linux-kernel - Re: [PATCH v6 33/52] memory: tegra20: Support interconnect framework

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: <4ff5d718-5e05-fc11-9795-7c3fef4525d3@gmail.com>
Date:   Wed, 28 Oct 2020 00:12:00 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Thierry Reding <thierry.reding@...il.com>
Cc:     Jonathan Hunter <jonathanh@...dia.com>,
        Georgi Djakov <georgi.djakov@...aro.org>,
        Rob Herring <robh+dt@...nel.org>,
        Michael Turquette <mturquette@...libre.com>,
        Stephen Boyd <sboyd@...nel.org>,
        Peter De Schrijver <pdeschrijver@...dia.com>,
        MyungJoo Ham <myungjoo.ham@...sung.com>,
        Kyungmin Park <kyungmin.park@...sung.com>,
        Chanwoo Choi <cw00.choi@...sung.com>,
        Mikko Perttunen <cyndis@...si.fi>,
        Viresh Kumar <vireshk@...nel.org>,
        Peter Geis <pgwipeout@...il.com>,
        Nicolas Chauvet <kwizart@...il.com>,
        Krzysztof Kozlowski <krzk@...nel.org>,
        linux-tegra@...r.kernel.org, linux-pm@...r.kernel.org,
        linux-kernel@...r.kernel.org, dri-devel@...ts.freedesktop.org,
        devicetree@...r.kernel.org
Subject: Re: [PATCH v6 33/52] memory: tegra20: Support interconnect framework

27.10.2020 23:22, Dmitry Osipenko пишет:
...
>>> +
>>> +	*agg_avg += avg_bw;
>>> +	*agg_peak = max(*agg_peak, peak_bw);
>>
>> I'm not very familiar with ICC, but shouldn't the aggregated peak value
>> be the sum of the current aggregated peak and the new peak bandwidth?
>> Currently you're selecting the maximum peak bandwidth across all
>> clients, so isn't that going to be too small if for whatever reason
>> multiple clients need peak bandwidth at the same time?

The current variant with max-peak selection should be okay since it
takes into account the competing ISO bandwidths of other devices by
overestimating the bandwidth.

For now we have only display ISO clients and it won't be a problem to
tune the algorithm later on if it won't work well for other ISO clients.