[PATCH] Add warning capabilities in LLVM (backend part), Take 2

[Quentin Colombet]

On Dec 11, 2013, at 5:07 AM, Renato Golin <renato.golin at linaro.org> wrote:

> On 11 December 2013 00:51, Philip Reames <listmail at philipreames.com> wrote:
>>> This should definitely be used for things like "ran out of registers for
>>> inline asm".  That should only fail the current function codegen for
>>> something like MCJIT, and not exit the process.
>> 
>> I strongly support Reid's stated objective of replacing fatal events with
>> clean error exits where possible.
> 
> +1
I agree.
Let us do that as a second step, especially because I guess some of fatal event may require a non-trivial amount of work or at least the knowledge of the code owner to be turned into a clean error.

-Quentin

> 
> --renato
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits