Request for code review - (ZEE patch : Redundant Zero extension elimination)

Sriraman Tallam tmsriram@google.com
Mon May 17 16:54:00 GMT 2010


On Mon, May 17, 2010 at 2:00 AM, Bernd Schmidt <bernds@codesourcery.com> wrote:
> On 05/12/2010 11:18 PM, Sriraman Tallam wrote:
>>> I know this most likely has been hashed before but why is this pass
>>> have SImode and DImode hardcoded?
>>> Can't you use the modes and then try a wider mode?  Or does that need
>>> a total rewrite of the pass?
>>
>> Yes, that requires modifying a large portion.
>
> I agree with Andrew.  Hardcoding the modes the way it's done in this
> pass really isn't up to the standards we should expect from new passes.
>  Please work on fixing this.

Will do. However, when I wrote this pass I had x86_64 in mind which
does the optimization only for that specific mode.

Since I have already submitted the patch, I will make a new patch for the fix.

>
>
> Bernd
>



More information about the Gcc-patches mailing list