Re: process max virtual memory limit in amd64?
On Tue, 2004-06-29 at 18:16 -0300, Peter Cordes wrote:
> On Fri, Jun 25, 2004 at 04:30:36PM +0300, Alexander Rapp wrote:
> > Anders Fugmann wrote:
[snip]
>
> pointers are 64bits on AMD64. It's just that the top 16bits are all zero,
> until AMD makes CPUs that do something with them. Only the kernel will have
> to change to accommodate that, though. User-space programs won't have to be
> re-compiled. So, code is already leaving space for the extra bits, and
> transmitting them over the bus. No savings there, just stuff inside the CPU
> that deals with addresses, as others have described.
>
>
> The no-recompiling point is true as long as you don't do anything perverse
> like using the high bits of addresses for your own purposes. I can't find
> it now, but I remember reading (maybe in the jargon file?) about an old IBM
> system where the upper bits of addresses were reserved for future use, but
> people started using them to hold info about what was being pointed to, or
> something. Just to pack more data into limited memory. When a new system
> came out in which those address bits weren't ignored anymore, the programs
> had to be re-written. So don't repeat that mistake on AMD64!
Early Macintosh programmers did that. Then, when the MacII, based
on the 68020, came out, apps that were not "32-bit clean" wouldn't
run on the new systems. A programmer that tries to be too clever
always winds up causing more harm than good....
--
-----------------------------------------------------------------
Ron Johnson, Jr. ron.l.johnson@cox.net
Jefferson, LA USA
"Today's public figures can no longer write their own speeches or
books, and there is some evidence that they can't read them
either."
Gore Vidal
"He was about as useful in a crisis as a sheep."
Dorothy Eden
Reply to: