The shared memory architecture doesn't eliminate copying the data across to the ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		hughw 3 months ago \| parent \| context \| favorite \| on: Prefix sum: 20 GB/s (2.6x baseline) The shared memory architecture doesn't eliminate copying the data across to the device. Edit: or back.

yogishbaliga 3 months ago [–]

If it is unified memory, CPU can access the result of GPU processing without copying it to CPU memory (theoretically)

hughw 3 months ago | [–]

If the CPU touches an address mapped to the GPU doesn't it fault a page into the CPU address space? I mean the program doesn't do anything special, but a page gets faulted in I believe.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact