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

kv-cache : better estimate of n_kv for multi-sequence batches (#15610)

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-08-27 13:55:12 +03:00
committed by GitHub
parent 1e7489745a
commit 1bded5a3b3
2 changed files with 15 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
src/llama-kv-cache.cpp
View File

@@ -771,8 +771,8 @@ llama_kv_cache::slot_info llama_kv_cache::find_slot(const llama_ubatch & ubatch,
GGML_ASSERT(ubatch.seq_id [s*n_tokens][0] == seq_id);
}
res.s0 = std::min<llama_seq_id>(res.s0, seq_to_stream[seq_id]);
res.s1 = std::max<llama_seq_id>(res.s1, seq_to_stream[seq_id]);
res.s0 = std::min<uint32_t>(res.s0, seq_to_stream[seq_id]);
res.s1 = std::max<uint32_t>(res.s1, seq_to_stream[seq_id]);
res.strm[s] = seq_to_stream[seq_id];
res.idxs[s].reserve(n_tokens);
@@ -964,11 +964,11 @@ bool llama_kv_cache::get_has_shift() const {
return result;
}
uint32_t llama_kv_cache::get_n_kv() const {
uint32_t llama_kv_cache::get_n_kv(const slot_info & sinfo) const {
uint32_t result = 0;
for (uint32_t s = 0; s < n_stream; ++s) {
const auto & cells = v_cells[s];
for (uint32_t s = 0; s < sinfo.n_stream(); ++s) {
const auto & cells = v_cells[sinfo.strm[s]];
result = std::max(std::min(cells.size(), std::max(n_pad, GGML_PAD(cells.used_max_p1(), n_pad))), result);
}
@@ -1985,8 +1985,7 @@ bool llama_kv_cache_context::apply() {
}
kv->apply_ubatch(sinfos[i_cur], ubatches[i_cur]);
n_kv = kv->get_n_kv();
n_kv = kv->get_n_kv(sinfos[i_cur]);
return true;
}

6
src/llama-kv-cache.h
View File

@@ -38,8 +38,8 @@ public:
using idx_vec_t = std::vector<uint32_t>;
// number of streams: ns = s1 - s0 + 1
llama_seq_id s0;
llama_seq_id s1;
uint32_t s0;
uint32_t s1;
std::vector<llama_seq_id> strm; // [ns]
std::vector<idx_vec_t> idxs; // [ns]
@@ -139,7 +139,7 @@ public:
// graph_build API
//
uint32_t get_n_kv() const;
uint32_t get_n_kv(const slot_info & sinfo) const;
// TODO: temporary
bool get_supports_set_rows() const;