ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-01-27 14:49:55 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

whisper : avoid whisper_model_data wrapper

Browse Source
This commit is contained in:
Georgi Gerganov 2023-11-11 11:46:54 +02:00
parent 66bb2e9401
commit 0867e696a7
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 8 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
whisper.cpp
View File

@ -518,10 +518,6 @@ struct whisper_kv_cache {
int n; // number of tokens currently in the cache
};
struct whisper_model_data {
ggml_backend_buffer_t buffer_main;
};
struct whisper_model {
e_model type = MODEL_UNKNOWN;
@ -556,11 +552,11 @@ struct whisper_model {
std::vector<whisper_layer_encoder> layers_encoder;
std::vector<whisper_layer_decoder> layers_decoder;
// context
// ggml context that contains all the meta information about the model tensors
struct ggml_context * ctx;
// the model backend data is read-only and can be shared between processors
struct whisper_model_data * data;
struct ggml_backend_buffer * buffer;
// tensors
int n_loaded;
@ -1283,8 +1279,6 @@ static bool whisper_model_load(struct whisper_model_loader * loader, whisper_con
// init backends
{
model.data = new whisper_model_data;
ggml_backend_t backend_gpu = NULL;
// initialize the backends
@ -1323,17 +1317,17 @@ static bool whisper_model_load(struct whisper_model_loader * loader, whisper_con
size_main += ggml_nbytes(t.second) + ggml_tensor_overhead();
}
model.data->buffer_main = ggml_backend_alloc_buffer(wctx.backend, size_main);
model.buffer = ggml_backend_alloc_buffer(wctx.backend, size_main);
WHISPER_LOG_INFO("%s: %8s buffer size = %8.2f MB\n", __func__, ggml_backend_name(wctx.backend), size_main / 1024.0 / 1024.0);
}
ggml_allocr * alloc_main = ggml_allocr_new_from_buffer(model.data->buffer_main);
ggml_allocr * alloc = ggml_allocr_new_from_buffer(model.buffer);
// allocate tensors in the backend buffers
{
for (const auto & t : model.tensors) {
ggml_allocr_alloc(alloc_main, t.second);
ggml_allocr_alloc(alloc, t.second);
}
}
@ -1455,7 +1449,7 @@ static bool whisper_model_load(struct whisper_model_loader * loader, whisper_con
}
}
ggml_allocr_free(alloc_main);
ggml_allocr_free(alloc);
wctx.t_load_us = ggml_time_us() - t_start_us;
@ -3198,10 +3192,9 @@ void whisper_free(struct whisper_context * ctx) {
if (ctx->model.ctx) {
ggml_free(ctx->model.ctx);
}
if (ctx->model.data) {
ggml_backend_buffer_free(ctx->model.data->buffer_main);
delete ctx->model.data;
if (ctx->model.buffer) {
ggml_backend_buffer_free(ctx->model.buffer);
}
whisper_free_state(ctx->state);