linux-kernel - Re: [PATCH 1/3] radix-tree: 'slot' can be NULL in radix_tree_next

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]

Date:	Wed, 10 Aug 2016 08:52:06 -0600
From:	Ross Zwisler <ross.zwisler@...ux.intel.com>
To:	Konstantin Khlebnikov <koct9i@...il.com>
Cc:	Ross Zwisler <ross.zwisler@...ux.intel.com>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-nvdimm@...ts.01.org,
	Andrey Ryabinin <aryabinin@...tuozzo.com>,
	Dmitry Vyukov <dvyukov@...gle.com>
Subject: Re: [PATCH 1/3] radix-tree: 'slot' can be NULL in
 radix_tree_next_slot()

On Wed, Aug 10, 2016 at 09:29:23AM +0300, Konstantin Khlebnikov wrote:
> On Tue, Aug 9, 2016 at 6:27 PM, Ross Zwisler
> <ross.zwisler@...ux.intel.com> wrote:
> > On Mon, Aug 08, 2016 at 10:21:39PM +0300, Konstantin Khlebnikov wrote:
> > <>
> >> NAK. This is fast path and it's already bloated.
> >> I want to revert most changes here and rework "multiorder" entries.
> >>
> >> Here you can find almost ready patchset for that
> >> https://github.com/koct9i/linux/commits/radix-tree
> >
> > Okay...are you okay with the second 2 patches in the series?  They stand
> > alone, and I believe are both good to have.
> 
> They looks good.

Cool, can I interpret that as an Acked-by for my v2? :)

> If you're worried about complicated paths - I'll add comment for
> *_next_slot() about that.

Yea, if we can document the exact ways in which we're protected from not
dereferencing a NULL 'slot', that would be great.  I think it essentially
boils down to:

1) For tagged iteration, if 'slot' is NULL then iter->tags must be cleared
2) For non-tagged iteration, it 'slot' is NULL then
   radix_tree_chunk_size(iter) must return 1 or less.