linux-kernel - Re: [PATCH] oom, trace: Remove ENUM evaluation of COMPACTION

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: <20170615074813.1ed10a07@gandalf.local.home>
Date:   Thu, 15 Jun 2017 07:48:13 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Michal Hocko <mhocko@...nel.org>
Cc:     Vlastimil Babka <vbabka@...e.cz>,
        David Rientjes <rientjes@...gle.com>,
        Johannes Weiner <hannes@...xchg.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] oom, trace: Remove ENUM evaluation of
 COMPACTION_FEEDBACK

On Thu, 15 Jun 2017 12:57:21 +0200
Michal Hocko <mhocko@...nel.org> wrote:

> >  #define COMPACTION_FEEDBACK             \
> >         EM(COMPACTION_FAILED,           "failed")       \
> >         EM(COMPACTION_WITHDRAWN,        "withdrawn")    \
> >         EMe(COMPACTION_PROGRESS,        "progress")
> > 
> > Which is still needed for the __print_symbolic() usage in the
> > trace_event. But it is not needed to be evaluated.
> > 
> > Removing the evaluation part removes the unnecessary evaluations of
> > numbers to numbers.  
> 
> I will be honest with you. Even if I understood how this voodoo mapping
> works I forgot everything. So if the COMPACTION_*->names keeps working
> I have no objection to the patch.

Yes it does. It you want to verify, simply boot the kernel without the
patch and do:

 # cat /sys/kernel/debug/tracing/events/*/*/format > ~/formats1

Add the patch, compile and boot that kernel.

 # cat /sys/kernel/debug/tracing/events/*/*/format > ~/formats2

then:

 # diff formats1 formats2

and they should be the same, with the exception that the ID:'s can
change.

Also, let me send a v2. Resending this patch (from the private one)
seems to broke the format of it.

-- Steve