tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-30 12:55:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

context : fix state io for memory-less contexts (#13470)

Browse Source 
ggml-ci
This commit is contained in:
Georgi Gerganov
2025-05-12 15:12:27 +03:00
committed by GitHub
parent 91159ee9df
commit 064cc596ac
1 changed files with 14 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/llama-context.cpp
View File

@ -1788,10 +1788,13 @@ size_t llama_context::state_read_data(llama_io_read_i & io) {
}
}
if (memory) {
LLAMA_LOG_DEBUG("%s: - reading KV self\n", __func__);
llama_kv_cache * kv_self = static_cast<llama_kv_cache *>(memory.get());
kv_self->state_read(io);
}
return io.n_bytes();
}
@ -1799,9 +1802,11 @@ size_t llama_context::state_read_data(llama_io_read_i & io) {
size_t llama_context::state_seq_write_data(llama_io_write_i & io, llama_seq_id seq_id) {
GGML_UNUSED(seq_id);
if (memory) {
llama_kv_cache * kv_self = static_cast<llama_kv_cache *>(memory.get());
kv_self->state_write(io, seq_id);
}
return io.n_bytes();
}
@ -1809,9 +1814,11 @@ size_t llama_context::state_seq_write_data(llama_io_write_i & io, llama_seq_id s
size_t llama_context::state_seq_read_data(llama_io_read_i & io, llama_seq_id seq_id) {
GGML_UNUSED(seq_id);
if (memory) {
llama_kv_cache * kv_self = static_cast<llama_kv_cache *>(memory.get());
kv_self->state_read(io, seq_id);
}
return io.n_bytes();
}