[ECOS] Re: RedBoot DHCP failure due to race condition.

[Grant Edwards <grant dot b dot edwards at gmail dot com>]

Comments are welcome on my second pass at rewritten BOOTP/DHCP support
for Redboot:

   http://www.panix.com/~grante/files/ecos/bootp.c 

Changes compard to the first version I posted:

 * Refactored things a little bit to reduce stack usage (only have one
   bootp packet buffer on the stack at a time instead of two).

 * Added verification of XID in received packets.

 * Minor clean ups

It's about 10% fewer lines of code that what's in CVS, and the
generated binary size on ARM is very slightly smaller (around 1%).

Changes from what's in CVS:

 * Fix race condition that was corrupting received IP config data.

 * Fix timeout/retry mechanism.

 * Add code to generate what should be a unique XID.

 * Add verification of XID in received packets.

 * Add verbose output compile-time option that logs tx/rx packets and
   state machine transitions (not controlled by CDL at this point --
   you have to edit the source file).

-- 
Grant Edwards               grant.b.edwards        Yow! I'm having an
                                  at               emotional outburst!!
                              gmail.com            


-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss