[llvm] r226746 - [canonicalization] Refactor how we create new stores into a helper

Duncan P. N. Exon Smith dexonsmith at apple.com
Wed Jan 21 18:55:29 PST 2015


> On 2015-Jan-21, at 18:23, Chandler Carruth <chandlerc at gmail.com> wrote:
> 
> 
> On Wed, Jan 21, 2015 at 6:18 PM, Duncan P. N. Exon Smith <dexonsmith at apple.com> wrote:
> > On 2015-Jan-21, at 16:39, Chandler Carruth <chandlerc at gmail.com> wrote:
> >
> >
> > On Wed, Jan 21, 2015 at 3:45 PM, Chandler Carruth <chandlerc at gmail.com> wrote:
> > [canonicalization] Refactor how we create new stores into a helper
> > function. This is a bit tidier anyways and will make a subsquent patch
> > simpler as I want to add another case to this combine.
> >
> > I didn't end up needing this, but it seems cleaner to leave it in. Let me know if anyone prefers the old code.
> 
> New code looks better to me.  Might be cleaner without the return value
> if you're not using it, though.
> 
> I lied, and will end up using it after all. Just need to move it in my next commit.
> 
> I'm still not using the return value, but it felt weird to not return it. I can definitely imagine a caller in the future wanting it, so just wanted to make the utility function useful to them if/when they arrive as it adds essentially no complexity.
> 
> Still, I'm happy to remove it if that's your preference. Just let me know.

Whatever :).  Looks like dead code (and a new caller could easily add
it back), but it's harmless, and I agree it looks kind of weird
without it.



More information about the llvm-commits mailing list