Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <9216dfe9-eea0-b0c0-3630-339a2b60aa2d@redhat.com>
Date:   Tue, 25 Apr 2023 16:59:40 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.7.1
Subject: Re: [PATCH v10 03/14] HP BIOSCFG driver - bioscfg
Content-Language: en-US
To:     Jorge Lopez <jorgealtxwork@gmail.com>
Cc:     Armin Wolf <W_Armin@gmx.de>, platform-driver-x86@vger.kernel.org,
        linux-kernel@vger.kernel.org, thomas@t-8ch.de
References: <20230419151321.6167-1-jorge.lopez2@hp.com>
 <20230419151321.6167-4-jorge.lopez2@hp.com>
 <38929a45-79de-964b-5d6f-cfa44099b35e@gmx.de>
 <6c05cc9e-815d-7a94-8b2d-f17fd5d47354@redhat.com>
 <CAOOmCE_W7_pgQT=8bQBi8_0pD74O6uen6-8p0Orvi1qjEcaHng@mail.gmail.com>
From:   Hans de Goede <hdegoede@redhat.com>
In-Reply-To: <CAOOmCE_W7_pgQT=8bQBi8_0pD74O6uen6-8p0Orvi1qjEcaHng@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi,

On 4/25/23 16:32, Jorge Lopez wrote:
> On Tue, Apr 25, 2023 at 5:34 AM Hans de Goede <hdegoede@redhat.com> wrote:
>>
>> Hi All,
>>
>> On 4/19/23 20:04, Armin Wolf wrote:
>>> Am 19.04.23 um 17:13 schrieb Jorge Lopez:
>>
>> <snip>
>>
>>>> +static int __init bioscfg_init(void)
>>>> +{
>>>> +    int ret = 0;
>>>> +    int bios_capable = wmi_has_guid(HP_WMI_BIOS_GUID);
>>>> +
>>>> +    if (!bios_capable) {
>>>> +        pr_err("Unable to run on non-HP system\n");
>>>> +        return -ENODEV;
>>>> +    }
>>>> +
>>>
>>> Currently, this driver will no get automatically loaded on supported hardware,
>>> something which would be quite beneficial for users to have.
>>> Since the HP_WMI_BIOS_GUID is already handled by the hp-wmi driver, maybe this
>>> driver (which also already implements a function similar to hp_wmi_perform_query())
>>> could register a platform device which is then used by this driver? This together
>>> with MODULE_DEVICE_TABLE() would allow for automatically loading the module on supported hardware.
>>
>> Both drivers can already co-exist since the old hp-wmi driver uses the old
>> wmi kernel functions and is not a "wmi_driver" so there is no need for
>> a platform_device for this driver to bind to since the wmi_device is
>> still free for it to bind to.
>>
>> This does indeed need a MODULE_DEVICE_TABLE() statement for
>> the bios_attr_pass_interface_id_table[] id-table. Note only for that
>> table, because the HP_WMI_BIOS_GUID is present on models which do
>> not support this and we don't want the module to auto-load there.
>>
> 
> I am considering adding a check for the  HP_WMI_SET_BIOS_SETTING_GUID
> and HP_WMI_BIOS_GUID during the initialization process.
> The driver will report a message If HP_WMI_SET_BIOS_SETTING_GUID is
> found or not.
> If HP_WMI_SET_BIOS_SETTING_GUID does not exist, the driver cannot
> change the BIOS settings.

Right, adding a check for  HP_WMI_BIOS_GUID to bioscfg_init()
sounds like a good idea.

That + add a a MODULE_DEVICE_TABLE() statement to
the bios_attr_pass_interface_id_table[] id-table.

Regards,

Hans