linux-kernel - Re: [PATCH v7 3/3] gpio: dwapb: add gpio-signaled acpi event support

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: <570B9996.10402@huawei.com>
Date:	Mon, 11 Apr 2016 20:33:26 +0800
From:	Jiang Qiu <qiujiang@...wei.com>
To:	Linus Walleij <linus.walleij@...aro.org>
CC:	Alexandre Courbot <gnurou@...il.com>,
	Mika Westerberg <mika.westerberg@...ux.intel.com>,
	Andy Shevchenko <andy.shevchenko@...il.com>,
	Alan Tull <delicious.quinoa@...il.com>,
	"Jamie Iles" <jamie@...ieiles.com>, <charles.chenxin@...wei.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-gpio@...r.kernel.org" <linux-gpio@...r.kernel.org>,
	"ACPI Devel Maling List" <linux-acpi@...r.kernel.org>,
	Linuxarm <linuxarm@...wei.com>
Subject: Re: [PATCH v7 3/3] gpio: dwapb: add gpio-signaled acpi event support

在 2016/4/8 16:26, Linus Walleij 写道:
> On Wed, Apr 6, 2016 at 9:08 AM, qiujiang <qiujiang@...wei.com> wrote:
>
>> This patch adds gpio-signaled acpi event support. It is used for
>> power button on hisilicon D02 board, an arm64 platform.
>>
>> The corresponding DSDT file is defined as follows:
>>  Device(GPI0) {
>>         Name(_HID, "HISI0181")
>>         Name(_ADR, 0)
>>         Name(_UID, 0)
>>
>>         Name (_CRS, ResourceTemplate ()  {
>>                 Memory32Fixed (ReadWrite, 0x802e0000, 0x10000)
>>                 Interrupt (ResourceConsumer, Level, ActiveHigh,
>>                 Exclusive,,,)  {344}
>>         })
>>
>>         Device(PRTa) {
>>                 Name (_DSD, Package () {
>>                 Package () {
>>                         Package () {"reg",0},
>>                         Package () {"snps,nr-gpios",32},
>>                         }
>>                 })
>>         }
>>
>>         Name (_AEI, ResourceTemplate () {
>>                 GpioInt(Edge, ActiveLow, ExclusiveAndWake,
>>                 PullUp, , " \\_SB.GPI0") {8}
>>         })
>>
>>         Method (_E08, 0x0, NotSerialized) {
>>                 Notify (\_SB.PWRB, 0x80)
>>         }
>> }
>>
>> Acked-by: Mika Westerberg <mika.westerberg@...ux.intel.com>
>> Reviewed-by: Andy Shevchenko <andy.shevchenko@...il.com>
>> Signed-off-by: qiujiang <qiujiang@...wei.com>
> Admittedly I'm an ACPI novice and need help with deciding
> about ACPI, but I mostly trust Mika to know these things right.
>
> About this:
>
>> +       /* Add GPIO-signaled ACPI event support */
>> +       if (pp->irq)
>> +               acpi_gpiochip_request_interrupts(&port->gc);
> It's weird to me that the driver already has a requested IRQ and
> everything, now it has to request it again from ACPI.
>
> When I look into the acpi_gpiochip_request_interrupts()
> I find it weird that it is void given how much can go wrong
> inside it. Should it not return an errorcode?
Just as Mika said, these are two different things:

platform_get_irq() requestedIRQ resource from interrupt subsystem and
create irq mapping, then gose ready for device, but dose not request
a handler immediately.

acpi_gpiochip_request_interrupts() parse the _AEI and _EVT object and 
result awareness of what GPIO pin is used.Then, install a event handler
for each pin by request this pp->irq.

If something gose wrong when acpi_gpiochip_request_interrupts() process,
GPIO itself can still works fine.

>> +               if (has_acpi_companion(dev) && pp->idx == 0)
>> +                       pp->irq = platform_get_irq(to_platform_device(dev), 0);
> As it was already fetched here and then later requested,
> we still have to call acpi_gpiochip_request_interrupts()
> further down the road? That is confusing to me, can you
> explain what is going on?
>
> Yours,
> Linus Walleij
>
> .
>