[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <522ecf24-8d9c-99d3-828d-069219739bd9@roeck-us.net>
Date: Thu, 16 Sep 2021 06:23:58 -0700
From: Guenter Roeck <linux@...ck-us.net>
To: Andy Shevchenko <andy.shevchenko@...il.com>,
Denis Pauk <pauk.denis@...il.com>
Cc: Bernhard Seibold <mail@...nhard-seibold.de>,
Pär Ekholm <pehlm@...holm.org>,
to.eivind@...il.com, "Artem S . Tashkinov" <aros@....com>,
Vittorio Roberto Alfieri <me@...toor.com>,
Sahan Fernando <sahan.h.fernando@...il.com>,
Andy Shevchenko <andriy.shevchenko@...el.com>,
Jean Delvare <jdelvare@...e.com>, linux-hwmon@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6 3/3] hwmon: (nct6775) Support access via Asus WMI.
On 9/16/21 1:34 AM, Andy Shevchenko wrote:
[ ... ]
>> + u32 args = bank | (reg << 8) | (val << 16);
>> + struct acpi_buffer input = { (acpi_size) sizeof(args), &args };
>> + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
>> + acpi_status status;
>> + union acpi_object *obj;
>> + u32 tmp = 0;
>
Ah, you changed the above to add the initialization.
Without knowing that, your comment below is a bit misleading.
Guenter
>> + obj = output.pointer;
>> + if (obj && obj->type == ACPI_TYPE_INTEGER)
>> + tmp = obj->integer.value;
>> +
>> + if (retval)
>> + *retval = tmp;
>> +
>> + kfree(obj);
>
>> + if (tmp == ASUSWMI_UNSUPPORTED_METHOD)
>
> This is uninitialized tmp in case when no obj, or obj is of the wrong type.
>
>> + return -ENODEV;
>> + return 0;
>> +#else
>> + return -EOPNOTSUPP;
>> +#endif
>> +}
>
Powered by blists - more mailing lists