Hans de Goede wrote:
> On 03-04-19 17:36, Andy Shevchenko wrote:
>> On Wed, Apr 3, 2019 at 6:31 PM Hans de Goede <hdegoede@xxxxxxxxxx> wrote:
>>> On 03-04-19 16:08, David Müller wrote:
>>
>>>> clk_data->base = pmc_regmap; /* offset is added by client */
>>>> clk_data->clks = pmc_data->clks;
>>>> + if (d) {
>>>> + clk_data->chk_critclks = true;
>>>> + pr_info("%s critclks quirk enabled\n", d->ident);
>>>> + }
>>>>
>>>> clkdev = platform_device_register_data(&pdev->dev,
>>>> "clk-pmc-atom",
>>>> PLATFORM_DEVID_NONE,
>>>
>>> Normally I would be fine with your choice to use dmi_first_match(),
>>> but since
>>> you need to respin anyway, can you change this to:
>>>
>>> if (dmi_check_system(critclk_systems)) {
>>> clk_data->chk_critclks = true;
>>> pr_info("%s critclks quirk enabled\n", d->ident);
>>> }
>>>
>>> That will allow you to drop the 'd' variable declaration all together
>>> and is
>>> somewhat cleaner IMHO.
>>
>> pr_info() relies on it AFAICS.
>
> Ah right I missed that, I would just do: pr_info("System is on critclks
> quirk list, critclks quirk enabled\n");
> but keeping the solution with ident is fine too.
If it is OK, I will keep the "dmi_first_match()" implementation.
Dave
