Re: Ia32
From: wolfgang kern (nowhere_at_nevernet.at)
Date: 08/26/04
- Next message: Herbert Kleebauer: "Re: Unions in Assembly Language"
- Previous message: wolfgang kern: "Re: Ia32"
- In reply to: Beth: "Re: Ia32"
- Next in thread: Beth: "Re: Ia32"
- Reply: Beth: "Re: Ia32"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Thu, 26 Aug 2004 19:37:55 +0200
Hi Beth,
a very well done collection, I may add a few comments.
Hi Luca,
A confusing jungle?
Yes, but don't hesitate to continue your attempts.
You are very young now, so you are allowed to almost forget about
historical (museal relicts) for your target in writing an Os which
may become part of the future soon.
But it wont hurt to know about things that already exist, even
they are marked as obsolete yet.
I think, if you start yet, you may skip all CPU's before x468,
or even before Intel P4 (pentium 4) and AMD K7.
The 'obligatory' backwards compatibilty will give you the
ability to program/control older chips anyway.
Writing an Os sounds easy, and in fact it is.
The very heavy steps are the huge amount of different hardware around,
and the often not well (or even not) documented hardware features.
But anyway, you are welcome to ask any question here or in
'news:alt.os.development'. I've got some experience by writing
my KESYS, and I'm willing and happy to share what I've gathered so far,
with a fast learning youngster, like you seem to be ;) :)
__
wolfgang
<Beth's list: (worth to repeat["nosnip" yet])>
| Intel:
| Pre-IA32:
| Intel 8086 1979 16-bit The original chip that started
| everything
| Intel 8088 1979 16-bit Cheap version of 8086 with 8-bit
| external bus
| Intel 80186 16-bit Never actually used in PCs (IBM
| skipped over it)
| Intel 80286 1982 16-bit Introduced "Protected Mode" (16MB
| address space)
IIRC the 8086/80186 was the so called XT-area, with DOS1.0/2.0/windoze2.0.
| IA32:
| Intel 80386 1985 32-bit First 32-bit x86 chip (4GB address
| space)
| Intel 80486 1989 32-bit FPU co-processor integrated into
| main CPU
| Intel Pentium 1993 32-bit Dual pipeline architecture
| Intel Pentium Pro 1995 32-bit 36-bit (PAE) addressing (64GB
| address space)
| Intel Pentium MMX 1997 32-bit Introduced SIMD "multimedia"
| extensions
| Intel Pentium II 1997 32-bit
| Intel Pentium II Xeon 1998 32-bit High-end "server" version of PII
| Intel Celeron 1998 32-bit Cheap version of chip to compete
| with AMD
| Intel Pentium III 32-bit More SIMD in SSE
| Intel Pentium III Xeon 32-bit High-end "server" version of PIII
| Intel Pentium IV 2000 32-bit "Hyperthreading" and SSE2
| Intel Pentium M 2003 32-bit Mobile version of PIV
| Post-IA32 (IA64):
| Intel Itanium 2001 64-bit New 64-bit architecture
| Intel Itanium 2 2004 64-bit
| [ Note: The reason why Intel switched from "80x86" names to "Pentium"
| names - and from "x86" to "IA" ("Intel Architecture") - was because,
| simply, you're not allowed to patent or trademark numbers in law...yup, so
| realising that they couldn't legally control how "80x86" was being used,
| they switched to calling things by _names_ - "Pentium" - rather than
| numbers to give them some legal control over their own processor
| names...although, "Pent" means "five", of course, and is meant to suggest
| '586...and I think it's simply a "marketing move" to call the later
| processors "Pentium II", "Pentium III", "Pentium IV" rather than "Sextium",
| "Septium" and "Octium" (although, in fact, a "Pentium III", despite the
| name, is NOT a "Septium" - 80786 - but still a further enhanced "Sextium" -
| 80686 - so there is, in fact, currently no "Octium", anyway ;) so that it
| was clear to people buying the chips - all the clueless "Great Unwashed"
| out there - that they were "the same thing but improved" CPUs rather than a
| completely different product...and the easiest way to make that clear is to
| re-use the same name - your "recognised product brand name" - but call it
| "Mark 2" / "Mark 3" and so forth ;) ]
| AMD:
| [ AMD were originally one of Intel's "contracted" CPU manufacturers to help
| Intel make their "production figures" and actually worked for Intel, as
| their official "2nd x86 producer"...AMD decided, though, to start making
| their own independent x86 designs with the AM486, separate from Intel
| (though retaining "compatibility" with Intel and IA32)...notoriously
| producing chips that were actually cheaper and faster than Intel's
| own...oops...interestingly, unlike Intel, AMD's 64-bit chips are "IA32 with
| 64-bit registers" rather than a brand new 64-bit design, meaning that AMD
| and Intel are beginning to "part company" with each other over the 64-bit
| stuff... ]
| IA32 (compatible):
| AMD AM486 32-bit 486 clone
add: AMD486DX,AMD486SX,AMD486DX2 clones
| AMD K5 1996 32-bit Pentium clone
| AMD K6 1997 32-bit Pentium clone (MMX included)
| AMD K6-II 1998 32-bit Introduced "3D Now!" instructions
| AMD K6-III 1999 32-bit L2 cache "on-chip" (big
| performance boost)
| AMD Athlon (K7) 1999 32-bit
| AMD Athlon (K750) 32-bit
| AMD Athlon "Thunderbird" 32-bit
| AMD Duron 32-bit Budget version of Athlon
| "Thunderbird"
| AMD Athlon XP 32-bit
| AMD Sempron 32-bit "Mobile" AMD chip
|
| Post-IA32 (AMD64; Though IA-32 compatible):
| AMD Athlon 64 64-bit
| AMD Opteron 64-bit
| Cyrix / VIA:
| (VIA bought Cyrix but release chips as IDT, who are actually Centaur
| Designs and they call their chips "WinChips"...confusing, huh? ;)
| [ Like AMD, Cyrix used to be an official "contracted" x86 CPU manufacturer
| to Intel to help them make enough chips to meet demands...and, like AMD,
| they decided to be "adventurous" with the 486 and actually develop their
| chips independently...and, again, they are "compatible" with the IA32
| (although, Cyrix were notorious for having the odd "problem" with
| compatibility and overheating problems here and there)...as noted, VIA
| bought up Cyrix and also bought up Centaur Designs (who release as IDT) but
| the Centaur Designs stuff won out... ]
| IA32 (compatible):
| Cyrix 486 32-bit 486 clone
| Cyrix 5x86 32-bit despite name, still 486 clone (DX)
| Cyrix 6x86 1995 32-bit Pentium clone
| Cyrix 6x86MX / MII 1997 32-bit Pentium MMX clone (renamed during
| takeovers)
|
| Cyrix MediaGX 32-bit Built-in controllers (targetting
| embedded use)
|
| IDT WinChip (C6) 1997 32-bit Pentium MMX clone
| IDT WinChip 2 1998 32-bit Pentium MMX clone (just an
| improved C6)
| These are the three "main" manufacturers of note...there are others...for
| instance, IBM have been known to make x86 chips too...but this list is long
| enough and you'll probably never meet any of the more "obscure"
| chips...and, well, compiling this list is actually more difficult than I
| thought it would be, as the information seems spread out all over the
| place...
|
| Although, the idea of a simple "table" of all this information isn't
| actually a bad idea...to be honest, I was slightly surprised that I
| couldn't locate a website that already does this on the web somewhere
| (well, if such a site exists, then I couldn't find it with Google, anyway
| ;)...
|
| Beth :)
- Next message: Herbert Kleebauer: "Re: Unions in Assembly Language"
- Previous message: wolfgang kern: "Re: Ia32"
- In reply to: Beth: "Re: Ia32"
- Next in thread: Beth: "Re: Ia32"
- Reply: Beth: "Re: Ia32"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
