two mem-cycles for Peek ???

Thread View: comp.sys.transputer

2 messages

[T] Back to Threads [V] Tree View [G] Group Articles

2 total messages Started by J. Tiemann Fri, 22 Jan 1993 14:07

#1892

Author: J. Tiemann
Date: Fri, 22 Jan 1993 14:07

13 lines
778 bytes

Hi,

our T425 reads a FIFO with a 'Peek' from a link.
Only every second value from the FIFO came back
from 'Peek'. The logic analyzer shows that the T425
executes the 'Peek' with two memory cycles. We
used different adresses with even/uneven boundaries.
Every 'Peek' occured with two memory cycles on
the specific address ...

Why use our 'Peek' two memory cycles?

Jens

Re: two mem-cycles for Peek ???

#1908

Author: Andy Rabagliati
Date: Tue, 26 Jan 1993 16:16

20 lines
1076 bytes

In article <1993Jan22.140715.26743@fokus.gmd.de> jet@fokus.gmd.de (J. Tiemann) writes:
>Hi,
>
>Every 'Peek' occured with two memory cycles on
>the specific address ...
>
>Why use our 'Peek' two memory cycles?
>

Peek /poke was originally put into the Transputer for diagnostic
purposes only.

If you peek an address that is non-word-aligned, the transputer would
have to do two memory accesses. Maybe the microcode always does two
accesses, to simplicate and add lightness.

Cheers,    Andy.
-- 
Amendment 2 information newsgroup co.politics.amend2.info - mailing list also.
andyr@wizzy.com          W.Z.I. Consulting (719)635-6099               pgp-2.1

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads