On Fri, 5 May 2017, Ed Maste wrote:
On 5 May 2017 at 10:31, Konstantin Belousov <kostik...@gmail.com> wrote:
On Fri, May 05, 2017 at 11:39:23PM +1000, Bruce Evans wrote:
Change "size" to "representation" if you want to allow for more than the
size changing.
I suggest you to negotiate a formulation with brooks@, who provided the
existing note.
If it's really undesirable to mention pointer integrity here, I think
the main point can still be made with a phrase like Bruce proposed
(but I think it's fine as it exists now).
No one except Cheri developers would know the Cheri meaning of "integrity
mechanism".
"Representation" is a general term that is fuzzy enough to work here.
The representation can be almost anything, so using this term reduces
the statement to a warning that it should not be assumed that pointers
have the same [binary] representation as unsigned long, although the
previous paragraph says that they [currently] do.
What about "In order to maximize compatibility with future changes in
pointer size or representation, manipulation..."?
Or just remove the statement that pointers have the same representation
as unsigned long, so that FreeBSD gurantees nothing more than the C
standard for pointers, or emphasize this.
Very briefly (to give some context), in CheriBSD and the CHERI CPU
pointers are 128 or 256 bits wide, and have a base address, length,
permissions, and other fields. Thus the size differs from existing
standard integer types, and a pointer's representation and even the
essence of what a pointer is differs from pointers in existing FreeBSD
architectures.
How does it restore the extra information when converting back from
intptr_t?
Bruce
_______________________________________________
svn-src-all@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-all
To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
