CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-04 09:32:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

batch : add n_used count (#14512)

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-07-04 09:04:59 +03:00
committed by GitHub
parent 499a8f5a78
commit c79184d2d1
6 changed files with 34 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/llama-kv-cache-unified-iswa.cpp
View File

@@ -113,6 +113,11 @@ llama_memory_context_ptr llama_kv_cache_unified_iswa::init_batch(llama_batch_all
ubatches.push_back(std::move(ubatch)); // NOLINT
}
if (balloc.get_n_used() < balloc.get_n_tokens()) {
// failed to find a suitable split
break;
}
auto sinfos_base = kv_base->prepare(ubatches);
if (sinfos_base.empty()) {
break;
@@ -144,6 +149,11 @@ llama_memory_context_ptr llama_kv_cache_unified_iswa::init_batch(llama_batch_all
ubatches.push_back(std::move(ubatch)); // NOLINT
}
if (balloc.get_n_used() < balloc.get_n_tokens()) {
// failed to find a suitable split
break;
}
auto sinfos_base = kv_base->prepare(ubatches);
if (sinfos_base.empty()) {
break;