llvm dev - Mar 2012 - [LLVMdev] Mangling of UTF-8 characters in symbol names

On Fri, Mar 30, 2012 at 21:22, Eli Friedman <eli.friedman at gmail.com>
wrote:
> Okay... then we should probably support that as well.  Might need to
> be a bit careful to make sure the assembly files work correctly.
>
> -Eli
>
You mean machine assembly and not IR, right?

Sean

On Sat, Mar 31, 2012 at 2:44 PM, Sean Hunt <scshunt at
csclub.uwaterloo.ca> wrote:
> On Fri, Mar 30, 2012 at 21:22, Eli Friedman <eli.friedman at
gmail.com> wrote:
>> Okay... then we should probably support that as well.  Might need to
>> be a bit careful to make sure the assembly files work correctly.
>>
>> -Eli
>>
>
> You mean machine assembly and not IR, right?
Err, yes... LLVM IR already has a scheme for handling this safely.

-Eli

On Sat, Mar 31, 2012 at 19:05, Eli Friedman <eli.friedman at gmail.com>
wrote:
> Err, yes... LLVM IR already has a scheme for handling this safely.
>
> -Eli
GCC's assembly output simply has the UTF-8 put in with no changes.

I think that we should do this by default, and if this causes
problems, come up with a setting (possibly in the target?) to control
how we encode Unicode on other platforms.

Sean