Subject: Re: Multiple faulting
To: Charles M. Hannum <firstname.lastname@example.org>
From: Bill Sommerfeld <email@example.com>
Date: 03/25/1999 12:39:45
> * Because uvm_fault_wire() and uvm_vslock() have no way to insure that
> the page is not going to take R/M emulation faults, it's actually
> possible for them to occur inside interrupt handlers, while munging
> I/O buffers. This isn't necessarily a problem, but the machine
> dependent code needs to take this into account and make sure it
> doesn't break anything.
Hmm. If this gets fixed (and maybe even if it doesn't),
uvm_map_pageable should perhaps be changed so it gets a fault type
If you have a read/write-mapped region of memory which is mapped with
MAP_PRIVATE but is read-mostly, write-seldom, it seems silly for
uvm_map_pageable to COW everything in it if you just want to wire it
because a driver wants to copy the contents elsewhere.