2012/4/18 R. Diez <[email protected]>

>
> > I don't think that was a very good idea. There might be scripts that
> points out
> > the old filename. Getting rid of warnings is good, but this might break
> stuff.
> > This patch was one of the reasons that I think we should put out a proper
> > release before we change more things.
> I think that's taking it too far, that hinders almost all progress.
>
> Scripts that rely on individual filenames somewhere down the CPU core's
> sources are already broken. And even if they are to remain broken in this
> way, fixing them in this particular case should be trivial.
>
> Regards,
>   rdiez
>

I wouldn't say that they are broken. It's just a different approach than
assuming that every file in a certain directory should be included. The
same approach is generally used in Makefiles for software, where you
specifically point out all source files. It also breaks (modelsim,
synplify, xst) project files for all people who aren't generating them
dynamically as we do in orpsoc.

Both approaches have benefits and drawbacks, and it's not the end of the
world, but I think it's important to point out that even smaller changes
might have consequences.

-- 
Olof Kindgren
______________________________________________
ORSoC
Website: www.orsoc.se
Email: [email protected]
______________________________________________
FPGA, ASIC, DSP - embedded SoC design
_______________________________________________
OpenRISC mailing list
[email protected]
http://lists.openrisc.net/listinfo/openrisc

Reply via email to