linux-kernel - Re: [PATCH 1/2] usb: gadget: r8a66597: Fix two possible sleep-in-atomic-context bugs in init

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:   Fri, 29 Jun 2018 15:21:19 +0200
From:   Robert Jarzmik <robert.jarzmik@...e.fr>
To:     Felipe Balbi <balbi@...nel.org>
Cc:     Jia-Ju Bai <baijiaju1990@...il.com>, gregkh@...uxfoundation.org,
        nicolas.ferre@...rochip.com, keescook@...omium.org,
        allen.lkml@...il.com, linux-usb@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/2] usb: gadget: r8a66597: Fix two possible sleep-in-atomic-context bugs in init_controller()

Felipe Balbi <balbi@...nel.org> writes:

> Hi,
>
> Robert Jarzmik <robert.jarzmik@...e.fr> writes:
>>>> And as bonus question, why is it better to have mdelay() calls in the driver ?
>>>
>>> As a bugfix, it's the smallest fix possible, right? Ideally, we wouldn't
>>> need either of them. Perhaps there's a bit which can be polled instead?
>> Ideally yes. Do you remember if a "threaded interrupt" might use msleep() ? I
>> seem to remember that they can, so won't that be another alternative ?
>
> yeah, unless, of course, you have a spinlock held. ;-)
Ah yes, unless that :)

I would have proposed to call the disconnect out of the spinlock path, but
looking at the r8a66592_usb_disconnect(), with its spinlock flip-flop, I loose
heart ...

And even if I still think no mdelay() should be used, because of the kernel
stall (and global uniprocessor stall), I won't argue anymore. After all, if you
let in the mdelay(), perhaps the maintainers will agree to review their
architecture and drop the locks or sleeps in interrupt context in a follow-up
patch, who knows ...

Cheers.

-- 
Robert