linux-kernel - Re: [PATCH 3/3] objtool: Support stack layout changes in alternatives

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: <20210114002148.fiwjiuhhigv6nzxb@treble>
Date:   Wed, 13 Jan 2021 18:21:48 -0600
From:   Josh Poimboeuf <jpoimboe@...hat.com>
To:     Miroslav Benes <mbenes@...e.cz>
Cc:     Juergen Gross <jgross@...e.com>, xen-devel@...ts.xenproject.org,
        linux-kernel@...r.kernel.org,
        Peter Zijlstra <peterz@...radead.org>,
        Shinichiro Kawasaki <shinichiro.kawasaki@....com>
Subject: Re: [PATCH 3/3] objtool: Support stack layout changes in alternatives

On Fri, Jan 08, 2021 at 09:15:43AM +0100, Miroslav Benes wrote:
> > That comment is indeed now obsolete.  I can squash something like so:
> > 
> > diff --git a/tools/objtool/check.c b/tools/objtool/check.c
> > index 81d56fdef1c3..ce67437aaf3f 100644
> > --- a/tools/objtool/check.c
> > +++ b/tools/objtool/check.c
> > @@ -958,21 +958,8 @@ static int add_call_destinations(struct objtool_file *file)
> >  }
> >  
> >  /*
> > - * The .alternatives section requires some extra special care, over and above
> > - * what other special sections require:
> > - *
> > - * 1. Because alternatives are patched in-place, we need to insert a fake jump
> > - *    instruction at the end so that validate_branch() skips all the original
> > - *    replaced instructions when validating the new instruction path.
> > - *
> > - * 2. An added wrinkle is that the new instruction length might be zero.  In
> > - *    that case the old instructions are replaced with noops.  We simulate that
> > - *    by creating a fake jump as the only new instruction.
> > - *
> > - * 3. In some cases, the alternative section includes an instruction which
> > - *    conditionally jumps to the _end_ of the entry.  We have to modify these
> > - *    jumps' destinations to point back to .text rather than the end of the
> > - *    entry in .altinstr_replacement.
> > + * The .alternatives section requires some extra special care over and above
> > + * other special sections because alternatives are patched in place.
> >   */
> >  static int handle_group_alt(struct objtool_file *file,
> >  			    struct special_alt *special_alt,
> 
> Looks good to me.

Thanks, I squashed this in.

If there are no objections, I'll go ahead and merge this set in -tip.
It doesn't have any direct dependencies on the Juergen's changes, and it
would be nice to have some of these changes in-tree, especially the fake
jump removal.

-- 
Josh