AMD EPYC 9xx5-Series Processors Compiler Options Quick Reference - AMD EPYC 9xx5-Series Processors Compiler Options Quick Reference

AOCC compiler (C/C++/Fortran)

Latest release: 5.2, May 2026

https://www.amd.com/en/developer/aocc.html

Action	Command
Architecture
Generate instructions that run on AMD 5^th Gen EPYC™ and AMD 5^th Gen Ryzen™	`-march=znver5`
Generate instructions supported in the given machine	`-march=native`
Optimization Levels
Disable all optimizations	`-O0`
Enable minimal level optimizations	`-O1`
Enable moderate level optimizations (Default from AOCC 4.1)	`-O2/-O`
Enable all optimizations that attempt to make programs run faster	`-O3`
Enable O3 with other aggressive optimizations that may violate strict compliance and precisions	`-Ofast`
Enable link time optimization	`-flto`
Enable advanced optimizations - improved variants of various scalar, vector, and loop transformations	`-zopt`
Enable advanced vector transformations	`-fvector-transform -mllvm -enable-strided-vectorization`
Enable loop transformations	`-floop-transform`
Enable advanced loop transformations	`-faggressive-loop-transform`
Enable memory layout optimizations	`-flto -fremap-arrays -mllvm -reduce-array-computations=3`
Enable function level optimizations	`-flto -fitodcalls -mllvm -function-specialize -flto -finline-recursion={1..4}`
Profile guided optimizations	`-fprofile-instr-generate` (1st invocation) `-fprofile-instr-use` (2nd invocation)
Enable use of OpenMP® directives	`-fopenmp`
Enable streaming stores to optimize memory bandwidth usage	`-fnt-store`
Other Options
Enable faster, less precise math operations (part of Ofast)	`-ffast-math` `-freciprocal-math`
OpenMP® threads and affinity (N number of cores)	`export OMP_NUM_THREADS=N` `export GOMP_CPU_AFFINITY="0-{N-1}"`
Link to AMD library	`-L/libm-install-dir/lib -lamdlibm -lm`
Enable vector library	`-fveclib=AMDLIBM -lamdlibm -lm`
Enable faster library	`-Ofast -ffastlib=AMDLIBM -lamdlibmfast -lamdlibm -lm`
For Fortran Workloads
Compile Fortran free form layout	`-ffree-form`

AMD Optimized Libraries

Latest release: 5.3, May 2026

https://www.amd.com/en/developer/aocl.html

AMD uProf (Performance & Power Profiler)

Latest release: 5.3, May 2026

https://www.amd.com/en/developer/uprof.html

GNU Compiler Collection

Recommended version: GCC 14.1 or later

Latest release: GCC 16.1, April 2026

http://gcc.gnu.org

glibC

Latest release: 2.43, January 2026

Recommendation: 2.38 or later

https://www.gnu.org/software/libc/

Binutils

Latest release: 2.46.0, February 2026

Recommendation: 2.42 or later

https://www.gnu.org/software/binutils/

Action	Command
Architecture
Generate instructions that run on AMD 5^th Gen EPYC™ and AMD 5^th Gen Ryzen™	`-march=znver5`
Generate instructions supported in the given machine	`-march=native`
Optimization Levels
Disable all optimizations (default)	`-O0`
Enable minimal level optimizations	`-O1/ -O`
Enable moderate level optimizations	`-O2`
Enable all optimizations that attempt to make programs run faster	`-O3`
Enable O3 with other aggressive optimizations that may violate strict compliance and precisions	`-Ofast`
Additional Optimizations
Enable link time optimizations	`-flto`
Enable unrolling	`-funroll-all-loops`
Generate memory preload instructions	`-fprefetch-loop-arrays`
Enable profile-guided optimizations	`-fprofile-generate` (1^st invocation) `-fprofile-use` (2^nd invocation)
Enable use of OpenMP® directives	`-fopenmp`
Other Options
Enable compiler to use IEEE FP comparisons	`-mieee-fp`
Enable faster, less precise math operations	`-ffast-math`
Compile Fortran free form layout	`-ffree-form`
OpenMP® threads and affinity (N number of cores)	`export OMP_NUM_THREADS=N` `export GOMP_CPU_AFFINITY="0-{N-1}"`
Link to AMD library	`-L/libm-install-dir/lib -lamdlibm -lm`

Microsoft® Visual Studio 2026

https://visualstudio.microsoft.com/downloads/

Action	Command
Architecture
Generate instructions that run on AMD 5^th Gen EPYC™ and AMD 5^th Gen Ryzen™	`/arch:[AVX\|AVX2\|AVX512]`
Optimize for 64-bit AMD processors	`/favor:AMD64`
Optimization Levels
Disable optimizations	`/Od`
Maximum optimizations (favor space)	`/O1` includes `/Ob2`
Maximum optimizations (favor speed)	`/O2` includes `/Ob2`
Enable inline expansion	`/Ob` (0/1/2/3)
[link.exe] Eliminate unreferenced function and/or data	`/OPT:REF`
[link.exe] Perform identical COMDAT folding	`/OPT:ICF`
Output an informational message for loops that are auto-vectorized	`/Qvec-report:[1\|2]`
Enable automatic parallelization of loops, used with #pragma loop() directive	`/Qpar`
Output an informational message for loops that are auto-parallelized	`/Qpar-report:[1\|2]`
Additional Optimizations
Maintain the precision for floating-point operations through proper rounding	`/fp:precise`
Optimize floating-point code for speed at the expense of floating point accuracy and correctness	`/fp:fast`
Whole Program Optimization (link-time code generation)	`/GL`
Enable Profile-guided optimizations	`/LTCG /GENPROFILE` (1^st invocation) `/LTCG /USERPROFILE` (2^nd invocation)
Enable OpenMP® Support	`/openmp:experimental` `/openmp:llvm`

Intel® oneAPI DPC++/C++ Compiler

Latest release: 2025.3.3, March 2026

http://software.intel.com

Action	Command
Architecture
Generate instructions that run on AMD 5^th Gen EPYC™ and AMD 5^th Gen Ryzen™	`-axCORE-AVX512`
Optimization Levels
Disable all optimizations	`-O0`
Speed optimization without code growth	`-O1`
Enable optimization for speed including vectorization	`-O2`
Enable O2 and aggressive loop transformations	`-O3`
Enable set of aggressive options to improve speed	`-Ofast`
Additional Optimizations
Set function inline level	`-inline-level=<value>`
Set maximum number of times to unroll loops	`-unroll[=n]`
Disable improved precision floating divides	`-no-prec-div`
Enable vectorization	`-vec`
Enable inter procedural optimizations (alias for -flto)	`-ipo`
Enable whole program link time optimization (LTO)	`-flto[=arg];` `arg: full (default), thin`
Enable use of OpenMP® directives	`-qopenmp`
Enable profile generated optimization	`-prof-gen` and `-prof-use`
Other Options
Enable floating point accuracy tunings	`-fp-model`
Compile Fortran free form layout	`-free`

AMD EPYC 9xx5-Series Processors Compiler Options Quick Reference - AMD EPYC 9xx5-Series Processors Compiler Options Quick Reference - 63857

AOCC Quick Reference Guide