linux-kernel - Re: [PATCH 2/7] perf: Create a symlink for a PMU

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: <1d77c148-7ea6-4631-d799-e8ec3d31f347@linux.intel.com>
Date:   Thu, 24 Jun 2021 10:49:42 -0700
From:   Andi Kleen <ak@...ux.intel.com>
To:     Greg KH <greg@...ah.com>, "Liang, Kan" <kan.liang@...ux.intel.com>
Cc:     peterz@...radead.org, mingo@...hat.com,
        linux-kernel@...r.kernel.org, eranian@...gle.com,
        namhyung@...nel.org, acme@...nel.org, jolsa@...hat.com
Subject: Re: [PATCH 2/7] perf: Create a symlink for a PMU

> That is not how sysfs and the driver model works, sorry.  You don't get
> to keep both names, otherwise sysfs would be even more of a mess than it
> currently is.  What happens if you need "another" name in the future?
> When do you stop

I don't see any scenario where we would ever need more than a single 
symlink.

I believe there is already precedent for this elsewhere.

>
> this isn't ok, please do it right.

I don't see what exactly are you proposing.

Are you proposing to break every perf script on a kernel update? Doesn't 
seem acceptable to me.

Or move the compatibility into the perf tool? That would require the 
users to both update the perf tool and the kernel. I suppose it would be 
possible, but it would be totally against the standard perf abstraction 
design. Besides there are other consumers of this than just the perf 
tool, so it could possibly have a wide impact.

-Andi