linux-kernel - Re: [PATCH v6 0/3] Cavium ARM64 uncore PMU support

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]

Message-ID: <20170623135607.qey6p3pt4fltvyas@pd.tnic>
Date:   Fri, 23 Jun 2017 15:56:07 +0200
From:   Borislav Petkov <bp@...en8.de>
To:     Jan Glauber <jglauber@...ium.com>
Cc:     Mark Rutland <mark.rutland@....com>,
        Will Deacon <will.deacon@....com>,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 0/3] Cavium ARM64 uncore PMU support

On Fri, Jun 23, 2017 at 03:01:25PM +0200, Jan Glauber wrote:
> Add support for various PMU counters found on the Cavium ThunderX and
> OcteonTx SoC.
> 
> The driver provides common "uncore" functions to avoid code duplication and
> support adding more device PMUs (like L2 cache) in the future.
> 
> Probe and removal of the PMUs is done by hooking into the ThunderX EDAC
> driver as this driver owns all the PCI devices containing the PMU counters.

... which is a very bad idea, of course. EDAC doesn't have anything to
do with performance measurements.

And I don't see why the EDAC driver should own any PCI devices -
EDAC simply uses them like anything else in the kernel so the PMU
functionality can simply use them too, independent from EDAC.

-- 
Regards/Gruss,
    Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.