CXL Type-3 device discovery, configuration in firmware and prepare ACPI tables...
summary
Firmware builds ACPI tables so the kernel sees CXL Type‑3 memory as nodes: ACPI HMAT/SRAT/CEDT prepared and per‑host‑bridge CHB objects with UIDs linking host bridges to memory windows. Firmware performs pre‑AP SCP PCIe enumeration and a new CXL DXE during DXE to read CDAT and configure interconnects; kernel validates topology via the existing CXL framework. Work runs on a Neoverse N2 FVP with CXL 2.0 port; current test uses a single root port, single Type‑3 endpoint and demonstrates enumeration but reports that HDM interleaving logic is not yet implemented.
AI-written summary · about
chapters (9)
- 0:00Introduction and context
- 1:59System objectives
- 3:38Host bridge and firmware
- 4:44CDT and configuration tables
- 6:45Hardware simulation platform
- 9:09System hardware architecture
- 12:23Firmware boot process
- 13:13Enumeration and discovery
- 22:30Q&A and closing remarks
related talks
39mCompute Express Link(CXL), the Next Generation Interconnect - Overview and the Status of Linux
29mCXL HDM-DB support for Linux
15mCXL Type-2 core support
26mCXL Memory Tiering for heterogenous computing
43mExpanding the Memory Limits with CXL Software Stack
42mMessing up Your NUMA Topology with CXL
22mKernel CXL Cache (safe) support
29mPlumbing challenges in Dynamic capacity device
