replica
/
llama.cpp
mirror of https://github.com/ggml-org/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

kv-cache : fix k-shift for multiple streams (#14742) 

ggml-ci
This commit is contained in:
Georgi Gerganov 2025-07-17 20:52:33 +03:00 committed by GitHub
parent 01612b7409
commit d6fb3f6b49
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/llama-kv-cache-unified.cpp
View File

@ -1261,7 +1261,7 @@ void llama_kv_cache_unified::set_input_k_shift(ggml_tensor * dst) const {
const auto & cells = v_cells[s]; const auto & cells = v_cells[s];
for (uint32_t i = 0; i < cells.size(); ++i) { for (uint32_t i = 0; i < cells.size(); ++i) {
data[i] = cells.is_empty(i) ? 0 : cells.get_shift(i); data[s*cells.size() + i] = cells.is_empty(i) ? 0 : cells.get_shift(i);
} }
} }
} }