Re: The Design of Design

Liste des GroupesRevenir à c arch 
Sujet : Re: The Design of Design
De : johnl (at) *nospam* taugh.com (John Levine)
Groupes : comp.arch
Date : 26. Apr 2024, 19:38:43
Autres entêtes
Organisation : Taughannock Networks
Message-ID : <v0gsbj$2cok$1@gal.iecc.com>
References : 1 2 3 4
User-Agent : trn 4.0-test77 (Sep 1, 2010)
According to Thomas Koenig  <tkoenig@netcologne.de>:
S/360 invented eight bit byte addressed memory with larger power of 2

Brooks wrote that the design was supposed to have been 32-bit
clean from the start, but that the people who implemented the BALR
instruction (which puts some bits of the PSW into the high-value
byte) didn't follow that guideline.  He blamed himself for not making
that sufficiently clear to all the design team.

Yup.  Even worse, the OS programmers were under extreme pressure
to save memory so in every data structure with address words,
they used the high byte for flags or other stuff.  So when they
went to 31 bit addressing, they needed new versions of all of
the control blocks.

I thought the PDP-10 was swell, but even if DEC had been able to
design and ship the Jupiter follow-on to the KL-10, its expanded
addressing was a kludge. It only provided addressing 8M words or about
32M bytes with no way to go past that.

I misread the manual. The extended addresses were 30 bits or about 4GB
which was plenty for that era, but the way they did it in 256K word
sections was still a kludge. In the original PDP-6/10 every
instruction could address all of memory. In extended mode you could
directly address only the current section, and everything else needed
an index register or an indirect address.

While this wasn't terribly hard, it did mean that any time you wanted
to change a program to run in extended mode you had to look at all the
code and check every instruction that did an address calculation,
which was tedious.

Reading
>
http://bitsavers.informatik.uni-stuttgart.de/pdf/dec/pdp10/KC10_Jupiter/ExtendedAddressing_Jul83.pdf
>
I concur that it was a kludge, but at least they seem to have
allowed for further extension by reserving a 1-1- bit pattern,
as an illegal indirect word.

Given that it could already address 4GB I don't think that would help, since anything
larger would need multi-word addresses which would be an even worse kludge.

However, one questions.  Designs like the PDP-10 or the UNIVAC
(from what I read on Wikipedia) had "registers" at certain
memory locations.  On the PDP-10, it even appears to have been
possible to run code in the first memory locations/registers.

Funny you should mention that. On the PDP-6/10, the registers were the
first 16 memory locations. There were no register to register
instructions since you used the regular instruction with a memory
address between 0 and 017. You could indeed run code in the registers
which was somewhat faster. I wrote a little multi-precision factorial
routine that ran in the registers.

It seems that the /360 was the first machine which put many
registers into a (conceptually) separate space, leaving them open
to implementing them either in memory or as faster logic.
>
Is that the case, or did anybody beat them to it?

On the PDP-6 and KA-10 the transistor registers were an extra cost
option, so you could order your machine either way. I believe that DEC
never shipped a machine without the fast registers since the speed
difference was so great.
--
Regards,
John Levine, johnl@taugh.com, Primary Perpetrator of "The Internet for Dummies",
Please consider the environment before reading this e-mail. https://jl.ly

Date Sujet#  Auteur
21 Apr 24 * The Design of Design128Thomas Koenig
21 Apr 24 +* Re: The Design of Design67John Levine
25 Apr 24 i`* Re: The Design of Design66Thomas Koenig
25 Apr 24 i `* Re: The Design of Design65Stephen Fuld
26 Apr 24 i  +* Re: The Design of Design56John Levine
26 Apr 24 i  i+* Re: The Design of Design2MitchAlsup1
26 Apr 24 i  ii`- Re: The Design of Design1John Levine
26 Apr 24 i  i+* Re: The Design of Design50Thomas Koenig
26 Apr 24 i  ii+- Re: The Design of Design1Stephen Fuld
26 Apr 24 i  ii+* Re: The Design of Design47John Levine
27 Apr 24 i  iii+* Re: The Design of Design4Thomas Koenig
27 Apr 24 i  iiii`* Re: PDP-10 addressing, was The Design of Design3John Levine
27 Apr 24 i  iiii `* Re: PDP-10 addressing, was The Design of Design2MitchAlsup1
27 Apr 24 i  iiii  `- Re: PDP-10 addressing, was The Design of Design1John Levine
30 Apr 24 i  iii`* Re: The Design of Design42MitchAlsup1
30 Apr 24 i  iii +* Re: The Design of Design40John Levine
1 May 24 i  iii i`* Re: The Design of Design39Tim Rentsch
1 May 24 i  iii i `* Re: architecture, The Design of Design38John Levine
2 May 24 i  iii i  +- Re: index architecture, The Design of Design1John Levine
2 May 24 i  iii i  +* Re: architecture, The Design of Design4Thomas Koenig
3 May 24 i  iii i  i+- Re: architecture, The Design of Design1MitchAlsup1
5 May 24 i  iii i  i`* Re: architecture, The Design of Design2Thomas Koenig
5 May 24 i  iii i  i `- Re: ancient 704 architecture, The Design of Design1John Levine
7 May 24 i  iii i  `* Re: architecture, The Design of Design32Tim Rentsch
7 May 24 i  iii i   +- Re: architecture, The Design of Design1Thomas Koenig
7 May 24 i  iii i   +* Re: architecture, The Design of Design28Michael S
7 May 24 i  iii i   i+* Re: architecture, The Design of Design2John Levine
8 May 24 i  iii i   ii`- Re: architecture, The Design of Design1John Levine
8 May 24 i  iii i   i+* Re: architecture, The Design of Design2Tim Rentsch
9 May 24 i  iii i   ii`- Re: architecture, The Design of Design1John Levine
8 May 24 i  iii i   i`* Re: architecture, The Design of Design23Thomas Koenig
8 May 24 i  iii i   i `* Re: architecture, The Design of Design22Michael S
8 May 24 i  iii i   i  `* Re: backward architecture, The Design of Design21John Levine
9 May 24 i  iii i   i   +* Re: backward architecture, The Design of Design2Lynn Wheeler
10 May 24 i  iii i   i   i`- Re: backward architecture, The Design of Design1Lynn Wheeler
9 May 24 i  iii i   i   `* Re: backward architecture, The Design of Design18Michael S
9 May 24 i  iii i   i    +* Re: backward architecture, The Design of Design14Thomas Koenig
9 May 24 i  iii i   i    i`* Re: backward architecture, The Design of Design13Michael S
9 May 24 i  iii i   i    i +* Re: backward architecture, The Design of Design2Anton Ertl
9 May 24 i  iii i   i    i i`- Re: backward architecture, The Design of Design1Anton Ertl
9 May 24 i  iii i   i    i +* Re: backward architecture, The Design of Design9Stephen Fuld
9 May 24 i  iii i   i    i i+* Re: backward architecture, The Design of Design2Michael S
9 May 24 i  iii i   i    i ii`- Re: backward architecture, The Design of Design1John Dallman
10 May 24 i  iii i   i    i i`* Re: backward architecture, The Design of Design6Tim Rentsch
10 May 24 i  iii i   i    i i `* Re: backward architecture, The Design of Design5Stephen Fuld
30 May 24 i  iii i   i    i i  `* Re: backward architecture, The Design of Design4Tim Rentsch
30 May 24 i  iii i   i    i i   `* Re: backward architecture, The Design of Design3Stephen Fuld
30 May 24 i  iii i   i    i i    `* Re: backward architecture, The Design of Design2Tim Rentsch
31 May 24 i  iii i   i    i i     `- Re: backward architecture, The Design of Design1John Levine
10 May 24 i  iii i   i    i `- Re: backward architecture, The Design of Design1Tim Rentsch
10 May 24 i  iii i   i    +- Re: backward architecture, The Design of Design1John Levine
10 May 24 i  iii i   i    `* Re: backward architecture, The Design of Design2Tim Rentsch
10 May 24 i  iii i   i     `- Re: backward architecture, The Design of Design1John Levine
7 May 24 i  iii i   `* Re: architecture, The Design of Design2Anton Ertl
8 May 24 i  iii i    `- Re: architecture, The Design of Design1Tim Rentsch
30 Apr 24 i  iii `- Re: The Design of Design1MitchAlsup1
30 Apr 24 i  ii`- Re: what's a register, The Design of Design1John Levine
26 Apr 24 i  i`* Re: The Design of Design3Stephen Fuld
26 Apr 24 i  i `* Re: The Design of Design2John Levine
26 Apr 24 i  i  `- Re: The Design of Design1Stephen Fuld
27 Apr 24 i  +* Re: The Design of Design7Thomas Koenig
27 Apr 24 i  i+- Re: The Design of Design1Stephen Fuld
27 Apr 24 i  i+* Re: The Design of Design2John Levine
27 Apr 24 i  ii`- Re: The Design of Design1Thomas Koenig
28 Apr 24 i  i`* Re: The Design of Design3Tim Rentsch
29 Apr 24 i  i `* Re: antitrust history, The Design of Design2John Levine
1 May 24 i  i  `- Re: antitrust history, The Design of Design1Tim Rentsch
29 Apr 24 i  `- Re: The Design of Design1Tim Rentsch
29 Apr 24 `* Re: The Design of Design60Tim Rentsch
1 May 24  `* Re: The Design of Design59Stephen Fuld
1 May 24   +* Re: JCL, The Design of Design3John Levine
1 May 24   i`* Re: JCL, The Design of Design2Stephen Fuld
1 May 24   i `- Re: JCL, The Design of Design1Stephen Fuld
1 May 24   +- Re: The Design of Design1MitchAlsup1
1 May 24   +- Re: The Design of Design1Thomas Koenig
7 May 24   `* Re: The Design of Design53Tim Rentsch
7 May 24    +* Re: The Design of Design45Stephen Fuld
7 May 24    i+- Re: The Design of Design1Thomas Koenig
7 May 24    i+* Re: The Design of Design33Stephen Fuld
7 May 24    ii+* Re: The Design of Design29Thomas Koenig
7 May 24    iii`* Re: The Design of Design28Stephen Fuld
7 May 24    iii +* Re: interative use, The Design of Design25John Levine
7 May 24    iii i+* Re: interative use, The Design of Design4MitchAlsup1
8 May 24    iii ii`* Re: third system syndrome, interactive use, The Design of Design3John Levine
8 May 24    iii ii `* Re: third system syndrome, interactive use, The Design of Design2Lynn Wheeler
9 May 24    iii ii  `- Re: third system syndrome, interactive use, The Design of Design1Lynn Wheeler
8 May 24    iii i`* Re: interative use, The Design of Design20Stephen Fuld
8 May 24    iii i `* Re: interative use, The Design of Design19John Levine
9 May 24    iii i  `* Re: interative use, The Design of Design18Stephen Fuld
10 May 24    iii i   `* Re: address architecture, not interactive use, The Design of Design17John Levine
10 May 24    iii i    +- Re: address architecture, not interactive use, The Design of Design1Stephen Fuld
11 May 24    iii i    `* Re: address architecture, not interactive use, The Design of Design15Thomas Koenig
11 May 24    iii i     +* Re: address architecture, not interactive use, The Design of Design3MitchAlsup1
12 May 24    iii i     i`* Re: address architecture, not interactive use, The Design of Design2Thomas Koenig
13 May 24    iii i     i `- Re: address architecture, not interactive use, The Design of Design1MitchAlsup1
11 May 24    iii i     `* Re: branch address architecture, not interactive use, The Design of Design11John Levine
12 May 24    iii i      `* Re: branch address architecture, not interactive use, The Design of Design10Thomas Koenig
13 May 24    iii i       `* Re: branch address architecture, not interactive use, The Design of Design9John Levine
13 May 24    iii i        `* Re: branch address architecture, not interactive use, The Design of Design8Thomas Koenig
13 May 24    iii i         `* Re: branch address architecture, not interactive use, The Design of Design7John Levine
13 May 24    iii i          +* Re: branch address architecture, not interactive use, The Design of Design4MitchAlsup1
14 May 24    iii i          +- Re: branch address architecture, not interactive use, The Design of Design1Thomas Koenig
25 May 24    iii i          `- Re: branch address architecture, not interactive use, The Design of Design1Anton Ertl
8 May 24    iii `* Re: The Design of Design2Thomas Koenig
10 May 24    ii`* Re: The Design of Design3Tim Rentsch
30 May 24    i`* Re: The Design of Design10Tim Rentsch
7 May 24    `* Re: The Design of Design7Thomas Koenig

Haut de la page

Les messages affichés proviennent d'usenet.

NewsPortal