Re: [git patches 1/2] warnings: attack valid cases spotted by warnings

[Roland Dreier <rdreier@cisco.com>]

 > Quite frankly, I don't quite understand where you get those enormous balls 
 > you have, that you can then talk about how ugly it is to just add a "= 0" 
 > that shuts up a compiler warning. That's the _least_ ugly part of the 
 > whole damn function!

The clanking when I walk annoys people in the office too...

But you're right.  It is stupid of me to make such a big deal about
this.  My excuse is that I've seen those warnings so many times and
actually given them more thought than they deserve, and I really felt
that Jeff's change makes the admittedly already ugly code a tiny
little bit worse.

 > Anyway, here's a totally untested cleanup that compiles but probably 
 > doesn't work, because I didn't check that I did the right thing with all 
 > the pointer arithmetic (ie when I change "wqe" to a real structure pointer 
 > instead of just a "void *", maybe I left some pointer arithmetic around 
 > that expected it to work as a byte pointer, but now really works on the 
 > whole structure size instead).

Given that you took the time to do this, I'll get the patch into a
working state and apply it.  And maybe split it into reviewable chunks
while I'm at it ;)

Thanks,
  Roland