Public/koboldcpp
1
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2026-09-02 11:01:20 +02:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit '6effcecd0bf3cb2209999cecfa297ed4d8523b5a' into concedo_experimental

Browse Source 
# Conflicts:
#	.github/workflows/build-msys.yml
#	.pi/gg/SYSTEM.md
#	common/CMakeLists.txt
#	examples/speculative-simple/speculative-simple.cpp
#	ggml/src/ggml-cpu/kleidiai/kleidiai.cpp
#	ggml/src/ggml-sycl/ggml-sycl.cpp
#	ggml/src/ggml-sycl/mmvq.cpp
#	scripts/ui-assets.cmake
#	src/llama-model.cpp
#	tools/cli/cli.cpp
#	tools/imatrix/imatrix.cpp
#	tools/quantize/quantize.cpp
#	tools/server/README.md
#	tools/ui/tests/stories/a11y/ChatScreenForm.a11y.stories.svelte
This commit is contained in:
Concedo
2026-06-07 17:10:58 +08:00
parent 5f39666705 6effcecd0b
commit 202aaef40b
152 changed files with 1188 additions and 878 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-kv-cache.cpp
+4 -4
View File
@@ -101,7 +101,7 @@ llama_kv_cache::llama_kv_cache(
GGML_ASSERT(kv_size % n_pad == 0);
const uint32_t n_layer_kv = hparams.n_layer_kv();
const uint32_t n_layer = hparams.n_layer_all;
// define a comparator for the buft -> ctx map to ensure that the order is well-defined:
struct ggml_backend_buft_comparator {
@@ -116,7 +116,7 @@ llama_kv_cache::llama_kv_cache(
auto it = ctx_map.find(buft);
if (it == ctx_map.end()) {
ggml_init_params params = {
/*.mem_size =*/ size_t(2u*(1 + n_stream)*n_layer_kv*ggml_tensor_overhead()),
/*.mem_size =*/ size_t(2u*(1 + n_stream)*n_layer*ggml_tensor_overhead()),
/*.mem_buffer =*/ NULL,
/*.no_alloc =*/ true,
};
@@ -164,7 +164,7 @@ llama_kv_cache::llama_kv_cache(
const bool is_mla = hparams.is_mla();
for (uint32_t il = 0; il < hparams.n_layer; il++) {
for (uint32_t il = 0; il < n_layer; il++) {
if (!hparams.has_kv(il)) {
// LLAMA_LOG_DEBUG("%s: layer %3d: does not have KV cache\n", __func__, il);
continue;
@@ -234,7 +234,7 @@ llama_kv_cache::llama_kv_cache(
if (reuse) {
LLAMA_LOG_DEBUG("%s: reusing layers:\n", __func__);
for (uint32_t il = 0; il < hparams.n_layer; il++) {
for (uint32_t il = 0; il < n_layer; il++) {
const int32_t il_reuse = reuse(il);
if (il_reuse < 0) {