ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)

Browse source 
* llama : refactor llama_context, llama_kv_cache, llm_build_context

ggml-ci

* graph : don't mutate the KV cache during defrag

ggml-ci

* context : reduce virtuals + remove test function

ggml-ci

* context : move interface implementation to source file + factory

ggml-ci

* graph : move KV cache build functions to llama_context impl

ggml-ci

* graph : remove model reference from build_pooling

ggml-ci

* graph : remove llama_model reference

ggml-ci

* kv_cache : provide rope factors

ggml-ci

* graph : rework inputs to use only unique_ptr, remove attn input abstraction

ggml-ci

* context : remove llama_context_i abstraction

ggml-ci

* context : clean-up

ggml-ci

* graph : clean-up

ggml-ci

* llama : remove redundant keywords (struct, enum)

ggml-ci

* model : adapt gemma3

ggml-ci

* graph : restore same attention ops as on master

ggml-ci

* llama : remove TODO + fix indent

ggml-ci
This commit is contained in:
Georgi Gerganov 2025-03-13 12:35:44 +02:00 • committed by GitHub
parent 2048b5913d
commit e0dbec0bc6
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
46 changed files with 13903 additions and 12190 deletions
Download patch file Download diff file Expand all files Collapse all files

28
examples/passkey/passkey.cpp
View file

@ -133,11 +133,11 @@ int main(int argc, char ** argv) {
const int ib = i/n_batch - 1;
const int bd = n_batch_grp*(n_grp - 1);
llama_kv_cache_seq_add (ctx, 0, n_past - n_batch, n_past, ib*bd);
llama_kv_cache_seq_div (ctx, 0, n_past - n_batch + ib*bd, n_past + ib*bd, n_grp);
llama_kv_cache_update (ctx);
llama_kv_self_seq_add (ctx, 0, n_past - n_batch, n_past, ib*bd);
llama_kv_self_seq_div (ctx, 0, n_past - n_batch + ib*bd, n_past + ib*bd, n_grp);
llama_kv_self_update (ctx);
n_past = llama_kv_cache_seq_pos_max(ctx, 0) + 1;
n_past = llama_kv_self_seq_pos_max(ctx, 0) + 1;
}
common_batch_clear(batch);
@ -167,12 +167,12 @@ int main(int argc, char ** argv) {
LOG_INF("%s: shifting KV cache with %d\n", __func__, n_discard);
llama_kv_cache_seq_rm (ctx, 0, n_keep , n_keep + n_discard);
llama_kv_cache_seq_add(ctx, 0, n_keep + n_discard, n_ctx, -n_discard);
//llama_kv_cache_defrag (ctx);
llama_kv_cache_update (ctx);
llama_kv_self_seq_rm (ctx, 0, n_keep , n_keep + n_discard);
llama_kv_self_seq_add(ctx, 0, n_keep + n_discard, n_ctx, -n_discard);
//llama_kv_self_defrag (ctx);
llama_kv_self_update (ctx);
n_past = llama_kv_cache_seq_pos_max(ctx, 0) + 1;
n_past = llama_kv_self_seq_pos_max(ctx, 0) + 1;
common_batch_clear(batch);
@ -198,12 +198,12 @@ int main(int argc, char ** argv) {
if (n_discard > 0) {
LOG_INF("%s: shifting KV cache with %d to free space for the answer\n", __func__, n_discard);
llama_kv_cache_seq_rm (ctx, 0, n_keep , n_keep + n_discard);
llama_kv_cache_seq_add(ctx, 0, n_keep + n_discard, n_ctx, -n_discard);
//llama_kv_cache_defrag (ctx);
llama_kv_cache_update (ctx);
llama_kv_self_seq_rm (ctx, 0, n_keep , n_keep + n_discard);
llama_kv_self_seq_add(ctx, 0, n_keep + n_discard, n_ctx, -n_discard);
//llama_kv_self_defrag (ctx);
llama_kv_self_update (ctx);
n_past = llama_kv_cache_seq_pos_max(ctx, 0) + 1;
n_past = llama_kv_self_seq_pos_max(ctx, 0) + 1;
}
}