linux-kernel - Re: [PATCH] genirq: Check irq disabled & masked states in irq

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, 31 May 2017 10:38:34 +0200 (CEST)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Brian Norris <briannorris@...omium.org>
cc:     jeffy <jeffy.chen@...k-chips.com>,
        LKML <linux-kernel@...r.kernel.org>, dianders@...omium.org,
        tfiga@...omium.org, Johannes Berg <johannes.berg@...el.com>
Subject: Re: [PATCH] genirq: Check irq disabled & masked states in
 irq_shutdown

On Tue, 30 May 2017, Brian Norris wrote:
> Sorry to respond to myself. Thomas, your reply to another mail in this
> series helped me to notice:
> 
> On Tue, May 30, 2017 at 04:19:58PM -0700, Brian Norris wrote:
> > Side note: for issues like the first problem above, I wonder why there
> > isn't a flag that once could pass to request_irq() that suggests the IRQ
> > should be initially disabled?
> 
> Is that what IRQ_NOAUTOEN is for?

Yes.

> > I know this wouldn't work for shared
> > interrupts (but request_irq() could reject that combination, no?)
> 
> Hehe, but then I see this, for example, when grepping around:
> 
> drivers/usb/dwc3/dwc3-omap.c:
> 
>         irq_set_status_flags(omap->irq, IRQ_NOAUTOEN);
>         ret = devm_request_threaded_irq(dev, omap->irq, dwc3_omap_interrupt,
>                                         dwc3_omap_interrupt_thread, IRQF_SHARED,
>                                         "dwc3-omap", omap);
> 
> IIUC, that's quite broken, no?

Indeed. Because the interrupt could have been requested by some other
driver already. In that case IRQ_NOAUTOEN has no effect at all.

We probably should check that in __setup_irq() and yell at people.

Thanks,

	tglx