All of lore.kernel.org
 help / color / mirror / Atom feed
* [linus:master] [ext4]  4b41deb896:  stress-ng.fsize.ops_per_sec 23.3% improvement
@ 2025-08-18  2:46 kernel test robot
  0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2025-08-18  2:46 UTC (permalink / raw)
  To: Baokun Li
  Cc: oe-lkp, lkp, linux-kernel, Theodore Ts'o, Jan Kara,
	Ojaswin Mujoo, Zhang Yi, linux-ext4, oliver.sang



Hello,

kernel test robot noticed a 23.3% improvement of stress-ng.fsize.ops_per_sec on:


commit: 4b41deb896e3d0417701759194f0765c06258b9c ("ext4: remove unnecessary s_md_lock on update s_mb_last_group")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master


testcase: stress-ng
config: x86_64-rhel-9.4
compiler: gcc-12
test machine: 192 threads 2 sockets Intel(R) Xeon(R) Platinum 8468V  CPU @ 2.4GHz (Sapphire Rapids) with 384G memory
parameters:

	nr_threads: 100%
	disk: 1SSD
	testtime: 60s
	fs: ext4
	test: fsize
	cpufreq_governor: performance



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


The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20250818/202508181007.a67b7a17-lkp@intel.com

=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/nr_threads/rootfs/tbox_group/test/testcase/testtime:
  gcc-12/performance/1SSD/ext4/x86_64-rhel-9.4/100%/debian-12-x86_64-20240206.cgz/igk-spr-2sp1/fsize/stress-ng/60s

commit: 
  f0374d8071 ("ext4: remove unnecessary s_mb_last_start")
  4b41deb896 ("ext4: remove unnecessary s_md_lock on update s_mb_last_group")

f0374d80711adf86 4b41deb896e3d0417701759194f 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
 3.486e+08 ±  4%     +20.0%  4.183e+08 ±  5%  cpuidle..time
      0.29 ± 32%    +215.9%       0.90 ± 27%  iostat.cpu.iowait
      6783            +5.6%       7160        vmstat.system.cs
    414835            +4.2%     432174        vmstat.system.in
      0.29 ± 33%      +0.6        0.93 ± 28%  mpstat.cpu.all.iowait%
      1.00 ±  5%      -0.1        0.89 ±  4%  mpstat.cpu.all.irq%
      0.04 ±  2%      +0.0        0.05 ±  3%  mpstat.cpu.all.soft%
   9298066 ±  4%     +23.5%   11484309 ±  2%  numa-numastat.node0.local_node
   9407338 ±  4%     +22.9%   11564393 ±  2%  numa-numastat.node0.numa_hit
   9564550 ±  3%     +19.6%   11436032 ±  5%  numa-numastat.node1.local_node
   9653705 ±  3%     +19.7%   11555011 ±  4%  numa-numastat.node1.numa_hit
   9362645 ±  3%     +23.6%   11568925 ±  2%  numa-vmstat.node0.numa_hit
   9253372 ±  4%     +24.2%   11488841 ±  2%  numa-vmstat.node0.numa_local
   9609252 ±  3%     +20.3%   11556685 ±  4%  numa-vmstat.node1.numa_hit
   9520097 ±  3%     +20.1%   11437766 ±  5%  numa-vmstat.node1.numa_local
      3564 ±  7%     +12.7%       4018 ±  6%  perf-c2c.DRAM.local
     31639 ±  7%     +17.7%      37225 ±  6%  perf-c2c.DRAM.remote
     38438 ±  8%     +32.8%      51055 ±  5%  perf-c2c.HITM.local
     17530 ±  8%     +28.1%      22465 ±  6%  perf-c2c.HITM.remote
     55969 ±  8%     +31.4%      73520 ±  5%  perf-c2c.HITM.total
     66700            +6.4%      70950        proc-vmstat.nr_slab_reclaimable
  19021176 ±  3%     +21.6%   23136506        proc-vmstat.numa_hit
  18822748 ±  3%     +21.9%   22937443        proc-vmstat.numa_local
  21083213 ±  3%     +23.5%   26028768        proc-vmstat.pgalloc_normal
  20599860 ±  4%     +24.3%   25614743        proc-vmstat.pgfree
    869734 ±  3%     -31.1%     599551 ± 11%  sched_debug.cpu.avg_idle.avg
    181478 ± 14%     -87.9%      21936 ± 23%  sched_debug.cpu.avg_idle.min
    926.00 ±  3%     +10.2%       1020 ±  3%  sched_debug.cpu.nr_switches.min
     38.67 ± 25%     -32.5%      26.08 ± 18%  sched_debug.cpu.nr_uninterruptible.max
      5.72 ± 15%     -20.4%       4.56 ±  8%  sched_debug.cpu.nr_uninterruptible.stddev
      1600 ±  4%     +23.1%       1970        stress-ng.fsize.SIGXFSZ_signals_per_sec
    409112 ±  4%     +23.3%     504442        stress-ng.fsize.ops
      6819 ±  4%     +23.3%       8408        stress-ng.fsize.ops_per_sec
     15.18 ±  4%     +47.3%      22.37 ±  2%  stress-ng.time.user_time
      4913 ±  3%    +239.6%      16687 ±  9%  stress-ng.time.voluntary_context_switches
      1.35 ±  3%      +6.1%       1.43        perf-stat.i.MPKI
 1.508e+10            -8.2%  1.385e+10 ±  4%  perf-stat.i.branch-instructions
      0.37            +0.1        0.46        perf-stat.i.branch-miss-rate%
  53745812 ±  2%     +13.9%   61238496 ±  3%  perf-stat.i.branch-misses
      6655 ±  2%      +6.1%       7059        perf-stat.i.context-switches
    302.05           +11.2%     335.91        perf-stat.i.cpu-migrations
      0.13            -5.5%       0.12 ±  3%  perf-stat.i.ipc
      1.34 ±  3%      +6.4%       1.42        perf-stat.overall.MPKI
      0.35            +0.1        0.44        perf-stat.overall.branch-miss-rate%
 1.482e+10            -8.3%  1.359e+10 ±  4%  perf-stat.ps.branch-instructions
  52582897 ±  2%     +13.7%   59787848 ±  2%  perf-stat.ps.branch-misses
      6516 ±  2%      +6.0%       6904        perf-stat.ps.context-switches
    295.57           +11.1%     328.32 ±  2%  perf-stat.ps.cpu-migrations
 4.006e+12            -5.9%  3.768e+12 ±  3%  perf-stat.total.instructions
      0.01 ±124%   +1076.9%       0.15 ± 64%  perf-sched.sch_delay.avg.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_dirty_inode.__mark_inode_dirty.generic_update_time
      0.04 ± 77%    +706.0%       0.31 ± 91%  perf-sched.sch_delay.avg.ms.__cond_resched.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64
      3.04 ±  7%     -42.0%       1.77 ± 71%  perf-sched.sch_delay.avg.ms.__cond_resched.shmem_inode_acct_blocks.shmem_alloc_and_add_folio.shmem_get_folio_gfp.shmem_write_begin
      0.16 ± 14%     +68.2%       0.27 ± 15%  perf-sched.sch_delay.avg.ms.exit_to_user_mode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
      0.00 ±103%   +1355.2%       0.07 ± 53%  perf-sched.sch_delay.avg.ms.io_schedule.bit_wait_io.__wait_on_bit_lock.out_of_line_wait_on_bit_lock
      2.99 ± 28%     -25.2%       2.24 ±  7%  perf-sched.sch_delay.avg.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.ext4_mb_initialize_context
      0.31 ±154%    +340.7%       1.39 ± 50%  perf-sched.sch_delay.avg.ms.schedule_timeout.khugepaged_wait_work.khugepaged.kthread
      0.81 ±145%    +164.4%       2.14 ± 66%  perf-sched.sch_delay.max.ms.__cond_resched.__ext4_journal_get_write_access.ext4_mb_mark_context.ext4_mb_clear_bb.ext4_remove_blocks
      0.02 ±138%   +3796.0%       0.65 ±103%  perf-sched.sch_delay.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_dirty_inode.__mark_inode_dirty.generic_update_time
      0.66 ±163%    +348.6%       2.97 ± 44%  perf-sched.sch_delay.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_ext_truncate.ext4_truncate.ext4_setattr
      1.59 ± 92%    +121.2%       3.51 ± 12%  perf-sched.sch_delay.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_truncate.ext4_setattr.notify_change
      0.07 ± 71%   +1499.3%       1.07 ±115%  perf-sched.sch_delay.max.ms.__cond_resched.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64
      3.74 ±  6%     +29.7%       4.85 ± 11%  perf-sched.sch_delay.max.ms.__cond_resched.down_read.ext4_map_blocks.ext4_alloc_file_blocks.isra
      3.16 ±  7%     +35.0%       4.26 ±  5%  perf-sched.sch_delay.max.ms.__cond_resched.down_read.unmap_mapping_range.truncate_pagecache.ext4_setattr
      0.46 ±159%    +331.6%       1.99 ± 53%  perf-sched.sch_delay.max.ms.__cond_resched.ext4_free_blocks.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space
      0.55 ±190%    +351.1%       2.47 ± 66%  perf-sched.sch_delay.max.ms.__cond_resched.kmem_cache_alloc_noprof.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf
      0.24 ± 28%     +63.9%       0.39 ± 24%  perf-sched.sch_delay.max.ms.__cond_resched.smpboot_thread_fn.kthread.ret_from_fork.ret_from_fork_asm
      1.71 ± 21%     +61.7%       2.76 ± 29%  perf-sched.sch_delay.max.ms.do_wait.kernel_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe
      0.01 ±107%   +6151.4%       0.36 ± 34%  perf-sched.sch_delay.max.ms.io_schedule.bit_wait_io.__wait_on_bit_lock.out_of_line_wait_on_bit_lock
      0.69 ±132%    +561.4%       4.55 ± 24%  perf-sched.sch_delay.max.ms.io_schedule.folio_wait_bit_common.__find_get_block_slow.find_get_block_common
      0.31 ±154%    +340.7%       1.39 ± 50%  perf-sched.sch_delay.max.ms.schedule_timeout.khugepaged_wait_work.khugepaged.kthread
      3570 ±  9%     +27.7%       4557 ±  7%  perf-sched.total_wait_and_delay.max.ms
      3570 ±  9%     +27.7%       4557 ±  7%  perf-sched.total_wait_time.max.ms
      0.95 ± 12%    -100.0%       0.00        perf-sched.wait_and_delay.avg.ms.__cond_resched.bdev_getblk.ext4_read_block_bitmap_nowait.ext4_read_block_bitmap.ext4_mb_mark_context
      1226 ±  5%     -79.6%     250.00 ± 19%  perf-sched.wait_and_delay.count.__cond_resched.__ext4_handle_dirty_metadata.ext4_mb_mark_context.ext4_mb_mark_diskspace_used.ext4_mb_new_blocks
    278.17 ± 18%    +333.4%       1205 ± 20%  perf-sched.wait_and_delay.count.__cond_resched.__ext4_mark_inode_dirty.ext4_dirty_inode.__mark_inode_dirty.ext4_setattr
      1465 ±  3%     -92.0%     117.67 ± 71%  perf-sched.wait_and_delay.count.__cond_resched.__find_get_block_slow.find_get_block_common.bdev_getblk.ext4_read_block_bitmap_nowait
    216.83 ±  5%    -100.0%       0.00        perf-sched.wait_and_delay.count.__cond_resched.bdev_getblk.ext4_read_block_bitmap_nowait.ext4_read_block_bitmap.ext4_mb_mark_context
     25.17 ±223%   +2141.1%     564.00 ± 23%  perf-sched.wait_and_delay.count.__cond_resched.down_write.ext4_setattr.notify_change.do_truncate
    271.33 ± 18%    +315.5%       1127 ± 23%  perf-sched.wait_and_delay.count.__cond_resched.down_write.ext4_truncate.ext4_setattr.notify_change
      1778 ±  3%     -77.4%     402.33 ± 10%  perf-sched.wait_and_delay.count.__cond_resched.ext4_mb_regular_allocator.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks
    759.17 ± 22%     +90.0%       1442 ± 17%  perf-sched.wait_and_delay.count.exit_to_user_mode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe.[unknown]
     20.19 ± 85%    +838.0%     189.37 ± 85%  perf-sched.wait_and_delay.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_dirty_inode.__mark_inode_dirty.ext4_setattr
     18.44 ± 46%     -71.0%       5.34 ± 70%  perf-sched.wait_and_delay.max.ms.__cond_resched.__find_get_block_slow.find_get_block_common.bdev_getblk.ext4_read_block_bitmap_nowait
     21.76 ± 64%    -100.0%       0.00        perf-sched.wait_and_delay.max.ms.__cond_resched.bdev_getblk.ext4_read_block_bitmap_nowait.ext4_read_block_bitmap.ext4_mb_mark_context
      7.06 ±223%   +2468.4%     181.45 ± 93%  perf-sched.wait_and_delay.max.ms.__cond_resched.down_write.ext4_setattr.notify_change.do_truncate
      2676 ± 14%     -72.6%     733.87 ±142%  perf-sched.wait_and_delay.max.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64
      0.04 ± 77%    +745.3%       0.33 ± 95%  perf-sched.wait_time.avg.ms.__cond_resched.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64
      0.48 ± 28%    +245.9%       1.65 ± 79%  perf-sched.wait_time.avg.ms.__cond_resched.down_read.ext4_map_blocks.ext4_alloc_file_blocks.isra
      3.75 ± 47%     -37.0%       2.36 ±  5%  perf-sched.wait_time.avg.ms.schedule_preempt_disabled.__mutex_lock.constprop.0.ext4_mb_initialize_context
      0.31 ±154%    +351.8%       1.42 ± 44%  perf-sched.wait_time.avg.ms.schedule_timeout.khugepaged_wait_work.khugepaged.kthread
      1.19 ±104%    +187.2%       3.43 ±  7%  perf-sched.wait_time.max.ms.__cond_resched.__alloc_frozen_pages_noprof.alloc_pages_mpol.folio_alloc_noprof.__filemap_get_folio
      0.81 ±145%    +164.4%       2.14 ± 66%  perf-sched.wait_time.max.ms.__cond_resched.__ext4_journal_get_write_access.ext4_mb_mark_context.ext4_mb_clear_bb.ext4_remove_blocks
     18.34 ± 93%    +932.7%     189.35 ± 85%  perf-sched.wait_time.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_dirty_inode.__mark_inode_dirty.ext4_setattr
      0.15 ±127%    +384.4%       0.71 ± 89%  perf-sched.wait_time.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_dirty_inode.__mark_inode_dirty.generic_update_time
      0.66 ±163%    +348.6%       2.97 ± 44%  perf-sched.wait_time.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_ext_truncate.ext4_truncate.ext4_setattr
      0.99 ±137%    +179.9%       2.78 ± 42%  perf-sched.wait_time.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_setattr.notify_change.do_truncate
      1.59 ± 92%    +121.2%       3.51 ± 12%  perf-sched.wait_time.max.ms.__cond_resched.__ext4_mark_inode_dirty.ext4_truncate.ext4_setattr.notify_change
     17.82 ± 52%     -70.1%       5.33 ± 55%  perf-sched.wait_time.max.ms.__cond_resched.__find_get_block_slow.find_get_block_common.bdev_getblk.ext4_read_block_bitmap_nowait
      0.07 ± 71%   +1499.3%       1.07 ±115%  perf-sched.wait_time.max.ms.__cond_resched.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64
      7.20 ±107%   +2430.6%     182.16 ± 79%  perf-sched.wait_time.max.ms.__cond_resched.down_read.ext4_map_blocks.ext4_alloc_file_blocks.isra
      3.16 ±  7%   +2622.2%      85.90 ±141%  perf-sched.wait_time.max.ms.__cond_resched.down_read.unmap_mapping_range.truncate_pagecache.ext4_setattr
      3.78 ±  6%   +4780.6%     184.48 ± 91%  perf-sched.wait_time.max.ms.__cond_resched.down_write.ext4_map_blocks.ext4_alloc_file_blocks.isra
      0.46 ±159%    +331.6%       1.99 ± 53%  perf-sched.wait_time.max.ms.__cond_resched.ext4_free_blocks.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space
      4.47 ± 36%   +2537.3%     117.99 ±123%  perf-sched.wait_time.max.ms.__cond_resched.ext4_journal_check_start.__ext4_journal_start_sb.ext4_ext_remove_space.ext4_ext_truncate
      0.55 ±190%    +351.1%       2.47 ± 66%  perf-sched.wait_time.max.ms.__cond_resched.kmem_cache_alloc_noprof.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf
      2668 ± 14%     -70.3%     793.19 ±127%  perf-sched.wait_time.max.ms.devkmsg_read.vfs_read.ksys_read.do_syscall_64
      4.94 ±169%   +2462.5%     126.53 ±116%  perf-sched.wait_time.max.ms.io_schedule.folio_wait_bit_common.__find_get_block_slow.find_get_block_common
      0.31 ±154%    +351.8%       1.42 ± 44%  perf-sched.wait_time.max.ms.schedule_timeout.khugepaged_wait_work.khugepaged.kthread
     29.76 ±100%     -18.9       10.89 ±223%  perf-profile.calltrace.cycles-pp.ext4_ext_remove_space.ext4_ext_truncate.ext4_truncate.ext4_setattr.notify_change
     29.65 ±100%     -18.8       10.84 ±223%  perf-profile.calltrace.cycles-pp.ext4_ext_rm_leaf.ext4_ext_remove_space.ext4_ext_truncate.ext4_truncate.ext4_setattr
     29.63 ±100%     -18.8       10.82 ±223%  perf-profile.calltrace.cycles-pp.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space.ext4_ext_truncate.ext4_truncate
     29.62 ±100%     -18.8       10.82 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space.ext4_ext_truncate
     29.92 ±100%     -18.8       11.15 ±223%  perf-profile.calltrace.cycles-pp.ext4_ext_truncate.ext4_truncate.ext4_setattr.notify_change.do_truncate
     30.90 ±100%     -18.4       12.49 ±223%  perf-profile.calltrace.cycles-pp.ext4_truncate.ext4_setattr.notify_change.do_truncate.do_ftruncate
     32.51 ±100%     -17.9       14.63 ±223%  perf-profile.calltrace.cycles-pp.ftruncate64
     32.50 ±100%     -17.9       14.62 ±223%  perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.ftruncate64
     32.50 ±100%     -17.9       14.62 ±223%  perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.ftruncate64
     32.48 ±100%     -17.9       14.62 ±223%  perf-profile.calltrace.cycles-pp.__x64_sys_ftruncate.do_syscall_64.entry_SYSCALL_64_after_hwframe.ftruncate64
     32.48 ±100%     -17.9       14.62 ±223%  perf-profile.calltrace.cycles-pp.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64.entry_SYSCALL_64_after_hwframe
     32.48 ±100%     -17.9       14.62 ±223%  perf-profile.calltrace.cycles-pp.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64.entry_SYSCALL_64_after_hwframe.ftruncate64
     32.47 ±100%     -17.9       14.60 ±223%  perf-profile.calltrace.cycles-pp.notify_change.do_truncate.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate
     32.47 ±100%     -17.9       14.61 ±223%  perf-profile.calltrace.cycles-pp.do_truncate.do_ftruncate.do_sys_ftruncate.__x64_sys_ftruncate.do_syscall_64
     32.46 ±100%     -17.9       14.60 ±223%  perf-profile.calltrace.cycles-pp.ext4_setattr.notify_change.do_truncate.do_ftruncate.do_sys_ftruncate
     16.57 ±100%     -15.6        0.99 ±223%  perf-profile.calltrace.cycles-pp.ext4_map_create_blocks.ext4_map_blocks.ext4_alloc_file_blocks.ext4_do_fallocate.ext4_fallocate
     16.54 ±100%     -15.6        0.98 ±223%  perf-profile.calltrace.cycles-pp.ext4_ext_map_blocks.ext4_map_create_blocks.ext4_map_blocks.ext4_alloc_file_blocks.ext4_do_fallocate
     16.49 ±100%     -15.5        0.96 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks.ext4_map_blocks.ext4_alloc_file_blocks
     16.71 ±100%     -15.5        1.26 ±223%  perf-profile.calltrace.cycles-pp.ext4_map_blocks.ext4_alloc_file_blocks.ext4_do_fallocate.ext4_fallocate.vfs_fallocate
     17.33 ±100%     -15.4        1.96 ±223%  perf-profile.calltrace.cycles-pp.fallocate64
     17.31 ±100%     -15.4        1.95 ±223%  perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.fallocate64
     17.31 ±100%     -15.4        1.95 ±223%  perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.fallocate64
     17.21 ±100%     -15.3        1.86 ±223%  perf-profile.calltrace.cycles-pp.ext4_do_fallocate.ext4_fallocate.vfs_fallocate.__x64_sys_fallocate.do_syscall_64
     17.18 ±100%     -15.3        1.85 ±223%  perf-profile.calltrace.cycles-pp.ext4_alloc_file_blocks.ext4_do_fallocate.ext4_fallocate.vfs_fallocate.__x64_sys_fallocate
     17.23 ±100%     -15.3        1.91 ±223%  perf-profile.calltrace.cycles-pp.__x64_sys_fallocate.do_syscall_64.entry_SYSCALL_64_after_hwframe.fallocate64
     17.23 ±100%     -15.3        1.91 ±223%  perf-profile.calltrace.cycles-pp.vfs_fallocate.__x64_sys_fallocate.do_syscall_64.entry_SYSCALL_64_after_hwframe.fallocate64
     17.23 ±100%     -15.3        1.90 ±223%  perf-profile.calltrace.cycles-pp.ext4_fallocate.vfs_fallocate.__x64_sys_fallocate.do_syscall_64.entry_SYSCALL_64_after_hwframe
     16.94 ±100%     -12.7        4.20 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_free_metadata.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space
     11.97 ±100%     -11.4        0.58 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_regular_allocator.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks.ext4_map_blocks
      9.83 ±100%      -6.8        3.06 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_mark_context.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space
      9.69 ±100%      -6.7        2.98 ±223%  perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_mb_mark_context.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf
      9.67 ±100%      -6.7        2.98 ±223%  perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_mb_mark_context.ext4_mb_clear_bb.ext4_remove_blocks
      8.41 ±100%      -6.3        2.07 ±223%  perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_mb_free_metadata.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf
      8.42 ±100%      -6.3        2.08 ±223%  perf-profile.calltrace.cycles-pp.ext4_try_merge_freed_extent.ext4_mb_free_metadata.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf
      8.40 ±100%      -6.3        2.06 ±223%  perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_try_merge_freed_extent.ext4_mb_free_metadata.ext4_mb_clear_bb.ext4_remove_blocks
      8.39 ±100%      -6.3        2.06 ±223%  perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_mb_free_metadata.ext4_mb_clear_bb.ext4_remove_blocks
      8.36 ±100%      -6.3        2.04 ±223%  perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_try_merge_freed_extent.ext4_mb_free_metadata.ext4_mb_clear_bb
      5.35 ±100%      -5.4        0.00        perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_mb_regular_allocator.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks
      5.33 ±100%      -5.3        0.00        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_mb_regular_allocator.ext4_mb_new_blocks.ext4_ext_map_blocks
      4.46 ±100%      -4.1        0.34 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_mark_diskspace_used.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks.ext4_map_blocks
      4.45 ±100%      -4.1        0.34 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_mark_context.ext4_mb_mark_diskspace_used.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks
      4.24 ±100%      -4.0        0.24 ±223%  perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_mb_mark_context.ext4_mb_mark_diskspace_used.ext4_mb_new_blocks.ext4_ext_map_blocks
      4.23 ±100%      -4.0        0.24 ±223%  perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_mb_mark_context.ext4_mb_mark_diskspace_used.ext4_mb_new_blocks
      3.58 ±100%      -3.4        0.13 ±223%  perf-profile.calltrace.cycles-pp.ext4_mb_complex_scan_group.ext4_mb_regular_allocator.ext4_mb_new_blocks.ext4_ext_map_blocks.ext4_map_create_blocks
      3.31 ±100%      -3.3        0.00        perf-profile.calltrace.cycles-pp.ext4_mb_use_best_found.ext4_mb_complex_scan_group.ext4_mb_regular_allocator.ext4_mb_new_blocks.ext4_ext_map_blocks
      3.28 ±100%      -3.3        0.00        perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_mb_use_best_found.ext4_mb_complex_scan_group.ext4_mb_regular_allocator.ext4_mb_new_blocks
      3.26 ±100%      -3.3        0.00        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_mb_use_best_found.ext4_mb_complex_scan_group.ext4_mb_regular_allocator
      2.66 ±100%      +0.8        3.46 ±223%  perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf
      2.67 ±100%      +0.8        3.46 ±223%  perf-profile.calltrace.cycles-pp._raw_spin_lock.ext4_mb_clear_bb.ext4_remove_blocks.ext4_ext_rm_leaf.ext4_ext_remove_space
     49.87 ±100%     -33.3       16.60 ±223%  perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
     49.86 ±100%     -33.3       16.60 ±223%  perf-profile.children.cycles-pp.do_syscall_64
     44.38 ±100%     -33.1       11.32 ±223%  perf-profile.children.cycles-pp._raw_spin_lock
     44.30 ±100%     -32.5       11.80 ±223%  perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
     29.76 ±100%     -18.9       10.89 ±223%  perf-profile.children.cycles-pp.ext4_ext_remove_space
     29.65 ±100%     -18.8       10.84 ±223%  perf-profile.children.cycles-pp.ext4_ext_rm_leaf
     29.63 ±100%     -18.8       10.82 ±223%  perf-profile.children.cycles-pp.ext4_mb_clear_bb
     29.63 ±100%     -18.8       10.82 ±223%  perf-profile.children.cycles-pp.ext4_remove_blocks
     29.92 ±100%     -18.8       11.16 ±223%  perf-profile.children.cycles-pp.ext4_ext_truncate
     30.90 ±100%     -18.4       12.49 ±223%  perf-profile.children.cycles-pp.ext4_truncate
     32.52 ±100%     -17.9       14.63 ±223%  perf-profile.children.cycles-pp.ftruncate64
     32.48 ±100%     -17.9       14.62 ±223%  perf-profile.children.cycles-pp.__x64_sys_ftruncate
     32.48 ±100%     -17.9       14.62 ±223%  perf-profile.children.cycles-pp.do_sys_ftruncate
     32.48 ±100%     -17.9       14.62 ±223%  perf-profile.children.cycles-pp.do_ftruncate
     32.47 ±100%     -17.9       14.60 ±223%  perf-profile.children.cycles-pp.notify_change
     32.47 ±100%     -17.9       14.61 ±223%  perf-profile.children.cycles-pp.do_truncate
     32.46 ±100%     -17.9       14.60 ±223%  perf-profile.children.cycles-pp.ext4_setattr
     16.57 ±100%     -15.6        0.99 ±223%  perf-profile.children.cycles-pp.ext4_map_create_blocks
     16.49 ±100%     -15.5        0.96 ±223%  perf-profile.children.cycles-pp.ext4_mb_new_blocks
     16.72 ±100%     -15.5        1.26 ±223%  perf-profile.children.cycles-pp.ext4_map_blocks
     16.64 ±100%     -15.4        1.22 ±223%  perf-profile.children.cycles-pp.ext4_ext_map_blocks
     17.33 ±100%     -15.4        1.96 ±223%  perf-profile.children.cycles-pp.fallocate64
     17.21 ±100%     -15.3        1.86 ±223%  perf-profile.children.cycles-pp.ext4_do_fallocate
     17.18 ±100%     -15.3        1.85 ±223%  perf-profile.children.cycles-pp.ext4_alloc_file_blocks
     17.23 ±100%     -15.3        1.91 ±223%  perf-profile.children.cycles-pp.vfs_fallocate
     17.23 ±100%     -15.3        1.91 ±223%  perf-profile.children.cycles-pp.__x64_sys_fallocate
     17.23 ±100%     -15.3        1.90 ±223%  perf-profile.children.cycles-pp.ext4_fallocate
     16.94 ±100%     -12.7        4.20 ±223%  perf-profile.children.cycles-pp.ext4_mb_free_metadata
     11.97 ±100%     -11.4        0.58 ±223%  perf-profile.children.cycles-pp.ext4_mb_regular_allocator
     14.28 ±100%     -10.9        3.40 ±223%  perf-profile.children.cycles-pp.ext4_mb_mark_context
      8.42 ±100%      -6.3        2.08 ±223%  perf-profile.children.cycles-pp.ext4_try_merge_freed_extent
      5.43 ±100%      -5.4        0.03 ±223%  perf-profile.children.cycles-pp.ext4_mb_use_best_found
      4.46 ±100%      -4.1        0.34 ±223%  perf-profile.children.cycles-pp.ext4_mb_mark_diskspace_used
      3.58 ±100%      -3.5        0.13 ±223%  perf-profile.children.cycles-pp.ext4_mb_complex_scan_group
     43.79 ±100%     -32.1       11.69 ±223%  perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath




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


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2025-08-18  2:46 UTC | newest]

Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-18  2:46 [linus:master] [ext4] 4b41deb896: stress-ng.fsize.ops_per_sec 23.3% improvement kernel test robot

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.