linux-kernel - Re: [PATCH] sysctl: Avoid open coded arithmetic in memory allocator functions

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: <20211023103158.GA4145@titan>
Date:   Sat, 23 Oct 2021 12:31:58 +0200
From:   Len Baker <len.baker@....com>
To:     Matthew Wilcox <willy@...radead.org>
Cc:     Len Baker <len.baker@....com>,
        Luis Chamberlain <mcgrof@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        Iurii Zaikin <yzaikin@...gle.com>,
        "Gustavo A. R. Silva" <gustavoars@...nel.org>,
        linux-hardening@...r.kernel.org, linux-fsdevel@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sysctl: Avoid open coded arithmetic in memory allocator
 functions

Hi Matthew,

On Sat, Oct 16, 2021 at 05:18:24PM +0100, Matthew Wilcox wrote:
> On Sat, Oct 16, 2021 at 05:28:28PM +0200, Len Baker wrote:
> > +static size_t new_dir_size(size_t namelen)
> > +{
> > +	size_t bytes;
> > +
> > +	if (check_add_overflow(sizeof(struct ctl_dir), sizeof(struct ctl_node),
> > +			       &bytes))
> > +		return SIZE_MAX;
> > +	if (check_add_overflow(bytes, array_size(sizeof(struct ctl_table), 2),
> > +			       &bytes))
> > +		return SIZE_MAX;
> > +	if (check_add_overflow(bytes, namelen, &bytes))
> > +		return SIZE_MAX;
> > +	if (check_add_overflow(bytes, (size_t)1, &bytes))
> > +		return SIZE_MAX;
> > +
> > +	return bytes;
> > +}
>
> I think this is overkill.  All these structs are small and namelen is
> supplied by the kernel, not specified by userspace.  It really complicates
> the code, and I don't see the advantage.
>
Ok, understood. I will send a v2 without this function.

Thanks for the review,
Len