linux-kernel - Re: [PATCH] net: smc91x: propagate irq return code

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: <56AFC632.8070507@cogentembedded.com>
Date:	Mon, 1 Feb 2016 23:55:14 +0300
From:	Sergei Shtylyov <sergei.shtylyov@...entembedded.com>
To:	Robert Jarzmik <robert.jarzmik@...e.fr>
Cc:	Nicolas Pitre <nico@...xnic.net>,
	"David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] net: smc91x: propagate irq return code

On 02/01/2016 11:41 PM, Robert Jarzmik wrote:

>>> The smc91x driver doesn't honor the probe deferral mechanism when the
>>> interrupt source is not yet available, such as one provided by a gpio
>>> controller not probed.

>>     What if 'ndev->irq' does equal 0?

> That's not possible AFAIR.

    Possible if of_irq_get() returns 0 (and it will on failure!).

> There was a discussion where Linus had stated that the irq is a cookie, and a 0
> value is "no interrupt", expcepting for the single case of a PC and its timer
> interrupt.

    I know, I know... and even on x86 it was never passed to request_irq(), 
only to setup_irq()...

> As we're not in that case, and up to my understanding, platform_get_irq() cannot
> return a 0 value, only a strictly negative or positive one.

    Wishful thinking...

> And yet, that test now looks weird to me. I think I'll respin the patch with a
> "if (ndev->irq < 0) {" instead of the "if (ndev->irq <= 0) {".

    Defeating Linus' PoV as a result... ;-)

> Cheers.

MBR, Sergei