linux-kernel - Re: [PATCH] arm64: Clear the stack

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: <20180719104130.egfevpo3ie4azaq6@lakrids.cambridge.arm.com>
Date:   Thu, 19 Jul 2018 11:41:30 +0100
From:   Mark Rutland <mark.rutland@....com>
To:     Laura Abbott <labbott@...hat.com>
Cc:     will.deacon@....com, catalin.marinas@....com, alex.popov@...ux.com,
        Kees Cook <keescook@...omium.org>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>,
        kernel-hardening@...ts.openwall.com,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        james.morse@....com
Subject: Re: [PATCH] arm64: Clear the stack

Hi,

On Tue, Jul 17, 2018 at 03:58:19PM -0700, Laura Abbott wrote:
> On 07/03/2018 05:14 AM, Mark Rutland wrote:
> > > It might be cleaner just to use on_accessible_stack and then another
> > > function to get the top of stack. This also might just be
> > > reimplementing what x86 already has? (Mark, Ard?)
> > It looks like we could build a get_stack_info() as they have.
> > 
> > We could probably clean up our stack traced atop of that, too.
> 
> So I spent some time looking at this and I'm not 100% clear
> if there would actually be much benefit to re-writing with
> get_stack_info. Most of that design seems to come from x86
> needing to handle multiple unwind options which arm64 doesn't
> need to worry about. Any rework ended up with roughly
> the same code without any notable benefit that I could see.
> It's possible I'm missing what kind of cleanup you're suggesting
> but I think just going with a tweaked version of on_accessible_stack
> would be fine.

I was mostly thinking that a struct stack_info with stack type
enumeration would also be helpful for ensuring that we terminated stack
traces when we had a loop.

I'll reply on your new thread.

Thanks,
Mark.