| 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: <202301242012.efecb3a0-oliver.sang@intel.com>
Date: Tue, 24 Jan 2023 21:02:42 +0800
From: kernel test robot <oliver.sang@...el.com>
To: Liam Howlett <liam.howlett@...cle.com>
CC: <oe-lkp@...ts.linux.dev>, <lkp@...el.com>,
Jirka Hladky <jhladky@...hat.com>,
Matthew Wilcox <willy@...radead.org>,
Liam Howlett <Liam.Howlett@...cle.com>, <linux-mm@...ck.org>,
<ying.huang@...el.com>, <feng.tang@...el.com>,
<zhengjun.xing@...ux.intel.com>, <fengwei.yin@...el.com>,
"maple-tree@...ts.infradead.org" <maple-tree@...ts.infradead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] maple_tree: Remove GFP_ZERO from kmem_cache_alloc() and
kmem_cache_alloc_bulk()
Greeting,
FYI, we noticed a 3.8% improvement of will-it-scale.per_process_ops due to commit:
commit: 180c117d46be304a08b14fb080010773faf50788 ("[PATCH] maple_tree: Remove GFP_ZERO from kmem_cache_alloc() and kmem_cache_alloc_bulk()")
url: https://github.com/intel-lab-lkp/linux/commits/Liam-Howlett/maple_tree-Remove-GFP_ZERO-from-kmem_cache_alloc-and-kmem_cache_alloc_bulk/20230106-000849
base: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git 41c03ba9beea760bd2d2ac9250b09a2e192da2dc
patch link: https://lore.kernel.org/all/20230105160427.2988454-1-Liam.Howlett@oracle.com/
patch subject: [PATCH] maple_tree: Remove GFP_ZERO from kmem_cache_alloc() and kmem_cache_alloc_bulk()
in testcase: will-it-scale
on test machine: 104 threads 2 sockets (Skylake) with 192G memory
with following parameters:
nr_task: 16
mode: process
test: mmap1
cpufreq_governor: performance
test-description: Will It Scale takes a testcase and runs it from 1 through to n parallel copies to see if the testcase will scale. It builds both a process and threads based test in order to see any differences between the two.
test-url: https://github.com/antonblanchard/will-it-scale
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.
=========================================================================================
compiler/cpufreq_governor/kconfig/mode/nr_task/rootfs/tbox_group/test/testcase:
gcc-11/performance/x86_64-rhel-8.3/process/16/debian-11.1-x86_64-20220510.cgz/lkp-skl-fpga01/mmap1/will-it-scale
commit:
41c03ba9be ("Merge tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost")
180c117d46 ("maple_tree: Remove GFP_ZERO from kmem_cache_alloc() and kmem_cache_alloc_bulk()")
41c03ba9beea760b 180c117d46be304a08b14fb0800
---------------- ---------------------------
%stddev %change %stddev
\ | \
2604393 +3.8% 2702107 will-it-scale.16.processes
162774 +3.8% 168881 will-it-scale.per_process_ops
2604393 +3.8% 2702107 will-it-scale.workload
1.254e+10 +2.6% 1.286e+10 perf-stat.i.branch-instructions
1.251e+10 +2.2% 1.279e+10 perf-stat.i.dTLB-loads
6.283e+09 +1.3% 6.367e+09 perf-stat.i.dTLB-stores
5.838e+10 +2.5% 5.987e+10 perf-stat.i.instructions
301.29 +2.2% 307.85 perf-stat.i.metric.M/sec
0.81 -2.1% 0.79 perf-stat.overall.cpi
6756492 -1.0% 6686622 perf-stat.overall.path-length
1.25e+10 +2.6% 1.282e+10 perf-stat.ps.branch-instructions
1.247e+10 +2.2% 1.275e+10 perf-stat.ps.dTLB-loads
6.263e+09 +1.3% 6.346e+09 perf-stat.ps.dTLB-stores
5.819e+10 +2.5% 5.967e+10 perf-stat.ps.instructions
1.76e+13 +2.7% 1.807e+13 perf-stat.total.instructions
2.31 ± 10% -1.0 1.29 ± 7% perf-profile.calltrace.cycles-pp.mas_preallocate.do_mas_align_munmap.__vm_munmap.__x64_sys_munmap.do_syscall_64
2.24 ± 10% -1.0 1.23 ± 7% perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.do_mas_align_munmap.__vm_munmap.__x64_sys_munmap
2.28 ± 11% -1.0 1.27 ± 8% perf-profile.calltrace.cycles-pp.mas_preallocate.mmap_region.do_mmap.vm_mmap_pgoff.do_syscall_64
2.24 ± 10% -1.0 1.23 ± 8% perf-profile.calltrace.cycles-pp.mas_alloc_nodes.mas_preallocate.mmap_region.do_mmap.vm_mmap_pgoff
1.52 ± 11% -0.8 0.67 ± 8% perf-profile.calltrace.cycles-pp.kmem_cache_alloc_bulk.mas_alloc_nodes.mas_preallocate.mmap_region.do_mmap
1.49 ± 11% -0.8 0.66 ± 7% perf-profile.calltrace.cycles-pp.kmem_cache_alloc_bulk.mas_alloc_nodes.mas_preallocate.do_mas_align_munmap.__vm_munmap
4.94 ± 10% -2.1 2.83 ± 8% perf-profile.children.cycles-pp.mas_alloc_nodes
4.61 ± 10% -2.0 2.57 ± 8% perf-profile.children.cycles-pp.mas_preallocate
1.96 ± 10% -1.8 0.21 ± 7% perf-profile.children.cycles-pp.memset_erms
3.09 ± 10% -1.7 1.36 ± 7% perf-profile.children.cycles-pp.kmem_cache_alloc_bulk
0.09 ± 13% -0.1 0.03 ±100% perf-profile.children.cycles-pp.vm_area_free
0.07 ± 10% +0.0 0.09 ± 11% perf-profile.children.cycles-pp.mas_wr_modify
0.22 ± 10% +0.1 0.28 ± 7% perf-profile.children.cycles-pp.__might_sleep
0.00 +0.2 0.22 ± 10% perf-profile.children.cycles-pp.mas_pop_node
1.89 ± 11% -1.7 0.21 ± 8% perf-profile.self.cycles-pp.memset_erms
0.72 ± 11% -0.3 0.37 ± 7% perf-profile.self.cycles-pp.kmem_cache_alloc_bulk
0.09 ± 13% -0.1 0.03 ±100% perf-profile.self.cycles-pp.vm_area_free
0.11 ± 7% +0.1 0.17 ± 11% perf-profile.self.cycles-pp.do_mas_munmap
0.00 +0.2 0.21 ± 9% perf-profile.self.cycles-pp.mas_pop_node
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
View attachment "config-6.2.0-rc2-00058-g180c117d46be" of type "text/plain" (166849 bytes)
View attachment "job-script" of type "text/plain" (7816 bytes)
View attachment "job.yaml" of type "text/plain" (5294 bytes)
View attachment "reproduce" of type "text/plain" (345 bytes)
Powered by blists - more mailing lists