Skip to content

llama : synchronize before get/set session data #6911

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 25, 2024
Merged

llama : synchronize before get/set session data #6911

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions llama.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16131,6 +16131,8 @@ struct llama_data_file_context : llama_data_context {
*
*/
static void llama_state_get_data_internal(struct llama_context * ctx, llama_data_context * data_ctx) {
llama_synchronize(ctx);

// copy rng
{
std::ostringstream rng_ss;
Expand Down Expand Up @@ -16283,6 +16285,8 @@ size_t llama_state_get_data(struct llama_context * ctx, uint8_t * dst) {

// Sets the state reading from the specified source address
size_t llama_state_set_data(struct llama_context * ctx, const uint8_t * src) {
llama_synchronize(ctx);

const uint8_t * inp = src;

// set rng
Expand Down Expand Up @@ -16587,6 +16591,8 @@ size_t llama_state_seq_get_size(struct llama_context* ctx, llama_seq_id seq_id)
}

static size_t llama_state_seq_get_data_internal(struct llama_context * ctx, llama_data_context & data_ctx, llama_seq_id seq_id) {
llama_synchronize(ctx);

const auto & kv_self = ctx->kv_self;
GGML_ASSERT(!kv_self.recurrent); // not implemented

Expand Down Expand Up @@ -16704,6 +16710,8 @@ size_t llama_state_seq_get_data(struct llama_context* ctx, uint8_t* dst, llama_s
}

size_t llama_state_seq_set_data(struct llama_context * ctx, const uint8_t * src, llama_seq_id dest_seq_id) {
llama_synchronize(ctx);

auto & kv_self = ctx->kv_self;
GGML_ASSERT(!kv_self.recurrent); // not implemented

Expand Down