Re: Tagged memory (was Re: Memory Upgrade)

["Alexei Pytel" <pytel@xxxxxxxxxx>]

But the tag is not loaded from memory by a separate instruction , is it ?
(Say, load pointre, then load tag, then test tag).
Hardware loads tag into condition register when pointer is loaded.
I guess this makes us both right (or both wrong :-) ) - it looks like
pointer handling is "hardware-assisted software".

Alexei Pytel


From: Alexei Pytel <pytel@us.ibm.com>
> > 2) the tagged pointer protection is not hardware, but is done in
> > software, so need not contribute to make memory more expensive.
>
> With all respect, the tagged pointer protection *is* hardware.
> It could in theory be done by software, but it would incur unacceptable
> overhead.
>

Well, the RISC code to check if a pointer is valid is:
LQ R6, addr  ;  load pointer into R6, R7. set bit in exception reg if no
tag
TXER nn ; trap if no tag

to me that is software. The trap can be disabled by replacing the TXER
instruction by a NOOP.

The overhead is the TXER instruction and is not unacceptable.



_______________________________________________
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@midrange.com
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/cgi-bin/listinfo/midrange-l
or email: MIDRANGE-L-request@midrange.com
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.