OSDev.wiki

IA32 Architecture Family: Difference between revisions

Page Discussion

IA32 Architecture Family (view source)

Revision as of 05:15, 21 February 2018

3,036 bytes removed ,  6 years ago
→‎Advanced Micro Device Intel-compatible Processors
[unchecked revision][unchecked revision]
VisualWikitext
 
(10 intermediate revisions by 4 users not shown)
Line 13:
! Release Date
! [[FPU|FPU (80x87)]]
! [[Protected Mode]]
! [[SMP]]
! [[MMX]]
Line 23 ⟶ 24:
! 8086
| 1978
| {{Maybe|Optional}}
| {{No}}
| {{No}}
| {{No}}
Line 34 ⟶ 36:
! 80186
| 1982
| {{Maybe|Optional}}
| {{No}}
| {{No}}
| {{No}}
Line 45 ⟶ 48:
! 80286
| 1982
| {{Maybe|Optional}}
| {{Maybe|16-bit only}}
| {{No}}
| {{No}}
Line 56 ⟶ 60:
! 80386
| 1985
| {{Maybe|Optional}}
| {{Yes}}
| {{No}}
| {{No}}
Line 67 ⟶ 72:
! 80486
| 1989
| {{Maybe|Optional}}
| {{Yes}}
| {{Yes}}
| {{No}}
Line 78 ⟶ 84:
! Pentium
| 1993
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 89 ⟶ 96:
! Pentium Pro
| 1995
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 100 ⟶ 108:
! Pentium MMX
| 1996
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 111 ⟶ 120:
! Pentium II
| 1997
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 122 ⟶ 132:
! Pentium II Xeon
| 1998
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 133 ⟶ 144:
! Pentium III
| 1999
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 144 ⟶ 156:
! Pentium III Xeon
| 1999
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 155 ⟶ 168:
! Pentium IV
| 2000
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 166 ⟶ 180:
! Pentium M
| 2003
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 177 ⟶ 192:
! Core
| 2003
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 188 ⟶ 204:
! Xeon 51xx
| 2006
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 199 ⟶ 216:
! Core 2
| 2006
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 210 ⟶ 228:
! Xeon 54xx
| 2007
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 221 ⟶ 240:
! Core i-series
| 2008
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 232 ⟶ 252:
! Atom 200/300
| 2008
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 243 ⟶ 264:
! Atom N-series
| 2008
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 254 ⟶ 276:
! Atom Z-series
| 2008
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 264 ⟶ 287:
 
==Advanced Micro Device Intel-compatible Processors==
AMD has been developing integrated circuits since the early 70's, they originally licensed the 80286 from Intel and branded it as the Am286. The company later went on to release its first Intel 386 clone, the Am386, in 1991.
The biggest competitor to Intel at this time (2004 August). They came into being slightly after Cyrix with a 5k86 (being a 486 compatible similar to the 5x86, don't confuse them) and then followed it up by a K6 processor. This one was faster than the Pentiums, and more popular than the Cyrix ones because they both didn't rate it (afaik), and they didn't overheat (as was claimed, untrue, for the Cyrixes).
 
The CPUID identifier string is "AuthenticAMD".
 
It is important to note that the "SSE" used by AMD and the "SSE" used by Intel are actuallymay not compatible,be notentirely fully at least (somebody verify this?). This causes lots of confusioncompatible.
 
{| {{wikitable}}
Line 275 ⟶ 298:
! Release Date
! [[FPU|FPU (80x87)]]
! [[Protected Mode]]
! [[SMP]]
! [[MMX]]
Line 283 ⟶ 307:
! [[EM64T]]/[[AMD64]]
! Notes
|-
! Am386
| 1991
| {{Maybe|Optional}}
| {{Yes}}
| {{No}}
| {{No}}
| {{No}}
| {{No}}
| {{No}}
| {{No}}
| {{No}}
| AMD's first clone of the 32-bit i386 architecture, FPU was optional.
|-
! Am486
Line 288 ⟶ 325:
| 1994
1995
| {{Yes}}
| {{Yes}}
| {{No}}
Line 296 ⟶ 334:
| {{No}}
| {{No}}
| AMD's 486 clone, 2x the cache size of most of Intel's 486 chips.
| AMD's 486 clone turned out a success for being on-par with Intel's processors, while being a lot cheaper. Eventually, the higher-clocked Am5x86 was the fastest processor to fit in on a 486-based motherboard. Together with its 4x multiplier and 2x the cache size of most Intel's 486 chips, it was able to outperform the low end of Pentium chips.
|-
! K5
| 1996
| {{Yes}}
| {{Yes}}
| {{No}}
| {{No}}
| {{No}}
| ?
| {{No}}
| {{No}}
| {{No}}
| {{No}}
| AMD's first try at a Pentium-compatible CPU. Not a huge success, mostly due to low clock speeds.
| AMD's first try at a Pentium-compatible CPU.
|-
! K6
| 1997
| {{Yes}}
| {{Yes}}
| {{No}}
| {{Yes}}
| {{No}}
| ?
| {{No}}
| {{No}}
| {{No}}
| {{No}}
| Actually designed by NexGen (who AMD took over), the K6 is a fully Pentium-compatible CPU. It outperformed the Pentium in several areas. One notable instruction was the LOOPcc instruction, which executed in 2 cycles compared to a Pentium's 18, causing problems.
| Actually designed by NexGen (taken over by AMD), the K6 is a fully Pentium-compatible CPU. One notable instruction was the LOOPcc instruction, which executed in 2 cycles compared to a Pentium's 18, causing timing problems.
|-
! K6-2
| 1998
| {{Yes}}
| {{Yes}}
| {{No}}
| {{Yes}}
| {{Yes}}
| ?
| ?
| {{No}}
| {{No}}
| {{No}}
| AMD had a lesson learned there, don't make your processor too fast in some instructions. Not that they were put off by that, they just added 16 wait states to the execution of the LOOPcc and thus caused it to slow to the speed of a Pentium. AMD didn't just do this however. They added a special case (speculation, might be coincidence) for the DEC (E)CX; Jcc combination, which is semantically equivalent with the LOOPcc instruction, but this semantic equivalence and the loop being faster on Intels caused the loop instruction to always be used. Nobody used the DEC/Jcc combo. They kept the original speed for this combo and specified in their optimization manuals that this was the preferred method over the loopcc instruction.
| {{No}}
| AMD added 16 wait states to the execution of the LOOPcc and thus caused it to slow to the speed of a Pentium. They added a special case (speculation, might be coincidence) for the DEC (E)CX; Jcc combination, which is semantically equivalent with the LOOPcc instruction; since LOOPcc was faster on Intels, nobody used the DEC/Jcc combo there. So AMD kept the original speed for this combo, and specified in their optimization manuals that this was the preferred method over the LOOPcc instruction.
 
ItThe K6-2 also featured a new technology, the 3DNOW! technology, which was "MMX using floating point numbers", and multiplexed (again) on the floating point registers. The K6-2 was quite popular, and scaled higher than the P1 ever did. It was largely compatible with the P2, but (afaik) not completely.
|-
! K6-3
| 1999
| {{Yes}}
| ?{{Yes}}
| {{No}}
| {{Yes}}
| {{Yes}}
| ?
| ?
| {{No}}
| {{No}}
| {{No}}
| They started this design off with the concept of not making it underpowered in any place, and to make it at least P2 compatible. It was fully P2 compatible.
| {{No}}
 
| This design was fully P2 compatible.
The K6-3 was not too popular, mainly because the K6-2 did very well and people didn't see why they should buy a more expensive K6-3 for the same amount of megahertz. This of course was a joke, same as it is to call a 2 GHz opteron slower than a 2.2 GHz celeron.
 
A little known fact about the K6-3 is that it is in fact an Athlon, minus a few instructions, and minus one very important piece. The K6-3 suffered from a bottleneck at the instruction decode unit (which converts the X86x86 instructions to native instructions). It could only handle 2 in a cycle, which it made during about 20-30% of the cycles for average software. For optimized software you could bring it to 100% easily, and still want another channel. This wasn't too weird, becauseWhile it did have 3 execution units of each type (ALU / MMX / loadstore), whichthey were not used much at all. Note that these units are units executingsince the nativeinstruction instructions,decode sounit making 3 of each iscould not a stupid idea. They needed a new front end, and of course a new copy of instructions fromkeep Intelup.
|-
! Athlon
| 1999
| {{Yes}}
| ?{{Yes}}
| {{No}}
| {{Yes}}
| {{Yes}}
Line 366 ⟶ 407:
! Athlon XP/MP
| 2001
| {{Yes}}
| {{Yes}}
| {{Maybe}}
Line 374 ⟶ 416:
| {{No}}
| {{No}}
| Athlon XP (..starting with Palomino) introduced SSE,. SMP capable chips were branded as Athlon MP.
|-
! Athlon 64
| 2003
| {{Yes}}
| {{Yes}}
| {{No}}
Line 390 ⟶ 433:
! Athlon 64 X2
| 2005
| {{Yes}}
| {{Yes}}
| {{Yes}}
Line 402 ⟶ 446:
! Phenom
| 2007
| {{Yes}}
| {{Yes}}
| ?
Line 409 ⟶ 454:
| SSE4a
| {{No}}
| {{Yes}}
|
|-
! Ryzen Series
| 2017
| {{Yes}}
| {{Yes}}
| {{Yes}}
| {{Yes}}
| {{No}}
| {{Yes}}
| SSE4.2
| {{Yes}}
| {{Yes}}
|
|-
|}
 
===Athlon (first try)===
 
The first models of the Athlon were distinct, they were the first time that a competitor to Intel actually had a faster processor, without Intel having a backup plan. It was poised against the PIII, which at that time was their top model and best-running one too. The Athlon beat them to the 1 GHz mark, and at that time the 1 GHz had become completely irrelevant. It just meant that they had a new size to mark their processors with. Intel missed the point here, and they did until very shortly ago. The GHz myth had been broken, the Athlon at 1.1 GHz was still faster than the PIII at 1.3 GHz, and people knew. They didn't go for a P3 if a faster athlon was available at a lower clock speed, and at a lower price.
 
===Athlon XP / MP / Duron (new style)===
 
AMD switched to a big offensive, trying to persuade the buyers to demand AMD CPU's instead of being OK with Intels. The new versions of these processors were all just a tad better than the previous one, could do a slight number of instructions more (the Athlons started with not even SSE1, and from model 6 (both Athlon and Duron) they supported it). The processors also advanced very slightly in each other direction, making each new type just a tad faster than the previous one. In the end of the GHz wars (past year, about) the fastest Athlon was running at 2.2 GHz, but outperformed the better half of the 3 GHz P4's.
 
===AMD64 based CPU's===
This is slightly offtopic here, but still quite relevant, since theseThese processors all support the entire IA32 family natively. AMD created a new processor, with 64-bit (actually 48-bit, but who notices those 16 bits?) memory addressing and 64-bit calculations, being very compatible with the old style CPU's. So compatible, that the core for 32-bit and 64-bit is essentially equalidentical, aside from the size of calculations and the support of a few encodings that were in effecteffectively redundant. They removed a few 1-byte opcodes (about 20 in total, including all 1-byte INC and 1-byte DEC instructions) to make placeroom for a new REX prefix. They modified itthe core to use 16 registers instead of 8, added a load of new names, got the old software working, and optimized the 32-bit performance to unprecedented levels. These CPU's outperform the P4 at any clock speed, in almost (1/20 programs not) any calculation-intensive program.
 
== Other CPU vendors making similar chips==
Line 429 ⟶ 479:
===Cyrix===
 
Cyrix was a well-known CPU vendor from the 386 years (and slightly before), up to the Pentium II times, when it more or less vanished inside Via. Via now uses the name as a CPU name (not making it clearer), but this section is about the original Cyrix CPU's. The processors supporting [[CPUID]] call it a "CyrixInstead".
 
==== Cyrix 387 ====
 
This isn't actually a processor, but is the most famous Cyrixan processorFPU. It was the fastest coprocessor to the 386 to be found, and was even very usable aside a 486-SX. These were the main line of money for Cyrix.
 
==== Cyrix 4x86 ====
 
==== Cyrix 4x86====
A processor that had the abilities of a 486. The first generation plugged into 386 sockets giving greater speeds without requiring extra hardware updates. Later editions could also be used on 486 motherboards.
 
==== Cyrix 5x86 ====
 
A processor that performed as a 486 and was socket-compatible. It features some of the Pentium's abilities, but not all (such as cmpxchg8b).
 
==== Cyrix 6x86 / M1 ====
This processor is, even though the name suggests otherwise, compatible with the 586 (Pentium). It didn't contain any of the MMX or PPro features but is nevertheless very nice. It performed slightly better per cycle compared to the Pentium Pro, and was thus given ratings. This was the time they were loathed for rating their processors, since the performance of floating point operations was lower than that of the Pentium Pro
 
This processor is, even though the name suggests otherwise, compatible with the 586 (Pentium). It didn't contain any of the MMX or PPro features. It performed slightly better per cycle compared to the Pentium Pro, and was thus given ratings. The performance of floating point operations was lower than that of the Pentium Pro.
==== Cyrix 6x86MX / M2====
 
Was a Pentium MMX compatible processor, also using ratings which gave it a bad name to start with. It was again socket-compatible to the Pentium MMX and the older Pentiums (without MMX). It supported a few features from the Pentium Pro, among which the very usable CMOVcc set. This however wasn't well known at the time, and nobody seemed to care.
==== Cyrix 6x86MX / M2 ====
 
Was a Pentium MMX compatible processor, also using ratings. It was again socket-compatible to the Pentium MMX and the older Pentiums (without MMX). It supported a few features from the Pentium Pro, among which the very usable CMOVcc set.
 
==== MediaGX ====
 
This was a chip that apart from the CPU, included several peripherals on-chip including graphics and audio devices.
This was a chip that, apart from the CPU, included several peripherals on-chip including graphics and audio devices.
 
The company was bought by National Semiconductor, who sold the trademark to VIA. The MediaGX was developed further and was eventually bought by AMD who marketed it as the Geode.
 
=== Rise Technologies ===
 
I'veA onlycompany heardreputedly about this company makingproducing Pentium-compatible chips, without MMX,. butLittle Idetail don'tis know any detailknown, but the CPUID identifier string. It just stuck. The string was "RiseRiseRise", or the same in all 3 dwords (making a search for it very easy).
 
==See Also==
Anonymous user
Retrieved from "https://osdev.wiki/wiki/Special:MobileDiff/18024...18035"