Hi Tom,
On 9/29/22 4:34 PM, Tom Rini <tr...@konsulko.com> wrote:
On Thu, Sep 22, 2022 at 03:53:46PM +0200, Marek Vasut wrote:
> On 9/22/22 15:45, Quentin Schulz wrote:
>> From: Quentin Schulz <quentin.sch...@theobroma-systems.com>
>>
>> One needs to set u-boot,dm-pre-reloc in a device Device Tree node so
>> that it is kept in SPL/TPL variant of the device tree. However, a device
>> is automatically probed only if it's flagged with DM_FLAG_PRE_RELOC,
>> regardless of u-boot,dm-pre-reloc DT property presence. This flag is
>> currently only set by drivers directly.
>>
>> However, this flag should be allowed to be set on a per-device basis
>> even if the driver does not hardcode it so that the device will be
>> probed in SPL/TPL regardless of the driver's defaults.
>>
>> This sets the DM_FLAG_PRE_RELOC for all U_BOOT_CLASS devices whose DT
>> node contains the u-boot,dm-pre-reloc property.
>>
>> Cc: Quentin Schulz <foss+ub...@0leil.net>
>> Signed-off-by: Quentin Schulz <quentin.sch...@theobroma-systems.com>
>> ---
>> drivers/core/lists.c | 3 +++
>> 1 file changed, 3 insertions(+)
>>
>> diff --git a/drivers/core/lists.c b/drivers/core/lists.c
>> index c49695b24f..86b75ffede 100644
>> --- a/drivers/core/lists.c
>> +++ b/drivers/core/lists.c
>> @@ -260,6 +260,9 @@ int lists_bind_fdt(struct udevice *parent, ofnode node,
struct udevice **devp,
>> ret);
>> return log_msg_ret("bind", ret);
>> } else {
>> + if (ofnode_pre_reloc(node))
>> + dev_or_flags(dev, DM_FLAG_PRE_RELOC);
>> +
>> found = true;
>> if (devp)
>> *devp = dev;
>
> Reviewed-by: Marek Vasut <ma...@denx.de>
>
> Tom, this does look like a bugfix for this release, but it would be good to
> have Simon look over this too.
Cycling back here, I would like to hear what Simon thinks here.
This patch is superseded by
https://lore.kernel.org/u-boot/20220922155326.3293815-1-foss+ub...@0leil.net/,
please abandon it.
Cheers,
Quentin