[PATCH][RFC] cache generation + separate FDI parser process

[Sergey Lapin]

David Zeuthen wrote:
> 
>  - remote the cache-test program for now as we now require the runner
well, cache_test app is intenede for unit testing a cache.
So it is mostly of developers' interest. So I'd vote to return it back, 
probably with an option to set cache file. Just IMHO.

> 
>  http://gitweb.freedesktop.org/?p=hal.git;a=commitdiff;h=de7ada8fd2a5675554d6723b9145afa4e719e8a0;hp=285b20f453bd4ce5f963a9818711c7ffd38437d7
> 
> Hopefully I didn't break too much! I'm curious, don't you
> use ./run-hald.sh to do development? It's very useful, by design it's
> made such that you can run hald from your local source tree without
> 'make install'.
Well, I'll test it today. Since some of my specific needs I test hald on 
  Nokia 770 device and qemu-system-arm. But I use scripts for 
valgrinding and minimal sanity checking.

> 
> Thanks again for the patch!
Thanks for fixing it to be better.

S.