[PATCH][ARM]Fix urem/srem of i64 for aeabi target (PR 17881)

[Weiming Zhao]

Hi Renato,

 

Thanks for your suggestions. I'll revisit the approach of changing ARM
backend.

 

Thanks,

Weiming

 

 

 

Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by
The Linux Foundation

 

From: Renato Golin [mailto:renato.golin at linaro.org] 
Sent: Wednesday, November 20, 2013 12:10 PM
To: Zhao
Cc: Tim Northover; llvm-commits
Subject: Re: [PATCH][ARM]Fix urem/srem of i64 for aeabi target (PR 17881)

 

On 20 November 2013 19:39, Weiming Zhao <weimingz at codeaurora.org> wrote:

Ping?

The patch won't affect other targets

 

Hi Weiming,

 

Regarding the original patch, I think hijacking the current function with
another, similar function, will make the code worse than it already is to
understand DivRem lowering. That confusion will affect all targets, and that
was the major concern with my first patch, which changed in a slightly more
intrusive, but also more clear way.

 

It'd be good if you could try to add these changes to the ARM back-end (like
I did), and upon failure (like I had), we'll have more reasons to try a more
generic approach (or refactor the generic approach to add this in the ARM
back-end without the same problems, though the problems I had were deep
inside SelectionDAG type legalization). It is also possible that you will
succeed, in which case we'll be able to get the patch across. I doubt anyone
now will want to re-factor the SelectionDAG type legalization...

 

Only after failing to get it through the ARM back-end that I'd re-consider
the generic approach. This time, involving as many people as possible to get
this over with once and for all. 

 

Don't take this personally, it's just an advice that this patch will not
pass review as it is, given the context.

 

cheers,

--renato

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131120/af2d68ae/attachment.html>