[PATCH] Enable basic debug output for the MSP430 backend.

[Iain Sandoe]

I'll investigate a few basic testcases, modelled on other existing stuff.
Iain

On 17 Sep 2014, at 18:26, David Blaikie wrote:

> 
> 
> On Wed, Sep 17, 2014 at 10:16 AM, Eric Christopher <echristo at gmail.com> wrote:
> Sure, have at. I'm surprised that's all you need, but it's an obviously correct incremental patch as well. If you had some way to test it that'd be great too ;)
> 
> Any particular reason (I know nothing about msp430 - perhaps there's something inherent there that I'm missing) this wouldn't be testable in the usual manner? (bitcode + llc with a fixed triple + dwarfdump)
>  
> 
> -eric
> 
> On Wed, Sep 17, 2014 at 10:06 AM, Iain Sandoe <iain at codesourcery.com> wrote:
> Hi Adrian, Eric,
> 
> It's sometimes useful to see what is needed to support backends with ints smaller than 32bit.
> 
> This is a patch to enable basic debug output on the msp430 in-tree back-end so that it can be used to investigate such support.
> 
> 1. The DWARF register numbers chosen have no particular rationale (but they match one GCC implementation, at least, for R0-R15)
> 
> 2. I have not implemented any of the CFI_ stuff
> 
> however, this does still allow examination of basic variable handling etc. (and the CFI stuff could be added as/when it becomes useful).  It's certainly enough to investigate DWARF handling of variable pieces.
> 
> cheers
> Iain
> 
> 
> 
> 
> 
> 
> 
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
> 
>