Re: test result code was Re: [Simulavr-devel] avrtest and simulavr vs simulavrxx

[Joerg Wunsch]

As Joel Sherrill wrote:

> This is WAY overkill if all you want is the exit status.

You are right, but you've been asking about what is missing when
comparing it to the old simulavr.  (Also, it's not only the exit
status that was taken that way, but also a message string, and if a
test failed, I usually explored the "core dump" to also verify other
stuff around in the SRAM copy.)

Don't get me wrong, the old simulavr was everything but perfect, and I
don't want to glorify it.  However, given that simulavrxx is already
using the same name of the executable, and given that only very few
things are missing when comparing it to the old one, offering complete
commandline compatibility sounds really nice to me, as in particular
someone calling it from within a script would get a really easy
transition path that way.

If there's a basic agreement to add a -C option to implement a
"compatibility coredump" feature, I'm even willing to develop a patch
for that.

Of course, by the time simulavrxx might be the only remaining
simulator that exists, script writers could adapt their script to use
more modern methods that are only offered by simulavrxx.  But my gut
feeling says me that this will take at least a year or so until this
might happen.

-- 
cheers, J"org               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/                        NIC: JW11-RIPE
Never trust an operating system you don't have sources for. ;-)