linux-kernel - Re: [PATCH] i2c: imx: Fix external abort on early interrupt

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:   Fri, 12 Jun 2020 15:09:15 +0200
From:   Krzysztof Kozlowski <krzk@...nel.org>
To:     Wolfram Sang <wsa@...nel.org>
Cc:     Marc Kleine-Budde <mkl@...gutronix.de>,
        Fabio Estevam <festevam@...il.com>,
        Sascha Hauer <s.hauer@...gutronix.de>,
        linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        Oleksij Rempel <linux@...pel-privat.de>,
        Oleksij Rempel <o.rempel@...gutronix.de>,
        NXP Linux Team <linux-imx@....com>,
        Pengutronix Kernel Team <kernel@...gutronix.de>,
        Shawn Guo <shawnguo@...nel.org>,
        linux-arm-kernel@...ts.infradead.org, linux-i2c@...r.kernel.org
Subject: Re: [PATCH] i2c: imx: Fix external abort on early interrupt

On Fri, Jun 12, 2020 at 03:00:03PM +0200, Wolfram Sang wrote:
> On Fri, Jun 12, 2020 at 02:18:06PM +0200, Marc Kleine-Budde wrote:
> > On 6/12/20 1:51 PM, Wolfram Sang wrote:
> > > 
> > >> This basically kills the concept of devm for interrupts. Some other
> > > 
> > > It only works when you can ensure you have all interrupts disabled (and
> > > none pending) in remove() or the error paths of probe() etc.
> > 
> > But when requesting the interrupt as shared the interrupt handler can get called
> > any time, even if you have disabled the IRQ source in your IP core....The shared
> > IRQ debug code tests this.
> 
> Yes, so you'd need something like
> 
> 	if (clks_are_off)
> 		return IRQ_NONE;

Maybe then:
	if (pm_runtime_enabled())?

The device structure should be valid at this point so the call should
work.


> 
> or skip devm_ for interrupts and handle it manually. (IIRC the input
> subsystem really frowns upon devm + irqs for such reasons)
> 
> D'accord?

I guess dream of managing every resource automatically is an utopia :)

Best regards,
Krzysztof