Hi Dinh,
(2014/10/02 20:38), Dinh Nguyen wrote:
> On 10/1/14, 5:18 PM, Thor Thayer wrote:
>>
>> On 10/01/2014 04:10 PM, Dinh Nguyen wrote:
>>>
>>> On 10/1/14, 4:07 PM, Thor Thayer wrote:
>>>> On 10/01/2014 12:13 PM, Dinh Nguyen wrote:
>>>>> On 10/1/14, 11:31 AM, tthayer@xxxxxxxxxxxxxxxxxxxxx wrote:
>>>>>> From: Thor Thayer <tthayer@xxxxxxxxxxxxxxxxxxxxx>
>>>>>>
>>> [...]
>>>>>> static void socfpga_cyclone5_restart(enum reboot_mode mode, const
>>>>>> char *cmd)
>>>>>> @@ -98,6 +101,13 @@ static void socfpga_cyclone5_restart(enum
>>>>>> reboot_mode mode, const char *cmd)
>>>>>> writel(temp, rst_manager_base_addr + SOCFPGA_RSTMGR_CTRL);
>>>>>> }
>>>>>> +static void __init socfpga_cyclone5_init(void)
>>>>>> +{
>>>>>> + of_platform_populate(NULL, of_default_bus_match_table,
>>>>>> + NULL, NULL);
>>>>> Why is this needed?
>>>>>
>>>>> Dinh
>>>> Hi Dinh.
>>>>
>>>> Are you asking why the of_platform_populate() is needed? If so, it is
>>>> used to kick off discovery of devices at the root of the tree.
>>> I was asking about of_platform populate(). This was removed in this
>>> commit:
>>>
>>> 8b5c18f05 ARM: l2c: socfpga: convert to generic l2c OF initialisation
>>>
>>> Just trying to understand what's the need to add it back?
>>>
>>> Dinh
>> It is used to kick off discovery of devices at the root of the tree.
>> This is needed when .init_machine was defined because we're not using
>> the default implementation (which called this function).
>>
>
> Then, can you please add a separate patch to fix up the removal of the call?
It seems that this patch does enough reasonable thing.
To initialize ECC on OCRAM, socfpga_init_ocram_ecc() must be invoked
while booting. In that case, he might need this function.
8b5c18f05 is to replace l2x0_of_init() with .l2c_aux_* fields, and
this patch has a different reason to call another init function in
.init_machine. In this case, it is natural to define .init_machine
handler in this patch.
IOW, both use same socfpga_cyclone5_init() for the function name,
but those have different purpose. I think it is not reviving the
old function, but just adding a new function which has same name.
Thank you,
--
Masami HIRAMATSU
Software Platform Research Dept. Linux Technology Research Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: masami.hiramatsu.pt@xxxxxxxxxxx
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
