| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <202411051526.913796e-oliver.sang@intel.com>
Date: Tue, 5 Nov 2024 15:59:00 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Linus Torvalds <torvalds@...ux-foundation.org>
CC: <oe-lkp@...ts.linux.dev>, <lkp@...el.com>, <linux-kernel@...r.kernel.org>,
Josh Poimboeuf <jpoimboe@...nel.org>, "Kirill A. Shutemov"
<kirill.shutemov@...ux.intel.com>, <ying.huang@...el.com>,
<feng.tang@...el.com>, <fengwei.yin@...el.com>, <oliver.sang@...el.com>
Subject: [linus:master] [x86/uaccess] 0fc810ae3a:
will-it-scale.per_process_ops 3.1% improvement
Hello,
kernel test robot noticed a 3.1% improvement of will-it-scale.per_process_ops on:
commit: 0fc810ae3ae110f9e2fcccce80fc8c8d62f97907 ("x86/uaccess: Avoid barrier_nospec() in 64-bit copy_from_user()")
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
testcase: will-it-scale
config: x86_64-rhel-8.3
compiler: gcc-12
test machine: 256 threads 2 sockets GENUINE INTEL(R) XEON(R) (Sierra Forest) with 128G memory
parameters:
nr_task: 100%
mode: process
test: eventfd1
cpufreq_governor: performance
Details are as below:
-------------------------------------------------------------------------------------------------->
The kernel config and materials to reproduce are available at:
https://download.01.org/0day-ci/archive/20241105/202411051526.913796e-oliver.sang@intel.com
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-12/performance/x86_64-rhel-8.3/process/100%/debian-12-x86_64-20240206.cgz/lkp-srf-2sp1/eventfd1/will-it-scale
commit:
14b7d43c5c ("Merge tag 'perf-tools-fixes-for-v6.12-2-2024-10-30' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools")
0fc810ae3a ("x86/uaccess: Avoid barrier_nospec() in 64-bit copy_from_user()")
14b7d43c5c068cb9 0fc810ae3ae110f9e2fcccce80f
---------------- ---------------------------
%stddev %change %stddev
\ | \
6.721e+08 +3.1% 6.929e+08 will-it-scale.256.processes
2625451 +3.1% 2706555 will-it-scale.per_process_ops
6.721e+08 +3.1% 6.929e+08 will-it-scale.workload
1.603e+11 +2.3% 1.64e+11 perf-stat.i.branch-instructions
0.81 -2.6% 0.79 perf-stat.i.cpi
8.18e+11 +2.6% 8.392e+11 perf-stat.i.instructions
1.24 +2.6% 1.27 perf-stat.i.ipc
0.81 -2.4% 0.79 perf-stat.overall.cpi
1.24 +2.5% 1.27 perf-stat.overall.ipc
1.597e+11 +2.3% 1.634e+11 perf-stat.ps.branch-instructions
8.152e+11 +2.6% 8.364e+11 perf-stat.ps.instructions
2.484e+14 +2.5% 2.545e+14 perf-stat.total.instructions
10.11 -2.6 7.50 perf-profile.calltrace.cycles-pp._copy_from_user.eventfd_write.vfs_write.ksys_write.do_syscall_64
13.86 -2.6 11.31 perf-profile.calltrace.cycles-pp.eventfd_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
16.80 -2.5 14.30 perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
19.08 -2.4 16.64 perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
21.97 -2.4 19.58 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
23.21 -2.4 20.86 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.write
52.90 -1.5 51.43 perf-profile.calltrace.cycles-pp.write
0.82 +0.0 0.83 perf-profile.calltrace.cycles-pp.syscall_exit_to_user_mode.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
0.60 +0.0 0.62 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_safe_stack.read
1.46 +0.0 1.49 perf-profile.calltrace.cycles-pp.rw_verify_area.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.08 +0.0 1.10 perf-profile.calltrace.cycles-pp.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
1.30 +0.0 1.32 perf-profile.calltrace.cycles-pp.testcase
0.60 +0.0 0.63 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_safe_stack.write
1.12 +0.0 1.16 perf-profile.calltrace.cycles-pp.read@plt
1.62 +0.0 1.66 perf-profile.calltrace.cycles-pp.fdget_pos.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
2.18 +0.0 2.23 perf-profile.calltrace.cycles-pp.entry_SYSRETQ_unsafe_stack.read
1.06 +0.1 1.11 perf-profile.calltrace.cycles-pp.x64_sys_call.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
2.18 +0.1 2.24 perf-profile.calltrace.cycles-pp.entry_SYSRETQ_unsafe_stack.write
2.39 +0.1 2.44 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.eventfd_write.vfs_write.ksys_write.do_syscall_64
1.61 +0.1 1.68 perf-profile.calltrace.cycles-pp.fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
1.72 +0.1 1.80 perf-profile.calltrace.cycles-pp.rw_verify_area.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.22 +0.1 2.31 perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.eventfd_read.vfs_read.ksys_read.do_syscall_64
9.44 +0.3 9.78 perf-profile.calltrace.cycles-pp._copy_to_iter.eventfd_read.vfs_read.ksys_read.do_syscall_64
16.77 +0.4 17.18 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.read
13.14 +0.5 13.62 perf-profile.calltrace.cycles-pp.eventfd_read.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
17.00 +0.6 17.61 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.write
21.07 +0.8 21.87 perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
23.41 +0.9 24.32 perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
26.39 +1.0 27.40 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
27.66 +1.1 28.72 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.read
57.88 +1.7 59.63 perf-profile.calltrace.cycles-pp.read
10.46 -2.6 7.86 perf-profile.children.cycles-pp._copy_from_user
14.06 -2.5 11.52 perf-profile.children.cycles-pp.eventfd_write
17.01 -2.5 14.50 perf-profile.children.cycles-pp.vfs_write
19.30 -2.4 16.86 perf-profile.children.cycles-pp.ksys_write
52.62 -1.5 51.12 perf-profile.children.cycles-pp.write
48.64 -1.4 47.27 perf-profile.children.cycles-pp.do_syscall_64
51.11 -1.3 49.81 perf-profile.children.cycles-pp.entry_SYSCALL_64_after_hwframe
0.60 +0.0 0.61 perf-profile.children.cycles-pp.read@plt
2.39 +0.0 2.43 perf-profile.children.cycles-pp.syscall_return_via_sysret
2.05 +0.0 2.09 perf-profile.children.cycles-pp.testcase
2.25 +0.0 2.30 perf-profile.children.cycles-pp.security_file_permission
1.53 +0.1 1.58 perf-profile.children.cycles-pp.entry_SYSCALL_64_safe_stack
2.19 +0.1 2.27 perf-profile.children.cycles-pp.x64_sys_call
3.24 +0.1 3.34 perf-profile.children.cycles-pp.rw_verify_area
3.24 +0.1 3.36 perf-profile.children.cycles-pp.fdget_pos
4.66 +0.2 4.81 perf-profile.children.cycles-pp._raw_spin_lock_irq
9.46 +0.3 9.80 perf-profile.children.cycles-pp._copy_to_iter
17.59 +0.4 18.00 perf-profile.children.cycles-pp.entry_SYSRETQ_unsafe_stack
13.24 +0.5 13.72 perf-profile.children.cycles-pp.eventfd_read
22.10 +0.7 22.78 perf-profile.children.cycles-pp.entry_SYSCALL_64
21.12 +0.8 21.93 perf-profile.children.cycles-pp.vfs_read
23.44 +0.9 24.35 perf-profile.children.cycles-pp.ksys_read
57.52 +1.8 59.32 perf-profile.children.cycles-pp.read
10.38 -2.6 7.80 perf-profile.self.cycles-pp._copy_from_user
0.66 +0.0 0.67 perf-profile.self.cycles-pp.ksys_write
0.45 +0.0 0.47 perf-profile.self.cycles-pp.security_file_permission
0.71 +0.0 0.73 perf-profile.self.cycles-pp.ksys_read
1.08 +0.0 1.10 perf-profile.self.cycles-pp.syscall_exit_to_user_mode
1.43 +0.0 1.46 perf-profile.self.cycles-pp.testcase
2.39 +0.0 2.43 perf-profile.self.cycles-pp.syscall_return_via_sysret
1.55 +0.0 1.59 perf-profile.self.cycles-pp.eventfd_read
0.96 +0.1 1.01 perf-profile.self.cycles-pp.rw_verify_area
1.48 +0.1 1.54 perf-profile.self.cycles-pp.entry_SYSCALL_64_safe_stack
1.85 +0.1 1.91 perf-profile.self.cycles-pp.do_syscall_64
2.16 +0.1 2.23 perf-profile.self.cycles-pp.entry_SYSCALL_64_after_hwframe
2.13 +0.1 2.20 perf-profile.self.cycles-pp.x64_sys_call
3.22 +0.1 3.34 perf-profile.self.cycles-pp.fdget_pos
4.48 +0.1 4.62 perf-profile.self.cycles-pp._raw_spin_lock_irq
7.71 +0.2 7.90 perf-profile.self.cycles-pp.write
7.94 +0.2 8.18 perf-profile.self.cycles-pp.read
6.12 +0.3 6.38 perf-profile.self.cycles-pp.vfs_read
9.38 +0.3 9.72 perf-profile.self.cycles-pp._copy_to_iter
10.36 +0.4 10.72 perf-profile.self.cycles-pp.entry_SYSCALL_64
17.54 +0.4 17.95 perf-profile.self.cycles-pp.entry_SYSRETQ_unsafe_stack
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
Powered by blists - more mailing lists