Advanced Matrix Extensions (AMX), also known as Intel Advanced Matrix Extensions (Intel AMX), are extensions to the x86

instruction set architecture In computer science, an instruction set architecture (ISA), also called computer architecture, is an abstract model of a computer. A device that executes instructions described by that ISA, such as a central processing unit (CPU), is called an ...

(ISA) for

microprocessor A microprocessor is a computer processor where the data processing logic and control is included on a single integrated circuit, or a small number of integrated circuits. The microprocessor contains the arithmetic, logic, and control circu ...

s from

Intel Intel Corporation is an American multinational corporation and technology company headquartered in Santa Clara, California, Santa Clara, California. It is the world's largest semiconductor chip manufacturer by revenue, and is one of the devel ...

and

Advanced Micro Devices Advanced Micro Devices, Inc. (AMD) is an American multinational semiconductor company based in Santa Clara, California, that develops computer processors and related technologies for business and consumer markets. While it initially manufa ...

(AMD) designed to work on

matrices Matrix most commonly refers to: * ''The Matrix'' (franchise), an American media franchise ** ''The Matrix'', a 1999 science-fiction action film ** "The Matrix", a fictional setting, a virtual reality environment, within ''The Matrix'' (franchis ...

to accelerate

artificial intelligence Artificial intelligence (AI) is intelligence—perceiving, synthesizing, and inferring information—demonstrated by machines, as opposed to intelligence displayed by animals and humans. Example tasks in which this is done include speech r ...

(AI) /

machine learning Machine learning (ML) is a field of inquiry devoted to understanding and building methods that 'learn', that is, methods that leverage data to improve performance on some set of tasks. It is seen as a part of artificial intelligence. Machine ...

(ML) -related workloads.

Extensions

AMX was introduced by Intel in June 2020 and first supported by Intel with the

Sapphire Rapids Sapphire Rapids is a codename for Intel's server (fourth generation Xeon Scalable) and workstation processors based on Intel 7. Sapphire Rapids was intended as part of the Eagle Stream server platform. In addition, it will be powering Aurora, a ...

microarchitecture In computer engineering, microarchitecture, also called computer organization and sometimes abbreviated as µarch or uarch, is the way a given instruction set architecture (ISA) is implemented in a particular processor. A given ISA may be imp ...

for

Xeon Xeon ( ) is a brand of x86 microprocessors designed, manufactured, and marketed by Intel, targeted at the non-consumer workstation, server, and embedded system markets. It was introduced in June 1998. Xeon processors are based on the same a ...

servers, released in January 2023. It introduced 2-dimensional registers called tiles upon which accelerators can perform operations. It is intended as an extensible architecture; the first accelerator implemented is called tile matrix multiply unit (TMUL). In Intel Architecture Instruction Set Extensions and Future Features revision 46, published in September 2022, a new AMX-FP16 extension was documented, planned for inclusion in the future

Granite Rapids Granite () is a coarse-grained (phaneritic) intrusive igneous rock composed mostly of quartz, alkali feldspar, and plagioclase. It forms from magma with a high content of silica and alkali metal oxides that slowly cools and solidifies undergro ...

processors. This extension adds support for half-precision floating-point numbers.

Tile matrix multiply unit

TMUL unit supports BF16 and

INT8 In computer science, an integer is a datum of integral data type, a data type that represents some range of mathematical integers. Integral data types may be of different sizes and may or may not be allowed to contain negative values. Integers are ...

input types. AMX-FP16 also adds support for

FP16 In computing, half precision (sometimes called FP16) is a binary floating-point computer number format that occupies 16 bits (two bytes in modern computers) in computer memory. It is intended for storage of floating-point values in applications wh ...

numbers. The register file consists of 8 tiles, each with 16 rows of size 64-byte (32 BF16/FP16 or 64 INT8 values). The only supported operation as for now is

matrix multiplication In mathematics, particularly in linear algebra, matrix multiplication is a binary operation that produces a matrix from two matrices. For matrix multiplication, the number of columns in the first matrix must be equal to the number of rows in the ...

C_ += \sum_^K A_B_.

Software support

* Compiler and assembler support **

LLVM LLVM is a set of compiler and toolchain technologies that can be used to develop a front end for any programming language and a back end for any instruction set architecture. LLVM is designed around a language-independent intermediate repre ...

initial support committed at 1 July 2020 **

GNU Assembler The GNU Assembler, commonly known as gas or as, is the assembler developed by the GNU Project. It is the default back-end of GCC. It is used to assemble the GNU operating system and the Linux kernel, and various other software. It is a part o ...

(GAS) initial support committed at 25 June 2020 ** GCC patches have been posted for version 11 and merged *Operating system support **

glibc The GNU C Library, commonly known as glibc, is the GNU Project's implementation of the C standard library. Despite its name, it now also directly supports C++ (and, indirectly, other programming languages). It was started in the 1980s by ...

support for detecting AMX feature in CPUs committed at 25 June 2020 **

Linux kernel The Linux kernel is a free and open-source, monolithic, modular, multitasking, Unix-like operating system kernel. It was originally authored in 1991 by Linus Torvalds for his i386-based PC, and it was soon adopted as the kernel for the GNU ...

support released in version 5.16

References

External links

Intel Intrinsics Guide

Wikichip: Advanced Matrix Extension (AMX) - x86
{{Multimedia extensions, state=uncollapsed X86 instructions SIMD computing AMD technologies