// the requests stop after a certain amount of time (keep_alive_s) of inactivity
dispatch_queue_t d_queue = dispatch_get_global_queue(QOS_CLASS_DEFAULT, 0);
dispatch_group_async(res->d_group, d_queue, ^{
- while (!atomic_load_explicit(&res->d_stop, memory_order_relaxed)) {
- if (atomic_load_explicit(&res->d_loop, memory_order_relaxed) > 0) {
- [res->lock lock];
+#if defined(GGML_METAL_HAS_RESIDENCY_SETS)
+ if (@available(macOS 15.0, iOS 18.0, tvOS 18.0, visionOS 2.0, *)) {
+ while (!atomic_load_explicit(&res->d_stop, memory_order_relaxed)) {
+ if (atomic_load_explicit(&res->d_loop, memory_order_relaxed) > 0) {
+ [res->lock lock];
- for (int i = 0; i < (int) res->data.count; ++i) {
- [res->data[i] requestResidency];
- }
+ for (int i = 0; i < (int) res->data.count; ++i) {
+ [res->data[i] requestResidency];
+ }
- atomic_fetch_sub_explicit(&res->d_loop, 1, memory_order_relaxed);
+ atomic_fetch_sub_explicit(&res->d_loop, 1, memory_order_relaxed);
- [res->lock unlock];
- }
+ [res->lock unlock];
+ }
- // half a second
- usleep(500 * 1000);
- }
+ // half a second
+ usleep(500 * 1000);
+ }
+ }
+#endif
});
return res;
return;
}
+ // note: if you hit this assert, most likely you haven't deallocated all Metal resources before exiting
GGML_ASSERT([rsets->data count] == 0);
atomic_store_explicit(&rsets->d_stop, true, memory_order_relaxed);
dev->rsets = nil;
}
-
- // --------------------------------------------------
-
// print MTL GPU family:
GGML_LOG_INFO("%s: GPU name: %s\n", __func__, dev->props.name);