[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20211105064159.GB17949@xsang-OptiPlex-9020>
Date: Fri, 5 Nov 2021 14:41:59 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Mickaël Salaün <mic@...ikod.net>
Cc: lkp@...ts.01.org, lkp@...el.com, ying.huang@...el.com,
feng.tang@...el.com, zhengjun.xing@...ux.intel.com,
fengwei.yin@...el.com, Al Viro <viro@...iv.linux.org.uk>,
Andrew Morton <akpm@...ux-foundation.org>,
Mickaël Salaün <mic@...ikod.net>,
Aleksa Sarai <cyphar@...har.com>, Andy Lutomirski <luto@...nel.org>,
Arnd Bergmann <arnd@...db.de>,
Casey Schaufler <casey@...aufler-ca.com>,
Christian Brauner <christian.brauner@...ntu.com>,
Christian Heimes <christian@...hon.org>,
Deven Bowers <deven.desai@...ux.microsoft.com>,
Dmitry Vyukov <dvyukov@...gle.com>,
Eric Biggers <ebiggers@...nel.org>,
Eric Chiang <ericchiang@...gle.com>,
Florian Weimer <fweimer@...hat.com>,
Geert Uytterhoeven <geert@...ux-m68k.org>,
James Morris <jmorris@...ei.org>, Jan Kara <jack@...e.cz>,
Jann Horn <jannh@...gle.com>, Jonathan Corbet <corbet@....net>,
Kees Cook <keescook@...omium.org>,
Lakshmi Ramasubramanian <nramas@...ux.microsoft.com>,
"Madhavan T . Venkataraman" <madvenka@...ux.microsoft.com>,
Matthew Garrett <mjg59@...gle.com>,
Matthew Wilcox <willy@...radead.org>,
Miklos Szeredi <mszeredi@...hat.com>,
Mimi Zohar <zohar@...ux.ibm.com>, Paul Moore <paul@...l-moore.com>,
Philippe Trébuchet <philippe.trebuchet@....gouv.fr>,
Scott Shell <scottsh@...rosoft.com>, Shuah Khan <shuah@...nel.org>,
Steve Dower <steve.dower@...hon.org>,
Steve Grubb <sgrubb@...hat.com>,
Thibaut Sautereau <thibaut.sautereau@....gouv.fr>,
Vincent Strubel <vincent.strubel@....gouv.fr>,
kernel-hardening@...ts.openwall.com, linux-api@...r.kernel.org,
linux-fsdevel@...r.kernel.org, linux-integrity@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-security-module@...r.kernel.org,
Mickaël Salaün <mic@...ux.microsoft.com>
Subject: [fs] a0918006f9: netperf.Throughput_tps -11.6% regression
Greeting,
FYI, we noticed a -11.6% regression of netperf.Throughput_tps due to commit:
commit: a0918006f9284b77397ae4f163f055c3e0f987b2 ("[PATCH v15 1/3] fs: Add trusted_for(2) syscall implementation and related sysctl")
url: https://github.com/0day-ci/linux/commits/Micka-l-Sala-n/Add-trusted_for-2-was-O_MAYEXEC/20211013-032533
patch link: https://lore.kernel.org/kernel-hardening/20211012192410.2356090-2-mic@digikod.net
in testcase: netperf
on test machine: 192 threads 4 sockets Intel(R) Xeon(R) Platinum 9242 CPU @ 2.30GHz with 192G memory
with following parameters:
ip: ipv4
runtime: 300s
nr_threads: 16
cluster: cs-localhost
test: TCP_CRR
cpufreq_governor: performance
ucode: 0x5003006
test-description: Netperf is a benchmark that can be use to measure various aspect of networking performance.
test-url: http://www.netperf.org/netperf/
please be noted we made out some further analysis/tests, as Fengwei mentioned:
==============================================================================
Here is my investigation result of this regression:
If I add patch to make sure the kernel function address and data address is
almost same even with this patch, there is almost no performance delta(0.1%)
w/o the patch.
And if I only make sure function address same w/o the patch, the performance
delta is about 5.1%.
So suppose this regression is triggered by different function and data address.
We don't know why the different address could bring such kind of regression yet
===============================================================================
we also tested on other platforms.
on a Cooper Lake (Intel(R) Xeon(R) Gold 5318H CPU @ 2.50GHz with 128G memory),
we also observed regression but the gap is smaller:
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase/ucode:
cs-localhost/gcc-9/performance/ipv4/x86_64-rhel-8.3/16/debian-10.4-x86_64-20200603.cgz/300s/lkp-cpl-4sp1/TCP_CRR/netperf/0x700001e
commit:
v5.15-rc4
a0918006f9284b77397ae4f163f055c3e0f987b2
v5.15-rc4 a0918006f9284b77397ae4f163f
---------------- ---------------------------
%stddev %change %stddev
\ | \
333492 -5.7% 314346 ± 2% netperf.Throughput_total_tps
20843 -4.5% 19896 netperf.Throughput_tps
but no regression on a 96 threads 2 sockets Ice Lake with 256G memory:
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase/ucode:
cs-localhost/gcc-9/performance/ipv4/x86_64-rhel-8.3/16/debian-10.4-x86_64-20200603.cgz/300s/lkp-icl-2sp1/TCP_CRR/netperf/0xb000280
commit:
v5.15-rc4
a0918006f9284b77397ae4f163f055c3e0f987b2
v5.15-rc4 a0918006f9284b77397ae4f163f
---------------- ---------------------------
%stddev %change %stddev
\ | \
555600 -0.1% 555305 netperf.Throughput_total_tps
34725 -0.1% 34706 netperf.Throughput_tps
Fengwei also helped review these results and commented:
I suppose these three CPUs have different cache policy. It also could be
related with netperf throughput testing.
If you fix the issue, kindly add following tag
Reported-by: kernel test robot <oliver.sang@...el.com>
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
sudo bin/lkp install job.yaml # job file is attached in this email
bin/lkp split-job --compatible job.yaml # generate the yaml file for lkp run
sudo bin/lkp run generated-yaml-file
# if come across any failure that blocks the test,
# please remove ~/.lkp and /lkp dir to run from a clean state.
=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase/ucode:
cs-localhost/gcc-9/performance/ipv4/x86_64-rhel-8.3/16/debian-10.4-x86_64-20200603.cgz/300s/lkp-csl-2ap3/TCP_CRR/netperf/0x5003006
commit:
v5.15-rc4
a0918006f9 ("fs: Add trusted_for(2) syscall implementation and related sysctl")
v5.15-rc4 a0918006f9284b77397ae4f163f
---------------- ---------------------------
%stddev %change %stddev
\ | \
354692 -11.6% 313620 netperf.Throughput_total_tps
22168 -11.6% 19601 netperf.Throughput_tps
2.075e+08 -11.6% 1.834e+08 netperf.time.voluntary_context_switches
1.064e+08 -11.6% 94086163 netperf.workload
0.27 ± 35% -0.1 0.22 ± 2% mpstat.cpu.all.usr%
2207583 -6.3% 2068413 vmstat.system.cs
3029480 ± 6% -23.3% 2324079 ± 7% interrupts.CAL:Function_call_interrupts
13768 ± 25% -35.6% 8872 ± 23% interrupts.CPU30.CAL:Function_call_interrupts
2014617 ± 16% -26.3% 1485200 ± 24% softirqs.CPU180.NET_RX
3.268e+08 -12.1% 2.874e+08 softirqs.NET_RX
287881 ± 2% +24.6% 358692 softirqs.TIMER
3207001 -9.6% 2899010 perf-sched.wait_and_delay.count.schedule_timeout.inet_csk_accept.inet_accept.do_accept
0.01 ± 15% +67.1% 0.01 ± 9% perf-sched.wait_time.avg.ms.preempt_schedule_common.__cond_resched.__release_sock.release_sock.sk_wait_data
0.02 ± 2% +23.3% 0.03 ± 21% perf-sched.wait_time.avg.ms.preempt_schedule_common.__cond_resched.aa_sk_perm.security_socket_accept.do_accept
0.01 +20.0% 0.01 perf-sched.wait_time.avg.ms.schedule_timeout.wait_woken.sk_wait_data.tcp_recvmsg_locked
63320 ± 2% -10.6% 56615 ± 2% slabinfo.sock_inode_cache.active_objs
1626 ± 2% -10.6% 1454 ± 2% slabinfo.sock_inode_cache.active_slabs
63445 ± 2% -10.6% 56722 ± 2% slabinfo.sock_inode_cache.num_objs
1626 ± 2% -10.6% 1454 ± 2% slabinfo.sock_inode_cache.num_slabs
49195 -3.2% 47624 proc-vmstat.nr_slab_reclaimable
4278441 -6.6% 3996109 proc-vmstat.numa_hit
4052317 ± 2% -7.4% 3751341 proc-vmstat.numa_local
4285136 -6.5% 4006356 proc-vmstat.pgalloc_normal
1704913 -11.4% 1511123 proc-vmstat.pgfree
9.382e+09 -10.1% 8.438e+09 perf-stat.i.branch-instructions
1.391e+08 -10.0% 1.252e+08 perf-stat.i.branch-misses
13.98 +2.2 16.20 perf-stat.i.cache-miss-rate%
87082775 +14.0% 99273064 perf-stat.i.cache-misses
2231661 -6.4% 2088571 perf-stat.i.context-switches
1.65 +8.6% 1.79 perf-stat.i.cpi
7.603e+10 -2.1% 7.441e+10 perf-stat.i.cpu-cycles
907.53 ± 2% -13.0% 789.92 ± 2% perf-stat.i.cycles-between-cache-misses
920324 ± 19% -20.3% 733572 ± 5% perf-stat.i.dTLB-load-misses
1.417e+10 -10.3% 1.271e+10 perf-stat.i.dTLB-loads
182445 ± 16% -57.6% 77419 ± 9% perf-stat.i.dTLB-store-misses
8.254e+09 -10.3% 7.403e+09 perf-stat.i.dTLB-stores
88.23 -1.7 86.52 perf-stat.i.iTLB-load-miss-rate%
96633753 -11.0% 85983323 perf-stat.i.iTLB-load-misses
12277057 +4.0% 12766535 perf-stat.i.iTLB-loads
4.741e+10 -10.2% 4.259e+10 perf-stat.i.instructions
0.62 -8.2% 0.57 perf-stat.i.ipc
0.40 -2.1% 0.39 perf-stat.i.metric.GHz
168.88 -10.1% 151.87 perf-stat.i.metric.M/sec
16134360 ± 2% +15.0% 18550862 perf-stat.i.node-load-misses
1576525 ± 2% +10.0% 1734370 ± 2% perf-stat.i.node-loads
10027868 -11.5% 8871598 perf-stat.i.node-store-misses
386034 ± 3% -16.0% 324290 ± 7% perf-stat.i.node-stores
13.15 +9.2% 14.36 perf-stat.overall.MPKI
13.97 +2.3 16.23 perf-stat.overall.cache-miss-rate%
1.60 +8.9% 1.75 perf-stat.overall.cpi
873.29 -14.2% 749.60 perf-stat.overall.cycles-between-cache-misses
0.00 ± 15% -0.0 0.00 ± 9% perf-stat.overall.dTLB-store-miss-rate%
88.73 -1.7 87.07 perf-stat.overall.iTLB-load-miss-rate%
0.62 -8.2% 0.57 perf-stat.overall.ipc
135778 +1.7% 138069 perf-stat.overall.path-length
9.351e+09 -10.1% 8.41e+09 perf-stat.ps.branch-instructions
1.387e+08 -10.0% 1.248e+08 perf-stat.ps.branch-misses
86797490 +14.0% 98949207 perf-stat.ps.cache-misses
2224197 -6.4% 2081616 perf-stat.ps.context-switches
7.578e+10 -2.1% 7.416e+10 perf-stat.ps.cpu-cycles
917495 ± 19% -20.3% 731365 ± 5% perf-stat.ps.dTLB-load-misses
1.412e+10 -10.3% 1.267e+10 perf-stat.ps.dTLB-loads
181859 ± 16% -57.6% 77179 ± 9% perf-stat.ps.dTLB-store-misses
8.227e+09 -10.3% 7.379e+09 perf-stat.ps.dTLB-stores
96313891 -11.0% 85700283 perf-stat.ps.iTLB-load-misses
12236194 +4.0% 12724086 perf-stat.ps.iTLB-loads
4.726e+10 -10.2% 4.245e+10 perf-stat.ps.instructions
16081690 ± 2% +15.0% 18490522 perf-stat.ps.node-load-misses
1571411 ± 2% +10.0% 1728755 ± 2% perf-stat.ps.node-loads
9995103 -11.5% 8842824 perf-stat.ps.node-store-misses
385193 ± 3% -16.0% 323588 ± 7% perf-stat.ps.node-stores
1.445e+13 -10.1% 1.299e+13 perf-stat.total.instructions
1.51 ± 7% -0.2 1.29 ± 7% perf-profile.calltrace.cycles-pp.smpboot_thread_fn.kthread.ret_from_fork
1.53 ± 7% -0.2 1.31 ± 7% perf-profile.calltrace.cycles-pp.ret_from_fork
1.53 ± 7% -0.2 1.31 ± 7% perf-profile.calltrace.cycles-pp.kthread.ret_from_fork
1.48 ± 7% -0.2 1.26 ± 7% perf-profile.calltrace.cycles-pp.rcu_core.__softirqentry_text_start.run_ksoftirqd.smpboot_thread_fn.kthread
1.49 ± 7% -0.2 1.27 ± 7% perf-profile.calltrace.cycles-pp.__softirqentry_text_start.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork
1.50 ± 7% -0.2 1.27 ± 7% perf-profile.calltrace.cycles-pp.run_ksoftirqd.smpboot_thread_fn.kthread.ret_from_fork
1.47 ± 7% -0.2 1.25 ± 7% perf-profile.calltrace.cycles-pp.rcu_do_batch.rcu_core.__softirqentry_text_start.run_ksoftirqd.smpboot_thread_fn
1.41 ± 7% -0.2 1.19 ± 7% perf-profile.calltrace.cycles-pp.kmem_cache_free.rcu_do_batch.rcu_core.__softirqentry_text_start.run_ksoftirqd
1.25 ± 7% -0.2 1.06 ± 7% perf-profile.calltrace.cycles-pp.obj_cgroup_uncharge_pages.kmem_cache_free.rcu_do_batch.rcu_core.__softirqentry_text_start
1.21 ± 7% -0.2 1.03 ± 7% perf-profile.calltrace.cycles-pp.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.rcu_do_batch.rcu_core
0.94 ± 7% -0.1 0.80 ± 7% perf-profile.calltrace.cycles-pp.page_counter_cancel.page_counter_uncharge.obj_cgroup_uncharge_pages.kmem_cache_free.rcu_do_batch
0.62 ± 7% +0.2 0.80 ± 9% perf-profile.calltrace.cycles-pp.tcp_rcv_state_process.tcp_child_process.tcp_v4_rcv.ip_protocol_deliver_rcu.ip_local_deliver_finish
1.51 ± 7% -0.2 1.29 ± 7% perf-profile.children.cycles-pp.smpboot_thread_fn
1.53 ± 7% -0.2 1.31 ± 7% perf-profile.children.cycles-pp.ret_from_fork
1.53 ± 7% -0.2 1.31 ± 7% perf-profile.children.cycles-pp.kthread
1.50 ± 7% -0.2 1.27 ± 7% perf-profile.children.cycles-pp.run_ksoftirqd
1.73 ± 6% -0.2 1.51 ± 5% perf-profile.children.cycles-pp._raw_spin_lock_bh
1.25 ± 5% -0.2 1.07 ± 6% perf-profile.children.cycles-pp.lock_sock_nested
1.03 ± 7% -0.1 0.88 ± 6% perf-profile.children.cycles-pp.native_queued_spin_lock_slowpath
0.83 ± 6% -0.1 0.72 ± 6% perf-profile.children.cycles-pp.sk_clone_lock
0.84 ± 6% -0.1 0.73 ± 6% perf-profile.children.cycles-pp.inet_csk_clone_lock
0.45 ± 8% -0.1 0.34 ± 6% perf-profile.children.cycles-pp.__tcp_get_metrics
0.70 ± 6% -0.1 0.60 ± 6% perf-profile.children.cycles-pp.percpu_counter_add_batch
0.52 ± 8% -0.1 0.42 ± 6% perf-profile.children.cycles-pp.tcp_get_metrics
0.72 ± 5% -0.1 0.62 ± 6% perf-profile.children.cycles-pp.sk_forced_mem_schedule
0.32 ± 7% -0.1 0.24 ± 7% perf-profile.children.cycles-pp.sk_filter_trim_cap
0.49 ± 7% -0.1 0.41 ± 8% perf-profile.children.cycles-pp.tcp_v4_destroy_sock
0.26 ± 7% -0.0 0.22 ± 8% perf-profile.children.cycles-pp.ip_finish_output
0.29 ± 6% -0.0 0.25 ± 9% perf-profile.children.cycles-pp.tcp_write_queue_purge
0.16 ± 10% -0.0 0.12 ± 8% perf-profile.children.cycles-pp.get_obj_cgroup_from_current
0.10 ± 8% -0.0 0.08 ± 6% perf-profile.children.cycles-pp.__destroy_inode
0.10 ± 8% -0.0 0.08 ± 6% perf-profile.children.cycles-pp.destroy_inode
0.10 ± 9% -0.0 0.08 ± 10% perf-profile.children.cycles-pp.sock_put
0.10 ± 10% -0.0 0.07 ± 8% perf-profile.children.cycles-pp.d_instantiate
0.08 ± 11% -0.0 0.06 ± 9% perf-profile.children.cycles-pp.kmem_cache_alloc_trace
0.11 ± 8% +0.0 0.15 ± 6% perf-profile.children.cycles-pp.__inet_lookup_listener
0.08 ± 9% +0.0 0.12 ± 8% perf-profile.children.cycles-pp.inet_lhash2_lookup
0.10 ± 7% +0.0 0.14 ± 7% perf-profile.children.cycles-pp.tcp_ca_openreq_child
0.08 ± 9% +0.0 0.13 ± 9% perf-profile.children.cycles-pp.tcp_newly_delivered
0.08 ± 6% +0.0 0.12 ± 9% perf-profile.children.cycles-pp.tcp_mtup_init
0.09 ± 8% +0.1 0.15 ± 6% perf-profile.children.cycles-pp.tcp_stream_memory_free
0.24 ± 6% +0.1 0.30 ± 8% perf-profile.children.cycles-pp.ip_rcv_core
0.06 ± 9% +0.1 0.12 ± 7% perf-profile.children.cycles-pp.tcp_push
0.11 ± 9% +0.1 0.17 ± 7% perf-profile.children.cycles-pp.tcp_synack_rtt_meas
0.00 ±412% +0.1 0.07 ± 14% perf-profile.children.cycles-pp.tcp_rack_update_reo_wnd
0.20 ± 8% +0.1 0.28 ± 6% perf-profile.children.cycles-pp.tcp_assign_congestion_control
0.34 ± 8% +0.1 0.42 ± 6% perf-profile.children.cycles-pp.tcp_init_metrics
0.14 ± 6% +0.1 0.22 ± 8% perf-profile.children.cycles-pp.tcp_sync_mss
0.33 ± 5% +0.1 0.41 ± 8% perf-profile.children.cycles-pp.inet_csk_route_req
0.31 ± 6% +0.1 0.40 ± 6% perf-profile.children.cycles-pp.inet_csk_route_child_sock
0.13 ± 8% +0.1 0.22 ± 6% perf-profile.children.cycles-pp.skb_entail
0.21 ± 6% +0.1 0.32 ± 7% perf-profile.children.cycles-pp.ip_rcv_finish_core
0.24 ± 5% +0.1 0.35 ± 7% perf-profile.children.cycles-pp.ip_rcv_finish
0.20 ± 7% +0.1 0.32 ± 5% perf-profile.children.cycles-pp.tcp_select_initial_window
0.14 ± 5% +0.1 0.26 ± 8% perf-profile.children.cycles-pp.secure_tcp_ts_off
0.45 ± 6% +0.1 0.58 ± 6% perf-profile.children.cycles-pp.tcp_finish_connect
0.23 ± 5% +0.1 0.35 ± 5% perf-profile.children.cycles-pp.tcp_parse_options
0.17 ± 7% +0.1 0.31 ± 6% perf-profile.children.cycles-pp.tcp_update_pacing_rate
0.20 ± 7% +0.1 0.35 ± 6% perf-profile.children.cycles-pp.tcp_openreq_init_rwin
0.27 ± 9% +0.1 0.42 ± 7% perf-profile.children.cycles-pp.tcp_connect_init
0.45 ± 7% +0.2 0.60 ± 5% perf-profile.children.cycles-pp.tcp_v4_init_sock
0.44 ± 7% +0.2 0.60 ± 6% perf-profile.children.cycles-pp.tcp_init_sock
0.23 ± 7% +0.2 0.39 ± 6% perf-profile.children.cycles-pp.tcp_schedule_loss_probe
0.35 ± 6% +0.2 0.57 ± 7% perf-profile.children.cycles-pp.inet_sk_rebuild_header
0.25 ± 9% +0.2 0.49 ± 7% perf-profile.children.cycles-pp.__tcp_select_window
0.35 ± 6% +0.3 0.61 ± 6% perf-profile.children.cycles-pp.tcp_ack_update_rtt
0.76 ± 5% +0.3 1.04 ± 6% perf-profile.children.cycles-pp.ip_route_output_flow
0.78 ± 6% +0.3 1.08 ± 6% perf-profile.children.cycles-pp.tcp_init_transfer
1.78 ± 6% +0.3 2.11 ± 6% perf-profile.children.cycles-pp.tcp_conn_request
1.07 ± 4% +0.4 1.44 ± 5% perf-profile.children.cycles-pp.ip_route_output_key_hash
1.02 ± 5% +0.4 1.40 ± 5% perf-profile.children.cycles-pp.ip_route_output_key_hash_rcu
2.02 ± 5% +0.5 2.50 ± 6% perf-profile.children.cycles-pp.tcp_ack
1.04 ± 7% +0.6 1.63 ± 7% perf-profile.children.cycles-pp.__sk_dst_check
1.18 ± 7% +0.7 1.86 ± 7% perf-profile.children.cycles-pp.ipv4_dst_check
5.95 ± 5% +0.9 6.87 ± 6% perf-profile.children.cycles-pp.tcp_v4_connect
1.02 ± 7% -0.2 0.87 ± 5% perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
0.44 ± 8% -0.1 0.34 ± 6% perf-profile.self.cycles-pp.__tcp_get_metrics
0.69 ± 6% -0.1 0.59 ± 6% perf-profile.self.cycles-pp.percpu_counter_add_batch
0.71 ± 5% -0.1 0.61 ± 6% perf-profile.self.cycles-pp.sk_forced_mem_schedule
0.32 ± 6% -0.1 0.26 ± 8% perf-profile.self.cycles-pp.ip_finish_output2
0.35 ± 7% -0.1 0.29 ± 5% perf-profile.self.cycles-pp.tcp_recvmsg_locked
0.15 ± 7% -0.0 0.12 ± 8% perf-profile.self.cycles-pp.exit_to_user_mode_prepare
0.17 ± 6% -0.0 0.14 ± 10% perf-profile.self.cycles-pp.__skb_clone
0.07 ± 5% -0.0 0.04 ± 43% perf-profile.self.cycles-pp.sk_filter_trim_cap
0.09 ± 9% -0.0 0.07 ± 6% perf-profile.self.cycles-pp.dequeue_task_fair
0.08 ± 7% -0.0 0.06 ± 8% perf-profile.self.cycles-pp.release_sock
0.07 ± 10% +0.0 0.09 ± 9% perf-profile.self.cycles-pp.tcp_create_openreq_child
0.11 ± 7% +0.0 0.15 ± 5% perf-profile.self.cycles-pp.tcp_connect
0.08 ± 9% +0.0 0.12 ± 8% perf-profile.self.cycles-pp.inet_lhash2_lookup
0.09 ± 9% +0.0 0.13 ± 6% perf-profile.self.cycles-pp.inet_csk_get_port
0.08 ± 10% +0.0 0.12 ± 8% perf-profile.self.cycles-pp.tcp_init_transfer
0.08 ± 9% +0.0 0.13 ± 8% perf-profile.self.cycles-pp.tcp_newly_delivered
0.07 ± 7% +0.0 0.12 ± 9% perf-profile.self.cycles-pp.tcp_mtup_init
0.35 ± 5% +0.1 0.40 ± 5% perf-profile.self.cycles-pp.__ip_queue_xmit
0.16 ± 7% +0.1 0.22 ± 6% perf-profile.self.cycles-pp.__inet_bind
0.09 ± 8% +0.1 0.15 ± 6% perf-profile.self.cycles-pp.tcp_stream_memory_free
0.24 ± 6% +0.1 0.30 ± 8% perf-profile.self.cycles-pp.ip_rcv_core
0.06 ± 9% +0.1 0.12 ± 6% perf-profile.self.cycles-pp.tcp_push
0.00 +0.1 0.07 ± 11% perf-profile.self.cycles-pp.tcp_rack_update_reo_wnd
0.23 ± 8% +0.1 0.30 ± 6% perf-profile.self.cycles-pp.ip_output
0.20 ± 8% +0.1 0.28 ± 5% perf-profile.self.cycles-pp.tcp_assign_congestion_control
0.10 ± 8% +0.1 0.18 ± 7% perf-profile.self.cycles-pp.tcp_v4_syn_recv_sock
0.09 ± 7% +0.1 0.17 ± 7% perf-profile.self.cycles-pp.tcp_openreq_init_rwin
0.07 ± 10% +0.1 0.16 ± 6% perf-profile.self.cycles-pp.tcp_v4_send_synack
0.13 ± 7% +0.1 0.22 ± 7% perf-profile.self.cycles-pp.tcp_sync_mss
0.12 ± 8% +0.1 0.20 ± 7% perf-profile.self.cycles-pp.skb_entail
0.18 ± 8% +0.1 0.27 ± 6% perf-profile.self.cycles-pp.ip_protocol_deliver_rcu
0.21 ± 5% +0.1 0.31 ± 6% perf-profile.self.cycles-pp.ip_rcv_finish_core
0.15 ± 9% +0.1 0.26 ± 6% perf-profile.self.cycles-pp.tcp_update_metrics
0.20 ± 8% +0.1 0.31 ± 5% perf-profile.self.cycles-pp.tcp_select_initial_window
0.12 ± 9% +0.1 0.25 ± 8% perf-profile.self.cycles-pp.tcp_connect_init
0.11 ± 8% +0.1 0.24 ± 8% perf-profile.self.cycles-pp.secure_tcp_ts_off
0.22 ± 5% +0.1 0.35 ± 5% perf-profile.self.cycles-pp.tcp_parse_options
0.13 ± 12% +0.1 0.27 ± 7% perf-profile.self.cycles-pp.tcp_init_metrics
0.17 ± 7% +0.1 0.30 ± 7% perf-profile.self.cycles-pp.tcp_update_pacing_rate
0.17 ± 10% +0.2 0.32 ± 6% perf-profile.self.cycles-pp.tcp_init_sock
0.18 ± 8% +0.2 0.35 ± 6% perf-profile.self.cycles-pp.tcp_schedule_loss_probe
0.42 ± 8% +0.2 0.62 ± 7% perf-profile.self.cycles-pp.tcp_write_xmit
0.25 ± 8% +0.2 0.48 ± 7% perf-profile.self.cycles-pp.__tcp_select_window
0.28 ± 8% +0.3 0.56 ± 5% perf-profile.self.cycles-pp.tcp_ack_update_rtt
0.71 ± 5% +0.4 1.09 ± 6% perf-profile.self.cycles-pp.ip_route_output_key_hash_rcu
1.17 ± 7% +0.7 1.84 ± 7% perf-profile.self.cycles-pp.ipv4_dst_check
netperf.Throughput_tps
22500 +-------------------------------------------------------------------+
| ...+...... ...+......+.....+.....|
22000 |.....+.. +.....+.....+.....+.....+.. |
| |
| |
21500 |-+ |
| |
21000 |-+ |
| |
20500 |-+ |
| |
| |
20000 |-+ |
| O O O O O |
19500 +-------------------------------------------------------------------+
netperf.Throughput_total_tps
360000 +------------------------------------------------------------------+
355000 |-+ ...+..... ...+..... ...+.. +.....|
|.....+.. +.....+.....+... +.. |
350000 |-+ |
345000 |-+ |
| |
340000 |-+ |
335000 |-+ |
330000 |-+ |
| |
325000 |-+ |
320000 |-+ |
| |
315000 |-+ O O O O O O O O O O |
310000 +------------------------------------------------------------------+
netperf.workload
1.08e+08 +----------------------------------------------------------------+
| ...+.....+..... ..+..... ...+.. +.....|
1.06e+08 |.....+.. +.....+.. +.. |
1.04e+08 |-+ |
| |
1.02e+08 |-+ |
| |
1e+08 |-+ |
| |
9.8e+07 |-+ |
9.6e+07 |-+ |
| |
9.4e+07 |-+ O O O O O O O O O O |
| |
9.2e+07 +----------------------------------------------------------------+
netperf.time.voluntary_context_switches
2.1e+08 +----------------------------------------------------------------+
|.....+.....+.....+.....+.....+....+..... ...+.. +.....|
2.05e+08 |-+ +.. |
| |
| |
2e+08 |-+ |
| |
1.95e+08 |-+ |
| |
1.9e+08 |-+ |
| |
| |
1.85e+08 |-+ O O O O O O O |
| O O O |
1.8e+08 +----------------------------------------------------------------+
0.006 +------------------------------------------------------------------+
| |
| |
0.0058 |-+ |
| |
| |
0.0056 |-+ |
| |
0.0054 |-+ |
| |
| |
0.0052 |-+ |
| |
| |
0.005 +------------------------------------------------------------------+
3.25e+06 +----------------------------------------------------------------+
|..... ...+.... ...+....+.....+.....+..... ...+.....|
3.2e+06 |-+ +.. . ...+.. +.. |
| +.. |
3.15e+06 |-+ |
3.1e+06 |-+ |
| |
3.05e+06 |-+ |
| |
3e+06 |-+ |
2.95e+06 |-+ |
| |
2.9e+06 |-+ O O O O O O O O O |
| O |
2.85e+06 +----------------------------------------------------------------+
[*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
lkp-icl-2sp1: 96 threads 2 sockets Ice Lake with 256G memory
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.
---
0DAY/LKP+ Test Infrastructure Open Source Technology Center
https://lists.01.org/hyperkitty/list/lkp@lists.01.org Intel Corporation
Thanks,
Oliver Sang
Download attachment "config-5.15.0-rc4-00001-ga0918006f928.old" of type "application/x-trash" (176725 bytes)
View attachment "job-script" of type "text/plain" (8126 bytes)
View attachment "job.yaml" of type "text/plain" (5549 bytes)
View attachment "reproduce" of type "text/plain" (1091 bytes)
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
