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

added oshmem backend to llama.cpp

Browse source
This commit is contained in:
ct-clmsn 2023-12-22 21:07:26 -05:00
parent eb0f775950
commit 71f4c96331
1 changed files with 38 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

38
llama.cpp
View file

@ -1549,6 +1549,11 @@ struct llama_context {
#ifdef GGML_USE_MPI #ifdef GGML_USE_MPI
ggml_mpi_context * ctx_mpi = NULL; ggml_mpi_context * ctx_mpi = NULL;
#endif #endif
#ifdef GGML_USE_OPENSHMEM
ggml_openshmem_context * ctx_oshmem = NULL;
#endif
}; };
// //
@ -6352,6 +6357,12 @@ static int llama_decode_internal(
ggml_mpi_graph_compute_pre(lctx.ctx_mpi, gf, n_layer); ggml_mpi_graph_compute_pre(lctx.ctx_mpi, gf, n_layer);
#endif #endif
#if GGML_USE_OPENSHMEM
const int64_t n_layer = hparams.n_layer;
ggml_openshmem_graph_compute_pre(lctx.ctx_oshmem, gf, n_layer);
#endif
#ifdef GGML_USE_METAL #ifdef GGML_USE_METAL
if (lctx.ctx_metal) { if (lctx.ctx_metal) {
ggml_metal_set_n_cb (lctx.ctx_metal, n_threads); ggml_metal_set_n_cb (lctx.ctx_metal, n_threads);
@ -6367,6 +6378,10 @@ static int llama_decode_internal(
ggml_mpi_graph_compute_post(lctx.ctx_mpi, gf, n_layer); ggml_mpi_graph_compute_post(lctx.ctx_mpi, gf, n_layer);
#endif #endif
#if GGML_USE_OPENSHEM
ggml_openshmem_graph_compute_post(lctx.ctx_oshmem, gf, n_layer);
#endif
// update the kv ring buffer // update the kv ring buffer
{ {
if (kv_self.has_shift) { if (kv_self.has_shift) {
@ -9256,12 +9271,21 @@ void llama_backend_init(bool numa) {
#ifdef GGML_USE_MPI #ifdef GGML_USE_MPI
ggml_mpi_backend_init(); ggml_mpi_backend_init();
#endif #endif
#ifdef GGML_USE_OPENSHMEM
ggml_openshmem_backend_init();
#endif
} }
void llama_backend_free(void) { void llama_backend_free(void) {
#ifdef GGML_USE_MPI #ifdef GGML_USE_MPI
ggml_mpi_backend_free(); ggml_mpi_backend_free();
#endif #endif
#ifdef GGML_USE_OPENSHMEM
ggml_openshmem_backend_free();
#endif
} }
int64_t llama_time_us(void) { int64_t llama_time_us(void) {
@ -9524,6 +9548,20 @@ struct llama_context * llama_new_context_with_model(
} }
#endif #endif
#ifdef GGML_USE_OPENSHMEM
ctx->ctx_oshmem = ggml_openshmem_init();
if (ggml_openshmem_pe(ctx->ctx_oshmem) > 0) {
// Enter a blocking eval loop with dummy input, letting rank=0 drive the process
// TODO: needs fix after #3228
GGML_ASSERT(false && "not implemented");
//const std::vector<llama_token> tmp(ctx->model.hparams.n_ctx, llama_token_bos(ctx));
//while (!llama_eval(ctx, tmp.data(), tmp.size(), 0, 0)) {};
llama_backend_free();
exit(1);
}
#endif
return ctx; return ctx;
} }