so you have invent unified memory for linux first because that’s the limitation today
replyFairly sure most iGPUs these days are zero-copy and can dynamically allocate memory so what does "unified memory" mean to you exactly? A wider bus would be nice but it's not exactly a groundbreaking new invention.
replyI was actually pretty far off:
reply> Unified memory in Linux creates a single address space accessible to both the CPU and GPU, eliminating the need to manually copy data between system RAM and video memory. It is enabled via NVIDIA's CUDA, AMD's ROCm/HIP, or generic kernel-level Heterogeneous Memory Management (HMM).
So it does exist and is available for platforms that matter.
It is interesting how apple claimed that "unified memory" is something special, and ppl believed them.
replyIntel and AMD had been doing this for years already, and had linux support for it from day 1.
Cool. Apple was the only one who managed to ship a consumer device with UMA and RDMA support. 2TB VRAM max over RDMA.
replyI think the REALLY cool thing about apple's shared memory implementation is the ultra-wide memory bus.
replyOtherwise, AMD is quite close to what Apple has, and Strix Halo is honestly incredible.
Not sure what RDMA brings to the table.
RDMA increases the inference performance by a significant percentage across devices connected via Thunderbolt 5.4x512 is like a 2TB machine.
replyOr just buy a MI300A server like https://www.servethehome.com/gigabyte-g383-r80-aap1-amd-inst...
reply"Just". And then GPUs, and RAM? And cooling? Will you really appreciate it when sitting right next to it?
replyRaspberry Pi and other SBCs, Android phones and practically all of the embedded devices with a display and microprocessor.
replyAll have unified memory. Linux runs just fine on all of those.
reply
Ah ok. I replied to ~45 minutes stale page.
reply