CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-28 08:31:25 +00:00
Code Issues Packages Projects Releases Wiki Activity

context : fix overflow when re-ordering huge outputs

Browse Source
This commit is contained in:
Francis Couture-Harpin
2025-08-04 22:01:28 -04:00
parent ec428b02c3
commit f16a843a38
1 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
src/llama-context.cpp
View File

@@ -1328,21 +1328,21 @@ uint32_t llama_context::output_reserve(int32_t n_outputs) {
}
void llama_context::output_reorder() {
const uint32_t n_vocab = model.vocab.n_tokens();
const uint64_t n_vocab = model.vocab.n_tokens();
const uint64_t n_embd = model.hparams.n_embd;
for (uint32_t s = 0; s < output_swaps.size(); ++s) {
const uint32_t i0 = output_swaps[s].i0;
const uint32_t i1 = output_swaps[s].i1;
for (size_t s = 0; s < output_swaps.size(); ++s) {
const uint64_t i0 = output_swaps[s].i0;
const uint64_t i1 = output_swaps[s].i1;
if (logits_size > 0) {
for (uint32_t k = 0; k < n_vocab; k++) {
for (uint64_t k = 0; k < n_vocab; k++) {
std::swap(logits[i0*n_vocab + k], logits[i1*n_vocab + k]);
}
}
if (embd_size > 0) {
for (uint32_t k = 0; k < n_embd; k++) {
for (uint64_t k = 0; k < n_embd; k++) {
std::swap(embd[i0*n_embd + k], embd[i1*n_embd + k]);
}
}