CXL 2: Pooling, sharing, and I/O-memory resources
CXL 2: Pooling, sharing, and I/O-memory resources
Posted May 19, 2022 18:00 UTC (Thu) by MattBBaker (guest, #28651)Parent article: CXL 2: Pooling, sharing, and I/O-memory resources
So if you want the kernel to manage this memory there will have to be two separate concepts, the memory the device owns and cannot be taken away, and memory that it does not own and subject to being ripped away. I'm half of the mind that a better model for this is to expose remote CXL devices in /dev/ and then mmap() that like a normal 'file'.
Posted May 19, 2022 22:35 UTC (Thu)
by ejr (subscriber, #51652)
[Link]
This all needs specified in terms of failure modes. It may be so specified; I haven't kept up.
Posted Jun 5, 2022 8:05 UTC (Sun)
by njs (subscriber, #40338)
[Link]
I have no idea how well or poorly this will extend to CXL failure modes, but it's at least precedented.
CXL 2: Pooling, sharing, and I/O-memory resources
CXL 2: Pooling, sharing, and I/O-memory resources