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

kv-cache : refactor + add llama_memory_state_i (#13746)

Browse source 
* kv-cache : simplify the "struct llama_kv_cache" interface

ggml-ci

* kv-cache : revert the (n_swa + n_ubatch) change (for next PR)

ggml-ci

* kv-cache : some comments

ggml-ci

* context : fix graph reserve for multiple sequences

ggml-ci

* kv-cache : fix typo [no ci]

* kv-cache : fix find_slot() logic for free slots

ggml-ci

* llama : add TODO for deprecating the defrag API in the future

* kv-cache : improve find_slot() using min/max seq pos info

ggml-ci

* llama : handle aborts and compute errors

ggml-ci

* memory : extract state into llama_memory_state

ggml-ci

* kv-cache : add comments

ggml-ci

* server : update batching logic to reset n_batch on successful decode

* server : upon full re-processing, remove the sequence from the cache

* kv-cache : add TODO for doing split_equal when split_simple fails

ggml-ci
This commit is contained in:
Georgi Gerganov 2025-05-31 10:24:04 +03:00 • committed by GitHub
parent eb3949938e
commit 12d0188c0d
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
14 changed files with 1304 additions and 655 deletions
Download patch file Download diff file Expand all files Collapse all files

25
src/llama-batch.h
View file

@ -11,15 +11,15 @@ struct llama_ubatch {
bool equal_seqs;
// TODO: whole_seqs for embeddings?
uint32_t n_tokens; // total tokens (n_seq_tokens * n_seqs)
uint32_t n_tokens; // total tokens (n_seq_tokens * n_seqs)
uint32_t n_seq_tokens; // tokens per sequence
uint32_t n_seqs;
llama_token * token; // [n_tokens]
float * embd; // [n_embd, n_tokens]
llama_pos * pos; // [n_tokens]
int32_t * n_seq_id; // [n_seqs]
llama_seq_id ** seq_id; // [n_seqs]
int32_t * n_seq_id; // [n_seqs] // TODO: remove, should belong to only 1 sequence
llama_seq_id ** seq_id; // [n_seqs] // TODO: become llama_seq_id * seq_id;
int8_t * output; // [n_tokens]
};
@ -49,13 +49,18 @@ struct llama_sbatch {
const llama_batch * batch = nullptr;
// buffers for the ubatch
std::vector<llama_token> ubatch_token;
std::vector<float> ubatch_embd;
std::vector<llama_pos> ubatch_pos;
std::vector<int32_t> ubatch_n_seq_id;
std::vector<llama_seq_id *> ubatch_seq_id;
std::vector<int8_t> ubatch_output;
// buffers for the ubatches
// TODO: very hacky, this needs a complete rework
struct ubatch_data {
std::vector<llama_token> token;
std::vector<float> embd;
std::vector<llama_pos> pos;
std::vector<int32_t> n_seq_id;
std::vector<llama_seq_id *> seq_id;
std::vector<int8_t> output;
};
std::vector<ubatch_data> udatas;
llama_ubatch reserve_ubatch(size_t n_ubatch, bool has_embd = false);