Instruction
WRGSBASE
Writes the GS base from a register; 64-bit mode, when the kernel set CR4.FSGSBASE.
- Extensions
- Forms
- 1 (1 Legacy)
- Data
- Intel XED v2026.08.23
WRGSBASE loads the base address of the GS segment from a general register; later GS-relative accesses add the new base. It is the GS counterpart of WRFSBASE, whose page covers the shared rules: #UD unless the processor has FSGSBASE and the kernel has set CR4.FSGSBASE, #GP(0) for a base that is not canonical, and a 32-bit form that clears the upper half. Zen 5 did the same for GS: a write of 8000000000000000H raised #GP and kept the old base, and ECX = 80001234H made the base 0000000080001234H, with FS untouched in every case. RDGSBASE reads the base back.
Without a system call
Linux’s documentation leaves GS free for programs, and WRGSBASE moves
it in one unprivileged instruction where arch_prctl(ARCH_SET_GS, ...)
needs a system call, which the same documentation says the kernel’s
configuration may disable. A runtime can thus give each thread, or
each task it schedules, its own block of data at gs:0 and switch
between them without entering the kernel.
The compiler does not see the dependency
GCC and Clang accept __seg_gs for GS-relative accesses, but nothing
ties those accesses to the WRGSBASE that sets their base. In our build
of a function that saves the base with _readgsbase_u64(), installs
another with _writegsbase_u64(p), reads *(long __seg_gs *)8 and
restores the old base, GCC 14 at -O2 emitted both WRGSBASEs first and
the load after them, so it read through the old base; Clang 23 kept
the load between them. A compiler barrier,
asm volatile("" ::: "memory"), after the first write and before the
second, or a volatile access, kept the order with GCC too. Code that
switches GS around an access should not rely on the compiler to keep
the two apart.
Where the kernel stands
The base a program writes is the one that a kernel entered through
SWAPGS finds in IA32_KERNEL_GS_BASE, swapped out of the
way, and puts back on the return. GCC 14 and Clang 23 expose
WRGSBASE as _writegsbase_u64() and _writegsbase_u32() with
-mfsgsbase. AMD’s Zen 5 guide lists WRGSBASE as microcoded.
Forms
| Instruction | Encoding | Requires |
|---|---|---|
WRGSBASE r32/64 64-bit mode only
WRGSBASE_GPRy | F3 0F AE /3 | FSGSBASE |
Operation
What the forms do, in pseudocode that our tests run against results recorded on real processors. How to read.
v = src.read() # r32: zero-extended
if not canonical(v):
fault(GP, 0)
gsbase = vRun on AMD Zen 5 (EPYC 9645): 4 cases.
Operands
| Operand | Kind | Action |
|---|---|---|
WRGSBASE r32/64 64-bit mode only | ||
r32/64 | explicit | read |
GS base | implicit | write |
Flags
None: XED records no flag that the forms test or change.
Attributes
What else XED records about the forms. How to read.
- All forms
- may abort a TSX transaction
Intrinsics
The C intrinsics of Clang 23.1.2 that compile to WRGSBASE, each with the form it compiles to and the Clang options it needs. The forms come from compiling every intrinsic and decoding the result with Intel XED; all intrinsics A–Z explains the method.
| Intrinsic | Compiles to |
|---|---|
_writegsbase_u32
void _writegsbase_u32(unsigned int __V) |
|
CPUID and processors
Each form belongs to an XED ISA set. A form can be used when the processor reports every CPUID bit of one of its ISA set's alternatives.
| ISA set | CPUID | Processors |
|---|---|---|
RDWRFSGS | FSGSBASE | Ivy Bridge, Haswell, Broadwell, Skylake, Comet Lake, Cannon Lake, Ice Lake (client), Tiger Lake, Alder Lake, Arrow Lake, Panther Lake, Nova Lake, Skylake-SP, Cascade Lake, Cooper Lake, Ice Lake (server), Sapphire Rapids, Emerald Rapids, Granite Rapids, Diamond Rapids, Goldmont, Goldmont Plus, Tremont, Snow Ridge, Lakefield, Sierra Forest, Clearwater Forest, Knights Landing, Knights Mill, Bulldozer, Piledriver, Zen, Zen+, Zen 2, Zen 3, Zen 4, Zen 5, Zen 6 |
More FSGSBASE instructions
Sources
- Intel XED v2026.08.23 (commit
0bcb6237345c): forms, encodings, operands, flags, exception classes, attributes, ISA sets, CPUID bits and chip model. - The CPU data of x86-64.net: Zen 3 to Zen 6, from GCC’s
-marchfeature sets and AMD’s Processor Programming References, which the page of each processor cites. - Clang's x86 intrinsic headers, LLVM 23.1.2 (commit
85ac56026243): the intrinsics, their signatures and doc comments, and the Clang of that release, which compiled each of them. - Intel 64 and IA-32 Architectures Software Developer's Manual, Volume 2 (325383-093, September 2026): Volume 2D, page 6-6 (WRFSBASE/
WRGSBASE) - Linux kernel documentation: Using FS and GS segments in user space applications
- AMD64 Architecture Programmer's Manual, Volume 3: General-Purpose and System Instructions (24594, rev. 3.36, March 2024), Wayback Machine copy of AMD's PDF
- Software Optimization Guide for the AMD Zen5 Microarchitecture (58455, rev. 1.00, August 2024), Wayback Machine copy of AMD's ZIP
The tables are derived from Intel XED, Copyright Intel Corporation, licensed under the Apache License 2.0; x86-64.net converted and reformatted them. The intrinsics and their signatures come from Clang, part of the LLVM Project, licensed under the Apache License 2.0 with LLVM Exceptions. The text is our own.