Linux kernel compilation times have significantly decreased due to new patches developed by Linux MM developer Lorenzo Stoakes with the assistance of AI and LLMs. These optimizations allow for faster build processes even on modest hardware configurations, removing previous bottlenecks in the defconfig x86-64 kernel build.
Testing on a high-performance workstation equipped with an AMD EPYC processor showed that the time required for a defconfig x86-64 kernel build dropped from over 22 seconds to 15 seconds. Notably, these results were achieved without the use of a RAMdisk, which is often employed to artificially accelerate compilation tasks.
The performance improvements also extend to more complex builds. Compiling a Linux kernel with allmodconfig, which includes thousands of drivers and subsystems, saw the build time reduced from 169 seconds to 134 seconds on the same workstation. Experts now anticipate that future hardware, such as AMD Zen 6 EPYC 9686F systems with MRDIMMs and PCIe Gen6 storage, could push compilation times below 10 seconds.
