:> [ML] It is possible to handle these cases in VM code, by
:> trapping on any access to the partial page, and allowing only those
:> accesses which are withing the originally requested range. Performance
:> would suck without end, though.
:
:Well it would only suck for access to that page right, which wouldn't be too
:bad unless the program in question thrashes out the partial pages.
:
:---
:Daniel O'Connor software and network engineer
I don't think this would be workable. It would fail utterly on small
files -- every single access would be trapped. The system must be
reasonably deterministic. Programmers have a reasonable expectation
that accesses to mmap'd areas are, for in-core pages, instantanious.
Otherwise they wouldn't bother to use mmap and instead would use read()
and write() :-)
-Matt
To Unsubscribe: send mail to majord...@freebsd.org
with "unsubscribe freebsd-hackers" in the body of the message
