linux-kernel - Re: perfmon trouble

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: <20180611164859.GA28292@bombadil.infradead.org>
Date:   Mon, 11 Jun 2018 09:48:59 -0700
From:   Matthew Wilcox <willy@...radead.org>
To:     Christoph Hellwig <hch@...radead.org>
Cc:     Al Viro <viro@...IV.linux.org.uk>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-fsdevel <linux-fsdevel@...r.kernel.org>,
        Stephane Eranian <eranian@...gle.com>
Subject: Re: perfmon trouble

On Mon, Jun 11, 2018 at 09:23:13AM -0700, Christoph Hellwig wrote:
> Is perfmon even in still in use?  If not it might be a good idea to
> drop it for good.

The problem is that even oprofile on ia64 depends on perfmon.

We have about four options, none of which are terribly appealing:

1. rm -rf arch/ia64.  Bit harsh to do this when hardware is still being
produced.

2. rm -rf arch/ia64's support for profiling completely.  Would anybody
complain?

3. Convert ia64 over to perf interface.  Who would do this work?

4. Fix the perfmon kernel and userspace to not assume this insane semantic.
Flag day, blah, blah.

Gentoo and Debian seem to be the only remaining living ports to Itanium
and at least the Debian one is very unofficial.