linux-kernel - Re: [PATCH v5 05/14] software node: clean up property_copy_string

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: <20191016170059.GE35946@dtor-ws>
Date:   Wed, 16 Oct 2019 10:00:59 -0700
From:   Dmitry Torokhov <dmitry.torokhov@...il.com>
To:     Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
Cc:     "Rafael J. Wysocki" <rafael@...nel.org>,
        Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
        Mika Westerberg <mika.westerberg@...ux.intel.com>,
        Linus Walleij <linus.walleij@...aro.org>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>,
        linux-acpi@...r.kernel.org, linux-kernel@...r.kernel.org,
        platform-driver-x86@...r.kernel.org
Subject: Re: [PATCH v5 05/14] software node: clean up
 property_copy_string_array()

On Wed, Oct 16, 2019 at 10:53:00AM +0300, Andy Shevchenko wrote:
> On Tue, Oct 15, 2019 at 11:12:11AM -0700, Dmitry Torokhov wrote:
> > On Tue, Oct 15, 2019 at 03:07:26PM +0300, Andy Shevchenko wrote:
> > > On Fri, Oct 11, 2019 at 04:07:12PM -0700, Dmitry Torokhov wrote:
> > > > Because property_copy_string_array() stores the newly allocated pointer in the
> > > > destination property, we have an awkward code in property_entry_copy_data()
> > > > where we fetch the new pointer from dst.
> > > 
> > > I don't see a problem in this function.
> > > 
> > > Rather 'awkward code' is a result of use property_set_pointer() which relies on
> > > data type.
> > 
> > No, the awkwardness is that we set the pointer once in
> > property_copy_string_array(), then fetch it in
> > property_entry_copy_data() only to set it again via
> > property_set_pointer().
> 
> Yes, since property_set_pointer is called independently
> on the type of the value.

We still call property_set_pointer() independently of the type of the
value even with this patch. The point is that we do not set the pointer
in property_copy_string_array(), so we only set the pointer once.

We used to have essentially for string arrays:

	copy data
	set pointer in dst
	get pointer from dst
	set pointer in dst

With this patch we have:

	copy data
	set pointer in dst

> 
> 
> > This is confising and awkward and I believe it
> > is cleaner for property_copy_string_array() to give a pointer to a copy
> > of a string array, and then property_entry_copy_data() use it when
> > handling the destination structure.
> 
> We probably need a 3rd opinion here.

I think I can still convince you ;)

Thanks.

-- 
Dmitry