n_tokens()) { // note: disallow with mtmd contexts for now
// https://github.com/ggml-org/llama.cpp/issues/17043
if (!mctx && n_past > 0 && n_past < slot.prompt.n_tokens()) {
const auto pos_min = llama_memory_seq_pos_min(llama_get_memory(ctx), slot.id); const auto pos_min = llama_memory_seq_pos_min(llama_get_memory(ctx), slot.id);
if (pos_min == -1) { if (pos_min == -1) {
SLT_ERR(slot, "n_past = %d, slot.prompt.tokens.size() = %d, seq_id = %d, pos_min = %d\n", n_past, (int) slot.prompt.tokens.size(), slot.id, pos_min); SLT_ERR(slot, "n_past = %d, slot.prompt.tokens.size() = %d, seq_id = %d, pos_min = %d\n", n_past, (int) slot.prompt.tokens.size(), slot.id, pos_min);