linux-kernel - Re: [PATCH 4/5] x86/ftrace: Add annotations for ftrace_arch_code_modify_prepare() and ftrace_arch_code_modify_post

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: <20200602231000.6pfrgdgm4vd7sbvn@ltop.local>
Date:   Wed, 3 Jun 2020 01:10:00 +0200
From:   Luc Van Oostenryck <luc.vanoostenryck@...il.com>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     Jules Irenge <jbi.octave@...il.com>, linux-kernel@...r.kernel.org,
        tglx@...utronix.de, paulmck@...nel.org, mingo@...hat.com,
        boqun.feng@...il.com, Borislav Petkov <bp@...en8.de>,
        "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@...nel.org>,
        "H. Peter Anvin" <hpa@...or.com>
Subject: Re: [PATCH 4/5] x86/ftrace: Add annotations for
 ftrace_arch_code_modify_prepare() and ftrace_arch_code_modify_post_process()

On Mon, Jun 01, 2020 at 03:46:47PM -0400, Steven Rostedt wrote:
> On Mon,  1 Jun 2020 19:45:51 +0100
> Jules Irenge <jbi.octave@...il.com> wrote:
> 
> > Sparse reports warnings
> > 
> > warning: context imbalance in ftrace_arch_code_modify_prepare()
> > 	- wrong count at exit
> > warning: context imbalance in ftrace_arch_code_modify_post_process()
> > 	- wrong count at exit
> > 
> > The root cause is that even if
> >  the annotations on the function are correct,
> > mutex do not support annotation

Yes.

> Wait what? This looks like either a bug in sparse, or we just remove the
> annotations. This just makes the code ugly, and looks silly.

The annotations added by commit
  074376ac0e1d ("ftrace/x86: Anotate text_mutex split between ...
are indeed wrong (because they don't match what the functions are
really doing / mutex operations have never been annotated).
The're also pointless since their prototypes are un-annotated.

-- Luc