linux-kernel - RE: Linux 5.19-rc8

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <2750dc08eb6c4e6dbbd638310f903bce@AcuMS.aculab.com>
Date:   Wed, 27 Jul 2022 07:46:43 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     'Yury Norov' <yury.norov@...il.com>,
        "Russell King (Oracle)" <linux@...linux.org.uk>
CC:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Dennis Zhou <dennis@...nel.org>,
        Guenter Roeck <linux@...ck-us.net>,
        Catalin Marinas <catalin.marinas@....com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Geert Uytterhoeven <geert@...ux-m68k.org>,
        "linux-m68k@...ts.linux-m68k.org" <linux-m68k@...ts.linux-m68k.org>
Subject: RE: Linux 5.19-rc8

From: Yury Norov
> Sent: 27 July 2022 02:34
...
> If the difference really concerns you, I'd suggest running the test
> several times to measure confidence intervals.

Or, do what I've been doing and get an accurate cycle count
for a single call and repeat about 10 times.
The first value is typically slow (loading I$), the
rest typically identical.
They can often even be matched to the expected value!
The fastest value is the one to quote!

On x86 you have to use the perf cycle counter, the tsc
is just useless.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)