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
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Thu, 16 May 2024 10:42:34 +0800
From: Oliver Sang <oliver.sang@...el.com>
To: Frederic Weisbecker <frederic@...nel.org>
CC: <oe-lkp@...ts.linux.dev>, <lkp@...el.com>, <linux-kernel@...r.kernel.org>,
	Thomas Gleixner <tglx@...utronix.de>, Anna-Maria Behnsen
	<anna-maria@...utronix.de>, <ying.huang@...el.com>, <feng.tang@...el.com>,
	<fengwei.yin@...el.com>, <oliver.sang@...el.com>
Subject: Re: [linus:master] [timers/migration]  61f7fdf8fd:
 netperf.Throughput_Mbps 2.7% improvement

hi, Frederic Weisbecker,

On Wed, May 15, 2024 at 04:14:14PM +0200, Frederic Weisbecker wrote:
> Le Wed, May 15, 2024 at 04:06:50PM +0800, kernel test robot a écrit :
> > 
> > 
> > Hello,
> > 
> > kernel test robot noticed a 2.7% improvement of netperf.Throughput_Mbps on:
> > 
> > 
> > commit: 61f7fdf8fd00ce33d30ca3fae8d643c0850ce945 ("timers/migration: Fix ignored event due to missing CPU update")
> > https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master
> 
> Nice, so this means the previously reported regression with netperf due to the
> new timer migration code is gone, right?
> 
> For reference: https://lore.kernel.org/all/202403011511.24defbbd-oliver.sang@intel.com/

I believe so.
previous report is for 7ee9887703, the scores are:

=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase:
  cs-localhost/gcc-12/performance/ipv4/x86_64-rhel-8.3/200%/debian-12-x86_64-20240206.cgz/300s/lkp-icl-2sp2/SCTP_STREAM/netperf

<---- (1)
  
57e95a5c4117dc6a 7ee988770326fca440472200c3e 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      4.10            -1.2%       4.05        netperf.ThroughputBoth_Mbps
      1049            -1.2%       1037        netperf.ThroughputBoth_total_Mbps
      4.10            -1.2%       4.05        netperf.Throughput_Mbps
      1049            -1.2%       1037        netperf.Throughput_total_Mbps


for this report, we just have same test's score as:

=========================================================================================
cluster/compiler/cpufreq_governor/ip/kconfig/nr_threads/rootfs/runtime/tbox_group/test/testcase:
  cs-localhost/gcc-13/performance/ipv4/x86_64-rhel-8.3/200%/debian-12-x86_64-20240206.cgz/300s/lkp-icl-2sp2/SCTP_STREAM/netperf

<---- (2)
  
cffaefd15a8f423c 61f7fdf8fd00ce33d30ca3fae8d 
---------------- --------------------------- 
         %stddev     %change         %stddev
             \          |                \  
      4.05            +1.1%       4.10        netperf.ThroughputBoth_Mbps
      1037            +1.1%       1049        netperf.ThroughputBoth_total_Mbps
      4.05            +1.1%       4.10        netperf.Throughput_Mbps
      1037            +1.1%       1049        netperf.Throughput_total_Mbps


from option lines (1)(2), only diff is compiler version which we updated to use
gcc-13 recently. but seems no impact in this case.

what we can see from 2 tables above are, 7ee9887703 just has same scores as
parent of 61f7fdf8fd (cffaefd15a), and 61f7fdf8fd just has same scores as parent
of 7ee9887703 (57e95a5c41). these mean the regression is just resolved.


> 
> Thanks.
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ