All of lore.kernel.org
 help / color / mirror / Atom feed
From: kernel test robot <oliver.sang@intel.com>
To: Jeff Layton <jlayton@kernel.org>
Cc: <oe-lkp@lists.linux.dev>, <lkp@intel.com>,
	<linux-nfs@vger.kernel.org>, <oliver.sang@intel.com>
Subject: [jlayton:delstid] [nfsd]  48ff66155e:  fsmark.app_overhead 56.0% regression
Date: Tue, 24 Dec 2024 22:35:48 +0800	[thread overview]
Message-ID: <202412242245.dbf6ea4b-lkp@intel.com> (raw)



Hello,

kernel test robot noticed a 56.0% regression of fsmark.app_overhead on:
(but no obvious diff for fsmark.files_per_sec)

commit: 48ff66155e25a3db70cb44b340e1e990349825b8 ("nfsd: implement OPEN_ARGS_SHARE_ACCESS_WANT_OPEN_XOR_DELEGATION")
https://git.kernel.org/cgit/linux/kernel/git/jlayton/linux.git delstid

testcase: fsmark
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory
parameters:

	iterations: 1x
	nr_threads: 1t
	disk: 1HDD
	fs: btrfs
	fs2: nfsv4
	filesize: 4K
	test_size: 40M
	sync_method: fsyncBeforeClose
	nr_files_per_directory: 1fpd
	cpufreq_governor: performance


In addition to that, the commit also has significant impact on the following tests:

+------------------+------------------------------------------------------------------------------------------------+
| testcase: change | fsmark: fsmark.app_overhead 35.9% regression                                                   |
| test machine     | 48 threads 2 sockets Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz (Ivy Bridge-EP) with 64G memory |
| test parameters  | cpufreq_governor=performance                                                                   |
|                  | disk=1SSD                                                                                      |
|                  | filesize=16MB                                                                                  |
|                  | fs2=nfsv4                                                                                      |
|                  | fs=ext4                                                                                        |
|                  | iterations=1x                                                                                  |
|                  | nr_directories=16d                                                                             |
|                  | nr_files_per_directory=256fpd                                                                  |
|                  | nr_threads=32t                                                                                 |
|                  | sync_method=fsyncBeforeClose                                                                   |
|                  | test_size=20G                                                                                  |
+------------------+------------------------------------------------------------------------------------------------+


If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <oliver.sang@intel.com>
| Closes: https://lore.kernel.org/oe-lkp/202412242245.dbf6ea4b-lkp@intel.com


Details are as below:
-------------------------------------------------------------------------------------------------->


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20241224/202412242245.dbf6ea4b-lkp@intel.com

=========================================================================================
compiler/cpufreq_governor/disk/filesize/fs2/fs/iterations/kconfig/nr_files_per_directory/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase:
  gcc-12/performance/1HDD/4K/nfsv4/btrfs/1x/x86_64-rhel-9.4/1fpd/1t/debian-12-x86_64-20240206.cgz/fsyncBeforeClose/lkp-icl-2sp6/40M/fsmark

commit: 
  1436c81cbe ("nfsd: handle delegated timestamps in SETATTR")
  48ff66155e ("nfsd: implement OPEN_ARGS_SHARE_ACCESS_WANT_OPEN_XOR_DELEGATION")

1436c81cbe9bef84 48ff66155e25a3db70cb44b340e 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      2837 ± 10%     +21.1%       3436 ±  9%  numa-meminfo.node1.PageTables
    709.56 ± 10%     +21.1%     859.52 ±  9%  numa-vmstat.node1.nr_page_table_pages
   1551575           +17.7%    1826031 ± 28%  proc-vmstat.pgfault
     33.29            +2.9%      34.26        boot-time.boot
     24.88            +2.0%      25.38        boot-time.dhcp
   3081774           +56.0%    4807945        fsmark.app_overhead
     18.53            +0.2%      18.57        fsmark.files_per_sec
     53434           -17.3%      44176        fsmark.time.voluntary_context_switches
      2827           -11.7%       2496 ± 10%  perf-stat.i.context-switches
      2822           -11.7%       2492 ± 10%  perf-stat.ps.context-switches
      2882           -12.0%       2537 ± 11%  vmstat.system.cs
      4080           -10.7%       3643 ± 13%  vmstat.system.in
      5.82 ±  4%      -0.5        5.30 ±  5%  perf-profile.calltrace.cycles-pp.tick_nohz_handler.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt
      6.83 ±  4%      -0.5        6.33 ±  4%  perf-profile.calltrace.cycles-pp.__hrtimer_run_queues.hrtimer_interrupt.__sysvec_apic_timer_interrupt.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
      1.67 ± 10%      -0.4        1.28 ± 11%  perf-profile.calltrace.cycles-pp.__rpc_execute.rpc_async_schedule.process_one_work.worker_thread.kthread
      1.67 ± 10%      -0.4        1.28 ± 11%  perf-profile.calltrace.cycles-pp.rpc_async_schedule.process_one_work.worker_thread.kthread.ret_from_fork
      0.64 ±  4%      -0.4        0.28 ±100%  perf-profile.calltrace.cycles-pp.sched_balance_find_src_group.sched_balance_rq.sched_balance_domains.handle_softirqs.__irq_exit_rcu
      1.16 ±  8%      -0.3        0.88 ± 18%  perf-profile.calltrace.cycles-pp.sched_balance_domains.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt.asm_sysvec_apic_timer_interrupt
      0.95 ±  9%      -0.2        0.70 ± 20%  perf-profile.calltrace.cycles-pp.sched_balance_rq.sched_balance_domains.handle_softirqs.__irq_exit_rcu.sysvec_apic_timer_interrupt
      0.96 ±  9%      +0.1        1.09 ±  9%  perf-profile.calltrace.cycles-pp.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
      0.94 ±  9%      +0.2        1.09 ±  9%  perf-profile.calltrace.cycles-pp.do_sys_openat2.__x64_sys_openat.do_syscall_64.entry_SYSCALL_64_after_hwframe
      0.79 ±  8%      +0.2        0.96 ± 13%  perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__vfork
      0.79 ±  8%      +0.2        0.96 ± 13%  perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__vfork
      0.78 ±  6%      +0.2        0.95 ± 13%  perf-profile.calltrace.cycles-pp.__x64_sys_vfork.do_syscall_64.entry_SYSCALL_64_after_hwframe.__vfork
      1.92 ± 24%      -0.7        1.20 ± 32%  perf-profile.children.cycles-pp.do_softirq
      7.32 ±  5%      -0.5        6.82 ±  3%  perf-profile.children.cycles-pp.__hrtimer_run_queues
      1.85 ± 14%      -0.5        1.35 ± 18%  perf-profile.children.cycles-pp.ip6_xmit
      1.92 ± 12%      -0.5        1.44 ± 14%  perf-profile.children.cycles-pp.inet6_csk_xmit
      1.75 ± 14%      -0.5        1.28 ± 20%  perf-profile.children.cycles-pp.ip6_finish_output2
      2.64 ±  4%      -0.5        2.18        perf-profile.children.cycles-pp.__rpc_execute
      1.45 ± 12%      -0.4        1.00 ± 19%  perf-profile.children.cycles-pp.__local_bh_enable_ip
      1.58 ± 14%      -0.4        1.16 ±  4%  perf-profile.children.cycles-pp.xprt_request_transmit
      1.59 ± 14%      -0.4        1.16 ±  4%  perf-profile.children.cycles-pp.xprt_transmit
      1.30 ± 15%      -0.4        0.88 ± 16%  perf-profile.children.cycles-pp.__netif_receive_skb_one_core
      1.33 ± 14%      -0.4        0.92 ± 13%  perf-profile.children.cycles-pp.__napi_poll
      1.50 ± 11%      -0.4        1.09 ± 14%  perf-profile.children.cycles-pp.net_rx_action
      1.32 ± 14%      -0.4        0.92 ± 13%  perf-profile.children.cycles-pp.process_backlog
      1.19 ± 14%      -0.4        0.79 ± 13%  perf-profile.children.cycles-pp.ip6_input_finish
      1.18 ± 14%      -0.4        0.78 ± 13%  perf-profile.children.cycles-pp.ip6_protocol_deliver_rcu
      1.59 ± 13%      -0.4        1.20 ±  5%  perf-profile.children.cycles-pp.call_transmit
      1.54 ± 14%      -0.4        1.15 ±  4%  perf-profile.children.cycles-pp.xs_tcp_send_request
      1.67 ± 10%      -0.4        1.28 ± 11%  perf-profile.children.cycles-pp.rpc_async_schedule
      1.10 ± 15%      -0.4        0.74 ± 14%  perf-profile.children.cycles-pp.tcp_v6_rcv
      0.86 ± 22%      -0.3        0.54 ± 12%  perf-profile.children.cycles-pp.tcp_v6_do_rcv
      0.81 ± 26%      -0.3        0.52 ± 10%  perf-profile.children.cycles-pp.tcp_rcv_established
      0.42 ± 19%      -0.2        0.24 ± 17%  perf-profile.children.cycles-pp.___perf_sw_event
      0.20 ± 35%      -0.2        0.03 ±100%  perf-profile.children.cycles-pp.__put_nfs_open_context
      0.34 ± 12%      -0.1        0.23 ± 22%  perf-profile.children.cycles-pp.free_pages_and_swap_cache
      0.35 ± 10%      -0.1        0.24 ± 18%  perf-profile.children.cycles-pp.__tlb_batch_free_encoded_pages
      0.05 ± 76%      +0.1        0.13        perf-profile.children.cycles-pp.ct_nmi_enter
      0.12 ± 36%      +0.1        0.21 ± 28%  perf-profile.children.cycles-pp.autoremove_wake_function
      0.78 ±  6%      +0.2        0.95 ± 13%  perf-profile.children.cycles-pp.__x64_sys_vfork
      1.08 ± 16%      +0.2        1.32 ± 10%  perf-profile.children.cycles-pp.tick_nohz_stop_tick
      1.09 ± 15%      +0.2        1.34 ± 10%  perf-profile.children.cycles-pp.tick_nohz_idle_stop_tick
      1.26 ± 12%      +0.3        1.53 ±  7%  perf-profile.children.cycles-pp.kernel_clone
      2.33 ±  8%      +0.3        2.66 ±  7%  perf-profile.children.cycles-pp.do_filp_open
      2.30 ±  7%      +0.3        2.64 ±  7%  perf-profile.children.cycles-pp.path_openat
      0.30 ± 28%      -0.1        0.16 ± 46%  perf-profile.self.cycles-pp.tick_nohz_handler
      0.12 ± 32%      +0.1        0.20 ± 26%  perf-profile.self.cycles-pp.__memcg_slab_free_hook
      0.05 ± 76%      +0.1        0.13        perf-profile.self.cycles-pp.ct_nmi_enter
      0.36 ± 33%      +0.2        0.54 ± 27%  perf-profile.self.cycles-pp.next_uptodate_folio


***************************************************************************************************
lkp-ivb-2ep2: 48 threads 2 sockets Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz (Ivy Bridge-EP) with 64G memory
=========================================================================================
compiler/cpufreq_governor/disk/filesize/fs2/fs/iterations/kconfig/nr_directories/nr_files_per_directory/nr_threads/rootfs/sync_method/tbox_group/test_size/testcase:
  gcc-12/performance/1SSD/16MB/nfsv4/ext4/1x/x86_64-rhel-9.4/16d/256fpd/32t/debian-12-x86_64-20240206.cgz/fsyncBeforeClose/lkp-ivb-2ep2/20G/fsmark

commit: 
  1436c81cbe ("nfsd: handle delegated timestamps in SETATTR")
  48ff66155e ("nfsd: implement OPEN_ARGS_SHARE_ACCESS_WANT_OPEN_XOR_DELEGATION")

1436c81cbe9bef84 48ff66155e25a3db70cb44b340e 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
 2.472e+08 ±  4%     +35.9%  3.361e+08 ±  4%  fsmark.app_overhead
     16.00            +0.0%      16.00        fsmark.files_per_sec
     65306 ±121%    +168.8%     175532 ±  4%  numa-meminfo.node1.AnonHugePages
    449.72 ±  8%     +24.4%     559.64 ± 16%  sched_debug.cpu.curr->pid.avg
      0.02 ± 37%     -75.3%       0.00 ±101%  perf-sched.sch_delay.avg.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
      0.61 ± 47%     -69.2%       0.19 ± 52%  perf-sched.sch_delay.max.ms.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
      3.57 ± 35%      -1.6        1.98 ± 24%  perf-profile.calltrace.cycles-pp.read
      3.35 ± 38%      -1.4        1.98 ± 24%  perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
      3.35 ± 38%      -1.4        1.98 ± 24%  perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.read
      1.10 ± 72%      +2.0        3.11 ± 14%  perf-profile.calltrace.cycles-pp.__mmap_region.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64
      1.37 ± 83%      +2.3        3.66 ± 38%  perf-profile.calltrace.cycles-pp.do_mmap.vm_mmap_pgoff.ksys_mmap_pgoff.do_syscall_64.entry_SYSCALL_64_after_hwframe
      3.57 ± 35%      -1.6        1.98 ± 24%  perf-profile.children.cycles-pp.read
      1.90 ± 60%      +2.4        4.28 ± 15%  perf-profile.children.cycles-pp.__mmap_region
      1.37 ± 83%      +2.5        3.92 ± 35%  perf-profile.children.cycles-pp.ksys_mmap_pgoff
      2.17 ± 57%      +2.7        4.84 ± 34%  perf-profile.children.cycles-pp.do_mmap
      2.17 ± 57%      +2.9        5.08 ± 33%  perf-profile.children.cycles-pp.vm_mmap_pgoff





Disclaimer:
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.


-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki


                 reply	other threads:[~2024-12-24 14:36 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=202412242245.dbf6ea4b-lkp@intel.com \
    --to=oliver.sang@intel.com \
    --cc=jlayton@kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=lkp@intel.com \
    --cc=oe-lkp@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.