Re: [PATCH] device property: set fwnode->secondary to NULL in fwnode_init()

[Danilo Krummrich]

On Wed May 6, 2026 at 1:57 PM CEST, Bartosz Golaszewski wrote:
> If a firmware node is allocated on the stack (for instance: temporary
> software node whose life-time we control) or on the heap - but using a
> non-zeroing allocation function - and initialized using fwnode_init(),
> its secondary pointer will contain uninitalized memory which likely will
> be neither NULL nor IS_ERR().

I see why secondary is generally more prone to this, but if the justification of
this change is to not rely on the caller to zero out the memory, then we might
just want to initialize all fields.

For instance, if the caller is allowed to not zero-initialize the memory then
having flags with a random value isn't correct either; all accessors are atomic
bitwise operations that never zero the whole field.