jdelony/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-06-27 23:50:20 -05:00
Code Issues Packages Projects Releases Wiki Activity

fix assert

Browse Source
This commit is contained in:
Xuan Son Nguyen 2026-06-20 19:32:47 +02:00
parent 64ec03d10b
commit bf36838ebd
1 changed files with 3 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tools/server/server-context.cpp
View File

@ -2661,12 +2661,11 @@ private:
int32_t off_next = 0;
int32_t n_batch = llama_n_batch(ctx_tgt);
for (int32_t off = 0; off < batch.size(); off = off_next) {
const int32_t n_tokens = std::min(n_batch, batch.size() - off);
try {
// TODO @ngxson : maybe handle n_batch == 1 here instead of inside decode()
const int32_t n_tokens = std::min(n_batch, batch.size() - off);
batch_view = batch.get_view(off, n_batch);
batch_view = batch.get_view(off, n_tokens);
bool ok = decode(n_batch, off, batch_view);
if (ok) {
@ -2686,7 +2685,7 @@ private:
}
try {
post_decode(n_batch, off, batch_view);
post_decode(n_tokens, off, batch_view);
} catch (const std::exception & e) {
SRV_ERR("post_decode() failed: %s\n", e.what());
abort_all_slots("post_decode() failed: " + std::string(e.what()));