linux-kernel - Re: Lack of suspend/resume/shutdown ordering between GPIO providers and consumers

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, 25 Apr 2018 11:29:40 -0700
From:   Florian Fainelli <f.fainelli@...il.com>
To:     Grygorii Strashko <grygorii.strashko@...com>,
        linux-kernel@...r.kernel.org, linux-gpio@...r.kernel.org,
        devicetree@...r.kernel.org, dmitry.torokhov@...il.com,
        jeffy.chen@...k-chips.com, enric.balletbo@...labora.com,
        josephl@...dia.com, opendmb@...il.com, rjw@...ysocki.net
Subject: Re: Lack of suspend/resume/shutdown ordering between GPIO providers
 and consumers

On 04/25/2018 11:06 AM, Grygorii Strashko wrote:
> 
> 
> On 04/24/2018 05:58 PM, Florian Fainelli wrote:
>> Hi Linus, Rafael, all
>>
>> Our GPIO controller driver: gpio-brcmstb.c has a shutdown callback which
>> gets invoked when the system is brought into poweroff aka S5. So far so
>> good, except that we also wish to use gpio_keys.c as a possible wake-up
>> source, so we may have a number of GPIO pins declared as gpio-keys that
>> allow the system to wake-up from deep slumber.
>>
>> Recently we noticed that we could easily get into a state where
>> gpio-brcmstb.c::brcmstb_gpio_shutdown() gets called first, and then
>> gpio_keys.c::gpio_keys_suspend() gets called later, which is too late to
>> have the enable_irq_wake() call do anything sensible since we have
>> suspend its parent interrupt controller before. This is completely
>> expected unfortunately because these two drivers are both platform
>> device instances with no connection to one another except via Device
>> Tree and the use of the GPIOLIB APIs.
> 
> You can take a look at device_link_add() and Co.

OK, though that requires a struct device references, so while I could
certainly resolve the device_node -> struct device that corresponds to
the GPIO provider , that poses a number of issues:

- not all struct device_node have a corresponding struct device
reference (e.g: clock providers, interrupt controllers, and possibly
other custom drivers), though in this case, they most likely do have one

- resolving a struct device associated with a struct device_node is
often done in a "bus" specific way, e.g: of_find_device_by_node(), so if
the GPIO provider is e.g: i2c_device, pci_device etc. etc. this might
not work that easily

I think this is what Dmitry just indicated in his email as well.

> 
> But it's little bit unclear what exactly you have issue with:
> - shutdown
> - suspend
> 
> above are different (at least as it was before) and gpio-brcmstb.c
>  brcmstb_gpio_shutdown() should not be called as part of suspend !?
> may be you mean brcmstb_gpio_suspend?

The issue exists with shutdown (through the use of "poweroff"), that is
confirmed, but I cannot see how it does not exist with any suspend state
as well, for the same reason that the ordering is not strictly enforced.
--
Florian