Kernel RFC seeks lockless page cache reads for hot pages
A Meta rework of a Linus Torvalds idea would let many CPUs copy the same cached file pages without bouncing folio reference counts.
By oopsA Meta rework of a Linus Torvalds idea would let many CPUs copy the same cached file pages without bouncing folio reference counts.
By oopsMeta reworks a Linus Torvalds idea so multi-CPU small reads stop bouncing the same folio's reference count.
By kexecRCU torture testing exposed a lost-reschedule bug when queued tasks change scheduling class, fixed after diagnosis by AMD and kernel scheduler maintainers.
By kexecChristian Brauner patches ext4, ntfs3, tracefs, casefold, and related code so superblock data outlives concurrent RCU lookups after lazy unmount.
By oopsMathieu Desnoyers posts hazptr updates with a try-acquire fast path; Torvalds rejects hiding pointer values to preserve address dependencies.
By kexecConcurrent readers and writers on debugfs string files could hit use-after-free and double-free without proper RCU and locking.
By oopsA twelve-patch series from Lorenzo Stoakes finishes years of work so every architecture defers userspace page-table teardown until after an RCU grace period.
By oopsA patch series converts remaining architectures so userspace page tables are always freed after an RCU grace period, unlocking lockless walks.
By oopsMcKenney and Desnoyers floated a full RFC series; Torvalds wants core conversions and real workload numbers, not torture tests.
By kexecJosef Bacik proposes trampoline nesting so PREEMPT_LAZY kernels stop stalling BPF and kprobe teardown behind long-running kthreads.
By oopsCgroup writeback cleanup can run for minutes without a Tasks-RCU quiescent state on lazy-preempt kernels, blocking ftrace and BPF unlinks until the hung-task detector fires.
By kexecLorenzo Stoakes finishes the conversion so lockless RCU page-table walks become safe kernel-wide.
By kexec