by Anton Shilov
03/11/2004 | 12:30 AM
Intel Corporation will shortly unveil the official naming for its 64-bit extension technology that enables Xeon and Pentium 4 E processors to address larger amounts of memory and run specially written 64-bit code.
The official brand-name for one of the key technologies found in Intel’s upcoming server, desktop and even mobile processors will be Intel Extended Memory 64 Technology (EM64T). Apart from the EM64T, the new platforms from Intel due in the second quarter of the year will sport DDR-II SDRAM (DDR2) memory, PCI Express interconnection bus as well as enhancements to the Hyper-Threading technology already found in Intel Pentium 4 and Intel Xeon microprocessors, eWeek web-site notes.
Unlike the branding may imply Intel’s extensions to the x86 architecture are not limited to improvements concerning large amounts of system memory. Intel’s approach to enhance the IA32 architecture is very similar to what its arch-rival AMD implemented into its Opteron and Athlon 64 central processing units.
Intel, just like AMD, added a special CPU mode called “64-bit sub-mode”, where 64-bit flat linear addressing, 8 new general-purpose registers (GPRs), 8 new registers for streaming SIMD extensions (SSE, SSE2 and SSE3) and 64-bit-wide GPRs are available along with instruction pointers. Similar to AMD’s 64-bit chips, Intel’s 64-bit extension technology can run in either legacy IA32 mode or IA32e mode. IA-32e mode is the mode a processor uses when running a 64-bit operating system. The IA32e mode consists of two sub-modes: 64-bit mode and compatibility mode, just like it is implemented in AMD64 architecture.
Even though there are some differences between Intel’s and AMD’s 64-bit extended processors’ capabilities, such as NX (non execute) bit featured in AMD64, generally speaking IA32e – now called Extended Memory 64 Technology – and AMD64 – also known as x86-64 – are compatible and are capable of running similar specially written 64-bit code.