Subject: Re: Add a MAP_ALIGNED flag for mmap(2).
To: Bill Studenmund <email@example.com>
From: Greywolf <firstname.lastname@example.org>
Date: 03/03/2003 10:36:15
On Sun, 2 Mar 2003, Bill Studenmund wrote:
[BS: While I see the purity of adding an extra argument, I like the idea of
[BS: using flags. We have a good number open, and using them would mean not
[BS: changing a lot of called routines.
[BS: I also like listing the alignment in terms of bytes, not pages, since it
[BS: means userland code doesn't care what the internal page size is.
Perhaps, then, if we have the flags, we should have
PMAP_ALIGN _arg_ is a byte alignment specification (currently
restricted to a power of two? Do any weird byte-
size architectures (still) exist?). The returned
region will be aligned to on a _arg_-byte boundary
PMAP_PALIGN Request page alignment. The returned region will
[BS: Also it
[BS: might help emulation if future systems raise the page size (like say we go
[BS: from 4k to 8k or 8k to 16k as part of a CPU transition) - we don't have to
[BS: keep track of the implicit page size.
I don't know enough about how pages work; can many owners/processes share
a page of memory or is it usually N:1 (pages:processes)?
NetBSD: The power to serve, also.