Re: [PATCH v17 RESEND] pvpanic: pvpanic device driver

[Paolo Bonzini <pbonzini@redhat.com>]

Il 06/05/2013 13:38, Marcelo Tosatti ha scritto:
> On Mon, May 06, 2013 at 11:39:35AM +0800, Hu Tao wrote:
>> On Fri, May 03, 2013 at 06:59:18PM -0300, Marcelo Tosatti wrote:
>>> On Fri, May 03, 2013 at 10:47:10AM +0800, Hu Tao wrote:
>>>> pvpanic device is a qemu simulated device through which guest panic
>>>> event is sent to host.
>>>>
>>>> Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
>>>> ---
>>>>  drivers/platform/x86/Kconfig   |   7 +++
>>>>  drivers/platform/x86/Makefile  |   2 +
>>>>  drivers/platform/x86/pvpanic.c | 115 +++++++++++++++++++++++++++++++++++++++++
>>>>  3 files changed, 124 insertions(+)
>>>>  create mode 100644 drivers/platform/x86/pvpanic.c
>>>>
>>>> diff --git a/drivers/platform/x86/Kconfig b/drivers/platform/x86/Kconfig
>>>> index 3338437..527ed04 100644
>>>> --- a/drivers/platform/x86/Kconfig
>>>> +++ b/drivers/platform/x86/Kconfig
>>>> @@ -781,4 +781,11 @@ config APPLE_GMUX
>>>>  	  graphics as well as the backlight. Currently only backlight
>>>>  	  control is supported by the driver.
>>>>  
>>>> +config PVPANIC
>>>> +	tristate "pvpanic device support"
>>>> +	depends on ACPI
>>>> +	---help---
>>>> +	  This driver provides support for pvpanic device, which is a qemu
>>>> +	  simulated device through which guest panic event is sent to host.
>>>> +
>>>>  endif # X86_PLATFORM_DEVICES
>>>> diff --git a/drivers/platform/x86/Makefile b/drivers/platform/x86/Makefile
>>>> index ace2b38..ef0ec74 100644
>>>> --- a/drivers/platform/x86/Makefile
>>>> +++ b/drivers/platform/x86/Makefile
>>>> @@ -51,3 +51,5 @@ obj-$(CONFIG_INTEL_OAKTRAIL)	+= intel_oaktrail.o
>>>>  obj-$(CONFIG_SAMSUNG_Q10)	+= samsung-q10.o
>>>>  obj-$(CONFIG_APPLE_GMUX)	+= apple-gmux.o
>>>>  obj-$(CONFIG_CHROMEOS_LAPTOP)	+= chromeos_laptop.o
>>>> +
>>>> +obj-$(CONFIG_PVPANIC)           += pvpanic.o
>>>> diff --git a/drivers/platform/x86/pvpanic.c b/drivers/platform/x86/pvpanic.c
>>>> new file mode 100644
>>>> index 0000000..81c95ec
>>>> --- /dev/null
>>>> +++ b/drivers/platform/x86/pvpanic.c
>>>> @@ -0,0 +1,115 @@
>>>> +/*
>>>> + *  pvpanic.c - pvpanic Device Support
>>>> + *
>>>> + *  Copyright (C) 2013 Fujitsu.
>>>> + *
>>>> + *  This program is free software; you can redistribute it and/or modify
>>>> + *  it under the terms of the GNU General Public License as published by
>>>> + *  the Free Software Foundation; either version 2 of the License, or
>>>> + *  (at your option) any later version.
>>>> + *
>>>> + *  This program is distributed in the hope that it will be useful,
>>>> + *  but WITHOUT ANY WARRANTY; without even the implied warranty of
>>>> + *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>>>> + *  GNU General Public License for more details.
>>>> + *
>>>> + *  You should have received a copy of the GNU General Public License
>>>> + *  along with this program; if not, write to the Free Software
>>>> + *  Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
>>>> + */
>>>> +
>>>> +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
>>>> +
>>>> +#include <linux/kernel.h>
>>>> +#include <linux/module.h>
>>>> +#include <linux/init.h>
>>>> +#include <linux/types.h>
>>>> +#include <acpi/acpi_bus.h>
>>>> +#include <acpi/acpi_drivers.h>
>>>> +
>>>> +MODULE_AUTHOR("Hu Tao <hutao@cn.fujitsu.com>");
>>>> +MODULE_DESCRIPTION("pvpanic device driver");
>>>> +MODULE_LICENSE("GPL");
>>>> +
>>>> +static int pvpanic_add(struct acpi_device *device);
>>>> +static int pvpanic_remove(struct acpi_device *device);
>>>> +
>>>> +static const struct acpi_device_id pvpanic_device_ids[] = {
>>>> +	{ "QEMU0001", 0},
>>>> +	{ "", 0},
>>>> +};
>>>> +MODULE_DEVICE_TABLE(acpi, pvpanic_device_ids);
>>>> +
>>>> +#define PVPANIC_PANICKED	(1 << 0)
>>>> +
>>>> +static acpi_handle handle;
>>>> +
>>>> +static struct acpi_driver pvpanic_driver = {
>>>> +	.name =		"pvpanic",
>>>> +	.class =	"QEMU",
>>>> +	.ids =		pvpanic_device_ids,
>>>> +	.ops =		{
>>>> +				.add =		pvpanic_add,
>>>> +				.remove =	pvpanic_remove,
>>>> +			},
>>>> +	.owner =	THIS_MODULE,
>>>> +};
>>>> +
>>>> +static void
>>>> +pvpanic_send_event(unsigned int event)
>>>> +{
>>>> +	union acpi_object arg;
>>>> +	struct acpi_object_list arg_list;
>>>> +
>>>> +	if (!handle)
>>>> +		return;
>>>> +
>>>> +	arg.type = ACPI_TYPE_INTEGER;
>>>> +	arg.integer.value = event;
>>>> +
>>>> +	arg_list.count = 1;
>>>> +	arg_list.pointer = &arg;
>>>> +
>>>> +	acpi_evaluate_object(handle, "WRPT", &arg_list, NULL);
>>>> +}
>>>
>>> Is it safe to call acpi_evaluate_object from a panic notifier? For
>>> example:
>>>
>>> - Has it been confirmed that no code invoked via acpi_evaluate_object can panic() ?
>>
>> Confirmed.
>>
>>> - acpi_ex_enter_interpreter grabs a mutex. Is that path ever used?
>>
>> Unfortunately yes. As I can tell, there are 2 places in the path to grab
>> a mutex: when searching the namespace for the method, and when executing
>> the method. I didn't find a non-blocking version of acpi_evaluate_object.
> 
> Then thats a bug because the pvpanic notifier is called with interrupts
> disabled.
> 
> Suppose its safer to use an OUT instruction?

Yeah, it probably is...  You can still use acpi_walk_resources to find
the port number instead of hard-coding 0x505...

Paolo