[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <56C316E8.4090002@huawei.com>
Date: Tue, 16 Feb 2016 20:32:40 +0800
From: Jiang Qiu <qiujiang@...wei.com>
To: Linus Walleij <linus.walleij@...aro.org>,
Mika Westerberg <mika.westerberg@...ux.intel.com>
CC: Alexandre Courbot <gnurou@...il.com>,
"linux-gpio@...r.kernel.org" <linux-gpio@...r.kernel.org>,
ACPI Devel Maling List <linux-acpi@...r.kernel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
<linuxarm@...wei.com>, <haifeng.wei@...wei.com>,
<charles.chenxin@...wei.com>
Subject: Re: [RFC PATCH] GPIO/ACPI: DesignWare: Add GPIO-signaled ACPI events
support for power button
在 2016/2/16 2:25, Linus Walleij 写道:
> Mika can you help out looking at this patch. Tell me if you need a copy
> of the whole patch, I'm not smart with ACPI.
>
> On Fri, Feb 5, 2016 at 7:25 AM, qiujiang <qiujiang@...wei.com> wrote:
>
>> This patch modifies the DesignWare GPIO controller driver to
>> support the GPIO-signaled ACPI Events. This is used for power
>> button function on ARM server.
>>
>> To make it work, the _AEI and _EVT object must be defined in
>> the corresponding GPIO driver's dsdt table in UEFI. At the same
>> time, ACPI daemon component is also necessary.
>>
>> Signed-off-by: qiujiang <qiujiang@...wei.com>
>
> (...)
>> --- a/drivers/gpio/gpio-dwapb.c
>> +++ b/drivers/gpio/gpio-dwapb.c
>> @@ -23,6 +23,11 @@
>> #include <linux/spinlock.h>
>> #include <linux/platform_data/gpio-dwapb.h>
>> #include <linux/slab.h>
>> +#include <linux/acpi.h>
>> +#include <linux/gpio.h>
>
> You should only need <linux/driver.h>
Here, I don't understand well. I used driver.h instead acpi.h and
gpio.h, but it came to compile failed.
>
>> +#include "gpiolib.h"
>
> I guess this is for some acpi_gpiochip* functions that ACPI GPIO
> drivers need like this:
>
Yes, this is for function the following acpi_gpiochip* function.
These functions were declared in the gpiolib.h but gpiolib-acpi.h.
>> + /* Add GPIO-signaled ACPI event support */
>> + if (pp->irq)
>> + acpi_gpiochip_request_interrupts(&(port->bgc.gc));
>
> Hm, maybe these should be in "gpiolib-acpi.h" or so.
>
> Overall the patch looks sane to me, but I need some ACPI
> person to tell.
>
> Yours,
> Linus Walleij
>
> .
>
Powered by blists - more mailing lists