vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix make bui;d errors

Browse source
This commit is contained in:
FSSRepo 2023-10-17 17:18:56 -04:00
parent ed0c11cb83
commit d2b1fac6c7
2 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

4
examples/server/server.cpp
View file

@ -922,8 +922,8 @@ struct llama_server_context
const int n_left = slot.n_past - params.n_keep - 1;
const int n_discard = n_left / 2;
llama_kv_cache_seq_rm (ctx, 0, params.n_keep + 1 , params.n_keep + n_discard + 1);
llama_kv_cache_seq_shift(ctx, 0, params.n_keep + 1 + n_discard, slot.n_past, -n_discard);
llama_kv_cache_seq_rm (ctx, slot.id, params.n_keep + 1 , params.n_keep + n_discard + 1);
llama_kv_cache_seq_shift(ctx, slot.id, params.n_keep + 1 + n_discard, slot.n_past, -n_discard);
for (size_t i = params.n_keep + 1 + n_discard; i < slot.cache_tokens.size(); i++)
{