linux-kernel - Re: [PATCH v3 1/7] clkdev: Hold clocks

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]

Date:   Wed, 10 Apr 2019 09:45:10 -0700
From:   Stephen Boyd <sboyd@...nel.org>
To:     "Vaittinen, Matti" <Matti.Vaittinen@...rohmeurope.com>,
        Russell King - ARM Linux admin <linux@...linux.org.uk>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "wens@...e.org" <wens@...e.org>,
        "mturquette@...libre.com" <mturquette@...libre.com>,
        "miquel.raynal@...tlin.com" <miquel.raynal@...tlin.com>,
        "jhugo@...eaurora.org" <jhugo@...eaurora.org>,
        "linux-clk@...r.kernel.org" <linux-clk@...r.kernel.org>,
        "jbrunet@...libre.com" <jbrunet@...libre.com>
Subject: Re: [PATCH v3 1/7] clkdev: Hold clocks_mutex while iterating clocks list

Quoting Russell King - ARM Linux admin (2019-04-09 01:57:17)
> On Tue, Apr 09, 2019 at 05:31:48AM +0000, Vaittinen, Matti wrote:
> > On Mon, 2019-04-08 at 23:21 +0100, Russell King - ARM Linux admin
> > wrote:
> > > On Mon, Apr 08, 2019 at 10:00:02AM -0700, Stephen Boyd wrote:
> > > > > 
> > > > 
> > > > Sure. I was thinking along the same lines after you asked.
> > > 
> > > This is rubbish.  The reason clk_find() doesn't take the lock is that
> > > you _need_ to hold the lock while you dereference the clk_lookup
> > > data.
> > 
> > I think we all agreed on this already. Stephen pointed out that the
> > current user(s) of clk_find() do _not_ dereference the pointer.
> 
> Which is actually another incorrect statement - clk_get_sys()
> dereferences it, but Stephen's patch does rearrange the code there.
> 

I've split this patch into two. One to fix the locking of the new
clk_find() user and the other to introduce a clk_find_hw() API that lets
me use it later on in this series. I'll resend the series with this
update.