cputime optimize for arm64

[dwh110]

What problem does this PR solve?

Issue Number: n/a

Problem Summary: cpugettime optimize for ARM64

What is changed and the side effects?

Changed:
detail::invariant_cpu_freq value get by CNTFRQ_EL0

Side effects:

• Performance effects:

• Breaking backward compatibility: 无

---

Check List:

• Please make sure your changes are compilable.
• When providing us with a new feature, it is best to add related tests.
• Please follow Contributor Covenant Code of Conduct.

[Copilot]

Pull request overview

This PR aims to optimize cpuwide_time_ns() on ARM64 by using the architectural counter frequency (CNTFRQ_EL0) to populate detail::invariant_cpu_freq, shifting initialization away from per-call logic.

Changes:

• On __aarch64__, route cpuwide_time_ns() to use the cycle-counter path instead of the clock_gettime path.
• Move initialization of detail::invariant_cpu_freq into a pre-main initializer, using CNTFRQ_EL0 on ARM64.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
src/butil/time.h	Changes cpuwide_time_ns() preprocessor gating to include ARM64 in the invariant-counter fast path and removes lazy init recursion.
src/butil/time.cpp	Adds a pre-main initializer to set detail::invariant_cpu_freq, reading CNTFRQ_EL0 on ARM64.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.