[PATCH] Clarify the pass name of X86AsmPrinter

[Jim Grosbach]

On Feb 15, 2013, at 4:46 PM, Eli Bendersky <eliben at google.com> wrote:

> On Fri, Feb 15, 2013 at 4:42 PM, Jim Grosbach <grosbach at apple.com> wrote:
>> Seems reasonable to me.
>> 
> 
> Thanks Jim. Do you think that changing this output based on the actual
> streamer would be interesting or would it be overdoing it? To be
> explicit, I'm referring to printing which kind of output is actually
> being emitted by examining the streamer attached to the AsmPrinter.

I think that's a bit overdoing it, honestly. This output is mainly just for debug output to be able to tell which passes are running in which order. It's probably better that it be static so it's easier to recognize rather than strictly speaking more accurate dynamically.

-Jim

> 
> Eli
> 
> 
>> -Jim
>> 
>> On Feb 15, 2013, at 4:41 PM, Eli Bendersky <eliben at google.com> wrote:
>> 
>>> Hello,
>>> 
>>> Currently the x86 AsmPrinter reports its name as a hardcoded "X86
>>> AT&T-Style Assembly Printer". This is somewhat misleading in printouts
>>> (such as -time-passes and others) when object-file emission is used
>>> (X86AsmPrinter emits either an object file or assembly via a specific
>>> MCStreamer).
>>> 
>>> This patch replaces this string (and a related comment) with something
>>> a bit more generic. A more audacious approach would be for the method
>>> to check which streamer it has and return a precise name. Note also
>>> that this isn't a x86-specific problem - x86 is just being used as an
>>> example here. A similar change can be applied to other targets.
>>> 
>>> PTAL,
>>> Eli
>>> <asmprinter-name.1.patch>
>>