linux-kernel - Re: [PATCH 00/21] liblockdep fixes for v4.12

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: <20170528091128.wv5fufpijbs36gkf@gmail.com>
Date:   Sun, 28 May 2017 11:11:28 +0200
From:   Ingo Molnar <mingo@...nel.org>
To:     "Levin, Alexander (Sasha Levin)" <alexander.levin@...izon.com>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "ben@...adent.org.uk" <ben@...adent.org.uk>,
        "a.p.zijlstra@...llo.nl" <a.p.zijlstra@...llo.nl>,
        "tglx@...utronix.de" <tglx@...utronix.de>
Subject: Re: [PATCH 00/21] liblockdep fixes for v4.12


* Levin, Alexander (Sasha Levin) <alexander.levin@...izon.com> wrote:

> On Fri, May 26, 2017 at 01:21:51PM +0200, Ingo Molnar wrote:
> > 
> > * Levin, Alexander (Sasha Levin) <alexander.levin@...izon.com> wrote:
> > 
> > > Right, and as you can see from this patchset where we added to
> > > tools/include/ when needed and removed from lib/lockdep/uinclude,
> > > liblockdep is slowly creeping the "right" way.
> > > 
> > > perf, like liblockdep, didn't finish the switch to exclusively use
> > > tools/include/ yet.
> > > 
> > > I can put more work into getting it done over the next few releases,
> > > but it's not something I see as a critical fix for the upcoming
> > > release.
> > 
> > Since liblockdep was broken for an extended period of time I'd really
> > like to see this fixed before I apply any more patches.
> 
> What does the build breakage has to do with converting the way we use headers?  
> There's no broken functionality as far as I can tell, so why is the header thing 
> defined as a "fix" to begin with?

liblockdep was essentially build-broken for almost a year.

I worry about and question liblockdep's general maintainability and as a 
maintainer of lockdep I can see only two options going forward: either it's 
cleaned up for good (going beyond what is needed to fix the build failures and 
warnings), or we remove it (with the option of reintroducing it if/when it's clean 
enough).

I.e. the quality bar has increased.

> I also don't see a single tools/ project to exclusively use tools/include at 
> this point.
> 
> Would you rather keep liblockdep broken for the next couple of months until this 
> is sorted? I really doubt I could get something (+ enough time to soak in -next) 
> for v4.13, so we're looking at v4.14 at the earliest.

No, if it does not improve then I'd rather remove it, and re-add it at a later 
date if/when it's clean enough.

Thanks,

	Ingo