netdev - Re: [PATCH] ptp: ixp46x: Fix error handling in ptp_ixp

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] [day] [month] [year] [list]

Date:   Sat, 27 Nov 2021 00:09:27 +0100
From:   Linus Walleij <linus.walleij@...aro.org>
To:     tangbin <tangbin@...s.chinamobile.com>
Cc:     davem@...emloft.net, kuba@...nel.org, arnd@...db.de,
        wanjiabing@...o.com, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] ptp: ixp46x: Fix error handling in ptp_ixp_probe()

On Fri, Nov 26, 2021 at 2:40 AM tangbin <tangbin@...s.chinamobile.com> wrote:
> On 2021/11/26 9:26, Linus Walleij wrote:
> > On Fri, Nov 26, 2021 at 2:10 AM Tang Bin <tangbin@...s.chinamobile.com> wrote:
> >
> >> In the function ptp_ixp_probe(), when get irq failed
> >> after executing platform_get_irq(), the negative value
> >> returned will not be detected here. So fix error handling
> >> in this place.
> >>
> >> Fixes: 9055a2f591629 ("ixp4xx_eth: make ptp support a platform driver")
> >> Signed-off-by: Tang Bin <tangbin@...s.chinamobile.com>
> > OK the intention is right but:
> >
> >> -           !ixp_clock.master_irq || !ixp_clock.slave_irq)
> >> +           (ixp_clock.master_irq < 0) || (ixp_clock.slave_irq < 0))
> > Keep disallowing 0. Because that is not a valid IRQ.
> >
> > ... <= 0 ...
>
> Please look at the function platform_get_irq() in the file
> drivers/base/platform.c,
>
> the example is :
>
>      * int irq = platform_get_irq(pdev, 0);
>
>      * if (irq < 0)
>
>      *        return irq;

In this case, reading the code is a bad idea. IRQ 0 is not valid,
and the fact that platform_get_irq() can return 0 does not make
it valid.

See:
https://lwn.net/Articles/470820/

Yours,
Linus Walleij