On Saturday 24 December 2016 19:08:54 Pavel Machek wrote:
> On Sat 2016-12-24 17:52:59, Pali Rohár wrote:
> > Before this patch driver generated random MAC address every time
> > when was doing initialization. And after that random MAC address
> > could be overwritten with fixed one if provided.
> 
> Before this patch, driver generated random MAC address every time it
> was initialized. After that random MAC address could be overwritten
> with fixed one, if provided.
> 
> > This patch changes order. First it tries to read fixed MAC address
> > and if it fails then driver generates random MAC address.
> 
> I don't quite get where the advantage is supposed to be. Is it that
> "use_eeprom" is set, but reading fails?

Random bytes are read from kernel only if random MAC address is needed. 
And in wl->mac_addr is always either invalid address or permanenent mac 
address which will be used. Without patch in wl->mac_addr can be random 
temporary address for some time...

> The only case where this helps is if wl1251_read_eeprom_mac()
> succeeds but reads invalid address.
> 
> > Signed-off-by: Pali Rohár <pali.ro...@gmail.com>
> 
> Acked-by: Pavel Machek <pa...@ucw.cz>

-- 
Pali Rohár
pali.ro...@gmail.com

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to