And? So you're going to take a syscall (mincore()?) hit before every file-backed mapping access to test if it'd incur a page fault, and try switch to another coroutine if it would?
Syscalls aren't free, especially today, and mmap() is already bringing significant overhead to the party.
If you've brought coroutines into the picture, you might as well schedule them using async IO completions and kick mmap() to the curb.
Syscalls aren't free, especially today, and mmap() is already bringing significant overhead to the party.
If you've brought coroutines into the picture, you might as well schedule them using async IO completions and kick mmap() to the curb.