linux-kernel - Re: [RESEND PATCH V3 3/8] perf/x86/intel: Support hardware TopDown metrics

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: <20190831092921.GK2369@hirez.programming.kicks-ass.net>
Date:   Sat, 31 Aug 2019 11:29:21 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Stephane Eranian <eranian@...gle.com>
Cc:     Andi Kleen <ak@...ux.intel.com>,
        "Liang, Kan" <kan.liang@...ux.intel.com>,
        Arnaldo Carvalho de Melo <acme@...nel.org>,
        Ingo Molnar <mingo@...hat.com>,
        LKML <linux-kernel@...r.kernel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Jiri Olsa <jolsa@...nel.org>,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>
Subject: Re: [RESEND PATCH V3 3/8] perf/x86/intel: Support hardware TopDown
 metrics

On Sat, Aug 31, 2019 at 02:13:05AM -0700, Stephane Eranian wrote:

> With PERF_METRICS, the delta is always since previous read. If you read
> frequently enough you do not lose precision.

You always loose precision, the whole fraction of 255 looses the
fractional part; consider:

255 * 1 / 8
31.87500000000000000000
255 * 7 / 8
223.12500000000000000000

31 * 8 / 255
.97254901960784313725
223 * 8 / 255
6.99607843137254901960

Now, we can make reconstruction use normal 'round-nearest' and then we'd
get 1 and 7 back, but there's always cases where you loose things.

Also, with 255 being an odd number, round-nearest is actually 'hard' :/