Course code for 'Optimize code with Arm Scalable Vector Extension (SVE)'

This repository provides example programs, reference implementations, tests, and benchmarks for Optimize code with Arm Scalable Vector Extension (SVE). Follow the course for prerequisites, environment setup, and exercise instructions, then use the module links to find the accompanying code.

Find your module

Module	Code
1	Integer vector operations
2	Run-length decoding
3	C-string length
4	Multi-value search with SVE2 MATCH
5	Sorted set intersection
6	Find and process matching values in order
7	Histogram updates with SVE and SVE2
8	JSON structural-character indexing
9	Bloom filter
10	Vector-similarity metrics
11	Base64 encode and decode
12	Squared-distance calculations in Rust
13	Arm SVE vectorization in Go

Module 0 introduces the course and shared setup; it has no separate code directory.

Using this repository

• Follow the lesson's environment and command instructions, then open the implementation in the corresponding module directory.
• Keep the complete repository together. C++ modules use the shared build infrastructure in course_support/. Configure and build from the repository root using the build, test, and benchmark guide.
• Build the Rust and Go examples separately using the commands in the build guide. Their toolchain and runtime requirements differ from the C++ tasks.
• Use the intrinsics overview to find examples of particular SVE operations. For performance measurements, use a Release build on native hardware with the required features; the build guide also describes QEMU for correctness checks.

Optional SVE and SIMD optimization guidance for AI agents supports working with the code.

For fixes and new tasks, see Contributing. Code is provided under the Clear BSD License.