📅 性能日报

每日 Linux 内核与性能优化资讯汇总。

最新日报

• sched-zone
  mm: optimize zone-device memmap initialization
• sched-xor-
  lib/raid/xor: x86: Add AVX-512 optimized xor_gen()
• sched-x86-
  x86/kcfi: Optimize call sequence
• sched-work
  workqueue: add a per-cpu backend for unbound pwqs（unbound 工作队列每-CPU 后端）
• sched-vmsc
  mm/vmscan: reduce lru_lock contention via vmstat-derived scan-balance cost
• sched-vma-
  mm: use VMA lock for kernel faults on user addresses
• sched-sync
  sched/fair: Let sync wakeups target the waker's core
• sched-stmm
  net: stmmac: Use pagepool instead of streaming buffers for Tx
• sched-soft
  mm: Use a folio in the softleaf_is_device_private path
• sched-sche
  sched_ext: Add bandwidth-limited rescue execution for stranded tasks
• sched-rtla
  rtla/osnoise: 新增 IPI 追踪，把跨 CPU 中断干扰从 IRQ 计数中拆出（v4）
• sched-remo
  mm: remove wait_on_page_writeback()
• sched-rcu-
  rcu: Make call_rcu() safe to call from any context
• sched-prox
  sched/proxy：优化 try_to_wake_up() — 代理调度唤醒热路径瘦身
• sched-prox
  sched/proxy：try_to_wake_up() 内联处理 return-migration — 代理调度唤醒路径去锁往返
• sched-pree
  preempt/irq：计数式中断开关与 NMI 位域迁出 — Rust SpinLockIrq 系列 v4 进展更新
• sched-per-
  mm: Unconditional per-VMA locks and cleanups
• sched-page
  mm: add fast path for AS_NO_DIRECT_MAP
• sched-nmi-
  preempt/irq：NMI 嵌套计数迁出与中断重调度热路径优化（Rust sync v7.3 系列）
• sched-mm-s
  mm/slab: introduce kmalloc_flags()

按月份浏览