LLVM 17 Lands Arm Neoverse-V2 Scheduling Model

Written by Michael Larabel in Arm on 15 June 2023 at 07:24 AM EDT. 2 Comments
For helping to ensure optimal performance of AArch64 binaries generated by LLVM/Clang for the Neoverse-V2 processor cores, LLVM 17 Git has received a proper Neoverse-V2 scheduling model.

To this point the Neoverse-V2 targeting within LLVM has relied on using the Neoverse-N2 scheduling model while in Git as of yesterday a proper V2 scheduling model based on information from the Neoverse-V2 software optimization guide has been merged.

This commit adds the optimized Neoverse-V2 scheduling model so that LLVM can better cater to this Neoverse core to be found in various future SoCs.

Neoverse-V2 Platform

Arm's Neoverse V-series is designed for maximum performance and optimal TCO. Arm announced the Neoverse-V2 last year and since then patches have begun flowing into the open-source compilers and other components to prepare for future Arm server hardware based on the Neoverse-V2 platform. Already within LLVM 16 and GCC 13 is the initial support while as shown by yesterday's work with the new dedicated scheduling model there are further refinements and optimizations to come.

The LLVM/Clang 17 stable compiler release should debut around September.
Related News
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week