[llvm-commits] [llvm] r120230 - in /llvm/trunk/tools/llvmc/src: Base.td.in Hooks.cpp

[Eric Christopher]

On Nov 29, 2010, at 6:20 AM, Mikhail Glushenkov wrote:

> 
> How difficult of an undertaking would that be for someone who never worked on a
> LLVM backend? I'm not very fond of the current solution, but backend hacking is
> an uncharted territory for me.
> 

This is probably an involved task - it'll probably involve doing things like separating out instruction itineraries from the processor description for ARM, making sure that things that call into the backend use the tuning information if it's requested as opposed to the architecture information, etc.

I don't know how hard it'll be, but it may be a bit of typing.

>> In particular the implementation that you have here won't work for any
>> other processor that I know of at the moment.
> 
> Yes, it is incomplete.

I don't think it's the right direction for the support either. I think what I outlined above may be a better route. I can try to get a better roadmap or list of things that need to be done for you if you're interested in going the way I'm thinking?

-eric