x86_64

View documentation for the original command

$ tldr setarch

Not a command; output of uname -m or arch on x86_64 systems.

x86_64 is not a standalone Linux command or executable. Instead, it is the standard identifier for the 64-bit extension of the x86 processor architecture, also known as AMD64 or Intel 64. In Linux environments, it appears as output from commands like uname -m, arch, or grep '^model name' /proc/cpuinfo.

This architecture, introduced by AMD in 2000, supports 64-bit virtual addressing (up to 2^48 bytes typically), 16 general-purpose registers (vs 8 in 32-bit), SIMD extensions like SSE/AVX, and backward compatibility with 32-bit x86 code via a legacy mode. Nearly all modern Linux distributions (e.g., Ubuntu, Fedora, Debian) use x86_64 as their default target for kernels, binaries, and packages.

In practice, x86_64 is embedded in ELF file headers (readelf -h shows "Machine: Advanced Micro Devices X86-64"), dynamic linker paths (/lib64/ld-linux-x86-64.so.2), and package naming (e.g., glibc-x86_64). Cross-compilation uses triplets like x86_64-linux-gnu-gcc. For emulation, QEMU provides qemu-system-x86_64. Multiarch support allows running 32-bit apps on x86_64 systems via i386 packages.

Detecting it helps in scripting: if [[ $(uname -m) == x86_64 ]]; then echo '64-bit'; fi. It dominates servers, desktops, and cloud (99%+ of Linux servers). Limitations include no native 128-bit support yet.

Not executable; mistaking for command may indicate confusion with architecture detection tools. 32-bit compatibility requires additional packages.

uname -m, arch, lscpu | grep Architecture, or grep flags /proc/cpuinfo | grep lm (long mode = x86_64 capable).

Use gcc -m64 or triplet x86_64-linux-gnu-gcc; check with gcc -dumpmachine.

AMD announced Opteron (K8) architecture in 2000 as AMD64; Linux kernel support merged 2004 (v2.6.11). Intel adopted as EM64T (2004), later Intel 64. Became de facto standard by 2010, replacing i686.

uname(1), arch(1), readelf(1), file(1), getconf(3)