[En-Nut-Discussion] temporary hack in the thread NicRx

[Tomohiro HARAIKAWA]

Hello,

  I want somebody releave my insomnia.  It might be a simple problem, 
but I have been trapped.

  Though I found a following comment in Thread(NicRx,arg),

  /*
   * This is a temporary hack. Due to a change in initialization,
   * we may not have got a MAC address yet. Wait until one has been
   * set.
   */

I couldn't recognize what case enables NicRx() thread before the MAC 
is set.

# NutRegisterDevice() calls dev->dev_init() (that is, NicInit()) and 
# NicInit() usually loads the MAC immediately.

  I can not suppose the case, except NicInit() => NutNetLoadConfig() 
fails or written MAC indicates 00:00:00:xx:xx:xx.

  Could anybody help?  I would like to know what will become of this 
hack in the future too.

  Thank you for advance,


Off-topic subject:

  Also, I have little concern about the fact 00:00:00:xx:xx:xx might 
be a valid MAC address which is associated to XEROX.

  Using FF:FF:FF:FF:FF:FF as a mark might be better, I think.  Since 
it indicates a broadcast MAC address, and is never associated to any 
network controller.  Or, inserting a flag in IFNET structure is well 
too.

---
Tomohiro Haraikawa
Faculty of Information, Shizuoka University
E-mail: hal at cs.inf.shizuoka.ac.jp