# Mojo equivalent for CUDA \_\_maxnreg\_\_

**URL:** <https://forum.modular.com/t/mojo-equivalent-for-cuda-maxnreg/2806>\
**Category:** Performance\
**Tags:** gpu\
**Created:** [March 4, 2026, 8:35pm UTC](https://forum.modular.com/t/mojo-equivalent-for-cuda-maxnreg/2806 "2026-03-04T20:35:39Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![massimim](https://sea1.discourse-cdn.com/flex001/user_avatar/forum.modular.com/massimim/32/938_2.png) [@massimim](https://forum.modular.com/u/massimim)\
**Post date:** [March 4, 2026, 8:35pm UTC](https://forum.modular.com/t/mojo-equivalent-for-cuda-maxnreg/2806/1 "2026-03-04T20:35:39Z")

</div>

Controlling register pressure on GPUs is critical for performance. CUDA exposes a per-kernel mechanism to enforce a maximum number of registers per thread on the compiler. Is there something equivalent in Mojo?

Reference: [5.4. C/C++ Language Extensions — CUDA Programming Guide](https://docs.nvidia.com/cuda/cuda-programming-guide/05-appendices/cpp-language-extensions.html#maximum-number-of-registers-per-thread)

Thank you.