Re: [SLUB] nfs_page cmpxchg_double_fail and perf lock perturbation on dual-socket NFS/RDMA

[Namhyung Kim]

Hello,

On Thu, Sep 17, 2026 at 05:29:59PM +0200, Peter Zijlstra wrote:
> On Thu, Sep 17, 2026 at 03:10:47PM +0100, Harry Yoo wrote:
> > ... now I realize we need to add PERFORMANCE EVENTS SUBSYSTEM folks 
> > as well ;-)
> > 
> > Hmm, sounds like perf lock is somehow triggering slab allocations
> > and interfering the workload.
> 
> perf should not have runtime memory allocation, its build to run from
> NMI context. It can allocate some memory on event creation and the like,
> but once its on, there should be no more allocations happening.

Right, perf lock record command will save the event info to a ring
buffer without allocation.  It may lose some events when the buffer is
full though.

Thanks,
Namhyung