Add butil::Seqlock and use it for TaskGroup CPU time stat

[chenBright]

What problem does this PR solve?

Issue Number: resolve

Problem Summary:

TaskGroup tracks per-group CPU time as a 128-bit stat (a packed
last_run_ns + task-type word and a cumulated_cputime_ns word) that a
worker updates while other threads (e.g. bvar sampling) read it
concurrently. This was implemented with AtomicInteger128, whose 128-bit
"atomic" load/store had no portable, guaranteed lock-free backing:

• On x86 it relied on wide SSE/AVX aligned loads/stores being atomic. This is
  not guaranteed by the ISA -- Intel and AMD do not officially promise 128-bit
  AVX load/store atomicity; it merely happens to hold on current microarchitectures
  (Skylake, Zen 2). Depending on it is relying on unspecified hardware behavior.
• On platforms without such a wide atomic it fell back to a mutex, which serializes
  readers and blocks them behind the writer -- exactly what a consistent-snapshot
  read is meant to avoid.

What is changed and the side effects?

Changed:

• Add butil::Seqlock (butil/synchronization/seqlock.h), a general sequence lock
  providing lock-free, consistent snapshot reads around a caller-owned atomic payload.
• Replace TaskGroup::AtomicInteger128 with an AtomicCPUTimeStat backed by
  butil::Seqlock<> over a CPUTimeStat payload accessed via relaxed atomics.

Side effects:

• Performance effects:

• Breaking backward compatibility:

---

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 introduces a general-purpose butil::Seqlock primitive to provide consistent, lock-free snapshot reads over a multi-word payload, and applies it to bthread::TaskGroup’s per-group CPU time statistic to avoid relying on non-portable 128-bit atomic behavior.

Changes:

• Add butil::Seqlock (single-writer and mutex-serialized multi-writer variants) plus a unit test suite.
• Replace TaskGroup’s prior 128-bit “atomic” CPU time stat mechanism with a seqlock-based implementation.
• Centralize cpu_relax()/barrier() in a new butil/processor.h and update bthread/processor.h to include it; add an ASan-only LSan suppression hook for a known benign leak in controller tests.

Reviewed changes

Copilot reviewed 10 out of 10 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
test/seqlock_unittest.cpp	Adds Seqlock unit tests (single-threaded and concurrent reader/writer scenarios).
test/Makefile	Registers seqlock_unittest.cpp in the Make-based test sources.
test/CMakeLists.txt	Registers seqlock_unittest.cpp in the CMake test sources.
test/BUILD.bazel	Registers seqlock_unittest.cpp in the Bazel test sources.
test/brpc_controller_unittest.cpp	Adds an ASan-only __lsan_default_suppressions hook to suppress a known shutdown leak report.
src/butil/synchronization/seqlock.h	Introduces butil::Seqlock built on a cacheline-aligned sequence counter.
src/butil/processor.h	New shared header defining cpu_relax() and barrier() macros.
src/bthread/task_group.h	Reworks CPU time stat storage to use Seqlock (replacing prior 128-bit “atomic” approach).
src/bthread/task_group.cpp	Removes the previous AtomicInteger128 implementation.
src/bthread/processor.h	Switches to including butil/processor.h instead of duplicating macros.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.