On 19.1.2018 13:37, Bjørn Mork wrote:
> Ivan Vecera <ivecera@xxxxxxxxxx> writes:
>
>> Missing prefix 'pn' in MODULE_ALIAS lines causes that the module is
>> not loaded automatically. The patch changes them so they match
>> the entries in mlxplat_dmi_table where BOARD_VENDOR and PRODUCT_NAME
>> are used.
>>
>> Fixes: 6613d18e9038 ("platform/x86: mlx-platform: Move module from arch/x86")
>> Cc: Vadim Pasternak <vadimp@xxxxxxxxxxxx>
>> Signed-off-by: Ivan Vecera <ivecera@xxxxxxxxxx>
>> ---
>> drivers/platform/x86/mlx-platform.c | 10 +++++-----
>> 1 file changed, 5 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/platform/x86/mlx-platform.c b/drivers/platform/x86/mlx-platform.c
>> index 504256c3660d..b1a5bbe4acf0 100644
>> --- a/drivers/platform/x86/mlx-platform.c
>> +++ b/drivers/platform/x86/mlx-platform.c
>> @@ -370,8 +370,8 @@ module_exit(mlxplat_exit);
>> MODULE_AUTHOR("Vadim Pasternak (vadimp@xxxxxxxxxxxx)");
>> MODULE_DESCRIPTION("Mellanox platform driver");
>> MODULE_LICENSE("Dual BSD/GPL");
>> -MODULE_ALIAS("dmi:*:*Mellanox*:MSN24*:");
>> -MODULE_ALIAS("dmi:*:*Mellanox*:MSN27*:");
>> -MODULE_ALIAS("dmi:*:*Mellanox*:MSB*:");
>> -MODULE_ALIAS("dmi:*:*Mellanox*:MSX*:");
>> -MODULE_ALIAS("dmi:*:*Mellanox*:MSN21*:");
>> +MODULE_ALIAS("dmi:*:pnMSN24*:*:rvnMellanoxTechnologies*:");
>> +MODULE_ALIAS("dmi:*:pnMSN27*:*:rvnMellanoxTechnologies*:");
>> +MODULE_ALIAS("dmi:*:pnMSB*:*:rvnMellanoxTechnologies*:");
>> +MODULE_ALIAS("dmi:*:pnMSX*:*:rvnMellanoxTechnologies*:");
>> +MODULE_ALIAS("dmi:*:pnMSN21*:*:rvnMellanoxTechnologies*:");
>
> Strange. Why do so many dmi drivers depend on such long and fragile
> lists of aliases instead of simply using
>
> MODULE_DEVICE_TABLE(dmi, ..);
>
> like most other drivers do? Is there something wrong with file2alias?
> If so, then I believe it would be better to fix that instead.
>
>
> Bjørn
>
Good note, will re-send.
Ivan
