LinuxLists.cc - [PATCH] ipc/mqueue: Use get_tree_nodev() in mqueue_get

2022-01-22 02:03:29

Subject: [PATCH] ipc/mqueue: Use get_tree_nodev() in mqueue_get_tree()

When running the stress-ng clone benchmark with multiple testing
threads, it was found that there were significant spinlock contention
in sget_fc(). The contended spinlock was the sb_lock. It is under
heavy contention because the following code in the critcal section
of sget_fc():

hlist_for_each_entry(old, &fc->fs_type->fs_supers, s_instances) {
if (test(old, fc))
goto share_extant_sb;
}

After testing with added instrumentation code, it was found that the
the benchmark could generate thousands of ipc namespaces with the
corresponding number of entries in the mqueue's fs_supers list where
the namespaces are the key for the search. This leads to excessive time
in scanning the list for a match.

Looking back at the mqueue calling sequence leading to sget_fc():

mq_init_ns()
=> mq_create_mount()
=> fc_mount()
=> vfs_get_tree()
=> mqueue_get_tree()
=> get_tree_keyed()
=> vfs_get_super()
=> sget_fc()

Currently, mq_init_ns() is the only mqueue function that will indirectly
call mqueue_get_tree() with a newly allocated ipc namespace as the
key for searching. As a result, there will never be a match with the
exising ipc namespaces stored in the mqueue's fs_supers list.

So using get_tree_keyed() to do an existing ipc namespace search is just
a waste of time. Instead, we could use get_tree_nodev() to eliminate
the useless search. By doing so, we can greatly reduce the sb_lock hold
time and avoid the spinlock contention problem in case a large number
of ipc namespaces are present.

Of course, if the code is modified in the future to allow
mqueue_get_tree() to be called with an existing ipc namespace instead
of a new one, we will have to use get_tree_keyed() in this case.

The following stress-ng clone benchmark command was run on a 2-socket
48-core Intel system:

./stress-ng --clone 32 --verbose --oomable --metrics-brief -t 20

The "bogo ops/s" increased from 5948.45 before patch to 9137.06 after
patch. This is an increase of 54% in performance.

Fixes: 935c6912b198 ("ipc: Convert mqueue fs to fs_context")
Signed-off-by: Waiman Long <[email protected]>
---
ipc/mqueue.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)

diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 5becca9be867..089c34d0732c 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -45,6 +45,7 @@

struct mqueue_fs_context {
struct ipc_namespace *ipc_ns;
+ bool newns; /* Set if newly created ipc namespace */
};

#define MQUEUE_MAGIC 0x19800202
@@ -427,6 +428,14 @@ static int mqueue_get_tree(struct fs_context *fc)
{
struct mqueue_fs_context *ctx = fc->fs_private;

+ /*
+ * With a newly created ipc namespace, we don't need to do a search
+ * for an ipc namespace match, but we still need to set s_fs_info.
+ */
+ if (ctx->newns) {
+ fc->s_fs_info = ctx->ipc_ns;
+ return get_tree_nodev(fc, mqueue_fill_super);
+ }
return get_tree_keyed(fc, mqueue_fill_super, ctx->ipc_ns);
}

@@ -454,6 +463,10 @@ static int mqueue_init_fs_context(struct fs_context *fc)
return 0;
}

+/*
+ * mq_init_ns() is currently the only caller of mq_create_mount().
+ * So the ns parameter is always a newly created ipc namespace.
+ */
static struct vfsmount *mq_create_mount(struct ipc_namespace *ns)
{
struct mqueue_fs_context *ctx;
@@ -465,6 +478,7 @@ static struct vfsmount *mq_create_mount(struct ipc_namespace *ns)
return ERR_CAST(fc);

ctx = fc->fs_private;
+ ctx->newns = true;
put_ipc_ns(ctx->ipc_ns);
ctx->ipc_ns = get_ipc_ns(ns);
put_user_ns(fc->user_ns);
--
2.27.0

2022-02-06 18:53:35

by kernel test robot

[permalink] [raw]

Subject: [ipc/mqueue] d5f04a939d: stress-ng.clone.ops_per_sec 178.9% improvement

Greeting,

FYI, we noticed a

commit: d5f04a939d3a0f900a url: Thanks,
Oliver Sang

178.9% improvement of stress-ng.clone.ops_per_sec due to commit:
6f03f0979826e1cf85e0c7 ("[PATCH] ipc/mqueue: Use get_tree_nodev() in mqueue_get_tree()")
hub.com/0day-ci/linux/commits/Waiman-Long/ipc-mqueue-Use-get_tree_nodev-in-mqueue_get_tree/20220122-012445">https://github.com/0day-ci/linux/commits/Waiman-Long/ipc-mqueue-Use-get_tree_nodev-in-mqueue_get_tree/20220122-012445
t.kernel.org/cgit/linux/kernel/git/torvalds/linux.git">https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git 2c271fe77d52a0555161926c232cd5bc07178b39
s://lore.kernel.org/lkml/20220121172315.19652-1-longman@redhat.com">https://lore.kernel.org/lkml/[email protected]
96 threads 2 sockets Intel(R) Xeon(R) Gold 6252 CPU @ 2.10GHz with 512G memory
------------------------------------------------------------------------>
href="https://github.com/intel/lkp-tests.git">https://github.com/intel/lkp-tests.git
# job file is attached in this email
--compatible job.yaml # generate the yaml file for lkp run
run generated-yaml-file
across any failure that blocks the test,
remove ~/.lkp and /lkp dir to run from a clean state.
===============================================================
ernor/kconfig/nr_threads/rootfs/sc_pid_max/tbox_group/test/testcase/testtime/ucode:
nce/x86_64-rhel-8.3/100%/debian-10.4-x86_64-20200603.cgz/4194304/lkp-csl-2sp7/clone/stress-ng/60s/0x5003102
tag 'gpio-fixes-for-v5.17-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux")
Use get_tree_nodev() in mqueue_get_tree()")
d5f04a939d3a0f900a6f03f0979
---------------------------
%stddev
\
617849 ? 3% stress-ng.clone.ops
9420 ? 3% stress-ng.clone.ops_per_sec
657675 ? 15% stress-ng.time.involuntary_context_switches
5233 ? 9% stress-ng.time.major_page_faults
+16.4% 183379 stress-ng.time.maximum_resident_set_size
+156.1% 12804320 ? 3% stress-ng.time.minor_page_faults
5777 ? 7% stress-ng.time.percent_of_cpu_this_job_got
3906 ? 7% stress-ng.time.system_time
403970 ? 11% stress-ng.time.voluntary_context_switches
+11.3% 437.52 ? 4% pmeter.Average_Active_Power
4383 ? 8% uptime.idle
+57.3% 1.243e+09 ? 28% cpuidle..time
+54.4% 2584371 ? 26% cpuidle..usage
19.80 ? 26% vmstat.cpu.id
+105.0% 15447853 ? 3% vmstat.memory.cache
-85.0% 412.60 ? 41% vmstat.procs.r
43589 ? 8% vmstat.system.cs
24.49 ? 21% mpstat.cpu.all.idle%
1.16 ? 2% mpstat.cpu.all.irq%
0.35 ? 20% mpstat.cpu.all.soft%
73.34 ? 7% mpstat.cpu.all.sys%
0.66 ? 15% mpstat.cpu.all.usr%
+259.3% 66550638 ? 6% numa-numastat.node0.local_node
+257.5% 68128168 ? 6% numa-numastat.node0.numa_hit
+200.5% 1627074 ? 6% numa-numastat.node0.other_node
+193.6% 70903687 ? 9% numa-numastat.node1.local_node
+193.3% 72448658 ? 9% numa-numastat.node1.numa_hit
+182.9% 1633668 ? 8% numa-numastat.node1.other_node
2273 ? 6% turbostat.Avg_MHz
81.53 ? 6% turbostat.Busy%
283587 ?138% turbostat.C1
17.48 ? 30% turbostat.CPU%c1
+9.4% 267.44 ? 2% turbostat.PkgWatt
+35.0% 209.58 ? 2% turbostat.RAMWatt
+19.2% 15613708 ? 2% meminfo.AnonPages
+91.5% 3.008e+09 ? 21% meminfo.Committed_AS
+19.1% 15666947 ? 2% meminfo.Inactive
+19.1% 15666796 ? 2% meminfo.Inactive(anon)
+172.7% 12694048 ? 4% meminfo.KReclaimable
+77.8% 322162 ? 11% meminfo.KernelStack
+101.8% 85160207 ? 3% meminfo.Memused
+88.2% 6488419 ? 21% meminfo.PageTables
+180.1% 13737584 ? 3% meminfo.Percpu
+172.7% 12694048 ? 4% meminfo.SReclaimable
+177.3% 31319072 meminfo.SUnreclaim
+175.9% 44013121 meminfo.Slab
+61.8% 871386 ? 5% meminfo.VmallocUsed
+101.6% 85664789 ? 3% meminfo.max_used_kB
+19.6% 3901214 ? 2% proc-vmstat.nr_anon_pages
5731 proc-vmstat.nr_anon_transparent_hugepages
-8.9% 11008352 proc-vmstat.nr_dirty_background_threshold
-8.9% 22043620 proc-vmstat.nr_dirty_threshold
-8.9% 1.108e+08 proc-vmstat.nr_free_pages
+19.6% 3914490 ? 2% proc-vmstat.nr_inactive_anon
+77.6% 321803 ? 11% proc-vmstat.nr_kernel_stack
+88.5% 1619159 ? 21% proc-vmstat.nr_page_table_pages
+178.5% 3158719 ? 4% proc-vmstat.nr_slab_reclaimable
+183.0% 7794905 proc-vmstat.nr_slab_unreclaimable
+19.6% 3914490 ? 2% proc-vmstat.nr_zone_inactive_anon
+221.2% 1.406e+08 ? 4% proc-vmstat.numa_hit
+222.1% 1.374e+08 ? 4% proc-vmstat.numa_local
+190.9% 3260023 ? 6% proc-vmstat.numa_other
+249.9% 1.602e+08 ? 5% proc-vmstat.pgalloc_normal
+171.3% 25946899 ? 3% proc-vmstat.pgfault
+253.4% 1.452e+08 ? 6% proc-vmstat.pgfree
+167.1% 2320512 ? 3% proc-vmstat.pgreuse
+23.2% 1908459 numa-vmstat.node0.nr_anon_pages
+23.2% 1910626 numa-vmstat.node0.nr_inactive_anon
+90.2% 159196 ? 8% numa-vmstat.node0.nr_kernel_stack
801653 ? 17% numa-vmstat.node0.nr_page_table_pages
+150.7% 1616972 ? 9% numa-vmstat.node0.nr_slab_reclaimable
+167.8% 3952279 ? 4% numa-vmstat.node0.nr_slab_unreclaimable
+23.2% 1910626 numa-vmstat.node0.nr_zone_inactive_anon
+187.9% 36796137 ? 7% numa-vmstat.node0.numa_hit
+188.3% 35903627 ? 7% numa-vmstat.node0.numa_local
915825 ? 7% numa-vmstat.node0.numa_other
+18.7% 1992262 ? 3% numa-vmstat.node1.nr_anon_pages
+18.7% 2003440 ? 3% numa-vmstat.node1.nr_inactive_anon
+74.3% 163512 ? 15% numa-vmstat.node1.nr_kernel_stack
+83.3% 833523 ? 25% numa-vmstat.node1.nr_page_table_pages
+215.6% 1567753 ? 8% numa-vmstat.node1.nr_slab_reclaimable
+203.3% 3896661 ? 6% numa-vmstat.node1.nr_slab_unreclaimable
+18.7% 2003439 ? 3% numa-vmstat.node1.nr_zone_inactive_anon
+140.7% 39442015 ? 10% numa-vmstat.node1.numa_hit
+140.7% 38523963 ? 10% numa-vmstat.node1.numa_local
969097 ? 8% numa-vmstat.node1.numa_other
+21.8% 7629560 numa-meminfo.node0.AnonPages
+30.0% 9649748 numa-meminfo.node0.AnonPages.max
+21.7% 7638275 numa-meminfo.node0.Inactive
+21.7% 7638156 numa-meminfo.node0.Inactive(anon)
+146.7% 6436691 ? 9% numa-meminfo.node0.KReclaimable
+85.2% 158524 ? 8% numa-meminfo.node0.KernelStack
+95.4% 43241986 ? 4% numa-meminfo.node0.MemUsed
+100.4% 3174269 ? 17% numa-meminfo.node0.PageTables
+146.7% 6436691 ? 9% numa-meminfo.node0.SReclaimable
+162.8% 15741566 ? 5% numa-meminfo.node0.SUnreclaim
+157.9% 22178258 ? 6% numa-meminfo.node0.Slab
+17.2% 7967292 ? 3% numa-meminfo.node1.AnonPages
+29.2% 11031952 ? 4% numa-meminfo.node1.AnonPages.max
+17.1% 8011736 ? 3% numa-meminfo.node1.Inactive
+17.1% 8011703 ? 3% numa-meminfo.node1.Inactive(anon)
+209.2% 6237124 ? 8% numa-meminfo.node1.KReclaimable
+72.3% 162789 ? 15% numa-meminfo.node1.KernelStack
+110.4% 41776408 ? 7% numa-meminfo.node1.MemUsed
+79.2% 3298697 ? 25% numa-meminfo.node1.PageTables
+209.2% 6237124 ? 8% numa-meminfo.node1.SReclaimable
+197.0% 15513463 ? 6% numa-meminfo.node1.SUnreclaim
+200.4% 21750587 ? 6% numa-meminfo.node1.Slab
19.31 ? 7% perf-stat.i.MPKI
+63.5% 1.305e+10 ? 3% perf-stat.i.branch-instructions
+77.8% 1.276e+08 ? 2% perf-stat.i.branch-misses
+192.0% 4.269e+08 ? 2% perf-stat.i.cache-misses
+177.8% 1.236e+09 ? 3% perf-stat.i.cache-references
51979 ? 2% perf-stat.i.context-switches
4.37 ? 7% perf-stat.i.cpi
-71.6% 780.61 ? 8% perf-stat.i.cycles-between-cache-misses
0.65 ? 17% perf-stat.i.dTLB-load-miss-rate%
+172.0% 1.023e+08 ? 11% perf-stat.i.dTLB-load-misses
+74.9% 1.629e+10 ? 3% perf-stat.i.dTLB-loads
0.17 ? 11% perf-stat.i.dTLB-store-miss-rate%
+206.9% 12366631 ? 11% perf-stat.i.dTLB-store-misses
+125.8% 6.399e+09 ? 2% perf-stat.i.dTLB-stores
81.12 ? 2% perf-stat.i.iTLB-load-miss-rate%
+85.8% 9692202 ? 4% perf-stat.i.iTLB-load-misses
+32.5% 1797130 ? 9% perf-stat.i.iTLB-loads
+71.2% 6.143e+10 ? 3% perf-stat.i.instructions
7018 ? 8% perf-stat.i.instructions-per-iTLB-miss
+66.6% 235.36 ? 8% perf-stat.i.major-faults
2.37 ? 5% perf-stat.i.metric.GHz
1669 perf-stat.i.metric.K/sec
+96.6% 354.71 ? 6% perf-stat.i.metric.M/sec
+55.4% 440879 ? 4% perf-stat.i.minor-faults
+181.7% 1.004e+08 ? 3% perf-stat.i.node-load-misses
+221.5% 52847803 ? 3% perf-stat.i.node-loads
58.36 perf-stat.i.node-store-miss-rate%
+151.5% 46987417 ? 2% perf-stat.i.node-store-misses
+128.6% 32120053 perf-stat.i.node-stores
+55.4% 441114 ? 4% perf-stat.i.page-faults
20.03 perf-stat.overall.MPKI
0.95 perf-stat.overall.branch-miss-rate%
3.91 ? 2% perf-stat.overall.cpi
-80.3% 557.63 ? 3% perf-stat.overall.cycles-between-cache-misses
0.62 ? 9% perf-stat.overall.dTLB-load-miss-rate%
0.19 ? 8% perf-stat.overall.dTLB-store-miss-rate%
84.37 perf-stat.overall.iTLB-load-miss-rate%
6791 ? 6% perf-stat.overall.instructions-per-iTLB-miss
0.26 ? 2% perf-stat.overall.ipc
64.31 perf-stat.overall.node-load-miss-rate%
59.05 perf-stat.overall.node-store-miss-rate%
+94.1% 1.182e+10 ? 6% perf-stat.ps.branch-instructions
+253.0% 1.117e+08 ? 6% perf-stat.ps.branch-misses
+430.0% 3.903e+08 ? 7% perf-stat.ps.cache-misses
+411.2% 1.113e+09 ? 6% perf-stat.ps.cache-references
43576 ? 8% perf-stat.ps.context-switches
94558 perf-stat.ps.cpu-clock
4381 ? 12% perf-stat.ps.cpu-migrations
+315.1% 92118347 ? 15% perf-stat.ps.dTLB-load-misses
+115.5% 1.473e+10 ? 6% perf-stat.ps.dTLB-loads
+442.9% 11084501 ? 14% perf-stat.ps.dTLB-store-misses
+272.2% 5.721e+09 ? 6% perf-stat.ps.dTLB-stores
+248.7% 8187423 ? 3% perf-stat.ps.iTLB-load-misses
+121.6% 1514271 ? 3% perf-stat.ps.iTLB-loads
+110.3% 5.558e+10 ? 6% perf-stat.ps.instructions
196.80 ? 4% perf-stat.ps.major-faults
372942 ? 4% perf-stat.ps.minor-faults
+391.5% 90779607 ? 6% perf-stat.ps.node-load-misses
+291.3% 50481745 ? 9% perf-stat.ps.node-loads
+430.2% 40902523 ? 4% perf-stat.ps.node-store-misses
+315.2% 28391507 ? 6% perf-stat.ps.node-stores
373138 ? 4% perf-stat.ps.page-faults
94558 perf-stat.ps.task-clock
+143.7% 3.838e+12 ? 6% perf-stat.total.instructions
16884 ? 6% softirqs.CPU0.RCU
9991 ? 6% softirqs.CPU0.SCHED
15671 ? 16% softirqs.CPU1.RCU
14124 ? 9% softirqs.CPU10.RCU
15777 ? 7% softirqs.CPU11.RCU
14485 ? 13% softirqs.CPU12.RCU
13663 ? 8% softirqs.CPU13.RCU
13847 ? 8% softirqs.CPU14.RCU
14682 ? 10% softirqs.CPU15.RCU
14522 ? 12% softirqs.CPU16.RCU
14334 ? 10% softirqs.CPU17.RCU
15925 ? 5% softirqs.CPU18.RCU
15968 ? 19% softirqs.CPU19.RCU
14870 ? 12% softirqs.CPU2.RCU
14742 ? 11% softirqs.CPU20.RCU
15078 ? 7% softirqs.CPU21.RCU
14698 ? 9% softirqs.CPU22.RCU
14659 ? 9% softirqs.CPU23.RCU
14844 ? 13% softirqs.CPU24.RCU
14018 ? 6% softirqs.CPU25.RCU
13935 ? 7% softirqs.CPU26.RCU
14372 ? 10% softirqs.CPU27.RCU
14373 ? 11% softirqs.CPU28.RCU
14727 ? 10% softirqs.CPU29.RCU
14718 ? 8% softirqs.CPU3.RCU
14143 ? 7% softirqs.CPU30.RCU
14390 ? 11% softirqs.CPU31.RCU
13783 ? 10% softirqs.CPU32.RCU
13998 ? 9% softirqs.CPU33.RCU
14052 ? 8% softirqs.CPU34.RCU
14304 ? 6% softirqs.CPU35.RCU
14193 ? 10% softirqs.CPU36.RCU
14084 ? 7% softirqs.CPU37.RCU
14252 ? 10% softirqs.CPU38.RCU
13821 ? 7% softirqs.CPU39.RCU
15083 ? 9% softirqs.CPU4.RCU
13730 ? 8% softirqs.CPU40.RCU
13502 ? 10% softirqs.CPU41.RCU
14212 ? 6% softirqs.CPU42.RCU
14146 ? 6% softirqs.CPU43.RCU
13760 ? 7% softirqs.CPU44.RCU
13953 ? 6% softirqs.CPU45.RCU
14186 ? 7% softirqs.CPU46.RCU
13645 ? 6% softirqs.CPU47.RCU
13857 ? 10% softirqs.CPU48.RCU
15006 ? 6% softirqs.CPU49.RCU
13983 ? 11% softirqs.CPU5.RCU
13877 ? 10% softirqs.CPU50.RCU
14042 ? 5% softirqs.CPU51.RCU
14391 ? 12% softirqs.CPU52.RCU
13586 ? 9% softirqs.CPU53.RCU
13900 ? 11% softirqs.CPU54.RCU
14170 ? 7% softirqs.CPU55.RCU
13948 ? 10% softirqs.CPU56.RCU
14401 ? 7% softirqs.CPU57.RCU
14156 ? 8% softirqs.CPU58.RCU
14695 ? 9% softirqs.CPU59.RCU
14432 ? 7% softirqs.CPU6.RCU
14257 ? 11% softirqs.CPU60.RCU
14234 ? 8% softirqs.CPU61.RCU
13990 ? 11% softirqs.CPU62.RCU
14332 ? 11% softirqs.CPU63.RCU
14814 ? 9% softirqs.CPU64.RCU
14751 ? 9% softirqs.CPU65.RCU
15267 ? 11% softirqs.CPU66.RCU
14539 ? 7% softirqs.CPU67.RCU
15070 ? 12% softirqs.CPU68.RCU
14628 ? 12% softirqs.CPU69.RCU
15117 ? 21% softirqs.CPU7.RCU
14941 ? 9% softirqs.CPU70.RCU
14563 ? 8% softirqs.CPU71.RCU
14089 ? 7% softirqs.CPU72.RCU
15351 ? 21% softirqs.CPU73.RCU
14152 ? 7% softirqs.CPU74.RCU
13937 ? 8% softirqs.CPU75.RCU
13933 ? 7% softirqs.CPU76.RCU
14414 ? 10% softirqs.CPU77.RCU
14509 ? 8% softirqs.CPU78.RCU
14096 ? 8% softirqs.CPU79.RCU
14387 ? 8% softirqs.CPU8.RCU
14736 ? 12% softirqs.CPU80.RCU
13878 ? 7% softirqs.CPU81.RCU
14125 ? 7% softirqs.CPU82.RCU
14172 ? 7% softirqs.CPU83.RCU
14140 ? 9% softirqs.CPU84.RCU
13934 ? 8% softirqs.CPU85.RCU
14808 ? 13% softirqs.CPU86.RCU
13963 ? 7% softirqs.CPU87.RCU
13824 ? 7% softirqs.CPU88.RCU
13758 ? 7% softirqs.CPU89.RCU
15536 ? 14% softirqs.CPU9.RCU
13866 ? 6% softirqs.CPU90.RCU
13816 ? 8% softirqs.CPU91.RCU
13628 ? 7% softirqs.CPU92.RCU
14027 ? 6% softirqs.CPU93.RCU
13863 ? 6% softirqs.CPU94.RCU
10098 ? 5% softirqs.CPU95.RCU
+264.7% 1376204 ? 8% softirqs.RCU
+59.9% 465360 ? 12% softirqs.SCHED
32389 ? 17% softirqs.TIMER
1.31 ? 18% perf-profile.calltrace.cycles-pp.fc_mount.mq_create_mount.mq_init_ns.copy_ipcs.create_new_namespaces
1.31 ? 18% perf-profile.calltrace.cycles-pp.vfs_get_tree.fc_mount.mq_create_mount.mq_init_ns.copy_ipcs
1.31 ? 18% perf-profile.calltrace.cycles-pp.vfs_get_super.vfs_get_tree.fc_mount.mq_create_mount.mq_init_ns
1.31 ? 18% perf-profile.calltrace.cycles-pp.sget_fc.vfs_get_super.vfs_get_tree.fc_mount.mq_create_mount
1.35 ? 17% perf-profile.calltrace.cycles-pp.mq_init_ns.copy_ipcs.create_new_namespaces.unshare_nsproxy_namespaces.ksys_unshare
1.35 ? 17% perf-profile.calltrace.cycles-pp.mq_create_mount.mq_init_ns.copy_ipcs.create_new_namespaces.unshare_nsproxy_namespaces
1.36 ? 17% perf-profile.calltrace.cycles-pp.copy_ipcs.create_new_namespaces.unshare_nsproxy_namespaces.ksys_unshare.__x64_sys_unshare
0.00 perf-profile.calltrace.cycles-pp._raw_spin_lock.sget_fc.vfs_get_super.vfs_get_tree.fc_mount
0.00 perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.sget_fc.vfs_get_super.vfs_get_tree
6.66 ? 21% perf-profile.calltrace.cycles-pp.create_new_namespaces.unshare_nsproxy_namespaces.ksys_unshare.__x64_sys_unshare.do_syscall_64
6.66 ? 21% perf-profile.calltrace.cycles-pp.unshare_nsproxy_namespaces.ksys_unshare.__x64_sys_unshare.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.67 ? 21% perf-profile.calltrace.cycles-pp.__x64_sys_unshare.do_syscall_64.entry_SYSCALL_64_after_hwframe.unshare
6.67 ? 21% perf-profile.calltrace.cycles-pp.ksys_unshare.__x64_sys_unshare.do_syscall_64.entry_SYSCALL_64_after_hwframe.unshare
6.68 ? 21% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.unshare
6.68 ? 21% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.unshare
6.69 ? 21% perf-profile.calltrace.cycles-pp.unshare
0.90 ? 19% perf-profile.calltrace.cycles-pp.process_one_work.worker_thread.kthread.ret_from_fork
0.92 ? 18% perf-profile.calltrace.cycles-pp.worker_thread.kthread.ret_from_fork
0.92 ? 16% perf-profile.calltrace.cycles-pp.rwsem_down_write_slowpath.prealloc_shrinker.alloc_super.sget_fc.vfs_get_super
0.84 ? 22% perf-profile.calltrace.cycles-pp.register_netdev.loopback_net_init.ops_init.setup_net.copy_net_ns
0.86 ? 22% perf-profile.calltrace.cycles-pp.loopback_net_init.ops_init.setup_net.copy_net_ns.create_new_namespaces
0.87 ? 21% perf-profile.calltrace.cycles-pp.__sock_create.inet_ctl_sock_create.tcp_sk_init.ops_init.setup_net
1.16 ? 28% perf-profile.calltrace.cycles-pp.ret_from_fork
1.16 ? 28% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork
1.07 ? 23% perf-profile.calltrace.cycles-pp.inet_ctl_sock_create.icmp_sk_init.ops_init.setup_net.copy_net_ns
1.21 ? 19% perf-profile.calltrace.cycles-pp.prealloc_shrinker.alloc_super.sget_fc.vfs_get_super.vfs_get_tree
1.09 ? 23% perf-profile.calltrace.cycles-pp.icmp_sk_init.ops_init.setup_net.copy_net_ns.create_new_namespaces
1.28 ? 18% perf-profile.calltrace.cycles-pp.alloc_super.sget_fc.vfs_get_super.vfs_get_tree.fc_mount
1.18 ? 21% perf-profile.calltrace.cycles-pp.inet_ctl_sock_create.tcp_sk_init.ops_init.setup_net.copy_net_ns
1.20 ? 21% perf-profile.calltrace.cycles-pp.tcp_sk_init.ops_init.setup_net.copy_net_ns.create_new_namespaces
2.01 ? 22% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.exit_mmap.mmput.do_exit.do_group_exit
2.02 ? 22% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.exit_mmap.mmput.do_exit.__x64_sys_exit
2.16 ? 23% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.__memcg_kmem_charge_page.__alloc_pages.pte_alloc_one.__pte_alloc
2.24 ? 37% perf-profile.calltrace.cycles-pp.do_wait.kernel_wait4.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.24 ? 37% perf-profile.calltrace.cycles-pp.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe.__waitpid
2.24 ? 37% perf-profile.calltrace.cycles-pp.kernel_wait4.__do_sys_wait4.do_syscall_64.entry_SYSCALL_64_after_hwframe.__waitpid
2.25 ? 37% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__waitpid
2.25 ? 37% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__waitpid
2.25 ? 37% perf-profile.calltrace.cycles-pp.__waitpid
2.29 ? 23% perf-profile.calltrace.cycles-pp.__memcg_kmem_charge_page.__alloc_pages.pte_alloc_one.__pte_alloc.copy_pte_range
2.54 ? 37% perf-profile.calltrace.cycles-pp.anon_vma_interval_tree_insert.anon_vma_clone.anon_vma_fork.dup_mmap.dup_mm
2.70 ? 24% perf-profile.calltrace.cycles-pp.__alloc_pages.pte_alloc_one.__pte_alloc.copy_pte_range.copy_page_range
2.76 ? 23% perf-profile.calltrace.cycles-pp.pte_alloc_one.__pte_alloc.copy_pte_range.copy_page_range.dup_mmap
2.84 ? 24% perf-profile.calltrace.cycles-pp.__pte_alloc.copy_pte_range.copy_page_range.dup_mmap.dup_mm
2.85 ? 33% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.__memcg_kmem_uncharge_page.free_pcp_prepare.free_unref_page
2.94 ? 33% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.__memcg_kmem_uncharge_page.free_pcp_prepare.free_unref_page.zap_huge_pmd
2.95 ? 33% perf-profile.calltrace.cycles-pp.__memcg_kmem_uncharge_page.free_pcp_prepare.free_unref_page.zap_huge_pmd.unmap_page_range
2.95 ? 33% perf-profile.calltrace.cycles-pp.free_pcp_prepare.free_unref_page.zap_huge_pmd.unmap_page_range.unmap_vmas
2.95 ? 21% perf-profile.calltrace.cycles-pp.tlb_flush_mmu.tlb_finish_mmu.exit_mmap.mmput.copy_process
2.96 ? 21% perf-profile.calltrace.cycles-pp.tlb_finish_mmu.exit_mmap.mmput.copy_process.kernel_clone
3.05 ? 33% perf-profile.calltrace.cycles-pp.free_unref_page.zap_huge_pmd.unmap_page_range.unmap_vmas.exit_mmap
3.55 ? 27% perf-profile.calltrace.cycles-pp.zap_huge_pmd.unmap_page_range.unmap_vmas.exit_mmap.mmput
3.82 ? 22% perf-profile.calltrace.cycles-pp.obj_cgroup_charge_pages.__memcg_kmem_charge_page.__alloc_pages.pte_alloc_one.copy_huge_pmd
4.00 ? 22% perf-profile.calltrace.cycles-pp.tlb_flush_mmu.tlb_finish_mmu.exit_mmap.mmput.do_exit
4.03 ? 21% perf-profile.calltrace.cycles-pp.__memcg_kmem_charge_page.__alloc_pages.pte_alloc_one.copy_huge_pmd.copy_page_range
4.63 ? 34% perf-profile.calltrace.cycles-pp.anon_vma_clone.anon_vma_fork.dup_mmap.dup_mm.copy_process
4.69 ? 21% perf-profile.calltrace.cycles-pp.release_pages.tlb_flush_mmu.tlb_finish_mmu.exit_mmap.mmput
5.01 ? 19% perf-profile.calltrace.cycles-pp.ops_init.setup_net.copy_net_ns.create_new_namespaces.unshare_nsproxy_namespaces
5.01 ? 19% perf-profile.calltrace.cycles-pp.setup_net.copy_net_ns.create_new_namespaces.unshare_nsproxy_namespaces.ksys_unshare
4.77 ? 21% perf-profile.calltrace.cycles-pp.__alloc_pages.pte_alloc_one.copy_huge_pmd.copy_page_range.dup_mmap
5.02 ? 19% perf-profile.calltrace.cycles-pp.copy_net_ns.create_new_namespaces.unshare_nsproxy_namespaces.ksys_unshare.__x64_sys_unshare
4.87 ? 21% perf-profile.calltrace.cycles-pp.pte_alloc_one.copy_huge_pmd.copy_page_range.dup_mmap.dup_mm
5.25 ? 21% perf-profile.calltrace.cycles-pp.copy_huge_pmd.copy_page_range.dup_mmap.dup_mm.copy_process
5.43 ? 22% perf-profile.calltrace.cycles-pp.page_counter_charge.obj_cgroup_charge_pages.__memcg_kmem_charge_page.__alloc_pages.pte_alloc_one
6.35 ? 24% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.mmput.do_exit.__x64_sys_exit
6.35 ? 24% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.mmput.do_exit.do_group_exit
6.61 ? 34% perf-profile.calltrace.cycles-pp.anon_vma_fork.dup_mmap.dup_mm.copy_process.kernel_clone
7.26 ? 23% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.copy_process.kernel_clone.__do_sys_clone
7.26 ? 23% perf-profile.calltrace.cycles-pp.mmput.copy_process.kernel_clone.__do_sys_clone.do_syscall_64
7.28 ? 23% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.copy_process.kernel_clone.__do_sys_clone3
7.28 ? 23% perf-profile.calltrace.cycles-pp.mmput.copy_process.kernel_clone.__do_sys_clone3.do_syscall_64
7.70 ? 22% perf-profile.calltrace.cycles-pp.page_remove_rmap.zap_pte_range.unmap_page_range.unmap_vmas.exit_mmap
9.48 ? 24% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.do_group_exit.__x64_sys_exit_group
9.50 ? 24% perf-profile.calltrace.cycles-pp.exit_mmap.mmput.do_exit.__x64_sys_exit.do_syscall_64
9.49 ? 24% perf-profile.calltrace.cycles-pp.mmput.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64
9.50 ? 24% perf-profile.calltrace.cycles-pp.mmput.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after_hwframe
9.82 ? 23% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.mmput.copy_process
9.89 ? 23% perf-profile.calltrace.cycles-pp.unmap_vmas.exit_mmap.mmput.copy_process.kernel_clone
12.63 ? 24% perf-profile.calltrace.cycles-pp.unmap_page_range.unmap_vmas.exit_mmap.mmput.do_exit
16.43 ? 25% perf-profile.calltrace.cycles-pp.dup_mmap.dup_mm.copy_process.kernel_clone.__do_sys_clone
16.45 ? 24% perf-profile.calltrace.cycles-pp.dup_mmap.dup_mm.copy_process.kernel_clone.__do_sys_clone3
16.50 ? 24% perf-profile.calltrace.cycles-pp.dup_mm.copy_process.kernel_clone.__do_sys_clone.do_syscall_64
16.53 ? 24% perf-profile.calltrace.cycles-pp.dup_mm.copy_process.kernel_clone.__do_sys_clone3.do_syscall_64
18.22 ? 39% perf-profile.calltrace.cycles-pp.do_exit.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
18.22 ? 39% perf-profile.calltrace.cycles-pp.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
18.22 ? 39% perf-profile.calltrace.cycles-pp.do_group_exit.__x64_sys_exit_group.do_syscall_64.entry_SYSCALL_64_after_hwframe
18.26 ? 40% perf-profile.calltrace.cycles-pp.do_exit.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after_hwframe
18.26 ? 40% perf-profile.calltrace.cycles-pp.__x64_sys_exit.do_syscall_64.entry_SYSCALL_64_after_hwframe
18.74 ? 23% perf-profile.calltrace.cycles-pp.copy_pte_range.copy_page_range.dup_mmap.dup_mm.copy_process
18.55 ? 24% perf-profile.calltrace.cycles-pp.zap_pte_range.unmap_page_range.unmap_vmas.exit_mmap.mmput
24.43 ? 22% perf-profile.calltrace.cycles-pp.copy_page_range.dup_mmap.dup_mm.copy_process.kernel_clone
25.28 ? 24% perf-profile.calltrace.cycles-pp.copy_process.kernel_clone.__do_sys_clone.do_syscall_64.entry_SYSCALL_64_after_hwframe
25.34 ? 24% perf-profile.calltrace.cycles-pp.copy_process.kernel_clone.__do_sys_clone3.do_syscall_64.entry_SYSCALL_64_after_hwframe
25.36 ? 24% perf-profile.calltrace.cycles-pp.kernel_clone.__do_sys_clone.do_syscall_64.entry_SYSCALL_64_after_hwframe.__clone
25.36 ? 24% perf-profile.calltrace.cycles-pp.__do_sys_clone.do_syscall_64.entry_SYSCALL_64_after_hwframe.__clone
25.39 ? 24% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.__clone
25.39 ? 24% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.__clone
25.43 ? 24% perf-profile.calltrace.cycles-pp.kernel_clone.__do_sys_clone3.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall
25.43 ? 24% perf-profile.calltrace.cycles-pp.__do_sys_clone3.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall
25.50 ? 24% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.syscall
25.50 ? 24% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.syscall
25.58 ? 24% perf-profile.calltrace.cycles-pp.__clone
25.65 ? 24% perf-profile.calltrace.cycles-pp.syscall
36.49 ? 39% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe
36.49 ? 39% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.31 ? 18% perf-profile.children.cycles-pp.fc_mount
1.31 ? 18% perf-profile.children.cycles-pp.vfs_get_tree
1.31 ? 18% perf-profile.children.cycles-pp.vfs_get_super
1.31 ? 18% perf-profile.children.cycles-pp.sget_fc
1.35 ? 17% perf-profile.children.cycles-pp.mq_init_ns
1.35 ? 17% perf-profile.children.cycles-pp.mq_create_mount
1.36 ? 17% perf-profile.children.cycles-pp.copy_ipcs
6.66 ? 21% perf-profile.children.cycles-pp.unshare_nsproxy_namespaces
6.67 ? 21% perf-profile.children.cycles-pp.create_new_namespaces
6.67 ? 21% perf-profile.children.cycles-pp.__x64_sys_unshare
6.67 ? 21% perf-profile.children.cycles-pp.ksys_unshare
6.69 ? 21% perf-profile.children.cycles-pp.unshare
16.12 ?118% perf-profile.children.cycles-pp._raw_spin_lock
21.09 ? 83% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
0.07 ? 20% perf-profile.children.cycles-pp.new_inode
0.07 ? 18% perf-profile.children.cycles-pp.inode_init_always
0.07 ? 18% perf-profile.children.cycles-pp.folio_unlock
0.07 ? 18% perf-profile.children.cycles-pp.__vunmap
0.07 ? 21% perf-profile.children.cycles-pp.setns
0.08 ? 24% perf-profile.children.cycles-pp.addrconf_init_net
0.08 ? 27% perf-profile.children.cycles-pp.schedule
0.08 ? 15% perf-profile.children.cycles-pp.free_work
0.36 ? 7% perf-profile.children.cycles-pp.update_process_times
0.08 ? 20% perf-profile.children.cycles-pp.__anon_vma_interval_tree_augment_rotate
0.08 ? 24% perf-profile.children.cycles-pp.proc_pid_make_inode
0.08 ? 22% perf-profile.children.cycles-pp.stress_set_oom_adjustment
0.09 ? 30% perf-profile.children.cycles-pp.__addrconf_sysctl_register
0.08 ? 26% perf-profile.children.cycles-pp.free_unref_page_list
0.08 ? 23% perf-profile.children.cycles-pp.__lookup_slow
0.36 ? 6% perf-profile.children.cycles-pp.tick_sched_handle
0.08 ? 23% perf-profile.children.cycles-pp.copy_creds
0.08 ? 20% perf-profile.children.cycles-pp.find_vma
0.08 ? 9% perf-profile.children.cycles-pp.drain_stock
0.09 ? 27% perf-profile.children.cycles-pp.copy_page
0.09 ? 27% perf-profile.children.cycles-pp.pick_next_task_fair
0.09 ? 4% perf-profile.children.cycles-pp.native_irq_return_iret
0.09 ? 19% perf-profile.children.cycles-pp.fput_many
0.39 ? 6% perf-profile.children.cycles-pp.tick_sched_timer
0.09 ? 22% perf-profile.children.cycles-pp.kmem_cache_alloc_node
0.09 ? 21% perf-profile.children.cycles-pp.rmqueue_bulk
0.10 ? 24% perf-profile.children.cycles-pp.devinet_init_net
0.10 ? 24% perf-profile.children.cycles-pp.__mmdrop
0.10 ? 40% perf-profile.children.cycles-pp.free_p4d_range
0.10 ? 40% perf-profile.children.cycles-pp.free_pgd_range
0.10 ? 13% perf-profile.children.cycles-pp.do_set_pte
0.11 ? 14% perf-profile.children.cycles-pp.refill_stock
0.11 ? 18% perf-profile.children.cycles-pp.inet6_net_init
0.59 perf-profile.children.cycles-pp.hrtimer_interrupt
0.11 ? 24% perf-profile.children.cycles-pp.get_obj_cgroup_from_current
0.45 ? 5% perf-profile.children.cycles-pp.__hrtimer_run_queues
0.12 ? 29% perf-profile.children.cycles-pp.apparmor_sk_alloc_security
0.11 ? 17% perf-profile.children.cycles-pp.pgd_alloc
0.60 perf-profile.children.cycles-pp.__sysvec_apic_timer_interrupt
0.13 ? 20% perf-profile.children.cycles-pp.__rb_erase_color
0.12 ? 26% perf-profile.children.cycles-pp._raw_write_lock_bh
0.13 ? 26% perf-profile.children.cycles-pp.security_sk_alloc
0.12 ? 14% perf-profile.children.cycles-pp.free_pcppages_bulk
0.12 ? 15% perf-profile.children.cycles-pp.mm_init
0.13 ? 32% perf-profile.children.cycles-pp.osq_unlock
0.13 ? 22% perf-profile.children.cycles-pp.find_idlest_group
0.13 ? 19% perf-profile.children.cycles-pp.get_mem_cgroup_from_objcg
0.13 ? 10% perf-profile.children.cycles-pp.preempt_schedule_common
0.13 ? 21% perf-profile.children.cycles-pp.uncharge_batch
0.15 ? 27% perf-profile.children.cycles-pp.insert_header
0.13 ? 23% perf-profile.children.cycles-pp.vma_interval_tree_remove
0.14 ? 21% perf-profile.children.cycles-pp.select_task_rq_fair
0.15 ? 22% perf-profile.children.cycles-pp.inode_init_once
0.14 ? 26% perf-profile.children.cycles-pp.schedule_tail
0.15 ? 22% perf-profile.children.cycles-pp.__mem_cgroup_uncharge_list
0.15 ? 9% perf-profile.children.cycles-pp.walk_component
0.15 ? 11% perf-profile.children.cycles-pp.link_path_walk
0.15 ? 23% perf-profile.children.cycles-pp.__pmd_alloc
0.16 ? 20% perf-profile.children.cycles-pp.__pud_alloc
0.18 ? 24% perf-profile.children.cycles-pp.kmem_cache_alloc_trace
0.17 ? 21% perf-profile.children.cycles-pp.wake_up_new_task
0.17 ? 19% perf-profile.children.cycles-pp.__mod_lruvec_page_state
0.17 ? 19% perf-profile.children.cycles-pp.__folio_memcg_unlock
0.18 ? 28% perf-profile.children.cycles-pp.alloc_vmap_area
0.18 ? 29% perf-profile.children.cycles-pp.__get_vm_area_node
0.18 ? 22% perf-profile.children.cycles-pp.dup_userfaultfd
0.18 ? 23% perf-profile.children.cycles-pp.page_counter_try_charge
0.19 ? 19% perf-profile.children.cycles-pp.wp_page_copy
0.20 ? 16% perf-profile.children.cycles-pp.stress_clone_child
0.21 ? 42% perf-profile.children.cycles-pp.smp_call_function_single
0.23 ? 20% perf-profile.children.cycles-pp.ipv4_mib_init_net
0.21 ? 23% perf-profile.children.cycles-pp.apparmor_socket_post_create
0.21 ? 23% perf-profile.children.cycles-pp.security_socket_post_create
0.21 ? 23% perf-profile.children.cycles-pp.lock_page_memcg
0.21 ? 42% perf-profile.children.cycles-pp.sync_rcu_exp_select_node_cpus
0.22 ? 8% perf-profile.children.cycles-pp.__cond_resched
0.22 ? 14% perf-profile.children.cycles-pp.__slab_free
0.23 ? 27% perf-profile.children.cycles-pp.__vmalloc_node_range
0.24 ? 19% perf-profile.children.cycles-pp.__might_resched
0.23 ? 16% perf-profile.children.cycles-pp.finish_task_switch
0.25 ? 15% perf-profile.children.cycles-pp.put_task_stack
0.25 ? 9% perf-profile.children.cycles-pp.__schedule
0.26 ? 16% perf-profile.children.cycles-pp.__mod_memcg_lruvec_state
0.29 ? 22% perf-profile.children.cycles-pp.allocate_slab
0.29 ? 22% perf-profile.children.cycles-pp.rmqueue
0.30 ? 22% perf-profile.children.cycles-pp.__rb_insert_augmented
0.29 ? 24% perf-profile.children.cycles-pp.expand_shrinker_info
0.28 ? 17% perf-profile.children.cycles-pp.mutex_spin_on_owner
0.28 ? 24% perf-profile.children.cycles-pp.__tlb_remove_page_size
0.31 ? 14% perf-profile.children.cycles-pp.do_filp_open
0.31 ? 14% perf-profile.children.cycles-pp.path_openat
0.34 ? 24% perf-profile.children.cycles-pp.unlink_file_vma
0.33 ? 14% perf-profile.children.cycles-pp.do_sys_open
0.33 ? 14% perf-profile.children.cycles-pp.do_sys_openat2
0.34 ? 15% perf-profile.children.cycles-pp.__list_del_entry_valid
0.30 ? 21% perf-profile.children.cycles-pp.unlock_page_memcg
0.36 ? 16% perf-profile.children.cycles-pp.open64
0.39 ? 23% perf-profile.children.cycles-pp.anon_vma_interval_tree_remove
0.35 ? 24% perf-profile.children.cycles-pp.memcpy_erms
0.40 ? 21% perf-profile.children.cycles-pp.___slab_alloc
0.34 ? 18% perf-profile.children.cycles-pp.next_uptodate_page
0.40 ? 21% perf-profile.children.cycles-pp.__slab_alloc
0.35 ? 21% perf-profile.children.cycles-pp.up_write
0.36 ? 25% perf-profile.children.cycles-pp.inet6_create
0.35 ? 19% perf-profile.children.cycles-pp.__get_free_pages
0.37 ? 16% perf-profile.children.cycles-pp.mod_objcg_state
0.43 ? 26% perf-profile.children.cycles-pp.__register_sysctl_table
0.40 ? 54% perf-profile.children.cycles-pp.irq_exit_rcu
0.50 ? 24% perf-profile.children.cycles-pp.memset_erms
0.42 ? 37% perf-profile.children.cycles-pp.dropmon_net_event
0.46 ? 24% perf-profile.children.cycles-pp.vm_normal_page
0.45 ? 69% perf-profile.children.cycles-pp._raw_spin_lock_irqsave
0.47 ? 23% perf-profile.children.cycles-pp.raw_notifier_call_chain
0.48 ? 33% perf-profile.children.cycles-pp.unregister_netdevice_many
0.48 ? 33% perf-profile.children.cycles-pp.cleanup_net
0.48 ? 33% perf-profile.children.cycles-pp.default_device_exit_batch
1.02 ? 22% perf-profile.children.cycles-pp.sysvec_apic_timer_interrupt
0.56 ? 22% perf-profile.children.cycles-pp.remove_vma
0.53 ? 28% perf-profile.children.cycles-pp.wait_consider_task
0.55 ? 22% perf-profile.children.cycles-pp.sock_alloc_inode
1.12 ? 20% perf-profile.children.cycles-pp.asm_sysvec_apic_timer_interrupt
0.56 ? 23% perf-profile.children.cycles-pp.down_write
0.62 ? 25% perf-profile.children.cycles-pp.pcpu_alloc
0.55 ? 88% perf-profile.children.cycles-pp.rcu_do_batch
0.61 ? 21% perf-profile.children.cycles-pp.try_charge_memcg
0.62 ? 17% perf-profile.children.cycles-pp.filemap_map_pages
0.57 ? 34% perf-profile.children.cycles-pp.__put_anon_vma
0.64 ? 18% perf-profile.children.cycles-pp.do_fault
0.58 ? 83% perf-profile.children.cycles-pp.rcu_core
0.90 ? 19% perf-profile.children.cycles-pp.process_one_work
0.63 ? 22% perf-profile.children.cycles-pp.sock_alloc
0.63 ? 22% perf-profile.children.cycles-pp.raw_unhash_sk
0.92 ? 18% perf-profile.children.cycles-pp.worker_thread
0.62 ? 81% perf-profile.children.cycles-pp.__softirqentry_text_start
0.67 ? 22% perf-profile.children.cycles-pp.alloc_inode
0.70 ? 23% perf-profile.children.cycles-pp.folio_memcg_lock
0.69 ? 22% perf-profile.children.cycles-pp.new_inode_pseudo
0.67 ? 22% perf-profile.children.cycles-pp.dup_task_struct
0.69 ? 22% perf-profile.children.cycles-pp.raw_hash_sk
0.71 ? 23% perf-profile.children.cycles-pp.icmpv6_sk_init
0.73 ? 24% perf-profile.children.cycles-pp.sk_prot_alloc
0.76 ? 22% perf-profile.children.cycles-pp.vm_area_dup
0.75 ? 24% perf-profile.children.cycles-pp.sk_alloc
0.78 ? 38% perf-profile.children.cycles-pp.release_task
0.78 ? 38% perf-profile.children.cycles-pp.wait_task_zombie
0.84 ? 22% perf-profile.children.cycles-pp.register_netdev
1.16 ? 28% perf-profile.children.cycles-pp.kthread
0.86 ? 22% perf-profile.children.cycles-pp.loopback_net_init
0.85 ? 44% perf-profile.children.cycles-pp.queued_read_lock_slowpath
0.96 ? 21% perf-profile.children.cycles-pp.prep_new_page
0.97 ? 21% perf-profile.children.cycles-pp.clear_page_erms
0.95 ? 18% perf-profile.children.cycles-pp.__handle_mm_fault
0.97 ? 21% perf-profile.children.cycles-pp.kernel_init_free_pages
0.91 ? 26% perf-profile.children.cycles-pp.__mutex_lock
1.00 ? 18% perf-profile.children.cycles-pp.handle_mm_fault
1.21 ? 19% perf-profile.children.cycles-pp.prealloc_shrinker
1.33 ? 21% perf-profile.children.cycles-pp.ret_from_fork
1.05 ? 9% perf-profile.children.cycles-pp.kmem_cache_free
1.28 ? 18% perf-profile.children.cycles-pp.alloc_super
1.15 ? 18% perf-profile.children.cycles-pp.do_user_addr_fault
1.09 ? 23% perf-profile.children.cycles-pp.icmp_sk_init
1.15 ? 18% perf-profile.children.cycles-pp.exc_page_fault
1.10 ? 22% perf-profile.children.cycles-pp.inet_create
1.26 ? 17% perf-profile.children.cycles-pp.asm_exc_page_fault
1.20 ? 21% perf-profile.children.cycles-pp.tcp_sk_init
1.29 ? 21% perf-profile.children.cycles-pp.get_page_from_freelist
1.46 ? 23% perf-profile.children.cycles-pp.obj_cgroup_charge
1.97 ? 38% perf-profile.children.cycles-pp.rwsem_spin_on_owner
2.22 ? 25% perf-profile.children.cycles-pp.free_swap_cache
2.36 ? 23% perf-profile.children.cycles-pp.__sock_create
2.29 ? 25% perf-profile.children.cycles-pp.free_pages_and_swap_cache
2.25 ? 37% perf-profile.children.cycles-pp.do_wait
2.25 ? 37% perf-profile.children.cycles-pp.__waitpid
2.25 ? 37% perf-profile.children.cycles-pp.__do_sys_wait4
2.25 ? 37% perf-profile.children.cycles-pp.kernel_wait4
2.47 ? 22% perf-profile.children.cycles-pp.kmem_cache_alloc
2.56 ? 37% perf-profile.children.cycles-pp.anon_vma_interval_tree_insert
2.77 ? 30% perf-profile.children.cycles-pp.unlink_anon_vmas
2.84 ? 24% perf-profile.children.cycles-pp.__pte_alloc
3.00 ? 22% perf-profile.children.cycles-pp.inet_ctl_sock_create
2.85 ? 23% perf-profile.children.cycles-pp.propagate_protected_usage
3.24 ? 32% perf-profile.children.cycles-pp.osq_lock
2.99 ? 18% perf-profile.children.cycles-pp.page_counter_cancel
3.28 ? 29% perf-profile.children.cycles-pp.free_pgtables
3.21 ? 21% perf-profile.children.cycles-pp.free_pcp_prepare
3.36 ? 21% perf-profile.children.cycles-pp.free_unref_page
3.41 ? 21% perf-profile.children.cycles-pp.__memcg_kmem_uncharge_page
3.65 ? 21% perf-profile.children.cycles-pp.zap_huge_pmd
3.98 ? 18% perf-profile.children.cycles-pp.obj_cgroup_uncharge_pages
4.06 ? 18% perf-profile.children.cycles-pp.page_counter_uncharge
4.64 ? 34% perf-profile.children.cycles-pp.anon_vma_clone
4.95 ? 41% perf-profile.children.cycles-pp.rwsem_down_write_slowpath
5.01 ? 19% perf-profile.children.cycles-pp.ops_init
5.01 ? 19% perf-profile.children.cycles-pp.setup_net
4.80 ? 21% perf-profile.children.cycles-pp.release_pages
5.02 ? 19% perf-profile.children.cycles-pp.copy_net_ns
5.26 ? 21% perf-profile.children.cycles-pp.copy_huge_pmd
5.31 ? 35% perf-profile.children.cycles-pp.queued_write_lock_slowpath
6.61 ? 34% perf-profile.children.cycles-pp.anon_vma_fork
6.99 ? 22% perf-profile.children.cycles-pp.tlb_finish_mmu
6.94 ? 22% perf-profile.children.cycles-pp.__memcg_kmem_charge_page
7.11 ? 23% perf-profile.children.cycles-pp.tlb_flush_mmu
7.33 ? 22% perf-profile.children.cycles-pp.page_counter_charge
7.65 ? 22% perf-profile.children.cycles-pp.pte_alloc_one
7.93 ? 22% perf-profile.children.cycles-pp.page_remove_rmap
8.08 ? 22% perf-profile.children.cycles-pp.__alloc_pages
8.07 ? 22% perf-profile.children.cycles-pp.obj_cgroup_charge_pages
18.24 ? 39% perf-profile.children.cycles-pp.__x64_sys_exit_group
18.24 ? 39% perf-profile.children.cycles-pp.do_group_exit
18.28 ? 39% perf-profile.children.cycles-pp.__x64_sys_exit
18.67 ? 24% perf-profile.children.cycles-pp.zap_pte_range
18.77 ? 23% perf-profile.children.cycles-pp.copy_pte_range
22.45 ? 23% perf-profile.children.cycles-pp.unmap_page_range
22.59 ? 23% perf-profile.children.cycles-pp.unmap_vmas
24.43 ? 22% perf-profile.children.cycles-pp.copy_page_range
25.37 ? 24% perf-profile.children.cycles-pp.__do_sys_clone
25.43 ? 24% perf-profile.children.cycles-pp.__do_sys_clone3
25.59 ? 24% perf-profile.children.cycles-pp.__clone
25.67 ? 24% perf-profile.children.cycles-pp.syscall
32.88 ? 24% perf-profile.children.cycles-pp.dup_mmap
33.03 ? 24% perf-profile.children.cycles-pp.dup_mm
33.52 ? 23% perf-profile.children.cycles-pp.exit_mmap
33.54 ? 23% perf-profile.children.cycles-pp.mmput
36.52 ? 39% perf-profile.children.cycles-pp.do_exit
50.63 ? 24% perf-profile.children.cycles-pp.copy_process
50.80 ? 24% perf-profile.children.cycles-pp.kernel_clone
20.95 ? 83% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
0.07 ? 20% perf-profile.self.cycles-pp.folio_unlock
0.08 ? 24% perf-profile.self.cycles-pp.unmap_vmas
0.08 ? 22% perf-profile.self.cycles-pp.__alloc_pages
0.08 ? 19% perf-profile.self.cycles-pp.__anon_vma_interval_tree_augment_rotate
0.08 ? 24% perf-profile.self.cycles-pp.do_wait
0.08 ? 22% perf-profile.self.cycles-pp.unmap_page_range
0.08 ? 22% perf-profile.self.cycles-pp.copy_page_range
0.09 ? 21% perf-profile.self.cycles-pp.___slab_alloc
0.09 ? 27% perf-profile.self.cycles-pp.copy_page
0.09 ? 22% perf-profile.self.cycles-pp.page_counter_try_charge
0.09 ? 18% perf-profile.self.cycles-pp.fput_many
0.09 ? 4% perf-profile.self.cycles-pp.native_irq_return_iret
0.09 ? 15% perf-profile.self.cycles-pp.zap_huge_pmd
0.10 ? 20% perf-profile.self.cycles-pp.get_obj_cgroup_from_current
0.10 ? 24% perf-profile.self.cycles-pp.anon_vma_fork
0.11 ? 22% perf-profile.self.cycles-pp.free_pages_and_swap_cache
0.11 ? 20% perf-profile.self.cycles-pp.__rb_erase_color
0.11 ? 22% perf-profile.self.cycles-pp.find_idlest_group
0.12 ? 26% perf-profile.self.cycles-pp._raw_write_lock_bh
0.12 ? 18% perf-profile.self.cycles-pp.get_mem_cgroup_from_objcg
0.12 ? 20% perf-profile.self.cycles-pp.__folio_memcg_unlock
0.13 ? 32% perf-profile.self.cycles-pp.osq_unlock
0.14 ? 22% perf-profile.self.cycles-pp.inode_init_once
0.13 ? 23% perf-profile.self.cycles-pp.vma_interval_tree_remove
0.14 ? 28% perf-profile.self.cycles-pp.obj_cgroup_charge_pages
0.14 ? 13% perf-profile.self.cycles-pp._raw_spin_lock_irqsave
0.16 ? 23% perf-profile.self.cycles-pp.unlink_anon_vmas
0.17 ? 18% perf-profile.self.cycles-pp.mod_objcg_state
0.17 ? 24% perf-profile.self.cycles-pp.lock_page_memcg
0.18 ? 22% perf-profile.self.cycles-pp.dup_userfaultfd
0.18 ? 22% perf-profile.self.cycles-pp.__tlb_remove_page_size
0.19 ? 44% perf-profile.self.cycles-pp.smp_call_function_single
0.20 ? 21% perf-profile.self.cycles-pp.copy_huge_pmd
0.22 ? 37% perf-profile.self.cycles-pp.rwsem_down_write_slowpath
0.21 ? 25% perf-profile.self.cycles-pp.apparmor_socket_post_create
0.22 ? 14% perf-profile.self.cycles-pp.__slab_free
0.23 ? 18% perf-profile.self.cycles-pp.__might_resched
0.27 ? 18% perf-profile.self.cycles-pp.kmem_cache_free
0.24 ? 20% perf-profile.self.cycles-pp.dup_mmap
0.24 ? 24% perf-profile.self.cycles-pp.__memcg_kmem_charge_page
0.25 ? 16% perf-profile.self.cycles-pp.__mod_memcg_lruvec_state
0.26 ? 24% perf-profile.self.cycles-pp.vm_area_dup
0.28 ? 23% perf-profile.self.cycles-pp.__rb_insert_augmented
0.27 ? 17% perf-profile.self.cycles-pp.mutex_spin_on_owner
0.34 ? 16% perf-profile.self.cycles-pp.__list_del_entry_valid
0.30 ? 21% perf-profile.self.cycles-pp.unlock_page_memcg
0.33 ? 20% perf-profile.self.cycles-pp.try_charge_memcg
0.35 ? 21% perf-profile.self.cycles-pp.kmem_cache_alloc
0.38 ? 23% perf-profile.self.cycles-pp.anon_vma_interval_tree_remove
0.35 ? 24% perf-profile.self.cycles-pp.memcpy_erms
0.34 ? 18% perf-profile.self.cycles-pp.next_uptodate_page
0.34 ? 22% perf-profile.self.cycles-pp.up_write
0.36 ? 25% perf-profile.self.cycles-pp.anon_vma_clone
0.41 ? 23% perf-profile.self.cycles-pp.vm_normal_page
0.49 ? 24% perf-profile.self.cycles-pp.memset_erms
0.41 ? 37% perf-profile.self.cycles-pp.dropmon_net_event
0.49 ? 24% perf-profile.self.cycles-pp.down_write
0.52 ? 28% perf-profile.self.cycles-pp.wait_consider_task
0.64 ? 19% perf-profile.self.cycles-pp._raw_spin_lock
0.69 ? 23% perf-profile.self.cycles-pp.folio_memcg_lock
0.73 ? 25% perf-profile.self.cycles-pp.queued_write_lock_slowpath
0.96 ? 21% perf-profile.self.cycles-pp.clear_page_erms
1.95 ? 38% perf-profile.self.cycles-pp.rwsem_spin_on_owner
2.16 ? 25% perf-profile.self.cycles-pp.free_swap_cache
2.54 ? 37% perf-profile.self.cycles-pp.anon_vma_interval_tree_insert
2.83 ? 23% perf-profile.self.cycles-pp.propagate_protected_usage
3.21 ? 33% perf-profile.self.cycles-pp.osq_lock
2.96 ? 18% perf-profile.self.cycles-pp.page_counter_cancel
4.51 ? 21% perf-profile.self.cycles-pp.release_pages
5.49 ? 22% perf-profile.self.cycles-pp.page_counter_charge
6.96 ? 22% perf-profile.self.cycles-pp.page_remove_rmap
9.75 ? 25% perf-profile.self.cycles-pp.zap_pte_range
15.38 ? 23% perf-profile.self.cycles-pp.copy_pte_range
estimated based on internal Intel analysis and are provided
purposes only. Any difference in system hardware or software
may affect actual performance.
Open Source Technology Center
.org/hyperkitty/list/lkp@lists.01.org">https://lists.01.org/hyperkitty/list/[email protected] Intel Corporation

Attachments:

(No filename) (65.38 kB)
config-5.16.0-11414-gd5f04a939d3a (177.28 kB)
job-script (8.31 kB)
job.yaml (5.63 kB)
reproduce (393.00 B)
Download all attachments