thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

metal : support virtual devices (#18919)

Browse source 
* metal : support virtual devices

* cont : manage buffer type context memory

* metal : add events

* cont : implement cpy_tensor_async
This commit is contained in:
Georgi Gerganov 2026-02-02 14:29:44 +02:00 • committed by GitHub
parent 6156ae5111
commit 6fdddb4987
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 508 additions and 127 deletions
Download patch file Download diff file Expand all files Collapse all files

1
src/llama-context.cpp
View file

@ -317,6 +317,7 @@ llama_context::llama_context(
auto dev_type = ggml_backend_dev_type(ggml_backend_get_device(backend.get()));
if (dev_type == GGML_BACKEND_DEVICE_TYPE_CPU) {
// ignore CPU backend
// TODO: should we ignore ACCEL types too?
continue;
}
auto * dev = ggml_backend_get_device(backend.get());