* Enforcing byte-wide access to memory
@ 2001-01-02 15:13 Nick Williams
2001-01-02 16:22 ` Robert Dewar
0 siblings, 1 reply; 2+ messages in thread
From: Nick Williams @ 2001-01-02 15:13 UTC (permalink / raw)
Hi there,
Is there any way of enforcing that accesses to a particular region of memory
be byte-wide?
More specifically, I am involved with development of a real-time system
which needs to read values from flash RAM, which can only be accessed
byte-at-a-time. I understand that representation clauses and use of the
Volatile pragma will cause the compiler to generate the appropriate code to
view this memory as an array of bytes...
However, a structured view of the data stored there is required (i.e. I want
to be able to have Address clauses on variables to access those variables as
stored on the flash RAM).
What is the expected behaviour of an Unchecked_Conversion in these
circumstances? Is there another way of achieving the desired effect?
Thanks.
Nick Williams.
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Enforcing byte-wide access to memory
2001-01-02 15:13 Enforcing byte-wide access to memory Nick Williams
@ 2001-01-02 16:22 ` Robert Dewar
0 siblings, 0 replies; 2+ messages in thread
From: Robert Dewar @ 2001-01-02 16:22 UTC (permalink / raw)
In article <3a51f063_2@nnrp1.news.uk.psi.net>,
"Nick Williams" <nwilliams@praxis-cs.co.uk> wrote:
> Hi there,
>
> Is there any way of enforcing that accesses to a particular
> region of memory be byte-wide?
No, this is definitely something that is in the implementation
dependent area. If you can isolate such references to a single
unit, then the cleanest thing is to use a machine code
insertion to ensure that the specific instruction you want
is generated.
Sent via Deja.com
http://www.deja.com/
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2001-01-02 16:22 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2001-01-02 15:13 Enforcing byte-wide access to memory Nick Williams
2001-01-02 16:22 ` Robert Dewar
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox