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

ggml : sync (abort callback, mul / add broadcast, fix alibi) (#2183)

Browse source
This commit is contained in:
Georgi Gerganov 2023-07-11 22:53:34 +03:00 • committed by GitHub
parent 5bf2a27718
commit 20d7740a9b
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
5 changed files with 173 additions and 72 deletions
Download patch file Download diff file Expand all files Collapse all files

11
ggml.h
View file

@ -201,8 +201,13 @@
#define GGML_MAX_NAME 48
#define GGML_DEFAULT_N_THREADS 4
#define GGML_EXIT_SUCCESS 0
#define GGML_EXIT_ABORTED 1
#define GGML_UNUSED(x) (void)(x)
#define GGML_ASSERT(x) \
do { \
if (!(x)) { \
@ -442,6 +447,10 @@ extern "C" {
// the `n_tasks` of nodes, 1:1 mapping to cgraph nodes
int n_tasks[GGML_MAX_NODES];
// abort ggml_graph_compute when true
bool (*abort_callback)(void * data);
void * abort_callback_data;
};
// computation graph
@ -1303,7 +1312,7 @@ extern "C" {
// ggml_graph_plan() has to be called before ggml_graph_compute()
// when plan.work_size > 0, caller must allocate memory for plan.work_data
GGML_API struct ggml_cplan ggml_graph_plan (struct ggml_cgraph * cgraph, int n_threads /*= GGML_DEFAULT_N_THREADS*/);
GGML_API void ggml_graph_compute(struct ggml_cgraph * cgraph, struct ggml_cplan * cplan);
GGML_API int ggml_graph_compute(struct ggml_cgraph * cgraph, struct ggml_cplan * cplan);
GGML_API void ggml_graph_reset (struct ggml_cgraph * cgraph);
// same as ggml_graph_compute() but the work data is allocated as a part of the context