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

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

[Harry Yoo]

... 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.

It could be because SLUB is merging nfs_page cache with some other
cache that perf uses.

Could you please check if it reproduces with slab_nomerge kernel
parameter?

On Thu, Sep 17, 2026 at 02:46:24PM +0100, Harry Yoo wrote:
> Hi Tim and Chuck, thanks for reporting this to linux-mm.
> Will take a look at this but let me Cc SLAB ALLOCATOR folks here first.
> 
> -- 
> Cheers,
> Harry / Hyeonggon
> 
> On Wed, Sep 16, 2026 at 11:22:27PM +0000, Tim Menninger wrote:
> > Chuck Lever suggested I bring this to linux-mm after we found some SLUB
> > measurements we could not account for while investigating an NFS/RDMA
> > throughput regression.
> > 
> > The original NFS discussion is here for context:
> > 
> >     https://lore.kernel.org/linux-nfs/0e2f9688-097c-4bb9-a7ec-82b41bdb3653@slotpi15m67/
> > 
> > The NFS regression itself has been separated from this issue. What
> > remains interesting here is the behavior of the nfs_page slab cache on
> > this machine, particularly the measurements with and without perf lock.
> > 
> > The system is:
> > 
> >     Intel Xeon Silver 4516Y+
> >     2 sockets
> >     24 cores/socket
> >     2 threads/core
> >     96 logical CPUs
> > 
> >     NUMA node0 CPUs: 0-23,48-71
> >     NUMA node1 CPUs: 24-47,72-95
> >
> > The workload is a high-throughput NFS/RDMA direct-read workload using
> > 1 MiB I/O, 160 threads, and iodepth 64.
> > 
> > The relevant debug options are all disabled:
> > 
> >     # CONFIG_KASAN is not set
> >     # CONFIG_PROVE_LOCKING is not set
> >     # CONFIG_LOCK_STAT is not set
> >     # CONFIG_DEBUG_SPINLOCK is not set
> >     # CONFIG_DEBUG_LIST is not set
> > 
> > All measurements below were collected on the unpatched base kernel:
> > 
> >     $ git rev-parse HEAD
> >     940de590b839f71d6dc846160534bf202401b8b7
> > 
> >     $ uname -r
> >     7.3.0-rc1-mainline+
> > 
> > The initial observation was a high apparent contention rate on the
> > nfs_page slab's list_lock. In 10-second perf-lock captures I was seeing
> > roughly 4.5M contended acquisitions and about 170-185 us of average
> > reported wait.
> > 
> > Chuck reproduced a similar acquisition rate on a single-node EPYC
> > system, but saw only about 7 ns average wait and fewer than 100
> > cmpxchg_double_fail events over a corresponding interval. He suggested
> > checking cmpxchg_double_fail because __slab_free() drops list_lock and
> > retries when the freelist cmpxchg fails.
> >
> > I repeated the measurements in three placement configurations:
> > 
> >     A. workload unpinned, CQs all on node0
> >     B. workload pinned to node0, CQs all on node0
> >     C. workload unpinned, CQs balanced across the nodes
> > 
> > Without perf lock, throughput is similar in all three:
> > 
> >     A. unpinned / CQs node0:       ~45.5 GB/s
> >     B. node0 pinned / CQs node0:   ~45.7 GB/s
> >     C. unpinned / balanced CQs:    ~45.5 GB/s
> > 
> > During the perf-lock captures, throughput is approximately 25 GB/s.
> > 
> > For each instrumented 10-second window I ran:
> > 
> >     sudo perf lock record -a -g -o "$D/perf-locks.data" -- sleep 10
> >     mpstat -P ALL 1 10
> > 
> > Before and after the same window I sampled the counters under:
> > 
> >     /sys/kernel/slab/nfs_page/
> > 
> > I also collected separate 10-second counter and mpstat windows under
> > the same workload configurations without perf lock.
> > 
> > The resulting slab counter deltas were:
> > 
> >                             A               B                 C
> >                      unpinned/node0    node0/node0   unpinned/balanced
> > 
> >     free_fastpath
> >     instrumented        655,917,745    457,936,140       328,670,030
> >     uninstrumented       34,906,233    119,984,226        59,469,931
> > 
> >     free_slowpath
> >     instrumented        236,735,724      7,720,917       270,716,506
> >     uninstrumented       85,039,753          2,814        60,145,870
> > 
> >     sheaf_flush
> >     instrumented         39,842,700     42,706,800         8,341,440
> >     uninstrumented        1,286,400     13,487,700           887,700
> > 
> >     barn_put_fail
> >     instrumented            663,994        711,745           139,037
> >     uninstrumented           21,464        224,762            14,765
> > 
> >     barn_get_fail
> >     instrumented          4,609,925        840,344         4,651,225
> >     uninstrumented        1,438,643        224,787         1,017,006
> > 
> >     cmpxchg_double_fail
> >     instrumented             70,524          8,077            28,062
> >     uninstrumented            7,471            713             1,503
> > 
> >     alloc_slowpath
> >     all cases                     0              0                 0
> > 
> > The SLUB counter profile changes substantially with perf lock despite
> > the lower NFS throughput, and the exact mix depends strongly on
> > placement.
> > 
> > The uninstrumented node0/node0 run also reproduces the barn/sheaf
> > relationship Chuck pointed out earlier. The nfs_page sheaf capacity is
> > 60:
> > 
> >     sheaf_flush / 60 = 13,487,700 / 60 = 224,795
> >     barn_put_fail                      = 224,762
> > 
> > The placement dependence of free_slowpath is also large. It falls from
> > about 85M events per 10 seconds in the unpinned/node0 case to 2,814 in
> > the node0/node0 case.
> > 
> > The reported perf-lock result, however, is similar across all three
> > placements:
> > 
> >                          contentions    total wait    average wait
> >     unpinned/node0         4,782,839      14.39 min      180.47 us
> >     node0/node0            4,582,012      14.13 min      185.05 us
> >     unpinned/balanced      4,572,196      12.81 min      168.14 us
> > 
> > I also revisited an inconsistency Chuck noticed in my earlier
> > measurements. Previously I had compared aggregate perf-lock wait from
> > one 10-second capture with CPU utilization measured during a different
> > window.
> > 
> > I now have paired 10-second mpstat samples for each placement, with and
> > without perf lock. The node values below are averages of the per-CPU
> > %idle values for the CPUs in each NUMA node:
> > 
> >                                   system-wide       node0       node1
> >                                   %idle             %idle       %idle
> > 
> >     A. unpinned / CQs node0
> >        uninstrumented                 31.46           8.0        54.6
> >        instrumented                    3.60           0.06        7.1
> > 
> >     B. node0 pinned / CQs node0
> >        uninstrumented                 84.26          69.2        99.4
> >        instrumented                   12.07          10.4        13.8
> > 
> >     C. unpinned / balanced CQs
> >        uninstrumented                 66.86          66.7        66.9
> >        instrumented                   14.83          19.0        10.6
> > 
> > This resolves the accounting inconsistency in my earlier measurements.
> > The large aggregate perf-lock wait and high idle percentage had come
> > from different windows. In the aligned samples, the system is much
> > busier during the perf-lock capture than in the corresponding
> > uninstrumented run.
> > 
> > I am still unsure how representative the reported ~170-185 us average
> > wait is of the uninstrumented workload.
> > 
> > The remaining number I am less sure how to interpret is
> > cmpxchg_double_fail. In the uninstrumented windows I see:
> > 
> >     unpinned / CQs node0:       7,471 / 10 sec
> >     node0 pinned / CQs node0:     713 / 10 sec
> >     unpinned / balanced CQs:    1,503 / 10 sec
> > 
> > compared with fewer than 100 in Chuck's test.
> > 
> > I understand that cmpxchg_double_fail counts failed slab freelist
> > updates rather than failed logical frees, so I am not sure what the
> > appropriate denominator is here. In particular, the node0/node0 case
> > still has 713 failures while sustaining full throughput and only 2,814
> > free_slowpath events over the interval.
> > 
> > My questions are:
> > 
> > 1. Do the uninstrumented cmpxchg_double_fail rates above look abnormal
> >    for this workload/topology, or are they within the range one would
> >    expect from this degree of concurrency and NUMA placement?
> > 
> > 2. Is there a less invasive way you would recommend measuring the
> >    nfs_page list_lock/freelist contention? I would like to distinguish
> >    the steady-state behavior from what is observed during the perf-lock
> >    capture.
> > 
> > Thanks,
> > Tim

-- 
Cheers,
Harry / Hyeonggon

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

[Peter Zijlstra]

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.

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

[Tim Menninger]

It reproduces with slab_nomerge.

I booted the same kernel with:

    $ cat /proc/cmdline
    ... slab_nomerge ...

Uninstrumented throughput remains in the high-45 to low-46 GB/s
range, as before. During perf lock it was around 30 GB/s in these
runs, versus roughly 25 GB/s previously.

The dominant lock measurements are essentially unchanged:

                         default             slab_nomerge
    unpinned/node0
      contentions        4,782,839             4,858,759
      total wait          14.39 min             14.26 min
      average wait       180.47 us             176.13 us

    node0/node0
      contentions        4,582,012             4,686,729
      total wait          14.13 min             14.34 min
      average wait       185.05 us             183.53 us

    unpinned/balanced
      contentions        4,572,196             4,544,677
      total wait          12.81 min             12.56 min
      average wait       168.14 us             165.84 us

The uninstrumented cmpxchg_double_fail deltas were:

                         default     slab_nomerge
    unpinned/node0          7,471          8,892
    node0/node0               713            457
    unpinned/balanced       1,503          1,504

The CPU-utilization change while perf lock is active also remains.
System-wide %idle for uninstrumented/instrumented windows was:

                           default                  slab_nomerge
                       uninstr.   instr.         uninstr.   instr.
    unpinned/node0      31.46%    3.60%           23.02%    2.15%
    node0/node0         84.26%   12.07%           84.62%   11.85%
    unpinned/balanced   66.86%   14.83%           66.58%   14.78%

Thanks,
Tim