linux-kernel - Re: [PATCH V7 07/14] perf/core: Add a new PERF_EV_CAP

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: <20200724145142.GC10769@hirez.programming.kicks-ass.net>
Date:   Fri, 24 Jul 2020 16:51:42 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Andi Kleen <ak@...ux.intel.com>
Cc:     kan.liang@...ux.intel.com, acme@...hat.com, mingo@...nel.org,
        linux-kernel@...r.kernel.org, jolsa@...nel.org, eranian@...gle.com,
        alexander.shishkin@...ux.intel.com, like.xu@...ux.intel.com
Subject: Re: [PATCH V7 07/14] perf/core: Add a new PERF_EV_CAP_COEXIST event
 capability

On Fri, Jul 24, 2020 at 07:39:08AM -0700, Andi Kleen wrote:
> > The consequence is that SLOTS must be the leader, is that really a
> > problem? You keep providing the {cycles, slots, metric-things} example,
> 
> Yes that's a problem. One (major) use case for topdown is to
> sample on lots of different events, but always create groups that
> also measure topdown metrics for that sampling period. For that we need
> other leaders.

Apparently the only problem here is the tool, the kernel doesn't care
which events in a group are sampling, there could be multiple.