Microsoft will support Intel's 64-bit extensions

Status
Not open for further replies.

Julio Franco

Posts: 9,092   +2,043
Staff member
Microsoft Corp. today announced at the Intel Developer Forum in San Francisco that its Windows® operating systems for 64-bit extended systems will be fully compatible with Intel Corp.’s newly announced processors with 64-bit extension technology.

Though it's still unclear whether Intel and AMD 64-bit implementations are fully compatible (if not the same), it comes as no surprise that Microsoft will cover Intel's back.
 
The Intel implementation ( dubbed IA-32e ) is not only fully compatible, it's an exact copy of the AMD design ( at 99.9% ).

Take a look at Intel's documentation -> Intel 64-bit Extension Technology

& AMD's documentation -> AMD64 Architecture Tech Docs

The same 64-bit instructions are available on both processors : SWAPGS, SYSCALL, SYSRET, CDQE, CMPSQ, LODSQ, MOVSQ, MOVZX et STOSQ.

There is one instruction in IA-32e that isn't present in AMD64 & that's CMPXCHG16B.

AMD register sheet

IMG0007451.gif


Intel register sheet

IMG0007452.gif


Notice any similarities ? ;)
 
Someone posted the differences between the two implementations. Taken from -> Intel vs AMD x86-64

Other than the standard IA-32 differences (eg. HT, SSE3, Intel Enhanced
SpeedStep, etc.), there are few differences between the implementations
of
IA-32e and AMD64. The software visible ones are:

Fast system calls:
Syscall/sysret is supported only in 64-bit mode (not in compatibility
mode). Sysenter/Sysexit is supported in both 64-bit and compatible
mode.

CPUID:
If you look at Table 2-8 of Volume 1, you will find IA-32e specific
things,
including, GenuineIntel, HT, SSE3, monitor/mwait, Intel Enhanced
SpeedStep,
and cmpxchg16b.

The function 8000_0001h doesn't duplicate standard-feature bits from
function 1 in EDX. It sets only the new features that are implemented.

MSRs:
Not all MSRs are architectural, and IA-32e does not implement SYSCFG,
TOP_MEM, TOP_MEM2, for example. MSR usage should be vendor specific
and
be guarded with CPUID.Model

Fast-FXSAVE/FXRSTOR:
IA-32e always saves all of the FP state on FXSAVE/FXRSTOR. Does not
support FXSAVE/FXRSTOR with reduced FP state.

Microcode Update:
IA-32e supports microcode update as the 32-bit mode does, as you
already
found the discussions in the mailing list.

NX (No-Execute) bit:
Initial implementation will not support the NX bit.

BSF/BSR when source is 0 & operand size is 32:
In 64-bit mode, the processor sets ZF, and the upper 32 bits of
the destination are undefined. Should always check the ZF or do not
use
32-bit operand size.

Near branch with 66H prefix:
As documented in PRM the behavior is implementation specific and
should
avoid using 66H prefix on near branches.

Not supported in IA-32e
=======================
3DNow instructions (including prefecthw or prefetch with the opcode 0f
0d)

Thanks,
Jun
 
EM64T : Intel Names 64-Bit Extension Technology for Xeon

Intel Corp. will call the 64-bit extension technology for its 32-bit Xeon processors Intel Extended Memory 64 Technology, or Intel EM64T.

Starting in the second quarter with its Nocona chip for two-processor systems, Intel will outfit its Xeons with the ability to run 64-bit as well as 32-bit applications, a move announced at the company's Intel Developer Forum in February.

ADVERTISEMENT

Intel, of Santa Clara, Calif., will follow Nocona with 64-bit extensions for Prescott—for one-way servers and workstations—and in 2005 with chips for systems running four or more processors.

The announcement last month made official what industry observers had suspected for several years—that despite the money and effort Intel put behind creating and marketing its 64-bit Itanium chip, it was working on another project designed to add extensions to its 32-bit processors.

Itanium systems can run 32-bit applications, but only through an emulation software layer, and at a lower level of performance than 64-bit software.

The new EM64T technology—originally code-named Yamhill, then later CT—will be attractive to users who currently are running 32-bit applications on Xeon-based systems and who want to gradually migrate into 64-bit computing. Itanium will continue to be a high-end product for customers with needs for more memory and higher performance, or who want to move from their Unix or mainframe environments.

The extensions also give Intel an answer to Advanced Micro Devices Inc.'s Opteron chips, which also run 32- and 64-bit x86 applications.

Most major OEMs, including Dell Inc., Hewlett-Packard Co. and IBM, have said they will support the new chips in their systems. But Opteron has a year's head start on Intel's extensions, and HP, IBM and Sun Microsystems Inc. are among the vendors that are offering or that plan to offer systems with those processors.

Intel officials said that the extensions are only part of a long list of features that they are looking to incorporate into all of the company's chips, including HyperThreading, PCI Express, DDR2 memory support, enhanced security and virtualization, through its Vanderpool technology.

Article @ eWeek.com
 
Status
Not open for further replies.
Back